Intel 05-2409-003 User Manual

Global Call API for Host Media Processing on Windows

Programming Guide

August 2006

05-2409-003

INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. Intel products are not intended for use in medical, life saving, life sustaining, critical control or safety systems, or nuclear facility applications.

Intel may make changes to specifications and product descriptions at any time, without notice.

This Global Call API for Host Media Processing on Windows Programming Guide as well as the software described in it is furnished under license and may only be used or copied in accordance with the terms of the license. The information in this manual is furnished for informational use only, is subject to change without notice, and should not be construed as a commitment by Intel Corporation. Intel Corporation assumes no responsibility or liability for any errors or inaccuracies that may appear in this document or any software that may be provided in association with this document.

Except as permitted by such license, no part of this document may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without the express written consent of Intel Corporation.

Dialogic, Intel, Intel logo, and Intel NetStructure are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

* Other names and brands may be claimed as the property of others.

Publication Date: August 2006

Document Number: 05-2409-003

Intel 1515 Route 10 Parsippany, NJ 07054

For Technical Support, visit the Intel Telecom Support Resources website at:

http://developer.intel.com/design/telecom/support

For Products and Services Information, visit the Intel Telecom and Compute Products website at:

http://www.intel.com/design/network/products/telecom

For Sales Offices and other contact information, visit the Buy Telecom Products page at:

http://www.intel.com/buy/networking/telecom.htm

Global Call API for HMP on Windows Programming Guide – August 2006

Revision History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

About This Publication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Applicability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Intended Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

How to Use This Publication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Related Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1 Product Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.1 Global Call Software Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.2 Global Call Feature Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.2.1 Call Control Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.2.2 Operation, Administration and Maintenance Features . . . . . . . . . . . . . . . . . . . . . 18

1.3 Global Call Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

1.3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

1.3.2 Global Call API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.4 Call Control Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

1.4.1 Starting Call Control Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

1.4.2 Call Control Library States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

1.5 Global Call Object Identifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

1.5.1 Line Device Identifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

1.5.2 Call Reference Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

1.5.3 Object Identifiers and Resource Sharing Across Processes . . . . . . . . . . . . . . . . 25

1.5.4 Target Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2 Programming Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.1 Programming Models Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.2 Asynchronous Mode Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.2.1 Asynchronous Model Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.2.2 Asynchronous Model with Event Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

2.2.3 Asynchronous with Windows Callback Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.2.4 Asynchronous with Win32 Synchronization Model . . . . . . . . . . . . . . . . . . . . . . . . 31

2.2.5 Extended Asynchronous Programming Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3 Call State Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.1 Call State Model Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.2 Basic Call Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.2.1 Basic Call States at the Inbound Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.2.2 Basic Call States at the Outbound Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.2.3 Basic Call States for Call Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.3 Basic Call Model Configuration Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.3.1 Call State Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.3.2 Call State Event Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.3.3 Call Acknowledgement Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.3.4 Call Proceeding Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

Global Call API for HMP on Windows Programming Guide – August 2006 3

Contents

3.4 Basic Call Control in Asynchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.4.1 Inbound Calls in Asynchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.4.2 Outbound Calls in Asynchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3.4.3 Call Termination in Asynchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

3.4.4 Handling Unsolicited Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

3.5 Advanced Call Control with Call Hold and Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

3.5.1 Advanced Call State Model Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

3.5.2 Advanced Call States for Hold and Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

3.5.3 Call Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

3.5.4 Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

4 Event Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.1 Overview of Event Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.2 Event Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.3 Blocked and Unblocked Event Handling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

4.4 Event Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

4.5 Events Indicating Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74

4.6 Masking Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

4.7 Event Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

5 Application Development Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

5.1 General Programming Tips. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

5.2 Tips for Programming Drop and Insert Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

5.3 Using Global Call with Digital Network Interface Boards . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5.3.1 Routing Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

5.3.2 Working with Flexible Routing Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

5.3.3 Handling Multiple Call Objects Per Channel in a Glare Condition . . . . . . . . . . . . . 83

6 Error Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

6.1 Error Handling Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

7Call Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

7.1 Call Progress Analysis when Using IP Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

7.2 Call Progress Analysis when Using Digital Network Interface Boards . . . . . . . . . . . . . . . . 87

7.2.1 Call Progress Analysis Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

7.2.2 Configuring Default Call Progress Analysis Parameters . . . . . . . . . . . . . . . . . . . . 88

7.2.3 Configuring Call Progress Analysis on a Per Call Basis . . . . . . . . . . . . . . . . . . . . 88

7.2.4 Setting Call Analysis Attributes on a Per Call Basis . . . . . . . . . . . . . . . . . . . . . . . 90

7.2.5 Configuring Call Progress Analysis on a Per Channel Basis. . . . . . . . . . . . . . . . .91

7.2.6 Setting Call Analysis Attributes on a Per Channel Basis . . . . . . . . . . . . . . . . . . . . 92

7.2.7 Customizing Call Progress Tones on a Per Board Basis . . . . . . . . . . . . . . . . . . . 92

7.3 Resource Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

7.4 Feature Transparency and Extension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

7.4.1 Feature Transparency and Extension Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 93

7.4.2 Technology-Specific Feature Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

7.4.3 Technology-Specific User Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

8 Alarm Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

8.1 Alarm Handling Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

8.1.1 Alarm Management System Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

8.2 Operation and Configuration of GCAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

4 Global Call API for HMP on Windows Programming Guide – August 2006

Contents

8.2.1 Generation of Events for Blocking Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

8.2.2 Generation of Alarm Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

8.2.3 Configuration of Alarm Properties and Characteristics . . . . . . . . . . . . . . . . . . . . 101

8.2.4 Starting and Stopping Alarm Transmission. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

8.2.5 Retrieving Alarm Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

8.3 Sample Alarm Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

8.3.1 Scenario 1: Application Notified of First and Last Blocking Alarm . . . . . . . . . . . 105

8.3.2 Scenario 2: Default Behavior for Alarm Notification . . . . . . . . . . . . . . . . . . . . . . 107

8.3.3 Scenario 3: Alarm Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

9 Real Time Configuration Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

9.1 Real Time Configuration Manager Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

9.2 RTCM Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

9.2.1 Customer Application Using Global Call RTCM . . . . . . . . . . . . . . . . . . . . . . . . . 111

9.2.2 Global Call RTCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

9.2.3 RTCM Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

9.3 Using RTCM Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

9.3.1 Parameter Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

9.3.2 Parameter Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

9.4 Getting and Setting Parameter Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

9.4.1 GC_PARM_BLK Data Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

9.4.2 Control Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

9.5 Handling RTCM Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

9.6 Configuration Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

9.7 Sample Scenarios Using the RTCM API Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

9.7.1 Getting or Setting GCLib Configuration in Synchronous Mode. . . . . . . . . . . . . . 119

9.7.2 Getting or Setting CCLib Configuration in Synchronous Mode. . . . . . . . . . . . . . 120

9.7.3 Getting or Setting Line Device Configuration in Synchronous Mode . . . . . . . . . 121

9.7.4 Setting Line Device Configuration in Asynchronous Mode . . . . . . . . . . . . . . . . . 123

9.7.5 Setting Board Device Configuration in Asynchronous Mode (IP Technology) . . 124

10 Handling Service Requests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

10.1 Service Request Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

10.2 Service Request Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

10.3 Service Request Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

10.4 General Service Request Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

11 Using Global Call to Implement Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

11.1 Introduction to Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

11.1.1 Blind Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

11.1.2 Supervised Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

11.2 Call Transfer State Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

12 Building Applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

12.1 Compiling and Linking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

12.1.1 Include Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

12.1.2 Required Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

12.1.3 Variables for Compiling and Linking Commands . . . . . . . . . . . . . . . . . . . . . . . . 140

12.1.4 Dynamically Loaded Libraries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

13 Debugging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Global Call API for HMP on Windows Programming Guide – August 2006 5

Contents

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

6 Global Call API for HMP on Windows Programming Guide – August 2006

Contents

Figures

1 Global Call Architecture for IP Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

2 Global Call Architecture for E1/T1 and ISDN Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3 Call Control Library States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

4 Basic Asynchronous Inbound Call State Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

5 Basic Asynchronous Inbound Call Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

6 Incoming Call Scenario with Call Proceeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

7 Call Acknowledgement and Call Proceeding Done at the Application Layer. . . . . . . . . . . . . . . 49

8 Call Proceeding Done by the Application Layer with Minimum Information Configured . . . . . . 50

9 Call Acknowledgement and Call Proceeding Done at Technology Call Control Layer . . . . . . . 51

10 Call Acknowledgement Done by the Technology Call Control Layer and Call Proceeding Done by

the Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

11 Basic Asynchronous Outbound Call State Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

12 Asynchronous Outbound Call Scenario. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

13 Asynchronous Outbound Call Scenario With Call Acknowledgement . . . . . . . . . . . . . . . . . . . . 59

14 Asynchronous Outbound Call Scenario With Overlap Sending . . . . . . . . . . . . . . . . . . . . . . . . . 60

15 Asynchronous Call Tear-Down State Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

16 User Initiated Asynchronous Call Termination Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

17 Network Initiated Asynchronous Call Termination Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

18 Call State Transitions for Hold and Retrieve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

19 Call State Model for Supervised and Unsupervised Transfers . . . . . . . . . . . . . . . . . . . . . . . . . 68

20 Call Termination by the Network or Application During a Transfer . . . . . . . . . . . . . . . . . . . . . . 69

21 Architectural Diagram of Alarm Management Components . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

22 Notification of First and Last Blocking Alarm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

23 Default Behavior for Alarm Notification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

24 Alarm Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

25 Relationship of Customer Application, Global Call RTCM, and RTCM Parameters . . . . . . . . 110

26 Run Time Configuration Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

27 Getting or Setting GCLib Configuration in Synchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . 119

28 Getting or Setting CCLib Configuration in Synchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . 120

29 Getting or Setting Line Device Configuration in Synchronous Mode. . . . . . . . . . . . . . . . . . . . 122

30 Setting Line Device Configuration in Asynchronous Mode (E1, T1 and ISDN Technology) . . 123

31 Setting Board Device Configuration in Asynchronous Mode (IP Technology). . . . . . . . . . . . . 124

32 Service Request Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

33 Generic Service Request Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

34 Blind Call Transfer (Unsupervised Transfer) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

35 Supervised Call Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

36 Call State Model for Blind Call Transfer at Party A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

37 Call State Model for Blind Transfer at Party B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

38 Call State Model for Supervised Transfer at Party A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

39 Call State Model for Supervised Transfer at Party B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

40 Call State Model for Supervised Transfer at Party C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

Global Call API for HMP on Windows Programming Guide – August 2006 7

Contents

8 Global Call API for HMP on Windows Programming Guide – August 2006

Contents

Tables

1 Call Control Library States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2 Supported Target Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3 Target Types and Target IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4 Target Object Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

5 Obtaining Target IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

6 Asynchronous Inbound Call State Transitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

7 Asynchronous Outbound Call State Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

8 Asynchronous Call Termination Call State Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

9 Unsolicited Events Requiring Signal Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

10 Handling Glare. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

11 Call Progress Analysis Settings and Possible Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

12 Comparison with Call Progress Analysis Using gc_SetParm( ). . . . . . . . . . . . . . . . . . . . . . . . . 92

13 Update Condition Flag and Global Call Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

14 New Global Call Transfer Call States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Global Call API for HMP on Windows Programming Guide – August 2006 9

Contents

10 Global Call API for HMP on Windows Programming Guide – August 2006

Revision History

This revision history summarizes the changes made in each published version of this document.

Document No. Publication Date Description of Revisions

05-2409-003 August 2006 Call Control Libraries section: Updated the library descriptions to identify the

technologies/protocols that each library supports.

Using Protocols (Flexible Routing) section: Removed incorrect reference to using the

DM3 PDK Manager.

Setting Call Analysis Attributes on a Per Call Basis section: Updated to indicate that

PAMD_QUAL2TMP is not supported and to provide a pointer to a related Tech Note.

Debugging chapter : Added reference to the “Runtime Trace Facility (RTF) Reference”

chapter in the Intel Dialogic System Diagnostics Guide.

05-2409-002 October 2005 Global change: Added support for protocols that can be run on the E1 or T1

interfaces provided by Intel NetStructure

Global change: Updates to recognize the Intel NetStructure

Starting Call Control Libraries section : Added note about loading only the required

call control libraries to keep the required memory footprint small.

Overlap Sending section : Explicitly mentioned ISDN in the list of technologies that do

not have messages to request more information.

Configuring Default Call Progress Analysis Parameters section: Added a note that

pre-connect call progress is enabled by default, regardless of the CPA setting in the CONFIG file

Real Time Configuration Management chapter : Fixed several references to

gc_util_insert_val( ) and gc_util_insert_ref( ) which should be gc_util_insert_parm_val( ) and gc_util_insert_parm_ref( ).

Supervised Transfers section: Updated the call termination figure and added note to

describe the unsolicited GCEV_CONNECTED event that is generated for a call when the new call being set up is terminated.

05-2409-001 April 2005 Initial version of document. Much of the information contained in this document was

previously published in the Global Call API for Windows Operating Systems Programming Guide, document number 05-1867-002.

Digital Network Interface boards.

brand.

Global Call API for HMP on Windows Programming Guide — August 2006 11

Revision History

12 Global Call API for HMP on Windows Programming Guide — August 2006

About This Publication

The following topics provide information about this publication:

• Purpose

• Intended Audience

• How to Use This Publication

• Related Information

Purpose

This publication provides guidelines for using the Global Call API to build computer telephony applications that require call control functionality. Such applications include, but are not limited to, Call Routing, Enhanced Services, Unified Messaging, Voice Messaging, LAN Telephony Services, Computer Telephony Services, Switching, PBX, Interactive Voice Response, Help Desk and Work Flow applications. This publication is a companion guide to the Global Call API Library Reference that provides details on the functions and parameters in the Global Call API library and the Global Call Technology Guides that provide IP-, E1/T1- and ISDN-specific information.

Host Media Processing (HMP) software performs media processing tasks on general-purpose servers based on Intel system, HMP performs like a virtual DM3 board to the customer application, but all media processing takes place on the host processor. In this document, the term “board” represents the virtual DM3 board, unless explictly noted otherwise. Intel NetStructure boards provide physical E1 and T1 interfaces for applications that require E1/T1 network connectivity.

architecture without the need for specialized hardware. When installed on a

Digital Network Interface

Applicability

This document is published for Intel NetStructure® Host Media Processing Software.

Intended Audience

This publication is written for the following audience:

• Distributors

• System Integrators

• Toolkit Developers

• Independent Software Vendors (ISVs)

• Value Added Resellers (VARs)

Global Call API for HMP on Windows Programming Guide — August 2006 13

About This Publication

• Original Equipment Manufacturers (OEMs)

How to Use This Publication

Refer to this publication after you have installed the hardware and the system software, which includes the Global Call software.

This publication assumes that you are familiar with the Windows operating system and the C programming language.

The information in this guide is organized as follows:

• Chapter 1, “Product Description” provides an overview of the Global Call development

software.

• Chapter 2, “Programming Models” describes the supported programming models in the

Windows environment.

• Chapter 3, “Call State Models” describes the call state models used by Global Call.

• Chapter 4, “Event Handling” describes how to handle Global Call events.

• Chapter 6, “Error Handling” describes the error handling facilities provided by Global Call.

• Chapter 5, “Application Development Guidelines” provides guidelines when developing

applications that use Global Call.

• Chapter 7, “Call Control” describes basic call control capabilities, resource routing and feature

extensions provided by Global Call.

• Chapter 8, “Alarm Handling” describes how Global Call can be used to handle alarms.

• Chapter 9, “Real Time Configuration Management” describes how Global Call can be used for

real time configuration of parameters associated with the interface.

• Chapter 10, “Handling Service Requests” describes the generic service request facility

provided by Global Call.

• Chapter 11, “Using Global Call to Implement Call Transfer” provides general information on

the implementation of unsupervised (blind) and supervised call transfer.

• Chapter 12, “Building Applications” provides guidelines for building applications that use the

Global Call software.

• Chapter 13, “Debugging” provides pointers to where technology-specific debugging

information can be obtained.

• The Glossary provides a definition of terms used in this guide.

Related Information

Refer to the following sources for more information:

• Global Call API Library Reference

• Global Call E1/T1 CAS/R2 Technology Guide

• Global Call ISDN Technology Guide

14 Global Call API for HMP on Windows Programming Guide — August 2006

About This Publication

• Global Call IP Technology Guide

• Standard Runtime Library API Programming Guide.

• Standard Runtime Library API Library Reference.

• The Release Update for your HMP software, which may include updates to this manual,

available on the Telecom Support Resources website at:

http://www.intel.com/design/network/products/telecom/software/index.htm

• http://developer.intel.com/design/telecom/support/ (for technical support)

• http://www.intel.com/design/network/products/telecom (for product information)

Global Call API for HMP on Windows Programming Guide — August 2006 15

About This Publication

16 Global Call API for HMP on Windows Programming Guide — August 2006

1.Product Description

This chapter describes the Global Call software. Topics include:

• Global Call Software Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

• Global Call Feature Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

• Global Call Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

• Call Control Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

• Global Call Object Identifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

1.1 Global Call Software Overview

Global Call development software provides a common signaling interface for network-enabled applications, regardless of the signaling protocol needed to connect to the local telephone network. The signaling interface provided by Global Call software facilitates the exchange of call control messages between the telephone network and virtually any network-enabled application. Global Call software enables developers to create applications that can work with signaling systems worldwide, regardless of the network to which the applications are connected. The Global Call software is ideal for high-density, network-enabled solutions, such as voice, data, and video applications, where the supported hardware and signaling technology can vary widely from country to country.

As an example, the signal acknowledgement or information flow required to establish a call may vary from country to country. Rather than requiring the application to handle low-level details, Global Call software offers a consistent, high-level interface to the user and handles each country's unique protocol requirements transparently to the application.

The Global Call software comprises three major components:

Global Call Application Programming Interface (API)

A common, extensible API providing network interfaces to higher levels of software. Application developers use API function calls in their computer telephony applications. The Global Call API is the preferred call control interface.

Call Control Libraries

A set of libraries that provide the interface between the Global Call API and the various network signaling protocols.

Global Call Protocols

Network signaling protocols, such as T1 Robbed Bit, E1 CAS, ISDN, QSIG, IP H.323 and SIP can be invoked by the Global Call API to facilitate call control.

Global Call API for HMP on Windows Programming Guide — August 2006 17

Product Description

1.2 Global Call Feature Categories

The Global Call development software provides many features allowing for the development of flexible and robust applications. The features fall into one of two main categories:

• Call Control Features

• Operation, Administration and Maintenance Features

1.2.1 Call Control Features

The Global Call development software provides the following call control features:

Basic Call Control

Includes basic call control features such as, the ability to make a call, detect a call, answer a call, release a call, etc. The implementation of these capabilities is based on the basic call state model, which is a common model for all network technologies. See Section 3.2, “Basic Call

Model” for more information on the basic call model.

Advanced Call Model

Defines the behavior for advanced features, such as hold and transfer. These capabilities are provided to support technologies and protocols that support such features, for example, Supervised Transfer. The implementation of these capabilities is based on a more advanced call state model. See Section 3.5, “Advanced Call Control with Call Hold and Transfer” for more information. The advanced call model applies only to E1/T1 and ISDN technologies, not IP technology, which uses a different scheme for features such as call transfer. See the Global Call IP Technology Guide.

Call Progress and Call Analysis

Provides the capabilities for handling pre-connect (Call Progress) information that reports the status of the call connection, such as, busy, no dial tone or no ringback, and post connect (Call Analysis) information that reports the destination party’s media type, for example, voice, answering machine, or fax modem. This information is determined by the detection of tones defined specifically for this purpose. See Section 7.2, “Call Progress Analysis when Using

Digital Network Interface Boards” for more information. The call progress and call analysis

feature applies only to E1/T1 and ISDN technologies, not IP technology.

Feature Transparency and Extension (FTE)

Provides the ability to extend the capabilities of Global Call to handle features that are specific to a particular technology so that those features are accessible via the Global Call interface. For example, for ISDN applications, Global Call supports supplementary services such as Overlap Send, Overlap Receive, Any Message, Any IE, and User-to-User messaging. See

Section 7.4, “Feature Transparency and Extension” for more information.

1.2.2 Operation, Administration and Maintenance Features

The Global Call development software provides the following features that facilitate the operation, administration and maintenance of Global Call applications:

Error Handling Functionality

When an error occurs, Global Call provides functions that enable an application to retrieve more information about the error. See Chapter 6, “Error Handling” for more information.

