ISDN Echocancellation Circuit for Terminal Applications
IEC-Q TE
PSB 21911 Version 5.2
PSF 21911 Version 5.2
Data Sheet11.97
DS 1
Page 2
PSB 21911
Revision History: Original Version: 11.97
Previous Releases: None
Page Subjects (changes since last revision)
For questions on technology, delivery and prices please con tact the Semiconductor
Group Offices in Germany or the Siemens Companies and Representatives worldwide:
see our webpage at http://www.siemens.de/Semiconductor/address/address.htm.
As far as patents or other rights of third parties are concerned, liability is only assumed for components, not for
applications, processes and circui ts implemented within component s or assemblies.
The information describes the type of component and s hall not be considered as assured ch aracteristics.
Terms of delivery and rights to change design reserved.
Due to tec hnical requirements components may contain dangerous substances. For information on the types in
question please contact your nearest Siemens Off ice, Semiconductor Group.
Siemens AG is an approved CECC manufacturer.
Packing
Please use the recycling operators known to you. We can also help you – get in touch with your nearest sales
office. By agreement we will take packing material back, if it is sorted. You must bear the costs of transport.
For pac king mat erial that is ret urned to us unsorte d or wh ich we a re not oblig ed to a ccept, we shal l have to in v oice
you for any costs incurred.
Components used in life-support devices or systems must be expressly authorized for such purpose!
Criti cal compon ents1 of the Semiconductor Group of Siemens AG, may only be used in life-support devices or
systems2 with the express written approval of the Semiconductor Group of Siemens AG.
1 A critical component is a component used in a life-support devi ce or system whose failure can reasonably be
expected t o ca use t he f ail u re o f th at l ife- s upp ort devi ce o r s ys tem, or to af fe ct i t s saf et y or ef fe ct ive ness of t hat
device or system.
2 Life support devices or systems are intended (a) to be implanted in the human body, or (b) to support and/or
maintain and sustain human life. If they fail, it is reasonable to assume that the health of the user may be endangered.
The PSB 21911, IEC-Q TE Version 5.2, is a specific derivative of the PEB 2091, IEC-Q
for terminal and small PBX applications. It features all necessary functions required for
NTs and terminal applications like PC add-on cards and terminal adapt ers .
In stand-alone mode the PSB 21911 IEC-Q TE Version 5.2 can be used fully pin
compatible to IEC-Q V4.4 and former vers ions. In µP mode it offers a parallel or serial
microprocessor interface.
The P
rocessor Interface (PI) of the IEC-Q TE V 5.2 establishes the access of a
microprocessor between U-interface and IOM-2. It’s main function is illustrated in
figure 1
.
FSC
R
IOM -2
R
IOM -2
UU
PI
ITS10193
Figure 1Stand-Alone Mode (left) and µP Mode (right)
In µP mode B ch annels, D channel, C/I codes and Monitor commands can either be
passed between the U-transceiver and IOM-2 directly, or they can be looped through the
µP via the PI. Any selection of "passed" or "looped" channels can be programmed via a
control register.
The µP-interface mode is en abl ed by setting the pin PMODE to "1". This pin w as not to
be connected in older versions of the IEC-Q. Its internal pul l down resistor selects the
stand-alone mode, if the pin is left open.
In stand-alone mode the IEC-Q TE is controlled exclusively vi a the IOM-2 interface and
mode selection pins.