18 Global Call API for HMP on Windows Programming Guide — August 2006

Product Description

Event Handling Functionality

Provides the ability to handle and process events, including the ability to disable and enable events and to retrieve event information. See Chapter 4, “Event Handling” for more information.

Global Call Alarm Management System (GCAMS)

Provides the ability to manage alarms. GCAMS provides Global Call applications with the ability to receive extensive alarm information that can be used to troubleshoot line problems. See Chapter 8, “Alarm Handling” for more information.

Real Time Configuration Management (RTCM)

Allows the modification of call control and protocol elements in real time, providing a single common user interface for configuration management. See Chapter 9, “Real Time

Configuration Management” for more information.

Global Call Service Request (GCSR)

Enables an application to send a request for a service to a remote device. Examples of the types of services that this feature supports are device registration, channel setup, call setup, information requests, or other kinds of requests that need to be made between two devices across the network. See Chapter 10, “Handling Service Requests” for more information.

Library Information Functions

Enables an application to get information about the call control libraries being used. See the Global Call API Library Reference for more information about these functions.

Debugging Facilities

Global Call provides powerful debugging capabilities for troubleshooting protocol-related problems, including the ability to generate a detailed log file. See the appropriate Global Call Technology Guide for information on the debugging facilities available when using Global Call with each technology.

1.3 Global Call Architecture

The Global Call development software architecture is based on the Intel® Dialogic® architecture that supports Host Media Processing (HMP) software and DM3 hardware. The architecture is described in the following topics:

• Overview

• Global Call API

1.3.1 Overview

Figure 1 shows a system-level view of the Global Call architecture for IP technology and Figure 2 shows the Global Call architecture for E1/T1 and ISDN technologies on DM3 hardware.

Global Call API for HMP on Windows Programming Guide — August 2006 19

Product Description

Figure 1. Global Call Architecture for IP Technology

Host Application

GlobalCall

Signaling

IP Network

Call Control

Host

NIC

Media

IP Network

H.323 or SIP

Call Control

Library

(IPT CCLib)

RTP/RTCP

Media

Control

Media

Routing

IP Media

Call Control

Library

(IPM CCLib)

IP Media

Resource

20 Global Call API for HMP on Windows Programming Guide — August 2006

Figure 2. Global Call Architecture for E1/T1 and ISDN Technologies

User Application

GlobalCall API

Product Description

Other

Dialogic

Libraries

1.3.2 Global Call API

Call Control Library

Device Driver Operating Systems

Firmware

Network Interface

DM3CC

Firmware

Network Interface

PSTN

The Global Call API is a call control API. Similar to other Intel Dialogic APIs (such as the Voice API), the Global Call API uses the Standard Runtime Library (SRL) API to deliver response events to its API commands. The Global Call API and other Intel Dialogic APIs form a family of APIs that use the underlying services provided by the SRL API.

The Global Call API provides a collection of functions supporting call control operations as well as functions to support operation, administration and maintenance tasks. See the Global Call API Library Reference for detailed information about each function.

Global Call API for HMP on Windows Programming Guide — August 2006 21

Product Description

1.4 Call Control Libraries

Each supported network technology requires a call control library to provide the interface between the network and the Global Call library. The call control libraries currently supported by the Global Call API for HMP are as follows:

GC_CUSTOM1_LIB

The first of two call control library place holders for custom call control libraries. Any thirdparty Global Call compatible call control library can be used as a custom library. The Global Call library supports up to two custom libraries.

GC_CUSTOM2_LIB

The second of two call control library place holders for custom call control libraries. Any third-party Global Call compatible call control library can be used as a custom library. The Global Call library supports up to two custom libraries.

GC_DM3CC_LIB

The call control library that controls access to network interfaces on Digital Network Interface boards. This library is used for call control using ISDN and CAS/R2MF (PDK protocols) signaling on Digital Network Interface boards.

GC_H3R_LIB

The call control library that controls access to IP network interfaces. This call control library supports IP H.323 and SIP protocols and is used in conjunction with GC_IPM_LIB.

GC_IPM_LIB

The call control library that provides access to IP media resources. This library is used for H323/SIP call control signaling and is used in conjunction with GC_H3R_LIB.

1.4.1 Starting Call Control Libraries

Call control libraries must be started before they can be used by the Global Call functions. The call control libraries are started when a gc_Start( ) function is issued. The gc_Start( ) function allows the selective starting of call control libraries where the application can specify if all the call control libraries are to be started or only specified libraries are to be started. The application can also start a custom call control library that is not supported by Global Call. See the Global Call API Library Reference for more information about the gc_Start( ) function.

Note: Invoking gc_Start(NULL) loads all call control libraries and consequently the memory footprint

includes memory that is allocated for all call control libraries. To reduce the memory footprint, selective loading of call control libraries is recommended. For more information and an example, see the gc_Start( ) function in the Global Call API Library Reference.

1.4.2 Call Control Library States

The initial state of all the call control libraries is the Configured state. When a call control library is successfully started, the library will be in the Available state. If the call control library fails to start, the library will be in the Failed state as shown in the diagram below. If the call control library is not started, it remains in the Configured state.

22 Global Call API for HMP on Windows Programming Guide — August 2006

Figure 3. Call Control Library States

CONFIGURED

gc_Start()

Product Description

Start

Successful

AVAILABLE FAILED

Start

Failed

Table 1 describes the different states of a call control library.

Table 1. Call Control Library States

State Description

Configured A library that is supported by Global Call is considered a configured library.

Available A library that has been successfully started is considered to be available for use by a

Global Call application.

Failed A library that has failed to start is considered to be unavailable for use by a Global Call

application.

Each configured call control library is assigned an ID number by Global Call. Each library also has a name in an ASCII string format. Library functions perform tasks such as converting a call control library ID to an ASCII name and vice-versa, determining the configured libraries, determining the available libraries, determining the libraries started and the libraries that failed to start, and other library functions.

The following functions are the call control library information functions. All the library functions are synchronous, thus they return without a termination event.

• gc_CCLibIDToName( )

• gc_CCLibNameToID( )

• gc_CCLibStatusEx( )

• gc_GetVer( )

See the Global Call API Library Reference for detailed information about these functions.

1.5 Global Call Object Identifiers

The Global Call API is call-oriented, that is, each call initiated by the application or network is assigned a Call Reference Number (CRN) for call control and tracking purposes. Call handling is independent of the line device over which the call is routed. Each line device or device group is assigned a Line Device Identifier (LDID) that enables the application to address any resource or

Global Call API for HMP on Windows Programming Guide — August 2006 23

Product Description

group of resources using a single device identifier. Certain features, such as Feature Transparency and Extension (FTE), Real Time Configuration Management (RTCM), and Global Call Service Request (GCSR) operate on a basic entity called a Global Call target object. Target objects are identified by a target type and a target ID.

The following topics provide more detailed information:

• Line Device Identifier

• Call Reference Number

• Object Identifiers and Resource Sharing Across Processes

• Target Objects

1.5.1 Line Device Identifier

A Line Device Identifier (LDID) is a unique logical number assigned to a specific resource (for example, a time slot) or a group of resources within a process by the Global Call library. Minimally, the LDID number will represent a network resource. For example, both a network resource and a voice resource are needed to process an R2 MFC dialing function. Using Global Call, a single LDID number is used by the application (or thread) to represent this combination of resources for call control.

An LDID number is assigned to represent a physical device(s) or logical device(s) that will handle a call, such as a network interface resource, when the gc_OpenEx( ) function is called. This identification number assignment remains valid until the gc_Close( ) function is called to close the line device.

When an event arrives, the application (or thread) can retrieve the LDID number associated with the event by using the linedev field of the associated METAEVENT structure. The LDID is retrieved using the gc_GetMetaEvent( ) or the gc_GetMetaEventEx( ) function.

1.5.2 Call Reference Number

A Call Reference Number (CRN) is a means of identifying a call on a specific line device. A CRN is created by the Global Call library when a call is requested by the application, thread or network.

With the CRN approach, the application (or thread) can access and control the call without any reference to a specific physical port or line device. CRNs are assigned to both inbound and outbound calls:

Inbound calls

The CRN is assigned via the gc_WaitCall( ) function. For more information on gc_WaitCall( ), see the Global Call API Library Reference.

Outbound calls

The CRN is assigned via the gc_MakeCall( ) function. For more information on this function, see the Global Call API Library Reference.

This CRN has a single LDID associated with it, for example, the line device on which the call was made. However, a single line device may have multiple CRNs associated with it (that is, more than

24 Global Call API for HMP on Windows Programming Guide — August 2006

Product Description

one call may exist on a given line). A line device can have a maximum of 20 CRNs associated with it. At any given instant, each CRN is a unique number within a process. After a call is terminated and the gc_ReleaseCallEx( ) function is called to release the resources used for the call, the CRN is no longer valid.

1.5.3 Object Identifiers and Resource Sharing Across Processes

The CRNs and LDIDs assigned by the Global Call API library can not be shared among multiple processes. These assigned CRNs and LDIDs remain valid only within the process invoked. That is, for call control purposes, you should not open the same physical device from more than one process, nor from multiple threads in a Windows environment. Unpredictable results may occur if this advice is not followed.

1.5.4 Target Objects

A target object provides a way of identifying a particular entity that is maintained by a specific software module. In API function calls, the target object is specified by a pair of parameters, the target_type and target_ID:

target_type

Identifies the kind of software module and the entity that it maintains. For example, the target type GCTGT_GCLIB_CHAN represents the Global Call Library and a channel entity that it maintains.

target_ID

Identifies the specific target object, such as a line device ID (LDID), which is generated by Global Call at runtime.

Table 2 shows the combinations of physical or logical entities and software module entities that can make up a target type (target_type).

Table 2. Supported Target Types

Software Module

GCLib SSSS

CCLib S S S S

Protocol SV SV SV

Firmware SV SV

S = Supported SV = Supported with Variances, see the appropriate Global Call Technology Guide for more information.

The possible software modules include:

• GCLib

• CCLib

• Protocol

Entity

System Network Interface Channel CRN

Global Call API for HMP on Windows Programming Guide — August 2006 25

Product Description

• Firmware

The possible entities include:

System

NIC for IP technology; all physical boards for E1, T1 and ISDN technologies

Network Interface

logical board or virtual board

Channel

time slot

CRN

call reference number

A target type (target_type) name is composed of the prefix, GCTGT, which stands for Global Call Target, a software module name, such as GCLIB, and an entity name, such as NETIF. For example, the target type GCTGT_GCLIB_NETIF, indicates that the desired target type is a network interface maintained by the Global Call library.

A target ID (target_ID) identifies the specific object that is located within the category defined by the target type (target_type). A target ID can be any of the following:

• line device ID (LDID)

• call reference number (CRN)

• Global Call library ID (GCGV_LIB)

• call control library ID (CCLib ID)

• protocol ID

The types and IDs for target objects are defined at the Global Call level. Table 3 shows the target types, as described in Table 2, with various target IDs to represent valid target objects.

Table 3. Target Types and Target IDs

Target Type Target ID Description

GCTGT_GCLIB_SYSTEM

GCTGT_CCLIB_SYSTEM †

GCTGT_GCLIB_NETIF Global Call Line device ID Network interface target object in Global

GCTGT_CCLIB_NETIF Global Call Line device ID Network interface target object in call

GCTGT_GCLIB_CHAN Global Call Line device ID Channel target object in Global Call library

GCTGT_CCLIB_CHAN Global Call Line device ID Channel target object in call control library

† For E1, T1 and ISDN technologies only. ‡ Target types that can only be used by functions issued in synchronous mode. If a function uses one of these target types in asynchronous mode, an error will be generated. The functions that can use these target types are gc_GetConfigData( ) (E1, T1 and ISDN technologies only), gc_SetConfigData( ), gc_ReqService( ), and gc_RespService( ).

‡ GCGV_LIB Global Call library module target object.

‡ CCLib ID Call control library module target object.

Call Library module.

control library module.

module.

26 Global Call API for HMP on Windows Programming Guide — August 2006

Table 3. Target Types and Target IDs (Continued)

Target Type Target ID Description

GCTGT_GCLIB_CRN Global Call CRN CRN target object in Global Call library

GCTGT_CCLIB_CRN Global Call CRN CRN target object in call control library

Target Object Availability

Except for the GCTGT_GCLIB_SYSTEM target object, all target IDs are generated or assigned by the Global Call API when the target object is created (for physical targets) or loaded (for software targets). Table 4 shows when a target object becomes available and when it becomes unavailable, depending on the target type.

Table 4. Target Object Availability

Product Description

module.

Target Type Target Object Available Target Object Unavailable

GCTGT_GCLIB_SYSTEM

GCTGT_CCLIB_SYSTEM †

GCTGT_GCLIB_CRN

GCTGT_CCLIB_CRN

GCTGT_GCLIB_NETIF

GCTGT_CCLIB_NETIF

GCTGT_GCLIB_CHAN

GCTGT_CCLIB_CHAN

† For E1, T1 and ISDN technologies only.

Retrieving Target IDs

Before the Global Call application can retrieve, update, or query the configuration data of a target object, it should obtain the target ID as shown in Table 5.

Table 5. Obtaining Target IDs

Target ID Procedure for Obtaining Target ID

GCGV_LIB After the call control library has been successfully started (that is, after the

Global Call Line Device ID † After a line device is opened, the CCLib ID and protocol ID (if applicable)

After gc_Start( ) After gc_Stop( )

After a call is created (gc_MakeCall( ) returns or GCEV_OFFERED is received)

After gc_OpenEx( ) After gc_Close( )

gc_Start( ) function is called), the target object’s CCLib ID can be obtained by calling the gc_CCLibNameToID( ) function.

associated with this line device can be obtained by the gc_GetConfigData( ) function with the set ID and parameter ID as (GCSET_CCLIB_INFO, GCPARM_CCLIB_ID) and (GCSET_PROTOCOL, GCPARM_PROTOCOL_ID).

After gc_ReleaseCallEx( )

Global Call API for HMP on Windows Programming Guide — August 2006 27

Product Description

Table 5. Obtaining Target IDs

Target ID Procedure for Obtaining Target ID

Global Call CRN After a call target object is created, its target object ID (that is, the Global Call

† For E1, T1 and ISDN technologies only.

CRN) will be an output of the gc_MakeCall( ) function or provided by the metaevent associated with the GCEV_OFFERED event.

28 Global Call API for HMP on Windows Programming Guide — August 2006

2.Programming Models

This chapter describes the programming models supported by Global Call. Topics include:

• Programming Models Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

• Asynchronous Mode Programming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.1 Programming Models Overview

The Global Call development software supports application development using asynchronous programming models. By usage, the asynchronous models are often said to use asynchronous mode. Asynchronous mode programming is introduced briefly in this chapter and described in more detail in the Standard Runtime Library API Programming Guide.

2.2 Asynchronous Mode Programming

Programming in asynchronous mode in Windows is described in the following topics:

• Asynchronous Model Overview

• Asynchronous Model with Event Handlers

• Asynchronous with Windows Callback Model

• Asynchronous with Win32 Synchronization Model

• Extended Asynchronous Programming Model

2.2.1 Asynchronous Model Overview

Asynchronous mode programming is characterized by the calling thread performing other processing while a function executes. At completion, the application receives event notification from the SRL and then the thread continues processing the call on a particular channel.

A function called in the asynchronous mode returns control immediately after the request is passed to the device driver and allows thread processing to continue. A termination event is returned when the requested operation completes, thus allowing the Intel Dialogic operation (state machine processing) to continue.

Caution: In general, when a function is called in asynchronous mode, and an associated termination event

exists, the gc_Close( ) function should not be called until the termination event has been received. In order to disable gc_WaitCall( ), gc_ResetLineDev( ) should be called. If this is not done, there are potential race conditions under which the application may crash with a segmentation fault.

Functions may be initiated asynchronously from a single thread and/or the completion (termination) event can be picked up by the same or a different thread that calls the sr_waitevt( )

Global Call API for HMP on Windows Programming Guide — August 2006 29

Programming Models

and gc_GetMetaEvent( ) functions. When these functions return with an event, the event information is stored in the METAEVENT data structure. The event information retrieved determines the exact event that occurred and is valid until the sr_waitevt( ) and gc_GetMetaEvent( ) functions are called again.

For Windows environments, the asynchronous models provided for application development also include:

• asynchronous model with event handlers

• asynchronous with Windows callback

• asynchronous with Win32 synchronization

• extended asynchronous programming

The asynchronous programming models are recommended for more complex applications that require coordinating multiple tasks. Asynchronous model applications typically run faster than synchronous models and require lower levels of system resources. Asynchronous models reduce processor loading because of the reduced number of threads inherent in asynchronous models and the elimination of scheduling overhead. Asynchronous models use processor resources more efficiently because multiple channels are handled in a single thread or in a few threads. See

Section 5.1, “General Programming Tips”, on page 77 for details. Of the asynchronous models, the

asynchronous with SRL callback model and the asynchronous with Windows callback model provide the tightest integration with the Windows message/event mechanism. Asynchronous model applications are typically more complex than corresponding synchronous model applications due to a higher level of resource management (that is, the number of channels managed by a thread and the tracking of completion events) and the development of a state machine.

After the application issues an asynchronous function, the application uses the sr_waitevt( ) function to wait for events on Intel Dialogic devices. All event coding can be accomplished using switch statements in the main thread. When an event is available, event information may be retrieved using the gc_GetMetaEvent( ) function. Retrieved event information is valid until the sr_waitevt( ) function is called again. The asynchronous model does not use event handlers to process events.

In this model, the SRL handler thread must be initiated by the application by setting the SR_MODELTYPE value to SR_STASYNC.

2.2.2 Asynchronous Model with Event Handlers

The asynchronous with event handlers model uses the sr_enbhdlr( ) function to automatically create the SRL handler thread. The application does not need to call the sr_waitevt( ) function since the thread created by the sr_enbhdlr( ) already calls the sr_waitevt( ) function to get events. Each call to the sr_enbhdlr( ) function allows the Intel Dialogic events to be serviced when the operating system schedules the SRL handler thread for execution.

Note: The SR_MODELTYPE value must not be set to SR_STASYNC because the SRL handler thread

must be created by the sr_enbhdlr( ) call. The event handler must not call the sr_waitevt( ) function or any synchronous Intel Dialogic function.

30 Global Call API for HMP on Windows Programming Guide — August 2006

Individual handlers can be written to handle events for each channel. The SRL handler thread can be used when porting applications developed for other operating systems.

2.2.3 Asynchronous with Windows Callback Model

The asynchronous with Windows callback model allows an asynchronous application to receive SRL event notification through the standard Windows message handling scheme. This model is used to achieve the tightest possible integration with the Windows messaging scheme. Using this model, the entire Intel Dialogic portion of the application could be run on a single thread. This model calls the sr_NotifyEvt( ) function once to define a user-specified application window handle and a user-specified message type. When an event is detected, a message is sent to the application window. The application responds by calling the sr_waitevt( ) function with a 0 timeout value. For Global Call events and optionally for non-Global Call events, the application must then call the gc_GetMetaEvent( ) function before servicing the event.

In this model, the SRL event handler thread must be initiated by the application by setting the SR_MODELTYPE value to SR_STASYNC. For detailed information on this programming model, see the Standard Runtime Library API Programming Guide.

Programming Models

2.2.4 Asynchronous with Win32 Synchronization Model

The asynchronous with Win32 synchronization model allows an asynchronous application to receive SRL event notification through standard Windows synchronization mechanisms. This model uses one thread to run all Intel Dialogic devices and thus requires a lower level of system resources than the synchronous model. This model allows for greater scalability in growing systems. For detailed information on this programming model, see the Standard Runtime Library API Programming Guide.

2.2.5 Extended Asynchronous Programming Model

The extended asynchronous programming model is basically the same as the asynchronous model except that the application uses multiple asynchronous threads, each of which controls multiple devices. In this model, each thread has its own specific state machine for the devices that it controls. Thus, a single thread can look for separate events for more than one group of channels. This model may be useful, for example, when you have one group of devices that provides fax services and another group that provides interactive voice response (IVR) services, while both groups share the same process space and database resources. The extended asynchronous model can be used when an application needs to wait for events from more than one group of devices and requires a state machine.

Because the extended asynchronous model uses only a few threads for all Intel Dialogic devices, it requires a lower level of system resources than the synchronous model. This model also enables using only a few threads to run the entire Intel Dialogic portion of the application.

Global Call API for HMP on Windows Programming Guide — August 2006 31

Programming Models

Whereas default asynchronous programming uses the sr_waitevt( ) function to wait for events specific to one device, extended asynchronous programming uses the sr_waitevtEx( ) function to wait for events specific to a number of devices (channels).

Note: Do not use the sr_waitevtEx( ) function in combination with either the sr_waitevt( ) function or

event handlers.

This model can run an entire application using only a few threads. When an event is available, the gc_GetMetaEventEx( ) function must be used to retrieve event-specific information. The values returned are valid until the sr_waitevtEx( ) function is called again. Event commands can be executed from the main thread through switch statements; the events are processed immediately.

The extended asynchronous model calls the sr_waitevtEx( ) function for a group of devices (channels) and polls for (waits for) events specific to that group of devices. In this model, the SRL event handler thread is not created (the SR_MODELTYPE value is set to SR_STASYNC) and the sr_enbhdlr( ) function in not used.

In the extended asynchronous model, functions are initiated asynchronously from different threads. A thread waits for events using the sr_waitevtEx( ) function. The event information can be retrieved using the gc_GetMetaEventEx( ) function. When this function returns, the event information is stored in the METAEVENT data structure.

Caution: When calling the gc_GetMetaEventEx( ) function from multiple threads, ensure that your

application uses unique thread-related METAEVENT data structures (thread local variables or local variables), or ensure that the METAEVENT data structure is not overwritten until all processing of the current event has completed.

The event information retrieved determines the exact event that occurred and is valid until the sr_waitevtEx( ) function returns with another event.

32 Global Call API for HMP on Windows Programming Guide — August 2006

3.Call State Models

This chapter describes the call state models provided by Global Call. Topics include the following:

• Call State Model Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

• Basic Call Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

• Basic Call Model Configuration Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

• Basic Call Control in Asynchronous Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

• Advanced Call Control with Call Hold and Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . 64

3.1 Call State Model Overview

Global Call maintains a generic call model from which technology-specific call models can be derived. Some technologies support only a subset of the complete call model. The call establishment and termination procedures are based on this call model. The following sections describe the call states associated with the basic call model and configuration options.

3.2 Basic Call Model

Each call received or generated by Global Call is processed through a series of states, where each state represents the completion of certain tasks or the current status of the call. Some states in the basic call model are optional and can be enabled or disabled selectively. Only the optional states can be enabled or disabled. Every technology or call control library has a default call state model consisting of all the states it can possibly support from the basic call model. If a state is disabled, all corresponding events are disabled. If a state is enabled, all corresponding events are enabled.

The call states change in accordance with the sequence of functions called by the application and the events that originate in the network and system hardware. The current state of a call can be changed by:

• Function call returns

• Termination events (indications of function completion)

• Unsolicited events

The states of the basic call model are described in the following sections:

• Basic Call States at the Inbound Interface

• Basic Call States at the Outbound Interface

• Basic Call States for Call Termination

Global Call API for HMP on Windows Programming Guide — August 2006 33

Call State Models

3.2.1 Basic Call States at the Inbound Interface

The basic inbound call states are as follows:

Null state (GCST_NULL)

This state indicates that no call is assigned to the channel (time slot or line). This is the initial state of a channel when it is first opened. This state is also reached when a call is released or after the channel is reset. A channel in this state is available for inbound calls after being initialized to receive incoming calls.

Call Detected (GCST_DETECTED)

An incoming call has been received but not yet offered to the application. In this state, the call is being processed, which typically involves waiting for more information or allocating a resource. Although the call is not yet offered to the application, this state is for informational purposes to reduce glare conditions since the application is aware of the presence of a call on the channel.

Call Offered (GCST_OFFERED)

This state exists for an incoming call when the user application has received a call establishment request but has not yet responded. The newly arrived inbound call is offered to the user application to be accepted, answered, rejected, etc. Call information is typically available at this time to be examined so that the application can determine the appropriate action to take with regards to the call.

Get More Information (GCST_GETMOREINFO)

This state exists for an incoming call when the network has received an acknowledgement of the call establishment request, which permits the network to send additional call information (if any) in the overlap mode. The application is waiting for more information, typically called party number digits. (This state is optional and may not be supported in all technologies. See the appropriate Global Call Technology Guide for information.) This state applies to E1, T1 and ISDN technologies only.

Call Routing (GCST_CALLROUTING)

This state exists for an incoming call when the user has sent an acknowledgement that all call information necessary to effect call establishment has been received. The acknowledgement can be sent from the Offered or the GetMoreInfo state if all the information has been received. This transition typically involves the sending of Call Routing tones or technology specific messages; for example, in the case of ISDN, a CALL_PROCEEDING message is sent. The application can now accept or answer the call. (This state is optional and may not be supported in all technologies. See the appropriate Global Call Technology Guide for information.)

Call Accepted (GCST_ACCEPTED)

This state indicates that the incoming call was offered and accepted by the application. The user on the inbound side has indicated to the calling party that the destination user is alerting or ringing but has not yet answered.

Call Connected (GCST_CONNECTED)

This is a common state that exists for an incoming call when the user has answered the call.

34 Global Call API for HMP on Windows Programming Guide — August 2006

3.2.2 Basic Call States at the Outbound Interface

The basic outbound call states are as follows:

Null state (GCST_NULL)

This state indicates that no call is assigned to the channel (time slot or line). This is the initial state of a channel when it is first opened. This state is also reached when a call is released or after the channel is reset. The channel in this state is available for making outbound calls.

Call Dialing (GCST_DIALING)

This state exists for an outgoing call when an outbound call request is made. The call signaling or message is in the process of being prepared for transfer or being transferred across the telephony network. In response, the remote side may request more information,acknowledge the call, accept the call or answer the call.

Send More Information (GCST_SENDMOREINFO)

This state exists for an outgoing call when the user has received an acknowledgement of the call establishment request that permits or requests the user to send additional call information to the network in overlap mode. The information, typically digits, is in the process of being prepared for transfer or being transferred across the telephony network (overlap sending or partial dialing). (This state is optional and may not be supported in all technologies. See the appropriate Global Call Technology Guide for information.) This state applies to E1, T1 and ISDN technologies only.

Call State Models

Call Proceeding (GCST_PROCEEDING)

This state exists for an outgoing call when the user has received an acknowledgement that all call information necessary to effect call establishment has been received and the call is proceeding. The remote side can now accept or answer the call. (This state is optional and may not be supported in all technologies. See the appropriate Global Call Technology Guide for information.)

Call Alerting (GCST_ALERTING)

This state exists for an outgoing call when the calling user has received an indication that remote user alerting has been initiated, typically ringing. The outbound call has been delivered to the remote party, which has not yet answered the call.

Call Connected (GCST_CONNECTED)

This is a common state that exists for an outgoing call when the user has received an indication that the remote user has answered the call. The calling and called parties are connected and the call is therefore active on the related call channel.

3.2.3 Basic Call States for Call Termination

The basic call termination states are as follows:

Call Disconnected (GCST_DISCONNECTED)

This state indicates that the remote party has disconnected the call. The remote party can disconnect the call prior to establishing a connection, that is, while the call setup is in progress. Thus, the call does not have to be in the connected state before it can be disconnected. The user must respond by dropping the call and releasing the internal resources allocated for the call.

Global Call API for HMP on Windows Programming Guide — August 2006 35

Call State Models

Call Idle (GCST_IDLE)

This state indicates that the local user has dropped the call. This may be a termination initiated by the local user or a response to the remote side disconnecting the call. While the call no longer exists, internal system resources committed to servicing the call are still present. The user must release these resources, as they are no longer required.

3.3 Basic Call Model Configuration Options

Depending on the specific technology, the following options are available for configuring the technology call control layer or the application:

Call State

If a state is disabled, the corresponding call state event is also disabled.

Call State Event

Call state transition events are masked so that the events are not generated.

Call Acknowledgement

An acknowledgement is sent to indicate to the remote side that the call has been received but more information is required to proceed with the call.

Call Proceeding

Call proceeding information is sent to the remote side when an incoming call is received and all the information required to proceed with the call is available.

Minimum Information

A minimum amount of destination address information, such as DNIS, is collected before the call is offered to the application.

3.3.1 Call State Configuration

Some states in the basic call model are optional and can be enabled or disabled selectively. Every technology or call control library has a default call state model consisting of all the states it can possibly support from the basic call model. If a state is disabled, the corresponding call state event will also be disabled. If a state is enabled, the event mask setting still determines which call state events are sent to the application.

This configuration can be done by issuing the gc_SetConfigData( ) function with a target_type of GCTGT_GCLIB_CHAN and a target_ID of a line device, and passing the appropriate set ID and parameter IDs. The set ID used in this context is GCSET_CALLSTATE_MSK and the relevant parameter IDs are:

GCACT_ADDMSK

Enable the call states specified in the value in addition to other states already enabled.

GCACT_SUBMSK

Disable all the call states specified in the value.

GCACT_SETMSK

Enable the call states specified in the value and disable other optional states that are already enabled.

36 Global Call API for HMP on Windows Programming Guide — August 2006

The GCACT_ADDMSK, GCACT_SUBMSK and GCACT_SETMSK parameter IDs can be assigned one of the following values (of type GC_VALUE_LONG), or an ORed combination of the values:

• GCMSK_ALERTING_STATE

• GCMSK_CALLROUTING_STATE (for E1, T1, and ISDN technologies only)

• GCMSK_DETECTED_STATE

• GCMSK_GETMOREINFO_STATE (for E1, T1, and ISDN technologies only)

• GCMSK_PROCEEDING_STATE

• GCMSK_SENDMOREINFO_STATE (for E1, T1, and ISDN technologies only)

See the Global Call API Library Reference for more information on the gc_SetConfigData( ) function.

3.3.2 Call State Event Configuration

Some call state transition events can be masked so that the events are not generated. Although an event may be masked, the corresponding call state transition can still take place. This configuration can be done by issuing the gc_SetConfigData( ) function with a target_type of GCTGT_GCLIB_CHAN and a target_ID of a line device, and passing the appropriate set ID and parm IDs.

Call State Models

The set ID used in this context is GCSET_CALLEVENT_MSK and the relevant parm IDs are:

GCACT_ADDMSK

Enable the notification of events specified in the value in addition to previously enabled events.

GCACT_SUBMSK

Disable notification of the events specified in the value.

GCACT_SETMSK

Enable the notification of events specified in the value and disables notification of any event not specified.

The GCACT_ADDMSK, GCACT_SUBMSK and GCACT_SETMSK parm IDs can be assigned one of the following values (of type GC_VALUE_LONG), or an ORed combination of the values:

• GCMSK_ALERTING

• GCMSK_DETECTED

• GCMSK_DIALING

• GCMSK_PROCEEDING

• GCMSK_REQMOREINFO (for E1, T1, and ISDN technologies only)

Note: Using the gc_SetConfigData( ) function with a target_ID of a board device to mask events for all

devices associated with a board is not supported. Call state events can be masked on a per line device basis only.

Global Call API for HMP on Windows Programming Guide — August 2006 37

Call State Models

See the Global Call API Library Reference for more information on the gc_SetConfigData( ) function.

3.3.3 Call Acknowledgement Configuration

Note: This functionality applies to E1, T1 and ISDN technologies only.

When an incoming call is received, an acknowledgement is typically sent to the remote side to indicate that the call was received. In some technologies, if the incoming call does not have sufficient information, this acknowledgement also indicates to the remote side that more information is required to proceed with the call (see Section 3.4.1.8, “Overlap Receiving” for more information). Either the technology call control layer or the application can be configured to send the acknowledgement. This configuration can be set by the application issuing the gc_SetConfigData( ) function. The set ID used in this context is GCSET_CALL_CONFIG and the relevant parm ID is:

GCPARM_CALLACK

Specify whether call acknowledgement is provided by the application or the technology call control layer.

The GCPARM_CALLACK parm ID can be assigned one of the following values (of type GC_VALUE_INT):

• GCCONTROL_APP (application controlled)

• GCCONTROL_TCCL (technology call control layer controlled)

See the Global Call API Library Reference for more information on the gc_SetConfigData( ) function.

3.3.4 Call Proceeding Configuration

When an incoming call is received and all the information required to proceed with the call is available, an indication that the call is proceeding is usually sent to the remote side for informational purposes. Either the technology call control layer or the application can be configured to send a call proceeding indication to the remote side. This can be done by issuing the gc_SetConfigData( ) function. The set ID used in this context is GCSET_CALL_CONFIG and the relevant parm ID is:

GCPARM_CALLPROC

Specify whether call proceeding indication is provided by the application or the technology call control layer.

The GCPARM_CALLPROC parm ID can be assigned one of the following values (of type GC_VALUE_INT):

• GCCONTROL_APP (application controlled)

• GCCONTROL_TCCL (technology call control layer controlled)

See the Global Call API Library Reference for more information on the gc_SetConfigData( ) function.

38 Global Call API for HMP on Windows Programming Guide — August 2006

Call State Models

3.4 Basic Call Control in Asynchronous Mode

This section describes and illustrates the basic call model and state transitions for call control in asynchronous mode. This section also describes the process for call establishment for both inbound and outbound calls and call termination in the asynchronous mode.

The procedures for establishing and terminating calls in the asynchronous mode are described in the following sections:

• Inbound Calls in Asynchronous Mode

• Outbound Calls in Asynchronous Mode

• Call Termination in Asynchronous Mode

Note: For E1, T1 and ISDN technologies, the Advanced Call Model includes call states associated with

holding, retrieving and transferring calls. See Section 3.5, “Advanced Call Control with Call Hold

and Transfer” for more information.

Caution: In general, when a function is called in asynchronous mode, and an associated termination event

exists, the gc_Close( ) function should not be called until the termination event has been received. Otherwise, the behavior is undefined.

3.4.1 Inbound Calls in Asynchronous Mode

This section describes how calls are established and shows call scenarios for asynchronous inbound calls. The following topics describe the processing of inbound calls in asynchronous mode:

• Inbound Calls in Asynchronous Mode Overview

• Channel Initialization

• Call Detection

• Call Offered

• Call Routing

• Call Acceptance

• Call Establishment

• Overlap Receiving (for E1, T1, and ISDN technologies only)

• Call Failure

• Abandoned Calls

• Inbound Call Scenarios in Asynchronous Mode

3.4.1.1 Inbound Calls in Asynchronous Mode Overview

Figure 4 illustrates a Basic Inbound Call Model, which shows the call states associated with establishing a call in asynchronous mode. All calls start from a Null state. The call establishment process for inbound calls is shown in Figure 4.

See Table 6, “Asynchronous Inbound Call State Transitions”, on page 41 for a summary of the call state transitions.

Global Call API for HMP on Windows Programming Guide — August 2006 39

Call State Models

Figure 4. Basic Asynchronous Inbound Call State Diagram

gc_WaitCall()

(called only once)

NULL

GCEV_DETECTED

(maskable)

GetMoreInfo

gc_ReqMoreInfo()

GCEV_MOREINFO

gc_AnswerCall()

GCEV_ANSWERED

CONNECTED

Legend:

Required

Optional

DETECTED

GCEV_OFFERED

gc_CallAck(MORE_INFO)

GCEV_MOREINFO

gc_CallAck(CALL_PROC)

GCEV_CALLPROC

gc_AcceptCall()

GCEV_ACCEPT

gc_AnswerCall()

GCEV_ANSWERED

GCEV_OFFERED

OFFERED

gc_CallAck(CALL_PROC)

GCEV_CALLPROC

CallRouting

gc_AcceptCall()

GCEV_ACCEPT

gc_AcceptCall()

GCEV_ACCEPT

ACCEPTED

gc_AnswerCall()

GCEV_ANSWERED

Note: In Figure 4, the GetMoreInfo state and all transitions to/from that state apply to E1, T1, and ISDN

technologies only.

40 Global Call API for HMP on Windows Programming Guide — August 2006

Table 6. Asynchronous Inbound Call State Transitions

Call State Models

State Description Previous/Next State

Accepted (GCST_ACCEPTED) Maskable

Call Routing (GCST_CALLROUTING)

Maskable

Connected (GCST_CONNECTED)

Not Maskable

Detected (GCST_DETECTED) Maskable

GetMoreInfo (GCST_GETMOREINFO) †

Maskable

Previous: Offered, GetMoreInfo †, CallRouting Next: GCEV_ANSWERED -> Connected state GCEV_DISCONNECTED -> Disconnected state GCEV_DROP CALL -> Idle state

Previous: Offered, GetMoreInfo †

Next: GCEV_ANSWERED -> Connected state GCEV_ACCEPT ->

Accepted state GCEV_DISCONNECTED ->

Disconnected state GCEV_DROPCALL -> Idle state

Previous: Accept, Offered, GetMoreInfo †,CallRouting, Dialing, SendMoreInfo †,Proceeding, Alerting

Next: GCEV_DISCONNECTED ->

Disconnected state GCEV_DROPCALL -> Idle

state

Previous: Null Next:

GCEV_OFFERED -> Offered state GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL -> Idle state

Previous: Offered Next:

GCEV_ANSWERED -> Connected state

GCEV_MOREINFO -> GetMoreInfo state

GCEV_ACCEPT -> Accepted state GCEV_CALLPROC -> CallRouting state GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL -> Idle state

† Applies to E1, T1 and ISDN technology only.

Valid Call State

Tra nsiti on

Call Transition Events

Functions

gc_AnswerCall( ), gc_DropCall( )

gc_AnswerCall( ), gc_AcceptCall( ), gc_DropCall( )

gc_DropCall( ) GCEV_DISCONNECTED

gc_DropCall( ) GCEV_DISCONNECTED,

gc_ReqMoreInfo( ), gc_CallAck( ) gc_AnswerCall( ), gc_AcceptCall( ), gc_DropCall( )

GCEV_DISCONNECTED, GCEV_DROPCALL or GCEV_ANSWERED

GCEV_DISCONNECTED, GCEV_DROPCALL, GCEV_ACCEPT or GCEV_ANSWERED

GCEV_DROPCALL

GCEV_DROPCALL, GCEV_OFFERED,

GCEV_DISCONNECTED, GCEV_DROPCALL, GCEV_ACCEPT, GCEV_ANSWERED, GCEV_MOREINFO or GCEV_CALLPROC

Global Call API for HMP on Windows Programming Guide — August 2006 41

Call State Models

Table 6. Asynchronous Inbound Call State Transitions

State Description Previous/Next State

Null (GCST_NULL) Not Maskable

Offered (GCST_OFFERED) Not Maskable

† Applies to E1, T1 and ISDN technology only.

Previous: Idle Next: gc_WaitCall( ) ->

Null state gc_ResetLineDev( ) ->

Null state GCEV_OFFERED -> Offered state GCEV_DETECTED ->

Detected state

Previous: Null, Detected Next:

GCEV_ANSWERED -> Connected state

GCEV_ACCEPT -> Accepted state GCEV_CALLPROC -> CallRouting state GCEV_MOREINFO -> GetMoreInfo state † GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL -> Idle state

Valid Call State

Transition Functions

gc_WaitCall( ) GCEV_DETECTED,

gc_CallAck( ), gc_AnswerCall( ), gc_AcceptCall( ), gc_DropCall( )

Call Transition Events

GCEV_OFFERED

GCEV_DISCONNECTED, GCEV_DROPCALL, GCEV_ACCEPT, GCEV_ANSWERED, GCEV_MOREINFO †, GCEV_CALLPROC

The following sections describe the asynchronous inbound call processes.

3.4.1.2 Channel Initialization

To establish calls, the following conditions must be met:

• The condition of the line device must be unblocked. When a channel is initially opened, the

initial condition of a line device is blocked. A “blocking” condition on a line device is indicated by the reception of a GCEV_BLOCKED event and an “unblocking” condition on a line device is indicated by the reception of a GCEV_UNBLOCKED event. The GCEV_BLOCKED and GCEV_UNBLOCKED events are sent as unsolicited events to the application in response to blocking alarms. GCEV_BLOCKED and GCEV_UNBLOCKED events are related to layer 1 alarms, as well as to channel states (service status in T1 ISDN, bit states in CAS). GCEV_BLOCKED and GCEV_UNBLOCKED are used as what might be termed flow-control events within the application. For more information on blocking alarms and the GCEV_BLOCKED and GCEV_UNBLOCKED events, see Section 4.3, “Blocked and

Unblocked Event Handling”. When the condition of the line device is unblocked, the line

device is ready for establishing calls.