Semiconductor Group811.97
Page 9
ISDN Echo cancellation Circuit fo r Terminal
PSB 21911
Applications
IEC-Q TE
Version 5.2CMOS
1.1Features
• ISDN U-transceiver with IO M-2 and optional m icro-
processor interface
• Compatible to NT modes and TE mode of
PEB 2091 IEC-Q V5.1
• Perfectly suited for terminal and TA applications
• U-interface (2B1Q) conform to ANSI T1.601, ETSI
ETR 080 and CNET ST/LAA/ELR/DNP/822:
–Meets all transmission requirements on all ANSI,
ETSI and CNET loops with margin
–Conform to British Telecom’s RC7355E
–Compliant with ETSI 10ms micro interruptio ns
• IOM-2 interface for connection of e.g. ISAC-S,
SICOFI-2/4TE, ARCOFI, ITAC, HSCX-TE, ISAR,
IPAC, 3PAC
• Pin compatible to version 4.4 in the P-LCC-44
package
P-LCC-44
T-QFP-64
In µP mode:
• Parallel or serial microprocessor interface and watchdog
• µP access to B-channels, D-channel and intercommunication channels
• µP access to IOM-2 Monitor-channels and C/I-channels
• Adjustable microcontroller clock source between 0.96MHz and 7.68MHz
• Selection between Bit clock (BCL) and Data clock (DCL)
• Supports synchronization of bas esta tions in cordl ess app licatio ns (e.g . RITL)
• Supports D-channel arbitration with ELIC linecard (e.g. PBX)
Figure 4Pin Configuration P-LCC-44 and T-QFP-64 Package (top view)
Semiconductor Group1211.97
Page 13
PSB 21911
PSF 21911
Pin Definitions and Functions
1.5Pin Definitions and Functions
The following tables group the pins according to their functions. They include pin name,
pin number, type and a brief description of the function.
Pin No.SymbolI/OFunction
P-LCC-44 T-QFP64
Standalone
µP
mode
Power Supply Pins
1, 27, 8, 12
V
DDd
V
DDd
I5 V
±
5% digital supply voltage
514GNDa1 GNDa1I0 V analog
7, 818, 19
921
V
V
DDa1
REF
V
V
DDa1
REF
I5 V ± 5% analog supply voltage
O
V
pin to buffer internally generated
REF
voltage with capacitor 100 nF vs GND
1326
V
DDa2
V
DDa2
I5 V ± 5% analog supply voltage
1630GNDa2 GNDa2I0 V analog
2341GNDdG NDdI0 V digital
Mode Selection Pins
310TSPI
Single Pulse Test Mode
CS
1835AUTOI
RST
I
O
For activation refer to table 3 on
page 27. When active, alternating
2.5 V pulses are issued in 1.5 ms
intervals. Tie to GND if not used.
Chip Select
(Multiplexed, demultiplexed and serial
modes): Low active.
Auto EOC Mode
Selection between auto- and
transparent mode for EOC channel
processing. (Automode = (1))
Reset output
(Multiplexed, demultiplexed and serial
modes): Low active.
Semiconductor Group1311.97
Page 14
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
2443GNDdI
2544GNDd I
Standalone
µP
mode
A0I
SMODEI
D7I/O
AD7I/O
GNDd
Must be connected t o GNDd in sta ndalone mode.
Address Bus Pin
(Demultiplexed mode)
Serial Mode Pin:
serial mode, SMODE = 0 enables the
multiplexed mode.
GNDd
Must be connected t o GNDd in sta ndalone mode.
Data Bus Pin
(Demultiplexed modes)
Address Data Bus Pin
SMODE = 1 selects
not used I(Serial mode) tie to GND.
3355MS0I
not used I(Multiplexed mode) tie to GND.
A3I
not used(Serial mode) tie to GND.
3558MS1I
not used I(Multiplexed mode) tie to GND.
A2I
CDOUTO
(Multiplexed mode)
Mode Selection 0
refer to table 2 on page 26.
Address Bus Pin
(Demultiplexed modes).
Mode Selection 1
refer to table 2 on page 26.
Address Bus Pin
(Demultiplexed modes).
Controller Data Out
CCLK determines the data rate.
CDOUT is "high Z" if no data is
transmitted.
Semiconductor Group1411.97
Page 15
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
3659MS2I
2847RES
Power Controlle r Interface P in s
Standalone
µP
mode
not used I(Multiplexed mode) tie to GND.
A1I
CDINI
RESI
Mode Selection 2
refer to table 2 on page 26.
Address Bus Pin
(Demultiplexed modes).
Controller Data In
CCLK determines the data rate.
Reset
Low active, must be (0) at least for
10 ns.
Refer also to table 3 on page 27 for test
modes invoked with this pin.
(Serial mode)
445PCD0I/O
(PU)
AD0I/O
D0I/O
not used I(Serial mode) tie to GND.
434PCD1I/O
(PU)
AD1I/O
D1I/O
Data Bus 0 of Power Controller
Interface
internal pull-up.
Address/Data Bus Pin
(Multiplexed mode)
Data Bus Pin
(Demultiplexed modes)
Data Bus 1 of Power Controller
Interface
Internal pull-up.
Address/Data Bus Pin
(Multiplexed mode)
Data Bus Pin
(Demultiplexed modes)
not used I(Serial mode) tie to GND.
Semiconductor Group1511.97
Page 16
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
423PCD2I/O
3962PCA0O
Standalone
µP
mode
(PU)
AD2I/O
D2I/O
not used I(Serial mode) tie to GND.
D5I/O
AD5I/O
Data Bus 2 of Power Controller
Interface
Internal pull-up.
Address/Data Bus Pin
(Multiplexed mode)
Data Bus Pin
(Demultiplexed modes)
Address bus 0 of Power Controller
Interface
Data Bus Pin
(Demultiplexed modes)
Address Data Bus Pin
(Multiplexed mode)
.
not used I(Serial mode) tie to GND.
3861PCA1O
D6I/O
AD6I/O
not used I(Serial mode) tie to GND.
412PCRD
D3I/O
AD3I/O
not used I(Serial mode) tie to GND.
O
Address bus 1 of Power Controller
Interface
Data Bus Pin
(Demultiplexed modes)
Address Data Bus Pin
(Multiplexed mode)
Power Controller Bus Read Request
Low active.
Data Bus Pin
(Demultiplexed modes)
Address/Data Bus Pin
(Multiplexed mode)
Semiconductor Group1611.97
Page 17
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
401PCWRO
1936INTI
Standalone
µP
mode
D4I/O
AD4I/O
not used I(Serial mode) tie to GND.
INT
O
Power Controller Bus Write Request
Low active.
Data Bus Pin
(Demultiplexed modes)
Address/Data Bus Pin
(Multiplexed mode)
Interrupt
Change-sensitive. After a change of
level has been detected the C/I code
“INT” will be issued on IOM. Tie to
GND if not used.
Interrupt Line
(Multiplexed, demultiplexed and serial
modes): Low active
.
3760DISSO
MCLKO
2138PS1PS1I
2239PS2PS2I
Disable Power Supply
This pin is set to ’1’ after receipt of
MON-0 LBBD in EOC auto-mode.
Microprocessor Clock Output
(Multiplexed, demultiplexed and serial
modes): provided with four
programmable clock rates: 7.68 MHz ,
3.84 MHz, 1.92 MHz and 0.96 MHz.
Power Status 1 (prima ry )
’1’ indicates primary pow er supply ok.
The pin level is identical to the
overhead bit ’PS1’ value.
Power Status 2 (secondary)
’1’ indicates secondary power supply
ok. The pin level is identical to the
overhead bit ’PS2’ value.
.
Semiconductor Group1711.97
Page 18
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
Miscellaneous Function Pins
1022XOUTXOUTO
1123XINXINI
1732DODI
Standalone
µP
mode
WR
Crystal OUT
To connect 15.36-MHz crystal.
Leave open if not used.
Crystal IN
To connect 15.36-MHz crystal or
external 15.36-MHz clock.
DOUT Open Drain
Select open drain with DOD = (1)
(external pull-up resi sto r requir ed) and
tristate with DOD = (0). See also
table 4 on page 27.
I
Write
(Siemens/Intel multiplexed and
demultiplexed modes): indicates a
write operation, active low.
R/W
not used I(Serial mode) tie to GND.
2951TPTPI
2037TP1I
ALEI
I
(PD)
(PD)
Read/Write
(Motorola demultiplexed mode):
indicates a read (high) or write (low)
operation.
Test Pin
Not available to user. Do not connect.
Internal pull-down resistor.
Test Pin 1
Not available to user. Do not connect.
Internal pull-down resistor.
Address Latch Enable
(Multiplexed mode): In the Siemens/
Intel µP interface modes a high
indicates an address on the AD0..3
pins which is latched with the falling
edge of ALE (see also page 39).
Semiconductor Group1811.97
Page 19
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
Standalone
µP
mode
ALEI
CCLKI
3254CLSCLSO
1224PMODEPMODEI
(PD)
Address Latch Enable
(Demultiplexed mode): ALE tied to
GND selects the Siemens/Intel type.