• The call state of the channel must be in the Null state. This is the initial call state of a line

device when it is first opened. This state is also reached when a call is released or after the channel is reset.

If the above conditions are met, the application or thread must issue a gc_WaitCall( ) function in the Null state to indicate readiness to accept an inbound call request on the specified line device. In the asynchronous mode, the gc_WaitCall( ) function must be called only once after the line device

42 Global Call API for HMP on Windows Programming Guide — August 2006

is opened using the gc_OpenEx( ) function. However, if the gc_ResetLineDev( ) function was issued, gc_WaitCall( ) must be reissued. In asynchronous mode, it is not necessary to issue gc_WaitCall( ) again after a call is released.

Note: After gc_WaitCall( ) is issued to wait for incoming calls on a line device, it is possible to use

gc_makeCall( ) to make an outbound calls on that line device.

3.4.1.3 Call Detection

The inbound call from the network is received on the line device specified in the gc_WaitCall( ) function, but the call has not been offered to the application. The technology call control layer typically sends an acknowledgement to the remote side. In some configurations, this acknowledgement can also be sent by the application when the call is offered. At this stage, the call is being processed, which typically involves allocating resources or waiting for more information. The GCEV_DETECTED event is generated, if enabled. If the GCEV_DETECTED event is generated, a new CRN is assigned to the incoming call. This event is for informational purposes to reduce glare conditions as the application is now aware of the presence of a call on the channel.

Note: When developing applications, if the GCEV_DETECTED event is not supported, a

GCEV_DISCONNECTED event is only received if the host application already received the GCEV_OFFERED event before the remote side disconnects.

Call State Models

3.4.1.4 Call Offered

When an incoming call is received in en-bloc mode, where all the information required is available, the call is offered to the application by generating an unsolicited GCEV_OFFERED event (equivalent to a “ring detected” notification). This GCEV_OFFERED event causes the call to change to the Offered state. In the Offered state, a CRN is assigned as a means of identifying the call on a specific line device. If a GCEV_DETECTED event was generated before the GCEV_OFFERED event, the same CRN is assigned as the one assigned when the GCEV_DETECTED event was generated.

When using E1, T1 and ISDN technology, if the incoming call does not have sufficient information, the call is offered to the application when all the required information is received. If the technology is configured to accept minimum information, the call is offered to the application when the specified minimum amount of information is received. In this case, the application must request additional information if required. See Section 3.4.1.8, “Overlap Receiving” for more information.

A call proceeding indication can be sent by the technology call control layer, or by the application by issuing the gc_CallAck(GCACK_SERVICE_PROC) function. Otherwise, the application can accept or answer the call by issuing the gc_AcceptCall( ) or gc_AnswerCall( ) functions, respectively.

Note: When developing applications, if the GCEV_DETECTED event is not supported, a

GCEV_DISCONNECTED event is only received if the host application already received the GCEV_OFFERED event before the remote side disconnects.

Global Call API for HMP on Windows Programming Guide — August 2006 43

Call State Models

3.4.1.5 Call Routing

After the call has been offered, a call proceeding indication can be sent to the remote party to indicate that all the information has been received and the call is now proceeding. This indication can be sent by the technology call control layer or by the application by issuing the gc_CallAck(GCACK_SERVICE_PROC) function. This stage typically involves routing the call to the destination exchange or party. An information call routing tone can be played at this point to inform the remote party that the call is routing.

3.4.1.6 Call Acceptance

If the application or thread is not ready to answer the call, a gc_AcceptCall( ) function is issued to indicate to the remote end that the call was received but not yet answered. This provides an interval during which the system can verify parameters, determine routing, and perform other tasks before connecting the call. A GCEV_ACCEPT event is generated when the gc_AcceptCall( ) function is successfully completed and the call changes to the Accepted state. The application can then answer the call by issuing the gc_AnswerCall( ) function.

3.4.1.7 Call Establishment

When the call is to be directly connected, such as to a voice messaging system, or if the application or thread is ready to answer the call, a gc_AnswerCall( ) function is issued to make the final connection. Upon answering the call, a GCEV_ANSWERED event is generated and the call changes to the Connected state. At this point, the call is connected to the called party and call charges begin.

3.4.1.8 Overlap Receiving

Note: This functionality applies to E1, T1 and ISDN technologies only.

After an incoming call has been received, the call is offered to the application based on the call acknowledgement configuration and the availability of information required for proceeding with the call. If the incoming call is in en-bloc mode where all the information required for processing the call is present, the call is offered to the application. Otherwise, the call is offered to the application based on the following configurations:

Call Acknowledgement

If the application is configured to send the call acknowledgement, the call is immediately offered to the application regardless of the amount of information available. The application can then request and collect more information as required. If the technology call control layer is configured to send the call acknowledgement, then the call is offered to the application based on the minimum amount of information specified.

Minimum Information Specified

If the incoming call does not have sufficient information, the call is offered to the application based on the amount of information required. If the technology is configured to accept minimum information, the call is offered to the application only after the specified minimum amount of information is received. Thereafter, the application can request and collect more information as required. If the technology is not configured to accept minimum information,

44 Global Call API for HMP on Windows Programming Guide — August 2006

Call State Models

then the call is offered to the application regardless of the amount of information available. The application can then request and collect more information as required.

The following sections describe various configurations operating in overlap receiving mode.

Scenario 1

In this scenario, the application is configured to acknowledge the incoming call and send a call proceeding indication after sufficient information has been received. When an incoming call is detected, the call is immediately offered to the application regardless of the amount of information available to proceed with the call.

When the call is in the Offered state (after the generation of the unsolicited GCEV_OFFERED event), the application sends an acknowledgement for the incoming call by issuing a gc_CallAck(GCACK_SERVICE_INFO) function. The application may selectively retrieve call information, such as Destination address and Origination address (caller ID) by issuing the gc_GetCallInfo( ) function. If more information is still required, the gc_ReqMoreInfo( ) function is issued to request more information. When the information is received, the GCEV_MOREINFO event is generated again. When all the required information is received, the application may send a call proceeding indication to the remote side by issuing the gc_CallAck( ) function. Otherwise, the application can choose to accept or answer the call.

Scenario 2

In this scenario, the technology call control layer is configured to acknowledge the incoming call and send a call proceeding indication after sufficient information has been received. When an incoming call is detected, the technology call control layer immediately sends an acknowledgement. If the minimum amount of information required is specified, then the call is offered to the application only after the minimum amount of information required is received. After the call is offered to the application, the address information can be retrieved to determine if more information is required. If more information is required, a gc_CallAck(GCACK_SERVICE_INFO) function must be issued. Since an acknowledgement was already sent out, nothing is sent to the remote side at this time. However, if the minimum amount of information is not specified, then the technology control layer requests and collects more information. After all the maximum amount of information expected is received, the technology control layer sends a call proceeding indication to the remote side. The call is then offered to the application, which can then accept or answer the call.

Scenario 3

In this scenario, the technology call control layer is configured to acknowledge the incoming call and the application is configured to send a call proceeding indication after sufficient information has been received. When an incoming call is detected, the technology call control layer immediately sends an acknowledgement. If the minimum amount of information required is specified, then the call is offered to the application only after the minimum amount of information required is received. Otherwise the call is immediately offered to the application.

When the call is in the Offered state (after generation of the unsolicited GCEV_OFFERED event), the application may selectively retrieve call information, such as the Destination and Origination

Global Call API for HMP on Windows Programming Guide — August 2006 45

Call State Models

address (caller ID) by issuing the gc_GetCallInfo( ) function. If more information is required, the application may also request more address information using the gc_CallAck(GCACK_SERVICE_INFO) function. Since an acknowledgement was already sent out, no acknowledgement is sent to the remote side at this time. When the additional information is received, the GCEV_MOREINFO event is generated. If more information is still required, the gc_ReqMoreInfo( ) function is issued to request more information. When the additional information is received, the GCEV_MOREINFO event is generated again. When all the required information is received, the application may send a call proceeding indication to the remote side by issuing the gc_CallAck(GCACK_SERVICE_PROC) function. Otherwise, the application can choose to accept or answer the call.

Scenario 4

In this scenario, the application is configured to acknowledge the incoming call and the technology call control layer is configured to send a call proceeding indication after sufficient information has been received. When an incoming call is detected, the call is offered to the application regardless of the amount of information available.

When the call is in the Offered state (after generation of the unsolicited GCEV_OFFERED event), the application sends an acknowledgement for the incoming call by issuing a gc_CallAck(GCACK_SERVICE_INFO). The application may selectively retrieve call information, such as Destination address and Origination address (caller ID) by issuing the gc_GetCallInfo( ) function. If more information is still required, the gc_ReqMoreInfo( ) function is issued to request more information. When the information is received, the GCEV_MOREINFO event is generated again. When all the required information is received, the technology call control layer sends a call proceeding indication to the remote side. The application may also attempt to send a call proceeding indication to the remote side in case the technology call control layer hasn’t done so. The application can then choose to accept or answer the call.

3.4.1.9 Call Failure

The following are various causes of call failures:

Call Rejection

From the Offered state, the application or thread may reject the call by issuing the

gc_DropCall( ) function followed by a gc_ReleaseCallEx( ) function (see the Global Call API Library Reference).

Forced Release (applies to E1, T1 and ISDN technologies only)

From the Accepted state, not all protocols support a forced release of the line, that is, issuing a gc_DropCall( ) function after a gc_AcceptCall( ) function. If a forced release is not supported and is attempted, the function will fail and an error will be returned. To recover, the application should issue the gc_AnswerCall( ) function followed by gc_DropCall( ) and gc_ReleaseCallEx( ) functions. However, any time a GCEV_DISCONNECTED event is received in the Accepted state, the gc_DropCall( ) function can be issued.

Task Fa i l u r e

If a call fails at any point in the call establishment process, that is, if a GCEV_TASKFAIL event is received by the application, the call stays in its current state. In most cases, the application needs to drop and release the call to return the line device to the Null state. However, in some cases, such as call failure due to a trunk error, the application needs to use

46 Global Call API for HMP on Windows Programming Guide — August 2006

the gc_ResetLineDev( ) function to reset the line device to the Null state. For more information, see the gc_DropCall( ), gc_ReleaseCallEx( ) and gc_ResetLineDev( ) function descriptions in the Global Call API Library Reference.

3.4.1.10 Abandoned Calls

During call establishment, the remote side may choose to hang up before call setup has been completed. The application must be capable of handling error conditions and the lack of complete information when requesting call information.

Note: The GCEV_DETECTED event is not supported. If the host application has not received a

GCEV_OFFERED event when the call is disconnected by the remote side, the host application will not receive any event. If the host application has already received a GCEV_OFFERED event, it receives a GCEV_DISCONNECTED event when the call is disconnected.

3.4.1.11 Inbound Call Scenarios in Asynchronous Mode

This section shows various asynchronous inbound call scenarios. For call scenarios used by a specific signaling protocol, check the Global Call Technology Guide for that technology.

Figure 5 shows a basic asynchronous call scenario for an incoming call.

Call State Models

Figure 5. Basic Asynchronous Inbound Call Scenario

GlobalCall

Application

GCEV_OFFERED

gc_GetCallInfo(DESTINATION_ADDRESS) gc_GetCallInfo(ORIGINATION_ADDRESS)

(Sufficient Information Received)

gc_AcceptCall()

GCEV_ACCEPTED

gc_AnswerCall()

GCEV_ANSWERED

Library/

Technology

Incoming Call

(All Information

Received)

Network

Alerting

Call Answered

Global Call API for HMP on Windows Programming Guide — August 2006 47

Call State Models

Figure 6 shows an asynchronous call scenario for an incoming call with call proceeding.

Figure 6. Incoming Call Scenario with Call Proceeding

GlobalCall

Application

GCEV_OFFERED

gc_GetCallInfo(DESTINATION_ADDRESS)

gc_GetCallInfo(ORIGINATION_ADDRESS)

(Sufficient Information Received)

gc_CallAck(GCACK_SERVICE_PROC)

GCEV_CALLPROC

Library/

Technology

Incoming Call

(All Information

Received)

Network

Call Proceeding

gc_AcceptCall()

gc_AnswerCall()

GCEV_ACCEPTED

GCEV_ANSWERED

Alerting

Call Answered

48 Global Call API for HMP on Windows Programming Guide — August 2006

Call State Models

Figure 7 shows an asynchronous call scenario for an incoming call with call acknowledgement and call proceeding controlled by the application. This scenario applies to E1, T1 and ISDN technologies only.

Figure 7. Call Acknowledgement and Call Proceeding Done at the Application Layer

GlobalCall

Application

Library/

Technology

Incoming Call

GCEV_OFFERED

Network

gc_GetCallInfo(DESTINATION_ADDRESS) gc_GetCallInfo(ORIGINATION_ADDRESS)

gc_CallAck(GCACK_SERVICE_INFO)

GCEV_MOREINFO

gc_GetCallInfo(DESTINATION_ADDRESS)

(New Information

gc_ReqMoreInfo(DESTINATION_ADDRESS)

GCEV_MOREINFO

gc_GetCallInfo(DESTINATION_ADDRESS)

(Sufficient Information Received)

gc_CallAck(GCACK_SERVICE_PROC)

GCEV_CALLPROC

gc_AcceptCall()

GCEV_ACCEPTED

gc_AnswerCall()

GCEV_ANSWERED

Acknowledgement

and Request for

More Address

Information

More Information

Buffered)

Call Proceeding

Alerting

Call Answered

Global Call API for HMP on Windows Programming Guide — August 2006 49

Call State Models

Figure 8 shows an asynchronous call scenario for an incoming call with call proceeding controlled by the application with the minimum information configuration. This scenario applies to E1, T1 and ISDN technologies only.

Figure 8. Call Proceeding Done by the Application Layer with Minimum Information

Configured

GlobalCall

Application

GCEV_DETECTED

GCEV_OFFERED

Library/

Technology

Incoming Call

More Information

(Minimum

Information

Received)

Network

Acknowledgement

and Request for

More Address

Information

gc_GetCallInfo(DESTINATION_ADDRESS)

gc_GetCallInfo(ORIGINATION_ADDRESS)

(New Information

gc_CallAck(GCACK_SERVICE_INFO)

GCEV_MOREINFO

gc_CallAck(GCACK_SERVICE_PROC)

GCEV_CALLPROC

gc_AcceptCall()

GCEV_ACCEPTED

gc_AnswerCall()

GCEV_ANSWERED

More Information

Buffered)

Call Proceeding

Alerting

Call Answered

50 Global Call API for HMP on Windows Programming Guide — August 2006

Call State Models

Figure 9 shows an asynchronous call scenario for an incoming call with call acknowledgement and call proceeding controlled by the call control layer.

Figure 9. Call Acknowledgement and Call Proceeding Done at Technology Call Control Layer

Application

Incoming Call

(Not Enough

GCEV_DETECTED

(All Information

GCEV_OFFERED

gc_GetCallInfo(DESTINATION_ADDRESS) /

gc_GetCallInfo(ORIGINATION_ADDRESS)

gc_AcceptCall()

GCEV_ACCEPTED

GlobalCall

Library/

Technology

Information

Received)

Network

Acknowledgement

and Request for

More Address

Information

More Information

Call Proceeding

Alerting

gc_AnswerCall()

GCEV_ANSWERED

Call Answered

Note: In Figure 9, the “Acknowledgement and Request for More Address Information” applies to E1, T1,

and ISDN technologies only.

Global Call API for HMP on Windows Programming Guide — August 2006 51

Call State Models

Figure 10 shows an asynchronous call scenario for an incoming call with call acknowledgement controlled by the call control layer and call proceeding controlled by the application.

Figure 10. Call Acknowledgement Done by the Technology Call Control Layer and Call

Proceeding Done by the Application

Application

Incoming Call

(Not Enough

GCEV_DETECTED

(All Information

GCEV_OFFERED

gc_GetCallInfo(DESTINATION_ADDRESS) /

gc_GetCallInfo(ORIGINATION_ADDRESS)

gc_CallAck(GCACK_SERVICE_PROC)

GCEV_CALLPROC

gc_AcceptCall()

GCEV_ACCEPTED

GlobalCall

Library/

Technology

Information

Received)

Network

Acknowledgement

and Request for

More Address

Information

More Information

Call Proceeding

Alerting

gc_AnswerCall()

GCEV_ANSWERED

Call Answered

Note: In Figure 10, the “Acknowledgement and Request for More Address Information” applies to E1,

T1, and ISDN technologies only.

3.4.2 Outbound Calls in Asynchronous Mode

This section describes how calls are established and shows call scenarios for asynchronous outbound calls. The following topics describe the processing of outbound calls in asynchronous mode:

• Outbound Calls in Asynchronous Mode Overview

• Channel Initialization

52 Global Call API for HMP on Windows Programming Guide — August 2006

• Call Dialing

• Call Proceeding

• Call Alerting

• Call Connected

• Overlap Sending (for E1, T1, and ISDN technologies only)

• Call Failure

• Outbound Call Scenarios in Asynchronous Mode

3.4.2.1 Outbound Calls in Asynchronous Mode Overview

Figure 11 illustrates a basic Outbound Call Model, which shows the call states associated with establishing a call in the asynchronous mode. All calls start from a Null state. The call establishment process for outbound calls is shown. Table 7 presents a summary of the outbound call state transitions.

Call State Models

Global Call API for HMP on Windows Programming Guide — August 2006 53

Call State Models

Figure 11. Basic Asynchronous Outbound Call State Diagram

Null

gc_MakeCall()

GCEV_DIALING

(maskable)

Dialing

GCEV_ALERTING

GCEV_PROCEEDING

GCEV_REQMOREINFO

gc_SendMoreInfo()

GCEV_SENDMOREINFO

gc_SendMoreInfo()

GCEV_SENDMOREINFO

GCEV_REQMOREINFO

Proceeding

GCEV_CONNECTED

Connected

Legend:

Required

Optional

GCEV_PROCEEDING

GCEV_ALERTING

GCEV_CONNECTED

SendMoreInfo

GCEV_ALERTING

Alerting

Note: In Figure 11, the SendMoreInfo state and all transitions to/from that state apply to E1, T1, and

ISDN technologies only.

54 Global Call API for HMP on Windows Programming Guide — August 2006

Table 7. Asynchronous Outbound Call State Transitions

Call State Models

State Previous/Next State

Alerting (GCST_ALERTING) Maskable

Dialing (GCST_DIALING)

Not Maskable

Null (GCST_NULL) Not Maskable

Proceeding (GCST_PROCEEDING)

Maskable

SendMoreInfo (GCST_SENDMOREINFO)

Maskable

Previous: Proceeding, Dialing, SendMoreInfo † Next: GCEV_CONNECTED -> Connected state GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL -> Idle state

Previous: Null Next:

GCEV_CONNECTED -> Connected state

GCEV_ALERTING -> Alerting (Delivered) state GCEV_PROCEEDING -> Proceeding state GCEV_REQMOREINFO -> SendMoreInfo state GCEV_SENDMOREINFO -> SendMoreInfo state GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL -> Idle state

Previous: Idle Next: gc_ResetLineDev( ) -> Null

GCEV_DIALING -> Dialing state GCEV_DETECTED -> Detected state

Previous: Dialing, SendMoreInfo

Next: GCEV_ALERTING ->

Alerting (Delivered) state GCEV_CONNECTED ->

Connected state GCEV_DISCONNECTED -> Disconnected state GCEV_DROPCALL ->

Idle state

Previous: Dialing

†

Next: GCEV_CONNECTED ->

Connected state GCEV_PROCEEDING -> Proceeding state. GCEV_DISCONNECTED ->

Disconnected state GCEV_DROPCALL ->

Idle state

†

† Applies to E1, T1 and ISDN technologies only.