ALE tied to VDD selects the Motorola
type.
Controller Data Clock
(Serial mode): Shifts data from (1) and
to (0) the device.
Clock Signal
A 7.68MHz clock, sync hronous to the
U-interface, is provided on this pin.
Processor Interface Enable
Setting PMODE to “1“ enables the
Processor Interface. Tie to GND or do
not connect to select stand-alone
mode. Internal pull down.
Disables the internal 6 ms Monitor
time-out when set to (1). Internal pulldown resistor.
Read
(Siemens/Intel multiplexed and
demultiplexed modes): indicate s a
read operation, active low.
Data Strobe
(Motorola demultiplexed mode):
indicates a data transfer, active low.
Semiconductor Group1911.97
Page 20
Pin Definitions and Functions
Pin No.SymbolI/OFunction
PSB 21911
PSF 21911
P-LCC-44 T-QFP64
®
IOM
-2 Pins
Standalone
µP
mode
3153DCLDCLO
3052FSCFSCO
2645DINDINI
Data Clock
Data clock output 512 or 1536 kHz
(table 2 on page 26). In µ P mode this
pin can be programmed to deliver a bit
clock (256 or 768 kHz).
Frame Synchronization Clock
The start of the B1-channel in ti me-slot
0 is marked. FSC = (1) for one DCLperiod indicates a superframe marker.
FSC = (1) for at least two DCL-periods
marks a standard frame.
Data In
Input of IOM-2 data synchronous to
DCL-clock (Data upstream direction).
2746DOUTDOUTO
U-Interface Pins
1529AINAINI
1428BINBINI
616AOUTAOUTO
413BOUTBOUTO
PU: internal pull-up resistor
Data Out
Output of IOM-2 data synchronous to
DCL-clock. Open drain or tristate
depending on bit/pin DOD (Data
Downstream direction).
Differential U-Interface Input
Connect to hybrid.
Differential U-Interface Input
Connect to hybrid.
Differential U-Interface Output
Connect to hybrid.
Differential U-Interface Output
Connect to hybrid.
PD: internal pull-down resistor
Semiconductor Group2011.97
Page 21
PSB 21911
PSF 21911
Microprocessor Bus Interface (Overview)
1.6Microprocessor Bus Inte rface (Overview )
The table below gives an overview of the different microprocess or bus m odes.
Due to the IOM-2 interface the IEC-Q TE can be combined with a variety of other devices
to fit in numerous applications. This chapter only shows some typical applications of the
IEC-Q TE.
1.7.1ISDN PC Adapter Card
An ISDN adapter card which supports the U-interface may be realized using the
IEC-Q TE together with the PSB 2113 3PAC (
and two B-channel HDLC controllers. Optionally, a PSB 2132 SICOFI2-TE can be
connected to provide two POTS interfaces. If an S-interface is required, the PSB 2115
IPAC can be used instead of the 3PAC.
figure 5
). The 3PAC provides a D-channel
Figure 5ISDN PC Adapter Card
Semiconductor Group2211.97
Page 23
PSB 21911
PSF 21911
System Integration
1.7.2ISDN Stand-Alone Terminal with POTS Interface
The IEC-Q TE can be integrated in a microcontroller based stand-alone terminal
(figure 6) that is connected to the communications interface of a PC. The PSB 2132
SICOFI-TE enables connect ion of analog terminals (e.g. telepho nes or fax) to its dual
channel POTS interface.
Figure 6ISDN Stand-Alone Terminal with POTS Interface
Semiconductor Group2311.97
Page 24
PSB 21911
PSF 21911
System Integration
1.7.3ISDN Feature Phone
An ISDN feature phone with U-interface can be built using the IEC-Q TE together with
the ARCOFI-SP and the ICC.
Figure 7ISDN Feature Phone
Semiconductor Group2411.97
Page 25
PSB 21911
PSF 21911
System Integration
1.7.4ISDN-Modem PC Card
The combination of the IEC-Q TE and a PSB 7115 I SAR 34 allows to build an ISDNmodem PC card .