Valid Call State

Transition

Call Transition Events

Functions

gc_DropCall( ) GCEV_DISCONNECTED,

gc_SendMoreInfo( ) † gc_DropCall( )

GCEV_DROPCALL GCEV_CONNECTED

GCEV_CONNECTED, GCEV_ALERTING, GCEV_REQMOREINFO GCEV_PROCEEDING, GCEV_DISCONNECTED, GCEV_DROPCALL

†,

†

gc_MakeCall( ) GCEV_DIALING

gc_DropCall( ) GCEV_DISCONNECTED,

gc_SendMoreInfo( ) gc_DropCall( )

GCEV_DROPCALL, GCEV_CONNECTED, GCEV_ALERTING

GCEV_DISCONNECTED, GCEV_DROPCALL, GCEV_PROCEEDING GCEV_CONNECTED

Global Call API for HMP on Windows Programming Guide — August 2006 55

Call State Models

The following sections describe the asynchronous outbound call processes, as shown in Figure 11,

“Basic Asynchronous Outbound Call State Diagram”, on page 54.

3.4.2.2 Channel Initialization

To establish calls, the following conditions must be met:

• The condition of the line device must be unblocked. When a channel is initially opened, the

Unblocked Event Handling”). When the condition of the line device is unblocked, the line

device is ready for establishing calls.

• The call state of the channel must be in the Null state. This is the initial call state of a line

device when it is first opened. This state is also reached when a call is released or after the channel is reset by issuing the gc_ResetLineDev( ) function.

If the above conditions are met, the application is ready to make outbound calls.

3.4.2.3 Call Dialing

To initiate an outbound call using the asynchronous mode, the application issues a gc_MakeCall( ) function that requests an outgoing call to be made on a specific line device. The gc_MakeCall( ) function returns immediately. and the call state transitions to the Dialing state. The GCEV_DIALING event is generated (if enabled) to indicate that the call has transitioned to the Dialing state. A CRN is assigned to the call being established on that line device. If the gc_MakeCall( ) function fails, the line device remains in the Null state. In this state, dialing information is sent to the remote side.

3.4.2.4 Call Proceeding

In the Dialing state, the remote side may indicate that all the information was received and the call is proceeding. In this case, the GCEV_PROCEEDING event is generated and the call transitions to the Proceeding state. The remote side may either accept or answer the call.

3.4.2.5 Call Alerting

If the remote end is not ready to answer the call, a GCEV_ALERTING event is generated. This event indicates that the called party has accepted but not answered the call and that the network is waiting for the called party to complete the connection. At this stage, the remote side is typically ringing. This GCEV_ALERTING event changes the call state to the Alerting state.

3.4.2.6 Call Connected

When the called party immediately accepts the call, such as a call directed to a FAX or voice messaging system, a GCEV_CONNECTED event is generated to indicate that the connection was

56 Global Call API for HMP on Windows Programming Guide — August 2006

established. This event changes the call to the Connected state. In the Connected state, the call is connected to the called party and call charges begin.

When the call is answered (the remote end makes the connection), a GCEV_CONNECTED event changes the call to the Connected state. In the Connected state, the call is connected to the called party and call charges begin. The GCEV_CONNECTED event indicates successful completion of the gc_MakeCall( ) function.

3.4.2.7 Overlap Sending

Note: This functionality applies to E1, T1 and ISDN technologies only.

In the Dialing state, if the remote side requests more information such as the destination address, the GCEV_REQMOREINFO event is generated and the call transitions to the SendMoreInfo state. The gc_SendMoreInfo( ) function is issued to send more information. If the remote side still requests more information, the GCEV_REQMOREINFO event is generated again. Once the remote side has received sufficient information, it indicates that the call is proceeding, and accepts or answers the call. Some technologies, such as ISDN and SS7, do not have any messages or signals to request more information. For such protocols, the application never gets the unsolicited GCEV_REQMOREINFO event. In this case, the application may call the gc_SendMoreInfo( ) function to send more information as it becomes available.

Call State Models

3.4.2.8 Call Failure

The following are two causes of call failures:

Call Rejection

When the remote end does not answer the call, a GCEV_DISCONNECTED event is generated. This event is also generated when an inbound call arrives while the application is setting up an outbound call, causing a “glare” condition. Unless the protocol specifies otherwise, the incoming call takes precedence over the outbound call. When an asynchronous gc_MakeCall( ) function conflicts with the arrival of an inbound call, all the resources need to be released for the outbound call. Subsequently, the GCEV_DISCONNECTED event is generated with a result value indicating that an inbound call took precedence. The gc_DropCall( ) function must be issued after the GCEV_DISCONNECTED event is received.

If a gc_MakeCall( ) function is issued while the inbound call is being set up, the gc_MakeCall( ) function fails. The inbound call event is held in the driver until the CRN of the outbound call is released using the gc_ReleaseCallEx( ) function. After release of the outbound CRN, the pending inbound call event is sent to the application. This behavior may be modified by the individual protocol specification.

Task Failure

If the gc_MakeCall( ) cannot be completed successfully, a GCEV_TASKFAIL event or a GCEV_DISCONNECTED event is sent to the application. The result value associated with the event indicates the reason for the event. If the GCEV_TASKFAIL event is sent, then a problem occurred when placing the call from the local end.

Global Call API for HMP on Windows Programming Guide — August 2006 57

Call State Models

3.4.2.9 Outbound Call Scenarios in Asynchronous Mode

This section shows various asynchronous outbound call scenarios. For call scenarios used for a specific signaling protocol, check the Global Call Technology Guide for that technology.

Figure 12 shows a basic asynchronous call scenario for outgoing calls.

Figure 12. Asynchronous Outbound Call Scenario

Application

gc_MakeCall ()

GCEV_DIALING

GCEV_ALERTING

GCEV_CONNECTED

GlobalCall

Library/

Technology

Outbound Call

Network

Alerting

Call Answered

58 Global Call API for HMP on Windows Programming Guide — August 2006

Call State Models

Figure 13 shows an asynchronous call scenario for outgoing calls with call acknowledgement.

Figure 13. Asynchronous Outbound Call Scenario With Call Acknowledgement

Application

gc_MakeCall ()

GCEV_PROCEEDING

GCEV_CONNECTED

GCEV_DIALING

GCEV_ALERTING

GlobalCall

Library/

Technology

Outbound Call

Network

Call Proceeding

Alerting

Call Answered

Global Call API for HMP on Windows Programming Guide — August 2006 59

Call State Models

Figure 14 shows an asynchronous call scenario for outgoing calls with overlap sending.

Note: This scenario applies to E1, T1 and ISDN technologies only.

Figure 14. Asynchronous Outbound Call Scenario With Overlap Sending

Application

gc_MakeCall ()

GCEV_DIALING

GCEV_REQMOREINFO

gc_SendMoreInfo (DESTINATION_ADDRESS)

GCEV_SENDMOREINFO

GCEV_PROCEEDING

GCEV_ALERTING

GlobalCall

Library/

Technology

Outbound Call

Network

Request More

Address Information

More Address

Information

Call Proceeding

Alerting

GCEV_CONNECTED

Call Answered

3.4.3 Call Termination in Asynchronous Mode

This section describes how calls are terminated and shows call scenarios for asynchronous call termination. The following topics describe call termination in asynchronous mode:

• Call Termination in Asynchronous Mode Overview

• User Initiated Termination

• Network Initiated Termination

• Call Release

• Call Termination Call Control Scenarios in Asynchronous Mode

60 Global Call API for HMP on Windows Programming Guide — August 2006

3.4.3.1 Call Termination in Asynchronous Mode Overview

Figure 15 illustrates the call states associated with call termination or call teardown in the asynchronous mode initiated by either a call disconnection or failure. See Table 8 for a summary of the call state transitions. A call can be terminated by the application or by the detection of a call disconnect from the network. Either of these terminations can occur at any point in the process of setting up a call and during any call state.

Figure 15. Asynchronous Call Tear-Down State Diagram

Note: In Figure 15, the GetMoreInfo and SendMoreInfo states apply to E1, T1 and ISDN technologies

only.

Table 8. Asynchronous Call Termination Call State Transitions

Call State Models

State Previous/Next State

Disconnected (GCEV_DISCONNECTED) Not maskable

Idle (GCST_IDLE) Not Maskable

† Applies to E1, T1 and ISDN technologies only.

Previous: Offered, Accepted, Connected, Dialing, SendMoreInfo †, Proceeding, Alerting, GetMoreInfo †, CallRouting Next: GCEV_DROPCALL -> Idle state

Previous: Offered, Accepted, Connected, Dialing, SendMoreInfo †,Proceeding, Alerting, GetMoreInfo †,CallRouting, Disconnected

Next: GCEV_RELEASECALL -> Null

3.4.3.2 User Initiated Termination

The application terminates a call by issuing a gc_DropCall( ) function that initiates disconnection of the call specified by the CRN. When the remote side responds by disconnecting the call, a GCEV_DROPCALL event is generated and causes a transition from the current call state to the Idle state. The user must then issue the gc_ReleaseCallEx( ) function to release all internal resources allocated for the call.

Valid Call State

Transition Functions

gc_DropCall( ) GCEV_DROPCALL

gc_ReleaseCallEx( ) GCEV_RELEASECALL

Call Transition Events

3.4.3.3 Network Initiated Termination

When a network call termination is initiated, an unsolicited GCEV_DISCONNECTED event is generated. This event indicates the call was disconnected at the remote end or an error was detected, which prevented further call processing. The GCEV_DISCONNECTED event causes the call state to change from the current call state to the Disconnected state. This event may be received during call setup or after a connection is requested. In the Disconnected state, the user issues the

gc_DropCall( ) function to disconnect the call. The gc_DropCall( ) function is equivalent to set hook ON. After the remote side is notified about the call being dropped, a GCEV_DROPCALL

event is generated causing the call state to change to the Idle state. In the Idle state, the

Global Call API for HMP on Windows Programming Guide — August 2006 61

Call State Models

gc_ReleaseCallEx( ) function must be issued to release all internal resources committed to servicing the call.

3.4.3.4 Call Release

Once in the Idle state, the call has been disconnected and the application must issue a gc_ReleaseCallEx( ) function to free the line device for another call. The gc_ReleaseCallEx( ) function releases all internal system resources committed to servicing the call. A GCEV_RELEASECALL event is generated and the call state transitions to the Null state.

3.4.3.5 Call Termination Call Control Scenarios in Asynchronous Mode

This section shows various asynchronous call termination call scenarios. For call scenarios used for a specific signaling protocol, check the Global Call Technology Guide for that technology.

Figure 16 shows an asynchronous user initiated call termination scenario.

Figure 16. User Initiated Asynchronous Call Termination Scenario

Application

gc_DropCall ()

GCEV_DROPCALL

gc_ReleaseCallEx()

GCEV_RELEASECALL

GlobalCall

Library/

Technology

Network

Disconnect

Call Disconnected

Figure 17 shows an asynchronous network initiated call termination scenario.

62 Global Call API for HMP on Windows Programming Guide — August 2006

Figure 17. Network Initiated Asynchronous Call Termination Scenario

Call State Models

Application

GCEV_DISCONNECTED

gc_DropCall()

GCEV_DROPCALL

gc_ReleaseCallEx ()

GCEV_RELEASECALL

GlobalCall

Library/

Technology

Disconnected

Network

Call Disconnected

Global Call API for HMP on Windows Programming Guide — August 2006 63

Call State Models

3.4.4 Handling Unsolicited Events

The application must handle unsolicited events in the synchronous mode, unless these events are masked or disabled. The gc_SetConfigData( ) function specifies the events that are enabled or disabled for a specified line device. This function sets the event mask associated with the specified line device. If an event bit in the mask is cleared, the event is disabled and not sent to the application.

The unsolicited events listed in Table 9 require a signal handler if they are enabled. Unsolicited events that cannot be masked must use a signal handler. All technology-specific unsolicited events also require a signal handler (see the appropriate Global Call Technology Guide for details). If any of these unsolicited events are not masked by the application and signal handlers are not defined, they are queued without being retrievable and memory problems are likely to occur.

Table 9. Unsolicited Events Requiring Signal Handlers

Event Default Setting Maskable

GCEV_ALERTING enabled yes

GCEV_PROCEEDING disabled yes

GCEV_DETECTED disabled yes

GCEV_BLOCKED enabled yes

GCEV_UNBLOCKED enabled yes

GCEV_DISCONNECTED enabled no

GCEV_TASKFAIL enabled no

3.5 Advanced Call Control with Call Hold and Transfer

Note: The advanced call model applies only to E1, T1 and ISDN technologies. It does not apply to IP

technology which uses a different scheme for features such as call transfer. See the Global Call IP Technology Guide for more information.

This section describes the advanced call state model. Topics include:

• Advanced Call State Model Overview

• Advanced Call States for Hold and Transfer

• Call Hold

• Call Transfer

3.5.1 Advanced Call State Model Overview

The advanced call model provides additional call control functionality over the basic call model, adding the ability to transfer calls, place calls on hold and retrieve calls on hold. This section provides brief descriptions of the API functions used to hold, retrieve, and transfer calls and

64 Global Call API for HMP on Windows Programming Guide — August 2006

describes the call state transitions that occur when the functions are used. This section also provides figures that illustrate the call state transitions for advanced call model functions.

Note: The hold, retrieve, and transfer functions are supported by particular protocols for the ISDN, E1

(PDKRT only) and T1 (PDKRT only) technologies. For more information, see the function descriptions in the Global Call API Library Reference and the appropriate Global Call Technology Guide.

3.5.2 Advanced Call States for Hold and Transfer

Two advanced call states are appended to the basic call model to support call hold and transfer. These advanced call states are as follows:

On-Hold State (GCST_ONHOLD)

A call must be in the Connected call state to be put on hold. When a call is put on hold, the remote party is often routed via the local switch or network to receive background music while temporarily suspended from conversing with the local party. The call remains on hold until the application retrieves the call, effectively re-transitioning it into the Connected, “conversational” state. The application may not issue a gc_MakeCall( ) or receive another call while a call is in the On-hold state.

There is no limit to the number of times a call may be placed in and retrieved from the On-hold state. In addition, either the called party or the calling party can put the call in the On-hold state.

The On-hold call state applies only to call scenarios where a single call is present on the specified channel. The On-hold call state does not apply to call transfer scenarios that use the On-Hold Pending Transfer call state instead.

Call State Models

On-Hold Pending Transfer State (GCST_ONHOLDPENDINGTRANSFER)

During a supervised call transfer, two calls are made accessible to the local channel. Both calls must be in the Connected call state. The call that is temporarily suspended from conversing is considered to be in the On-hold Pending Transfer call state. This call is often routed via the local switch or network to receive background music while awaiting completion of the call transfer.

Both the suspended call and the currently active call may be swapped at any time so that the call that was in the On-hold Pending Transfer state is now actively connected, while the former active call is placed in the On-hold Pending Transfer state. There is no limit to the number of times two calls may be swapped between the On-hold Pending Transfer and “Connected” states. The completion of the call transfer is independent of which call is active or on hold.

3.5.3 Call Hold

The advanced call model allows the application to place a call on hold. The Global Call API provides the following functions to place a call on hold and, subsequently, to retrieve the call on hold:

gc_HoldCall( )

place a call on hold

gc_RetrieveCall( )

retrieve a call from hold

Global Call API for HMP on Windows Programming Guide — August 2006 65

Call State Models

The gc_HoldCall( ) function places an active call in the On-hold (GCST_ONHOLD) state. The gc_RetrieveCall( ) function retrieves the call from the GCST_ONHOLD state and returns it to the

Connected (GCST_CONNECTED) state.

Figure 18 illustrates the transition between call states when a call is put on hold and then retrieved.

Figure 18. Call State Transitions for Hold and Retrieve

gc_HoldCall()

state(ch1) = Connected

gc_RetrieveCall()

Calls in the On-hold state must be returned to the Connected state before they can be dropped. Calls are dropped following the Basic Call scenario. See Section 3.4, “Basic Call Control in

Asynchronous Mode” for more information.

3.5.4 Call Transfer

This section describes the different types of call transfer. Topics include:

• Call Transfer Overview

• Supervised Transfers

• Unsupervised Transfers

3.5.4.1 Call Transfer Overview

There are two types of call transfers:

Supervised transfers

the person transferring the call stays on the line, announces the call, and consults with the party to whom the call is being transferred before the transfer is completed

state(ch1) = Onhold

Unsupervised transfers

the call is sent without any consultation or announcement by the person transferring the call. Unsupervised transfers are also known as one-step transfers or blind transfers

Note: The the call transfer implementations described in this section are common to a number of different

technologies. However, not all technologies support these implementations and some technologies have technology-specific implementations for call transfer. See the appropriate Global Call Technology Guide for technology-specific information on call transfer.

66 Global Call API for HMP on Windows Programming Guide — August 2006

Supervised transfers use the following Global Call API functions:

gc_SetupTransfer( )

initiates a supervised transfer

gc_CompleteTransfer( )

completes a supervised transfer

gc_SwapHold( )

switches between the consultation call and the call pending transfer

Unsupervised transfers use the following Global Call API function:

gc_BlindTransfer( )

initiates and completes an unsupervised (one-step) transfer

3.5.4.2 Supervised Transfers

A supervised transfer begins with a successful call to the gc_SetupTransfer( ) function. The following steps describe how the transfer is completed:

1. Successful call to the gc_SetupTransfer( ) function changes the state of the original call to the

GCST_ONHOLDPENDINGTRANSFER state.

2. A consultation CRN is allocated with the initial state of GCST_DIALTONE and is returned by

the gc_SetupTransfer( ) function.

3. The gc_MakeCall( ) function is called to establish a connection on the consultation call. The CRN returned by gc_MakeCall( ) is the same CRN as was returned by gc_SetupTransfer( ).

4. The consultation call proceeds similarly to a singular outbound call proceeding through the GCST_DIALING and GCST_ALERTING (if enabled) call states. (See Section 3.4, “Basic

Call Control in Asynchronous Mode” for more information.)

5. If the consultation call is successfully established, the state of the consultation call changes to the GCST_CONNECTED state, and the state of the original call remains unchanged.

6. While the consultation call is in the GCST_CONNECTED state, the gc_SwapHold( ) function may be used to switch between the call pending transfer and the consultation call.

7. A call to the gc_CompleteTransfer( ) function transfers the original call to the consultation call and internally drops both channels.

8. The states of the original and the consultation call both change to the GCST_IDLE state upon receipt of the GCEV_COMPLETETRANSFER event.

9. The application must call gc_ReleaseCallEx( ) for both of the calls to release the resources allocated for both channels.

Call State Models

Note: The consultation call may be terminated at any point in the process by the application or by the

detection of a call disconnect from the network.

The call state transitions that occur during a supervised transfer are shown in Figure 19 (which also shows the call state transitions for an unsupervised transfer).

Global Call API for HMP on Windows Programming Guide — August 2006 67

Call State Models

Figure 19. Call State Model for Supervised and Unsupervised Transfers

INBOUND CALL OUTBOUND CALL

gc_ReleaseCallEx() (Call 1) gc_ReleaseCallEx() (Call 2)

gc_ReleaseCallEx() (Call 1)

gc_WaitCall()

GCEV_DETECTED

(maskable)

Detected (Call 1)

GCEV_OFFERED

Null (Call 1)

gc_WaitCall()

GCEV_OFFERED

gc_MakeCall()

Idle (Call 1)

gc_BlindTransfer()

gc_CompleteTransfer()

Connected (Call 1)

OnHoldPendingTransfer (Call 2)

gc_SwapHold()

GCEV_SWAPHOLD

Note: *

Indicates that Call 2 does not apply in a blind transfer.

gc_SwapHold()

OnHoldPendingTransfer (Call 1)

Connected (Call 2)

gc_CompleteTransfer()

Offered (Call 1)

gc_AcceptCall()

Accepted (Call 1)

gc_AnswerCall()gc_AnswerCall()

GCEV_COMPLETETRANSFER)

Idle (Call 1)

Idle (Call 2)*

gc_CompleteTransfer()

Dialing (Call 1)

GCEV_ALERTING

(maskalbe)

Alerting (Call 1)

Completion of

gc_MakeCall()

Connected (Call 1)

gc_CompleteTransfer()

GCEV_CONNECTED

gc_SetupTransfer()

sr_waitevt()

GCEV_SETUPTRANSFER)

OnHoldPendingTransfer (Call 1)

Dialtone (Call 2)

gc_MakeCall()

OnHoldPendingTransfer (Call 1)

Dialtone (Call 2)

GCEV_ALERTING

(maskable)

OnHoldPendingTransfer (Call 1)

Alerting (Call 2)

If the network or application terminates a call during a transfer, the call state transitions are as shown in Figure 20.

68 Global Call API for HMP on Windows Programming Guide — August 2006

Figure 20. Call Termination by the Network or Application During a Transfer

Call State Models

TERMINATED

NETWORK

(from any of the states

shown in the box below)

OnHoldPendingtransfer (Call 1)

Dialing (Call 2)

OnHoldPendingtransfer (Call 1)

Alerting (maskable) (Call 2)

OnHoldPendingtransfer (Call 1)

Connected (Call 2)

Connected (Call 1)

OnHoldPendingtransfer (Call 2)

Note:

This can be Call 1 or Call 2 depending on which call is currenly active, that is, not in an OnHoldPendingTransfer state.

Disconnected (Call 1)

state unchanged (Call 2)

state unchanged (Call 1)

Disconnected (Call 2)

gc_DropCall (Call 2)*

GCEV_DROPCALL (Call 2)

GCEV_CONNECTED (Call 1)

Connected (Call 1)

Idle (Call 2)

gc_ReleaseCallEx (Call 2)

Connected (Call 1)

Null (Call 2)

TERMINATED

APPLICATION

(from any of the states

shown in the box below)

OnHoldPendingtransfer (Call 1)

Dialing (Call 2)

OnHoldPendingtransfer (Call 1)

Alerting (maskable) (Call 2)

OnHoldPendingtransfer (Call 1)

Connected (Call 2)

Connected (Call 1)

OnHoldPendingtransfer (Call 2)

gc_DropCall (Call 2)*

GCEV_DROPCALL (Call 2)

GCEV_CONNECTED (Call 1)

Note: In Figure 20, when gc_DropCall( ) is issued, an unsolicited GCEV_CONNECTED event is

received for call 1 transitioning it back to the Connected state.

3.5.4.3 Unsupervised Transfers

In an unsupervised transfer, a successful call to the gc_BlindTransfer( ) function transfers the call in a single step, without any consultation or announcement by the person transferring the call. Internally, the currently connected call is placed on hold, the new party is dialed, and, finally, the connection to both parties is relinquished. When the application receives the GCEV_BLINDTRANSFER event, the original call enters the GCST_IDLE state. At this point the application must call gc_ReleaseCallEx( ) for the call to release the allocated resources.

Once the new party is dialed, the control and responsibility for the results of the transfer, whether successfully connected or not, lie totally with the remote party once the transfer is relinquished. Only one call is controlled by the application as the transfer is initiated internally via the protocol.

Unsupervised transfers do not provide call progress results for the transfer nor do they support terminating the transfer at any point via the gc_DropCall( ) function.

Global Call API for HMP on Windows Programming Guide — August 2006 69

Call State Models

Figure 19 illustrates the call state transitions that occur in an unsupervised transfer, which basically includes only:

• The transition of Call 1 from the Connected to the Idle state (invoked by the

• The transition of Call 1 from the Idle to the Null state (invoked by the gc_ReleaseCallEx( )

gc_BlindTransfer( ) function)

function).

70 Global Call API for HMP on Windows Programming Guide — August 2006

4.Event Handling

This chapter describes how Global Call handles events generated in the call state model. Topics include:

• Overview of Event Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

• Event Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

• Blocked and Unblocked Event Handling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

• Event Retrieval. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

• Events Indicating Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

• Masking Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

• Event Handlers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

4.1 Overview of Event Handling

The Global Call protocol handler continuously monitors the line device for events from the network. As each call is processed through its various states, corresponding events are generated and passed to the application. An overview of Global Call event categories is provided in this chapter. Specific event definitions are described in the Global Call API Library Reference. See the appropriate Global Call Technology Guide for technology-specific event information.

4.2 Event Categories

The events that can occur when using the Global Call API are divided into the following categories:

Termination

Events returned after the termination of a function. Termination events apply to asynchronous programming only.

Notification

Events that are requested by the application and provide information about a function call. Notification events apply to synchronous and asynchronous programming.

Unsolicited

Events triggered by, and providing more information about, external events. Unsolicited events apply to synchronous and asynchronous programming.

See the Global Call API Library Reference for detailed information about each event and the appropriate Global Call Technology Guide for any technology-specific event information.

Global Call API for HMP on Windows Programming Guide — August 2006 71

Event Handling

4.3 Blocked and Unblocked Event Handling

Global Call uses the concept of blocked and unblocked conditions for line devices. By default, when the gc_OpenEx( ) function is used to open a line device, the line device is in a blocked condition meaning that the application can not perform call related functions on the line device, such as waiting for a call or making a call. The application must wait for the GCEV_UNBLOCKED event before waiting for a call or making a call.

Note: Since, by default, the line device is initially in the blocked condition, the application does not

receive an initial GCEV_BLOCKED event.

Circumstances can occur, such as a blocking layer 1 (physical) alarm or the remote side going out of service, that cause a line device to move to a blocked condition. When this happens, the application receives a GCEV_BLOCKED event. When the line device is in the blocked condition, the application can only perform a small subset of the valid functions for line devices. The functions common to all interface technologies and that can be used while a line device is in the blocked condition are:

• gc_DropCall( )

• gc_ReleaseCall( ) (applies to E1, T1 and ISDN technologies only)

• gc_ReleaseCallEx( )

• gc_Close( )

• Functions related to alarm processing and retrieving alarm information, for example,

gc_AlarmName( )

• Functions related to error processing, for example, gc_ErrorInfo( )

• Functions related to event processing, for example, gc_ResultInfo( ), gc_GetMetaevent( )and

gc_GetMetaeventEx( )

• Functions related to retrieving information about the call control libraries, for example,

gc_CCLibIDToName( )

• gc_AttachResource( ) and gc_Detach( )

As indicated in the list above, the application may drop and release calls while a line device is in the blocked condition, but it should not do so in response to the GCEV_BLOCKED event. If a call is active, typically a GCEV_DISCONNECTED event arrives either just before or just after the GCEV_BLOCKED event, at which point the application should drop and release the call indicated by the GCEV_DISCONNECTED event.

Note: The Global Call term blocked does not refer to the signaling bits indicating a blocked condition as

defined in some network interface technologies, although the line device may move to a blocked condition as a consequence of the signaling bits indicating a blocked condition.

At some point, the application may receives a GCEV_UNBLOCKED event, indicating that the conditions blocking a line device have been removed and the line device has now returned to the unblocked condition. The application can once again use any valid function on the line device.

The reception of the GCEV_BLOCKED and GCEV_UNBLOCKED events may be disabled using the gc_SetConfigData( ) function. The default is that these events are enabled. However, disabling the reception of these events is not recommended since the application will not be notified of these

72 Global Call API for HMP on Windows Programming Guide — August 2006

critical events. In addition, if the GCEV_BLOCKED event is disabled, some functions will fail with a reason of EGC_INVALIDSTATE, which may cause confusion. For more information on blocking alarms and the GCEV_BLOCKED and GCEV_UNBLOCKED events, see Section 8.2.1,

“Generation of Events for Blocking Alarms”, on page 99.

Note: A GCEV_UNBLOCKED event will be generated when opening a virtual board device. A

GCEV_BLOCKED event will also be generated if there are blocking alarms on the virtual board and the corresponding GCEV_UNBLOCKED event will be generated when the blocking alarms clear. The application must be prepared to handle these events.

4.4 Event Retrieval

All events are retrieved using the current Standard Runtime Library (SRL) event retrieval mechanisms (see the Standard Runtime Library API Programming Guide for details), including event handlers. The gc_GetMetaEvent( ) function or, for Windows extended asynchronous models, the gc_GetMetaEventEx( ) function, maps the current SRL event into a metaevent. A metaevent is a data structure that explicitly contains the information describing the event. This data structure provides uniform information retrieval among all call control libraries.

For Global Call events, the structure contains Global Call related information (CRN and line device) used by the application. For events that are not Global Call events, the device descriptor, the event type, a pointer to variable length event data, and the length of the event data are available through the METAEVENT structure. Since all the data associated with an event is accessible via the METAEVENT structure, no additional SRL calls are required to access the event data.

Event Handling

The LDID associated with an event is available from the linedev field of the METAEVENT. If the event is related to a CRN, that CRN is available from the crn field of the METAEVENT; if the crn field of the METAEVENT is 0, then the event is not a call-related event.

The METAEVENT structure also includes an extevtdatap field which contains a pointer to more information about the event. The memory pointed to by the extevtdatap field should be treated as read-only and should not be altered and/or freed.

The application should issue a gc_DropCall( ) function before issuing the gc_ReleaseCallEx( ) function. Failure to issue this function could result in one or more of the following problems:

• memory problems due to memory being allocated and not being released

• a blocking condition

• events sent to the previous user of a CRN that could be processed by a later user of the CRN

with unexpected results

The reason for an event can be retrieved using the gc_ResultInfo( ) function. The information returned uniquely identifies the cause of the event.

Global Call API for HMP on Windows Programming Guide — August 2006 73

Event Handling

4.5 Events Indicating Errors

Events that explicitly provide error indications are as follows:

GCEV_TASKFAIL

Received when an API function call fails

When this events is received, the application should call gc_ResultInfo( ) immediately after the event arrives to determine the reason for the event. The data structure associated with gc_ResultInfo( ) can contain reason information provided by Global Call and additional reason information provided by the underlying call control library. See the Global Call API Library Reference for more information.

4.6 Masking Events

Some events are maskable. See the gc_SetConfigData( ) function description in the Global Call API Library Reference for specific information regarding enabling and disabling events.

4.7 Event Handlers

An event handler is a user-defined function called by the SRL to handle a specific event that occurs on a specified device.

Typically, in a Windows environment, processing events within a thread or using a separate thread to process events tends to be more efficient than using event handlers. However, if event handlers are used, such as when an application is being ported from Linux, then you must use the asynchronous with SRL callback model.

The following guidelines apply to using event handlers:

• more than one handler can be enabled for an event. The SRL calls all specified handlers when

the event is detected

• handlers can be enabled or disabled from any thread

• general handlers can be enabled to handle all events on a specific device

• a handler can be enabled to handle any event on any device

• synchronous functions cannot be called from a handler

By default, when the sr_enbhdlr( ) function is first called, a thread internal to the SRL is created to service the application-enabled event handlers. This SRL handler thread exists as long as one handler is still enabled. The creation of this internal SRL event handler thread is controlled by the SR_MODELTYPE value of the SRL sr_setparm( ) function. The SRL handler thread should be:

• enabled when using the asynchronous with SRL callback model. Enable the SRL event

handler thread by not specifying the SR_MODELTYPE value (default is to enable) or by setting this value to SR_MTASYNC (do not specify SR_STASYNC).

74 Global Call API for HMP on Windows Programming Guide — August 2006

Event Handling

• disabled when using an application-handler thread wherein a separate event handler thread is

created within the application that calls the sr_waitevt( ) and gc_GetMetaEvent( ) functions. For an application-handler model, use the asynchronous with SRL callback model but set the SR_MODELTYPE value to SR_STASYNC to disable the creation of the internal SRL event handler thread.

Note: An application-handler thread must not call any synchronous functions.

See the Standard Runtime Library API Programming Guide for the hierarchy (priority) order in which event handlers are called.

Global Call API for HMP on Windows Programming Guide — August 2006 75

Event Handling

76 Global Call API for HMP on Windows Programming Guide — August 2006

5.Application Development

Guidelines

This chapter provides some tips when developing programs using Global Call. Topics include:

• General Programming Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

• Tips for Programming Drop and Insert Applications . . . . . . . . . . . . . . . . . . . . . . . . . . 78

• Using Global Call with Digital Network Interface Boards . . . . . . . . . . . . . . . . . . . . . . 79

5.1 General Programming Tips

The following tips apply when programming with Global Call:

• When using Global Call functions, the application must use the Global Call handles (that is,

the line device ID and CRN) to access Global Call functions. Do not substitute a network, voice or media device handle for the Global Call line device ID or CRN. If the application needs to use a network, voice or media device handle for a specific network or voice library call, for example, nr_scroute( ) (for E1, T1 or ISDN technologies only) or dx_play( ) (all technologies), you must use the gc_GetResourceH( ) to retrieve the network, voice or media device handle, associated with the specified Global Call line device. The gc_GetResourceH( ) function is only needed if the voice or media resource is associated with a Global Call line device. If a voice resource is not part of the Global Call line device, the device handle returned from the dx_open( ) call should be used.

• Do not access the underlying call control libraries directly. All access must be done using the

Global Call library, that is, using Global Call (gc_) functions.

• Do not call any network library (dt_) function directly from your application that may affect

the state of the line or the reporting of events, for example, dt_settssig( ), dt_setevtmsk( ), or others.

• The GCEV_BLOCKED and the GCEV_UNBLOCKED events are line related events, not call

related events. These events do not cause the state of a call to change.

• Before exiting an application:

– Drop and release all active calls, using the gc_DropCall( ) and gc_ReleaseCallEx( )

functions.

– Close all open line devices, using the gc_Close( ) function.

– Stop the application, using the gc_Stop( ) function

• Before issuing gc_DropCall( ), you must use the dx_stopch( ) function to terminate any

application-initiated voice functions, such as dx_play( ) or dx_record( ).

• In Windows environments, although asynchronous models are more complex than the

synchronous model, asynchronous programming is recommended for more complex applications that require coordinating multiple tasks. Asynchronous programming can handle multiple channels in a single thread. In contrast, synchronous programming requires separate

Global Call API for HMP on Windows Programming Guide — August 2006 77

Application Development Guidelines

threads. Asynchronous programming uses system resources more efficiently because it handles multiple channels in a single thread. Asynchronous models let you program complex applications easily, and achieve a high level of resource management in your application by combining multiple voice channels in a single thread. This streamlined code reduces the system overhead required for inter process communication and simplifies the coordination of events from many devices.

• In Windows environments, when calling the gc_GetMetaEventEx( ) function from multiple

threads, ensure that your application uses unique thread-related METAEVENT data structures or ensure that the METAEVENT data structure is not written to simultaneously.

5.2 Tips for Programming Drop and Insert Applications

Note: This section applies to E1, T1 and ISDN technologies only.

To the Global Call application, signaling is made available to the application as follows:

• Signaling information is passed to the Global Call application in the form of call control

events; for example, line answer is passed as a GCEV_ANSWERED event.

• Signaling, such as line busy, is available to the application as an EGC_BUSY error code or a

GCRV_BUSY result value; line no answer is available as an EGC_NOANSWER error code or GCRV_NOANSWER result value.

• Signaling such as a protocol error, an alerting event, a fast busy, an undefined telephone

number, or network congestion are all returned to the application as an EGC_BUSY error code or a GCRV_BUSY result value.

• Protocols without acknowledgement, for example, non-backward CAS signaling protocols,

generate a GCEV_DISCONNECTED event with an EGC_BUSY error code or a GCRV_BUSY result value when timeout or protocol errors occur during dialing.

For a drop and insert application in which the calling party needs to be notified of the exact status of the called party’s line, the following approach may be used:

• Upon receipt of an incoming call from a calling party, issue a gc_MakeCall( ) function on the

outbound line to the called party.

• After dialing completes on the outbound line, the application should drop the dialing resource,

turn off call progress, and connect the inbound line to the outbound line so that the calling party can hear the tones returned on the outbound line. These tones provide positive feedback to the calling party as to the status of the called party’s line.

• If the status of the called party’s line is such that the call cannot be completed, the calling party

hangs up and the application can then drop the call and release the resources used. Otherwise, when the call is answered, a GCEV_CONNECTED event will be received.

When call progress is being used, after dialing completes, the call progress software looks for ringback or voice on the outbound line. When ringback is detected, a GCEV_ALERTING event is generated. When voice is detected, a GCEV_ANSWERED event is generated. An unacceptable amount of time may lapse before either of these events is generated while the calling party is waiting for a response that indicates the status of the call. Thus, for drop and insert applications, call progress should be disabled as soon as dialing completes and the inbound and outbound lines

78 Global Call API for HMP on Windows Programming Guide — August 2006

Application Development Guidelines

connected so as to provide the calling party with immediate outbound line status and voice cutthrough.

For a drop and insert application in which a call cannot be completed, the application can simulate and return a busy tone or a fast busy (redial) tone to the calling party. Typically, this condition occurs when a GCEV_DISCONNECTED event is generated due to a timeout or a protocol error during dialing or due to R2 backward signaling indicating a busy called party’s line, equipment failure, network congestion or an invalid telephone number.

When a call cannot be completed because the called party’s line is busy:

1. Use a tone or voice resource to generate a busy tone (60 ipm [impulses per minute]) or to record a busy tone.

2. Connect the busy tone to the calling party’s line or play back the recorded busy tone file.

3. Drop and release the calling party’s line when a GCEV_DISCONNECTED event is received.

When a call cannot be completed because of equipment failure, network congestion or an invalid telephone number:

1. Use a tone or voice resource to generate a fast busy tone (120 ipm) or to record a fast busy tone.

2. Connect the fast busy tone to the calling party’s line or play back the recorded fast busy tone file.

3. Drop and release the calling party’s line when a GCEV_DISCONNECTED event is received.

For voice function information, see the Voice API Library Reference for your operating system.

5.3 Using Global Call with Digital Network Interface Boards

The HMP software can be used in conjunction with Intel NetStructure® Digital Network Interface boards that provide physical E1 and T1 interfaces. These Digital Network Interface boards are based on the DM3 architecture. Global Call supports the development of applications that use these boards. The following topics provide guidelines for using Global Call with Digital Network Interface boards:

• Routing Overview

• Working with Flexible Routing Configurations

• Handling Multiple Call Objects Per Channel in a Glare Condition

5.3.1 Routing Overview

The HMP software supports flexible routing configurations. With flexible routing, the resource devices (voice/fax/media) and network interface devices are independent, which allows exporting and sharing of the resources.

Global Call API for HMP on Windows Programming Guide — August 2006 79

Application Development Guidelines

5.3.2 Working with Flexible Routing Configurations

The following topics provide more information about using Global Call with Digital Network Interface boards that use the flexible routing configuration:

• Determining Channel Capabilities (Flexible Routing)

• Using Device Handles (Flexible Routing)

• Multi-Threading and Multi-Processing (Flexible Routing)

• Initializing an Application (Flexible Routing)

• Initializing Global Call (Flexible Routing)

• Device Initialization Hint (Flexible Routing)

• Using Protocols (Flexible Routing)

5.3.2.1 Determining Channel Capabilities (Flexible Routing)

Digital Network Interface boards support three different types of voice devices:

• E1 CAS compatible

• T1 CAS compatible

• ISDN compatible

The E1 CAS compatible is a superset of T1 CAS compatible, and the T1 CAS compatible is a superset of ISDN compatible.

When using Global Call, only certain DM3 network interface devices can be associated with certain other DM3 voice devices using gc_OpenEx( ) or gc_AttachResource( ). Attaching DM3 devices together depends on the network protocol used and voice device capabilities. Specifically:

• A DM3 ISDN network device can be attached to any DM3 voice device.

• A DM3 T1 CAS network device must be attached to a T1 CAS compatible DM3 voice device.

• A DM3 E1 CAS network device must be attached to an E1 CAS compatible DM3 voice

device.

An application can query the capabilities of a device using the dx_getfeaturelist( ) function which includes information about the front end supported, meaning ISDN, TI CAS, or R2/MF. See the Voice API Library Reference for more information about the dx_getfeaturelist( ) function.

When using Global Call, if a voice device is not CAS or R2/MF capable, it cannot be attached (either in the gc_OpenEx( ) function or when using the gc_AttachResource( ) function) to a network interface device that has CAS or R2/MF loaded. Likewise, if a voice device is not routable, it cannot be used in a gc_AttachResource( ) call.

While a network interface protocol cannot be determined programmatically, the dx_getfeaturelist( ) function provides a programmatic way of determining voice capability so that the application can make decisions.

80 Global Call API for HMP on Windows Programming Guide — August 2006

Application Development Guidelines

5.3.2.2 Using Device Handles (Flexible Routing)

When using Digital Network Interface boards, application performance may be a consideration when opening and closing devices using Global Call. If an application must use Global Call to dynamically open and close devices as needed, it can impact the application’s performance. One way to avoid this is to open all devices during application initialization and keep them open for the duration of the application, closing them only at the end.