Figure 8ISDN-Modem PC Card
Semiconductor Group2511.97
Page 26
PSB 21911
PSF 21911
Operating Modes
2Functional Description
2.1Operating Modes
The default configuration after pow er-on or external reset depends on t he state of the
PMODE pin. The cases µP mode and stand-alone mode have to be distinguished:
µP mode (
PMODE
= VDD)
In µP mode a microprocessor inte rface gives access t o the IOM-2 channel r egi st ers as
well as configuration registers. The operating mode is selected via bits STCR:MS0-MS2
according to
table 2
. The STCR register is described on page 119.
Test modes Send Single Pulses, Quiet Mode or Data Through are invoked via the
corresponding C/I channel command (page 75) or via bits STCR:TM1-2 (
Stand-alone mode (
PMODE
= GND)
table 3
).
In stand-alone mode the operating mode is selected via pin strapping according to
table 2
. It is possible to change the mode of a device during operation (e.g. for test
purposes) if the mode change is followed by a reset.
The test modes Send Single Pulses (SSP), Quiet Mode (QM) and Data Through (DT)
are invoked via the corresponding C/I channel command (page 75) or via pins RES
TSP (
table 3
).
and
Table 2Modes of Operation (µP and Stand-Alone Mode)
1 DCL-period high-phase of FSC at superframe position
2 DCL-periods high-phase of FSC at normal position
2)
CLS-clock signal not available while device is in power-down
Page 27
Table 3Test Modes
PSB 21911
PSF 21911
Operating Modes
Test-Mode
Master-Reset
1)
Send Single-Pulses
Data-Through
3)
2)
Bit TM1/
Pin RES
00
11
01
Bit TM2/
Pin TSP
Normal10
1)
Used for Quiet Mode and Return Loss measurements
2)
Used for Pulse Mask measurements
3)
Used for Insertion Loss, Power Spect ral Density and Tot al Power measurements
Table 4DOUT Driver Modes
ModePin
RES
Pin
1)
TSP
2)
Pin /
Bit
DOD
Pin DOUT Output Driver
ValueDOUT in
active time
slot
DOUT in
passive time
slot
Pin-Reset 00x0lowint. p u ll-up
1int. pull-up
Normal
(Tristate)
Normal
3)
(Open Drain
1)
In stand-alone mode and µP mode
2)
Only in stand-alone mode. In µP mode the output driver of pin DOUT is selected via bit DOD in the ADF2
regist er
3)
External pull-up resistors required (typ.1 kΩ)
)
1000lowhigh Z
1high
1010lowfloating
1floating
Semiconductor Group2711.97
Page 28
2.2Devi ce Arch itecture
In µP mode the following interfaces and functional blocks are used:
• IOM-2 interfaceseepp. 30
• Microprocessor interfacepp. 39, 81, 112
• U-transceiverpp. 40
• Clock G ener ationpp. 111
• Resetpp. 93
• Factory Test Unit
PSB 21911
PSF 21911
Device Architecture
Figure 9IEC-Q TE Device Architecture (µP Mode)
Semiconductor Group2811.97
Page 29
PSB 21911
PSF 21911
Device Architecture
In stand-alone mode the following interfaces and functional bloc ks are use d:
• Mode Selectionseepp. 26
• IOM-2 interfacepp. 30
• IOM-2 configurationpp. 36, 38
• U-transceiverpp. 40
• Clock Generationpp. 111
• Resetpp. 93
• Power Controller Interfacepp. 94
• Factory Test Unit
Figure 10 IEC-Q TE Device Architecture (Stand-Alone Mode)
Semiconductor Group2911.97
Page 30
PSB 21911
PSF 21911
IOM®-2 Interface
2.3IOM®-2 Interface
The IOM-2 interface is used to interconnect telecommunication ICs. It provides a
symmetrical full-dupl ex communicat ion link, containing us er dat a, control/program ming
and status channels. The structure used follows the 2B + 1D-channel structure of ISDN.
The ISDN user data rate of 144 kbit/s (B1 + B2 + D) is transmitted in both directions over
the interface.
The IOM-2 interface is a generalization and enhancement of the IOM-1 interface.