5.3.2.3 Multi-Threading and Multi-Processing (Flexible Routing)

When using Digital Network Interface boards, the R4 APIs support multi-threading and multiprocessing with some restrictions on multi-processing as follows:

• One specific channel can only be opened in one process at a time. There can, however, be

multiple processes accessing different sets of channels. In other words, ensure that each process is provided with a unique set of devices to manipulate.

• If a channel was opened in process A and then closed, process B is then allowed to open the

same channel. However, since closing a channel is an asynchronous operation when using R4, there is a small gap between the time when the xx_close( ) function returns in process A and the time when process B is allowed to open the same channel. If process B opens the channel too early, things could go wrong. For this reason, this type of sequence should be avoided.

5.3.2.4 Initializing an Application (Flexible Routing)

A device must first be opened in order to obtain its handle, which can then be used to access the device functionality. Since applications use Global Call for call control (that is, for call setup and tear-down), all Intel network interface devices must be opened using the gc_OpenEx( ) function.

Note: When call control is not required, such as with ISDN NFAS, dt_open( ) can be used to open DM3

network interface devices.

Once the call has been established, voice and or data streaming should be done using the Voice API. Functions such as dx_playiottdata( ), dx_reciottdata( ), and dx_dial( ) can be used. Of course, in order to do so, the voice device handle must be obtained.

5.3.2.5 Initializing Global Call (Flexible Routing)

This scenario is one where an application uses Digital Network Interface boards in a flexible routing configuration. When initializing an application to use boards based on the DM3 architecture, you must use Global Call to handle the call control.

Take note of the following flexibility that exists for the gc_OpenEx( ) function when opening a Global Call line device on Digital Network Interface boards:

• Due to the nature of the DM3 architecture, the protocol name is irrelevant at the time of

opening the Global Call line device; that is, the protocol name is ignored. Also, when using R4 with boards based on the DM3 architecture, all protocols are bi-directional. You do not need to dynamically open and close devices to change the direction of the protocol.

• It is not necessary to specify a voice device name when opening a Global Call line device. If

you specify the voice device name, the network interface device is automatically associated

Global Call API for HMP on Windows Programming Guide — August 2006 81

Application Development Guidelines

with the voice device (they are attached and routed on the TDM bus). If you do not specify the voice device name when you open the Global Call line device, you can separately open a voice device, and then attach and route it to the network interface device.

For boards that use the DM3 architecture in a flexible routing configuration, only the network device name is required.

The following procedure shows how to initialize Global Call when using Digital Network Interface boards.

Note: In Windows, use the sr_getboardcnt( ) function with the class name set to DEV_CLASS_DTI and

DEV_CLASS_VOICE to determine the number of network and voice boards in the system, respectively. In Linux, use SRL device mapper functions to return information about the structure of the system. For information on these functions, see the Standard Runtime Library API Library Reference.

1. Start/Initialize Global Call using gc_Start( ).

2. Use gc_OpenEx( ) to open a Global Call line device.

• Specify the network interface device name and the protocol name in the devicename

parameter, as in the following example:

":N_dtiB1T1:P_ISDN"

• Alternatively, specify the network interface device name, the voice device name, and the

protocol name in the devicename parameter, as in the following example:

":N_dtiB1T1:V_dxxxB1C1:P_ar_r2_io"

3. Obtain the voice channel device handle.

• Open a voice channel device (for example, dxxxB1C1) with dx_open( ) to get its handle.

• Alternatively, if you specified the voice device name in the devicename parameter in

step 2, use gc_GetResourceH( ), with a resourcetype of GC_VOICEDEVICE, to get the handle.

4. Attach the voice and network interface devices.

• Use gc_AttachResource( ) to attach the voice resource and the network interface line

device.

• Alternatively, if you specified the voice device name in the devicename parameter in

step 2, the voice and network interface devices are attached by nature of the gc_OpenEx( ), so no action is necessary for this step.

5. Use gc_GetResourceH( ), with a resourcetype of GC_NETWORKDEVICE, to obtain the

network interface time slot device handle that is associated with the line device.

6. Set up TDM bus full duplex routing between the network interface device and voice device.

• Use nr_scroute(FULL DUPLEX).

• Alternatively, if you specified the voice device name in the devicename parameter in

step 2, the network interface device and voice device are automatically routed on the TDM bus by nature of the gc_OpenEx( ).

Repeat steps 2 to 6 for all Global Call device line devices.

82 Global Call API for HMP on Windows Programming Guide — August 2006

Application Development Guidelines

5.3.2.6 Device Initialization Hint (Flexible Routing)

In some applications, when xx_open( ) functions (Global Call, Voice, Fax) are issued asynchronously, it may cause slow device-initialization performance. Fortunately, you can avoid this particular problem quite simply by reorganizing the way the application opens and then configures devices. The recommendation is to do all xx_open( ) functions for all channels before proceeding with the next function. For example, you would have one loop through the system devices to do all the xx_open( ) functions first, and then start a second loop through the devices to configure them, instead of doing one single loop where an xx_open( ) is immediately followed by other API functions on the same device. With this method, by the time all xx_open( ) commands are completed, the first channel will be initialized, so you won’t experience problems.

This change is not necessary for all applications, but if you experience poor initialization performance, you can gain back speed by using this hint.

5.3.2.7 Using Protocols (Flexible Routing)

For ISDN protocols, the protocol to use is determined at board initialization time and not when opening a Global Call device. Protocol parameters are configured in the CONFIG file before the firmware is downloaded to the board. If a protocol is specified in the devicename parameter of the gc_OpenEx( ) function when opening a device, it is ignored.

For T1/E1 CAS/R2MF protocols, the protocol to use for a trunk is selected using the “Trunk Configurator” feature of the configuration manager (DCM). Protocol files are provided with the system software in the \data directory under the Dialogic home directory. A protocol can be configured by changing the parameter values in the corresponding Country Dependent Parameter (CDP) file located in the \data directory. See the Global Call Country Dependent Parameters (CDP) for PDK Protocols Configuration Guide for details on the parameters that can be changed for each protocol. If a protocol is specified in the devicename parameter of the gc_OpenEx( ) function when opening a device, it is ignored.

5.3.3 Handling Multiple Call Objects Per Channel in a Glare Condition

When using Digital Network Interface boards, Global Call supports the handling of multiple call objects per channel in a glare condition. An application running on bi-directional circuits is capable of handling two CRNs on a single line device, where one call can be in an Idle state, while the other call is in Active state. For example, a glare condition occurs when a call has been dropped but not released and an inbound call is detected as indicated in Table 10. In order to avoid a long delay in processing the inbound call, the Global Call library does not wait for the outbound call to be released before notifying the application of the inbound call.

Table 10. Handling Glare

Application Global Call Library

gc_MakeCall(CRN1) -->

<-- GCEV_DISCONNECTED(CRN1)

gc_DropCall(CRN1) -->

Global Call API for HMP on Windows Programming Guide — August 2006 83

Application Development Guidelines

Table 10. Handling Glare

Application Global Call Library

gc_AcceptCall(CRN2) -->

gc_ReleaseCallEx(CRN1) -->

Alternatively, the application can just respond to events using their associated CRN, simply performing a gc_ReleaseCallEx( ) upon reception of any GCEV_DROPCALL event whether the CRN is the active one or not. Using this procedure, the application only needs to store one CRN per line device.

<-- GCEV_OFFERED(CRN2)

<-- GCEV_DROPCALL(CRN1)

84 Global Call API for HMP on Windows Programming Guide — August 2006

6.Error Handling

The chapter describes the error handling capabilities provided by Global Call. Topics include the following:

• Error Handling Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

6.1 Error Handling Overview

When an error occurs during execution of a function, one of the following occurs:

• The function returns with a value < 0

• The unsolicited error event, GCEV_TASKFAIL, is sent to the application

Call control libraries supported by the Global Call API may have a larger set of error codes than those defined in the gcerr.h header file. The call control library error values are available using the

gc_ErrorInfo( ) function, which retrieves Global Call and call control library information. To

retrieve the information, this function must be called immediately after the Global Call function failed. This function returns a result value associated directly with the Global Call and call control library.

The gc_ResultInfo( ) function retrieves information about solicited and unsolicited events when a Global Call application gets an expected or unexpected event. To retrieve the information, the gc_ResultInfo( ) function must be called immediately after a Global Call event arrives and before the next event returns Global Call and call control library information related to the last Global Call function call. To process an error, this function must be called immediately after an event is returned to the application. For example, if an alarm occurs while making an outbound call, a GCEV_DISCONNECTED event is sent to the application with a result value indicating an alarm on the line. The GCEV_BLOCKED event is also generated with a result value that also indicates an alarm on the line. See the appropriate Global Call Technology Guide for information on specific protocol errors.

If an error occurs during execution of an asynchronous function, a termination event, such as the GCEV_GETCONFIGDATA_FAIL (E1, T1 and ISDN technologies only), or GCEV_SETCONFIGDATA_FAIL (all technologies) event is sent to the application. No change of state is triggered by this event. If events on the line require a state change, this state change occurs as described in Section 3.4.3, “Call Termination in Asynchronous Mode”, on page 60. When an error occurs during a protocol operation, the error event is placed in the event queue with the error value that identifies the error. Upon receiving a GCEV_TASKFAIL event, the application can retrieve the reason for the failure using the gc_ResultInfo( ) function.

An unsolicited GCEV_ERROR event can be received if an internal component fails. The gc_ResultInfo( ) function can be used to determine the reason for the event. Valid reasons are any of the Global Call reasons (error code or result values) or a call control library-specific reason (see the appropriate Global Call Technology Guide).

Global Call API for HMP on Windows Programming Guide — August 2006 85

Error Handling

86 Global Call API for HMP on Windows Programming Guide — August 2006

7.Call Control

This chapter describes Global Call capabilities relating to call control. Topics include:

• Call Progress Analysis when Using IP Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

• Call Progress Analysis when Using Digital Network Interface Boards . . . . . . . . . . . . 87

• Resource Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

• Feature Transparency and Extension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

7.1 Call Progress Analysis when Using IP Technology

When using IP technology, typically packetized messages are used to convey call analysis information. See the Global Call IP Technology Guide for more information.

7.2 Call Progress Analysis when Using Digital Network Interface Boards

Note: This section applies to E1, T1 and ISDN technologies only.

When using Intel NetStructure consistent method of pre-connect call progress and post-connect call analysis across E1/T1 CAS, and ISDN protocols. The level of support that Global Call provides is described in the following topics:

• Call Progress Analysis Definition

• Configuring Default Call Progress Analysis Parameters

• Configuring Call Progress Analysis on a Per Call Basis

• Setting Call Analysis Attributes on a Per Call Basis

• Configuring Call Progress Analysis on a Per Channel Basis

• Setting Call Analysis Attributes on a Per Channel Basis

• Customizing Call Progress Tones on a Per Board Basis

Digital Network Interface boards, Global Call provides a

7.2.1 Call Progress Analysis Definition

Pre-connect call progress determines the status of a call connection, that is, busy, no dial tone, no ringback, etc., and can also include the frequency detection of Special Information Tones (SIT), such as an operator intercept. Post-connect call analysis determines the destination party’s media type, that is, voice, fax, or answering machine. The term call progress analysis (CPA) is used to refer to call progress and call analysis collectively.

Global Call API for HMP on Windows Programming Guide — August 2006 87

Call Control

7.2.2 Configuring Default Call Progress Analysis Parameters

Call Progress Analysis (CPA) is characterized by parameters such as CaSignalTimeout (the maximum time to wait to detect a call progress tone), CaAnswerTimeout (the maximum time that call analysis will wait for ringback to stop), and others that define CPA behavior. Depending on the technology you are using, the default values of CPA parameters may be configurable in the CONFIG file corresponding to the board. If this is the case, the required information is documented in the corresponding Global Call Technology Guide.

Note: When a voice resource has been attached (using either gc_OpenEx( ) or gc_AttachResource( )),

by default, the DM3 host runtime library enables the detection of BUSY, RINGING, and SIT tone (that is, pre-connect call progress), even if Call Progress Analysis (CPA) is disabled in the CONFIG file. A user who does not want pre-connect call progress, must explicitly use the gc_SetConfigData( ) function to disable CPA on that line device. Alternatively, the user can attach the voice resource after the call is connected.

7.2.3 Configuring Call Progress Analysis on a Per Call Basis

To specify call progress analysis behavior, use the gc_MakeCall( ) function with an associated GC_PARM_BLK (accessible via the GC_MAKECALL_BLK and GCLIB_MAKECALL_BLK structures) containing the CCSET_CALLANALYSIS parameter set ID and the CCPARM_CA_MODE parameter ID with one or more of the following bitmask values ORed together:

GC_CA_BUSY

Pre-connect busy tone detection

GC_CA_RINGING

Pre-connect ringback tone detection

GC_CA_SIT

Pre-connect special information tone (SIT) detection

GC_CA_FAX

Post-connect fax detection

GC_CA_PVD

Post-connect positive voice detection (PVD)

GC_CA_PAMD

Post-connect positive answering machine detection (PAMD)

While the CCPARM_CA_MODE bitmask offers complete flexibility in terms of the selected options, not all option combinations make sense. For this reason, the following defines, that can also be used as values to the CCPARM_CA_MODE parameter ID, identify the most logical and traditionally used option combinations:

GC_CA_DISABLE

Call progress and call analysis disabled

GC_CA_PREONLY

Busy and Ringing enabled

88 Global Call API for HMP on Windows Programming Guide — August 2006

GC_CA_PREONLY_SIT

Busy, Ringing and SIT enabled

GC_CA_POSTONLY_PVD

Fax and PVD enabled

GC_CA_POSTONLY_PVD_PAMD

Fax, PVD and PAMD enabled

GC_CA_ENABLE_PVD

Busy, Ringing, and SIT enabled; fax and PVD enabled

GC_CA_ENABLE_ALL

Busy, Ringing, and SIT enabled; fax, PVD and PAMD enabled

These options correspond closely to call progress and call analysis options available when using the Voice API as indicated in Table 11. See the “Call Progress Analysis” chapter in the Voice API Programming Guide.

Table 11. Call Progress Analysis Settings and Possible Results

Call Control

CCPARM_CA_MODE Setting

GC_CA_DISABLE DISABLE

GC_CA_PREONLY DX_OPTDIS

GC_CA_PREONLY_SIT DX_OPTNOCON or DX_OPTEN

GC_CA_POSTONLY_PVD DX_PVDENABLE

GC_CA_POSTONLY_PVD_PAMD DX_PAMDENABLE

GC_CA_ENABLE_PVD DX_PVDOPTNOCON or DX_PVDOPTEN

GC_CA_ENABLE_ALL DX_PAMDOPTEN

Equivalent ca_intflg Setting in DX_CAP Structure

When Using Voice API

When an option that enables call progress is selected, a GCEV_DISCONNECTED event can be received. The gc_ResultInfo( ) function can be used to get more information about the event. The possible cause values (the gcValue field in the associated GC_INFO structure) that can be retrieved are:

GCRV_BUSY

Busy

GCRV_NOANSWER

No Answer

GCRV_CEPT

SIT, Operator Intercept

GCRV_UNALLOCATED

SIT, Vacant Circuit, non-registered number

GCRV_CONGESTION

SIT, No Circuit Found or SIT, Reorder, system busy

Global Call API for HMP on Windows Programming Guide — August 2006 89

Call Control

See the Global Call API Library Reference for more information about the gc_ResultInfo( ) function.

When an option that enables call analysis is selected, a GCEV_MEDIADETECTED event can be received. The gc_GetCallInfo( ) function can be used to determine the type of detection (by setting the info_id function parameter to CONNECT_TYPE). The valuep function parameter indicates the connect type when the function completes. Typical values in this context are:

GCCT_FAX

Fax detection

GCCT_PVD

Positive voice detection (PVD)

GCCT_PAMD

Positive answering machine detection (PAMD)

See the Global Call API Library Reference for more information about the gc_GetCallInfo( ) function.

7.2.4 Setting Call Analysis Attributes on a Per Call Basis

Certain call analysis attributes can be configured on a per call basis using the gc_MakeCall( ) function with an associated GC_PARM_BLK (accessible via the GC_MAKECALL_BLK and GCLIB_MAKECALL_BLK structures) that contains the CCSET_CALLANALYSIS parameter set ID and one of the following parameter IDs:

CCPARM_CA_PAMDSPDVAL

Positive answering machine detection (PAMD) speed value. Quick or full evaluation of answering machine detection. Possible values are:

• PAMD_FULL – Full evaluation of response.

• PAMD_QUICK – Quick look at connection characteristics.

• PAMD_ACCU – Recommended setting. Does the most accurate evaluation detecting live

voice as accurately as PAMD_FULL, but is more accurate than PAMD_FULL (although slightly slower) in detecting an answering machine. Use PAMD_ACCU when accuracy is more important than speed. This is the default value.

CCPARM_CA_NOANSR

No Answer. The length of time (in 10 ms units) to wait after the first ringback before deciding that the call is not answered. Possible values are in the range 0 to 65535. The default value is

3000.

CCPARM_CA_NOSIG

Continuous No Signal. The maximum amount of silence (in 10 ms units) allowed immediately after cadence detection begins. If exceeded, a no ringback is returned. Possible values are in the range 0 to 65535. The default value is 4000.

CCPARM_CA_PAMDFAILURE

PAMD Fail Time. The maximum time (in 10 ms units) to wait for positive answering machine detection (PAMD) or positive voice detection (PVD) after a cadence break. Possible values are in the range 0 to 65535. The default value is 800.

90 Global Call API for HMP on Windows Programming Guide — August 2006

Call Control

CCPARM_CA_PAMD_QTEMP

PAMD Qualification Template. Specifies which PAMD template to use. Possible values are:

• PAMD_QUAL1TMP – First predefined qualification template. This is the default value.

• -1 – No qualification template

Setting CCPARM_CA_PAMD_QTEMP to a value of PAMD_QUAL2TMP is not supported.

Note: The CCPARM_CA_PAMD_QTEMP parameter can also be set to a qualification template ID that

is defined in the CONFIG file.

CCPARM_CA_PVD_QTEMP

PVD Qualification Template. Specifies which PVD template to use. Possible values are:

• PAMD_QUAL1TMP – First predefined qualification template. This is the default value.

• -1 – No qualification template

Setting CCPARM_CA_PVD_QTEMP to a value of PAMD_QUAL2TMP is not supported.

Note: The CCPARM_CA_PVD_QTEMP parameter can also be set to a qualification template ID that is

defined in the CONFIG file.

By default, qualification template parameters are set to the most common values. However, it is possible to tune these parameters in the CONFIG file as described in Technical Note 030 available on the Customer Support web site at:

http://resource.intel.com/telecom/support/tnotes/tnbyos/2000/tn030.htm. The technical note is not

written specifically for HMP, but the same principle applies.

Note: DM/IP boards use a slightly different version of the PVD/PAMD qualification templates; the

values are adjusted for gain loss. CONFIG files for DM/IP boards do include PVD/PAMD qualification templates.

7.2.5 Configuring Call Progress Analysis on a Per Channel Basis

Global Call also supports the setting of call progress analysis parameters on a per channel basis. When call progress analysis parameters are set on a per channel basis, the parameter settings apply to all calls made on that channel (line device).

To specify call progress analysis behavior on a per channel basis, use the gc_SetConfigData( ) function. The relevant function parameters and values in this context are:

target_type

GCTGT_CCLIB_CHAN

target_id

the line device

target_datap

a pointer to a GC_PARM_BLK structure that contains the following parameter set ID and parameter IDs:

• SetId – CCSET_CALLANALYSIS

• ParmId – CCPARM_CA_MODE that can take any of the values described in

Section 7.2.3, “Configuring Call Progress Analysis on a Per Call Basis”, on page 88.

In earlier releases, when using CAS PDK protocols, it was possible to specify call progress and call analysis on a per channel basis using the gc_SetParm( ) function to enable or disable the

Global Call API for HMP on Windows Programming Guide — August 2006 91

Call Control

GCPR_CALLPROGRESS and GCPR_MEDIADETECT parameters. See the Global Call E1/T1 CAS/R2 Technology User’s Guide for more information.

Table 12 shows how the CCPARM_CA_MODE values correspond to the GCPR_CALLPROGRESS and GCPR_MEDIADETECT parameters. This table is provided as a convenience for users that have previously used the gc_SetParm( ) method and now wish to use the greater flexibility provided by gc_MakeCall( ) with the CCPARM_CA_MODE parameter.

Table 12. Comparison with Call Progress Analysis Using gc_SetParm( )

GCPR_CALLPROGRESS GCPR_MEDIADETECT Equivalent CCPARM_CA_MODE Value

GCPV_DISABLE GCPV_DISABLE GC_CA_DISABLE

GCPV_DISABLE GCPV_ENABLE GC_CA_POSTONLY_PVD_PAMD

GCPV_ENABLE GCPV_DISABLE GC_CA_PREONLY_SIT

GCPV_ENABLE GCPV_ENABLE GC_CA_ENABLE_ALL

Note: The gc_SetConfigData( ) method of setting call progress analysis on a per channel basis is an

enhancement over using the gc_SetParm( ) with the GCPR_MEDIADETECT and/or GCPR_CALLPROGRESS parameters. Applications should not use both the gc_SetConfigData( ) method and the gc_SetParm( ) method on the same line device. If both methods are used, the gc_SetConfigData( ) method takes precedence.

7.2.6 Setting Call Analysis Attributes on a Per Channel Basis

In addition to enabling and disabling call progress analysis on a per channel basis, certain call analysis attributes can be configured on a per channel basis using the gc_SetConfigData( ) function. The relevant function parameter values in this context are:

target_type

GCTGT_CCLIB_CHAN

target_id

the line device

target_datap

a pointer to a GC_PARM_BLK structure that contains the following parameter set ID and parameter IDs:

• SetId – CCSET_CALLANALYSIS

• ParmId – Any of the values described in Section 7.2.4, “Setting Call Analysis Attributes

on a Per Call Basis”, on page 90.

7.2.7 Customizing Call Progress Tones on a Per Board Basis

When using Digital Network Interfaceboards, an application can create, delete and query call progress tones on a per board device basis using the dx_createtone( ), dx_deletetone( ), and dx_querytone( ) functions and the associated TONE_DATA structure in the Voice API. See the Voice API Programming Guide for more information.

92 Global Call API for HMP on Windows Programming Guide — August 2006

7.3 Resource Routing

The Global Call routing functions use the device handles of resources such as a voice channel, a media resource, or a network time slot. The gc_GetResourceH( ) function can be used to obtain the network, media and voice device handles, associated with the specified line device.

The gc_GetResourceH( ) function, with a resourcetype of GC_MEDIADEVICE returns the media device handle for the specified line device.

The gc_GetResourceH( ) function, with a resourcetype of GC_NETWORKDEVICE returns the network device handle for the specified line device.

The gc_GetResourceH( ) function, with a resourcetype of GC_VOICEDEVICE, returns the voice device handle only if the specified line device has a voice, media, or tone resource associated with it, for example, if a voice channel was specified in the gc_OpenEx( ) function devicename parameter, or if the voice channel was subsequently attached to the line device and has remained attached to that line device.

Refer to the appropriate Global Call Technology Guide for technology-specific information on routing resources when using the gc_OpenEx( ) function to specify a voice or media resource, or when using the gc_AttachResource( ) function to associate a voice or media resource with a Global Call line device.

Call Control

7.4 Feature Transparency and Extension

Global Call Feature Transparency and Extension (FTE) provides a common interface to multiple network interface specific libraries for features that are abstracted across multiple call control libraries (see Figure 1, “Global Call Architecture for IP Technology”, on page 20 and Figure 2,

“Global Call Architecture for E1/T1 and ISDN Technologies”, on page 21). FTE is described in the

following topics:

• Feature Transparency and Extension Overview

• Technology-Specific Feature Access

• Technology-Specific User Information

7.4.1 Feature Transparency and Extension Overview

FTE is comprised of a number of Global Call functions. These functions provide the flexibility to extend the generic Global Call API to access all technology or protocol-specific features unique to any given network interfaces that were formerly only accessible via their native technology call control libraries. Thus, all technology-specific features may be accessible from the application solely via the singular Global Call library interface, thereby alleviating the need to access these call control libraries directly via additional APIs.

Global Call API for HMP on Windows Programming Guide — August 2006 93

Call Control

The Global Call API functions provided for FTE are:

gc_Extension( )

provides a generic interface extensible for technology-specific features

gc_GetUserInfo( ) (for E1, T1 and ISDN technologies only)

retrieves technology-specific user information for the specified line device

gc_SetUserInfo( )

permits technology-specific user information to be defined for the specified line device or call

Note: The gc_SetUserInfo( ) function is not supported for a board device.

7.4.2 Technology-Specific Feature Access

The gc_Extension( ) function provides a single common interface to access various technologyspecific features supported by underlying call control libraries.

This Global Call function utilizes an extension function identifier (ext_id) to specify the feature. The associated technology’s Global Call Technology Guide for each call control library lists all the supported extension function identifiers (ext_id values) and the associated features that are accessible via the gc_Extension( ) function (if any).

By specifying the associated parameter identifiers (also described in the associated technology’s Global Call Technology Guide), and either the target line device or a specific call, those features unique to the subject technology may be utilized entirely using the Global Call API. Without FTE support, a Global Call application requiring this feature support would also have to be written to the specific call control API in addition to the Global Call API.

For example, in an ISDN platform the application may use the gc_Extension( ) function to set D or B channel states. As the concept of B and D channels is ISDN specific and inherently foreign to other protocols, without FTE support, the application would have to link directly with the ISDN call control library then call the required ISDN library functions cc_SetBChanState( ) or cc_SetDChanState( ).

The gc_Extension( ) function may be supported in either asynchronous mode, synchronous mode or both depending on the call control library.

If the gc_Extension( ) function is supported and called in synchronous mode, the relevant information parameters returned in the GC_PARM_BLK buffer must be processed or copied prior to the next Global Call function call. The reason for this is that the GC_PARM_BLK buffer will be deallocated within Global Call in a subsequent function call.

If the gc_Extension( ) function is supported and called in asynchronous mode, relevant information may be returned via the call control library via GCEV_EXTENSIONCMPLT termination event and its referenced extension block structure, EXTENSIONEVTBLK. The EXTENSIONEVTBLK structure contains technology-specific information and is referenced via the extevtdatap pointer in the METAEVENT structure associated with the GCEV_EXTENSIONCMPLT event. See the Global Call API Library Reference for more information about these structures.

94 Global Call API for HMP on Windows Programming Guide — August 2006

The gc_Extension( ) function can also be used to transmit information to the remote endpoint. In this case, while the application at the local end point receives a GCEV_EXTENSIONCMPLT, the application at the remote end point will receive an unsolicited GCEV_EXTENSION notification event from the network with the transmitted information. The EXTENSIONEVTBLK structure contains the transmitted information and is referenced via the extevtdatap pointer in the METAEVENT structure associated with the GCEV_EXTENSION event.

The application at the local end point may also receive an unsolicited GCEV_EXTENSION event with information from the network.

It is important to note that the EXTENSIONEVTBLK structure referenced in the GCEV_EXTENSION event has a persistence only until the next call of gc_GetMetaEvent( ). In other words, any information contained or referenced in the associated EXTENSIONEVTBLK structure must be either processed or copied in the application, or risk having the memory space containing the actual information lost on the next gc_GetMetaEvent( ) call.

7.4.3 Technology-Specific User Information

The gc_GetUserInfo( ) (E1, T1 and ISDN technologies only) and gc_SetUserInfo( ) (all technologies) functions permits the application to retrieve and configure user information for the specified line device that is transmitted to or received from the remote side. The actual content and format of the user information is technology- or protocol-specific, or both. Refer to the associated technology’s Global Call Technology Guide for details on the format of the user information supported and the proper usage of the gc_GetUserInfo( ) and gc_SetUserInfo( ) functions.

Call Control

One typical application of the gc_SetUserInfo( ) and gc_GetUserInfo( ) functions is on an ISDN platform where it is desired to transmit and receive user-to-user information elements in each incoming and outgoing message.

In the case of gc_SetUserInfo( ), user information is transmitted to the remote side embedded in a protocol-specific message. The duration flag is used to specify the persistence of the information. Using the duration flag, the user information may be specified to persist as long as the current or next call, or for all calls (including the current call). When the duration is specified to be all calls on the specified line device, the user information is valid and utilized for all calls until the device is eventually closed via gc_Close( ).

In the case of gc_GetUserInfo( ), the user information is retrieved from an already received protocol-specific message that has been received from the remote side. Note that the user information parameters returned from the call control library in the GC_PARM_BLK buffer must be processed or copied prior to the next Global Call function call. The reason for this is that the GC_PARM_BLK buffer will be deallocated within Global Call in a subsequent function call.

Global Call API for HMP on Windows Programming Guide — August 2006 95

Call Control

96 Global Call API for HMP on Windows Programming Guide — August 2006

8.Alarm Handling

This chapter describes the Global Call Alarm Management System (GCAMS). Topics include the following:

• Alarm Handling Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

• Operation and Configuration of GCAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

• Sample Alarm Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

8.1 Alarm Handling Overview

Global Call alarms originate from alarm source objects (ASO). An alarm source object can be a network library, a call control library, or it can reside within a call control library. Some alarm source objects are for internal Global Call use only and are not available to the application.

There are basically two sources of Global Call alarms:

• Layer 1 alarms (physical alarms)

• “Logical” alarms, such as remote side out of service, or layer 2 or layer 3 out of service

The portion of the Global Call call control library that manages alarms is called the Global Call Alarm Management System (GCAMS). GCAMS is initialized automatically when Global Call is started.

GCAMS provides Global Call applications with the ability to receive extensive alarm information. Some of the ways this information can be used include:

• Managing the network

• Troubleshooting hardware

• Monitoring line quality

• Working with the central office to solve line problems

• Generating status reports

• Modifying alarm source object properties and characteristics based on alarm history

• Manual handling of alarms for drop and insert applications.

The following sections describe the components and operation of GCAMS.

8.1.1 Alarm Management System Components

The alarm management system is made up of several components, including GCAMS. The other components are the customer application’s alarm management system (AMS), and the alarm source objects (ASOs). ASOs can either reside within a call control library (cclib) or separate from

Global Call API for HMP on Windows Programming Guide — August 2006 97

Alarm Handling

a call control library. Figure 21 illustrates the relationship between the alarm management system components.

Figure 21. Architectural Diagram of Alarm Management Components

CUSTOMER APPLICATION

Operation and Configuration Subsystem

Customer AMS

GlobalCall

Operation and Configuration Subsystem

GCAMS

CALL CONTROL LIBRARY

ASO (optional)

Network Interface

ASO

Network

Interface

The customer application is responsible for configuring the behavior of GCAMS, including the designation of which alarms are blocking, which alarms the application wants to be notified of, and controlling the flow of alarms to the application. For more information, see Section 8.2.3,

“Configuration of Alarm Properties and Characteristics”, on page 101.

GCAMS acts as an interface between the customer application and the alarm source objects. GCAMS passes requests from the application to the ASOs, processes application configuration requests, and processes ASO alarm events. GCAMS also maintains a database of the current configuration attributes by alarm source object and line device. In addition, GCAMS implements the ASOs that are common across multiple technologies. For more on the operation and configuration of GCAMS, see Section 8.2, “Operation and Configuration of GCAMS”, on page 99.

The final components of the alarm management system are the ASOs. ASOs are responsible for generating alarm events when alarms occur and then clear. If configured to do so, ASOs are also

98 Global Call API for HMP on Windows Programming Guide — August 2006

responsible for starting and stopping the transmission of alarms and setting and getting alarm parameters, such as timing parameters.

8.2 Operation and Configuration of GCAMS

The primary functions of GCAMS are as follows:

• Generation of Events for Blocking Alarms

• Generation of Alarm Events

• Configuration of Alarm Properties and Characteristics

• Starting and Stopping Alarm Transmission (E1, T1 and ISDN technologies only)

• Retrieving Alarm Data

8.2.1 Generation of Events for Blocking Alarms

Global Call alarms are classified as either blocking or non-blocking. Blocking alarms are alarms that cause the application to become blocked and potentially generate a GCEV_BLOCKED event when the alarm is set (the “alarm on” condition is detected). Subsequently, all blocking alarms generate a GCEV_UNBLOCKED event when the alarm clears (the “alarm off” condition is detected). Non-blocking alarms are alarms that do not cause the application to become blocked and do not generate a GCEV_BLOCKED or GCEV_UNBLOCKED event when the alarm is set or clears.

Alarm Handling

Note: The gc_SetAlarmConfiguration( ) function can be used to change which alarms are blocking and

which alarms are not blocking for a given alarm source object. To retrieve the status of the current alarm configuration, use gc_GetAlarmConfiguration( ). For more on changing the configuration of alarm source objects, see Section 8.2.3, “Configuration of Alarm Properties and

Characteristics”, on page 101.

The GCEV_BLOCKED and GCEV_UNBLOCKED events are unsolicited events that are sent in addition to other Global Call events. The blocked and unblocked events do not require any application-initiated action. The blocked event is generated only for the first blocking condition detected. Subsequent blocking conditions on the same line device will not generate additional blocked events. Until all blocking conditions are cleared, the line device affected by the blocking condition (that is, the line device that received the GCEV_BLOCKED event) cannot generate or accept calls. When the line device has completely recovered from the blocking condition a GCEV_UNBLOCKED event is sent.

When a blocking condition occurs while a call is in progress or connected, any calls on the line device that is in the blocked condition are treated in the same manner as if a remote disconnection occurred: an unsolicited GCEV_DISCONNECTED event is sent to the application and the call changes to the Disconnected state. The result value retrieved for the event will indicate the reason for the disconnection, for example, an alarm condition occurred. Result values are retrieved by calling the gc_ResultInfo( ) function, see Section 4.4, “Event Retrieval”, on page 73. The GCEV_BLOCKED event is also sent to the application to indicate that a blocking condition occurred; the gc_ResultInfo( ) function can be called to retrieve the reason for the GCEV_BLOCKED event, as well.

Global Call API for HMP on Windows Programming Guide — August 2006 99

Alarm Handling

The GCEV_BLOCKED and GCEV_DISCONNECTED events may arrive in any order. When the blocking condition(s) clears, an unsolicited GCEV_UNBLOCKED event is sent to the application indicating complete recovery from the blocking condition.

When a blocking condition occurs while a line device is in the Null, Disconnected, or Idle state, only the GCEV_BLOCKED event is sent since there is no call to disconnect. The call state does not change when a GCEV_BLOCKED or GCEV_UNBLOCKED event is sent to the application.

Note: In the asynchronous mode, if a gc_WaitCall( ) function is pending when a GCEV_UNBLOCKED

event is generated, the gc_WaitCall( ) function does not need to be reissued.

The GCEV_BLOCKED and GCEV_UNBLOCKED events are generated for blocking alarms at the logical or virtual board or trunk level and the channel level:

Logical or Virtual Board (IP technology) or Trunk (E1, T1 or ISDN technology) Level

When the Global Call API recognizes a blocking alarm on condition at the logical or virtual board or trunk level, a GCEV_BLOCKED event is generated for the logical or virtual board or trunk device, assuming that the device is open. A GCEV_BLOCKED event is also generated for all time slots currently open on the logical or virtual board or trunk device, assuming that the application is currently unblocked. The application will receive a GCEV_BLOCKED event only for the first alarm on condition for a particular line device.

When the Global Call API recognizes a blocking alarm off condition at the logical or virtual board or trunk level, a GCEV_UNBLOCKED event is generated for the logical or virtual board or trunk device, assuming that the device is open. A GCEV_UNBLOCKED event is also generated for all time slots currently open on the logical or virtual board or trunk device, assuming there are no other blocking conditions on the line device. The application will receive a GCEV_UNBLOCKED event only for the last “alarm off” condition for a particular line device.

Channel Level

When the Global Call API recognizes a blocking alarm on condition at the channel level, a GCEV_BLOCKED event is generated for the channel, assuming that the application is currently unblocked. The application will receive a GCEV_BLOCKED event only for the first alarm on condition for the line device.

When the Global Call API recognizes a blocking alarm off condition at the channel level, a GCEV_UNBLOCKED event is generated for the time slot, assuming there are no other blocking conditions on the line device. The application will receive a GCEV_UNBLOCKED event only for the last alarm off condition for the line device.

Note: When using Global Call with Intel NetStructure

only a the trunk level. An alarm that occurs on a trunk applies to all channels on that trunk.

8.2.2 Generation of Alarm Events

The GCEV_ALARM event can be generated by both blocking and non-blocking alarms. Blocking alarms are alarms that generate GCEV_BLOCKED and GCEV_UNBLOCKED events when the alarms set and clear. GCEV_ALARM events are for information purposes only and do not cause any channel state or call state changes.

In order for the GCEV_ALARM event to be returned by the application, the notify attribute for the specified alarm source object must be set to “on” via the gc_SetAlarmConfiguration( ) function.

Digital Network Interface boards, alarms apply

100 Global Call API for HMP on Windows Programming Guide — August 2006

Intel 05-2409-003 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Global Call API for Host Media Processing on Windows

Contents

Figures

Tables

Revision History

About This Publication

1.Product Description

1.1 Global Call Software Overview

1.2 Global Call Feature Categories

1.2.1 Call Control Features

1.2.2 Operation, Administration and Maintenance Features

1.3 Global Call Architecture

1.3.1 Overview

1.3.2 Global Call API

1.4 Call Control Libraries

1.4.1 Starting Call Control Libraries

1.4.2 Call Control Library States

1.5 Global Call Object Identifiers

1.5.1 Line Device Identifier

1.5.2 Call Reference Number

1.5.3 Object Identifiers and Resource Sharing Across Processes

1.5.4 Target Objects

2.Programming Models

2.1 Programming Models Overview

2.2 Asynchronous Mode Programming

2.2.1 Asynchronous Model Overview

2.2.2 Asynchronous Model with Event Handlers

2.2.3 Asynchronous with Windows Callback Model

2.2.4 Asynchronous with Win32 Synchronization Model

2.2.5 Extended Asynchronous Programming Model

3.Call State Models

3.1 Call State Model Overview

3.2 Basic Call Model

3.2.1 Basic Call States at the Inbound Interface

3.2.2 Basic Call States at the Outbound Interface

3.2.3 Basic Call States for Call Termination

3.3 Basic Call Model Configuration Options

3.3.1 Call State Configuration

3.3.2 Call State Event Configuration

3.3.3 Call Acknowledgement Configuration

3.3.4 Call Proceeding Configuration

3.4 Basic Call Control in Asynchronous Mode

3.4.1 Inbound Calls in Asynchronous Mode

3.4.1.1 Inbound Calls in Asynchronous Mode Overview

3.4.1.2 Channel Initialization

3.4.1.3 Call Detection

3.4.1.4 Call Offered

3.4.1.5 Call Routing

3.4.1.6 Call Acceptance

3.4.1.7 Call Establishment

3.4.1.8 Overlap Receiving

3.4.1.9 Call Failure

3.4.1.10 Abandoned Calls

3.4.1.11 Inbound Call Scenarios in Asynchronous Mode

3.4.2 Outbound Calls in Asynchronous Mode

3.4.2.1 Outbound Calls in Asynchronous Mode Overview

3.4.2.2 Channel Initialization

3.4.2.3 Call Dialing

3.4.2.4 Call Proceeding

3.4.2.5 Call Alerting

3.4.2.6 Call Connected

3.4.2.7 Overlap Sending

3.4.2.8 Call Failure

3.4.2.9 Outbound Call Scenarios in Asynchronous Mode

3.4.3 Call Termination in Asynchronous Mode

3.4.3.1 Call Termination in Asynchronous Mode Overview

3.4.3.2 User Initiated Termination

3.4.3.3 Network Initiated Termination

3.4.3.4 Call Release

3.4.3.5 Call Termination Call Control Scenarios in Asynchronous Mode

3.4.4 Handling Unsolicited Events

3.5 Advanced Call Control with Call Hold and Transfer

3.5.1 Advanced Call State Model Overview

3.5.2 Advanced Call States for Hold and Transfer

3.5.3 Call Hold

3.5.4 Call Transfer