®
2.3.1IOM
The IOM-2 interface comprises two clock lines for synchronization and two data lines.
Data is carried over Data Upstream (DU) and Data Downstream (DD) signals. The
downstream and upstream di rection are always defined with respect to the exchan ge.
Downstream refers to information flow from the exchange to the subscriber and
upstream vice versa respectively. The IOM-2 Interface Specification describes open
drain data lines with external pull-up resistors. However, if operation is logically point-topoint, tristate operation is possible as well.
-2 Frame Structure
The data is clocked by a Data Clock (DCL) that operates at twice the data rate. Frames
are delimited by an 8-kHz Frame Synchronization Clock (FSC). Incoming data is
sampled on every second falling edge of the DCL cl ock.
Figure 11 IOM®-2 Clocks and Data Lines
Within one FSC period 32 bit or 96 bit are transmitted, corresponding to DCL
frequencies of 512 kHz or 1.536 MHz.
Two optimized IOM-2 timing modes exist:
– NT mode for NT1 applications
– TE mode for terminal and intelligent NT applications
Semiconductor Group3011.97
Page 31
PSB 21911
PSF 21911
IOM®-2 Interface
NT or TE mode is selected via pins MS0-2 in stand-alone m ode and via bits M S0-2 in
µP mode. Both the NT and TE mode utilize the same basic frame and clocking structure,
but differ in the number and usage of the individual channels.
®
Figure 12 Basic Channel Structure of IOM
Each frame consists of
-2
• two 64 kbit/s channels B1 and B2
• the Monitor channel for transferring maintenance informa tion
• two bits for the 16 kbit/s D-channel
• four command/indication (C/I) bits for controlling of layer-1 functions (U- and Stransceiver).
• two bits MR and MX for the handshake procedure in the Monitor channel
2.3.1.1TE Mode Frame Structure
In TE mode the IEC- Q T E provide s a data clock DCL with a frequency of 1536 kHz. As
a consequence the IOM-2 interface provides three channels each with a nominal data
rate of 256 kbit/s.
• Channel 0 contains 144 kbit/s (for 2B+D) plus Monitor and Command/Indication
channels for the layer-1 device.
• Channel 1 contains two 64-kbit/s intercommunication channels plus Monitor and
Command/Indication channels for other IOM-2 devices.
• Channel 2 is used for IOM bus arbitration (access to the TIC bus). Only the Command/
Indication bits are used in channel 2.
Figure 13 Definition of the IOM®-2 Frame in TE Mode
C/I0 in IOM
–
®
-2 Channel 0:
DU / DDDDC/I4C/I3C/I2C/I1MRMX
D:
C/I:
two bits for the 16 kbit/s D-channel
The four command/indication (C/I) bits are used for control of the U-
transceiver (activation/deactivation and additional control
functions).
MR, MX:
C/I1 in IOM
–
®
two bits MR and MX for the handshake in the Monitor channel 0
-2 Channel 1:
DU / DDC/I6C/I5C/I4C/I3C/I2C/I1MRMX
C/I1 to C/I6
are used for control of a transceiver or an other device in IOM-2
channel 1 (activation/deactivation and additional control functions).
MR, MX:
Semiconductor Group3211.97
two bits MR and MX for handshake in the Monitor channel 1
Page 33
PSB 21911
PSF 21911
IOM®-2 Interface
C/I2 in IOM
–
®
-2 Channel 2:
DU11BACTBA2TBA1TBA011
DDEES/GA/B1111
E:
BAC-bit
S/G-bit
D-echo bits
(Bus ACcessed). When the TIC bus is occupied the BAC-bit is low.
(Stop/Go), available to a connected HDLC controller to determine if
it can access the D-channel (S/G = 1: stop, S/G = 0: go).
A/B-bit
(available/blocked), supplementary bit for D-channel control.
(A/B = 1: D-channel available, A/B = 0: D-channel blocked).
TBA0-2:
TIC Bus Address
2.3.1.2NT Mode Structure
In NT mode the IEC-Q TE provides a data clock DCL with a frequency of 512 kHz. As a
consequence the IOM-2 interface provides only one channel with a nominal data rate of
256 kbit/s.
• Channel 0 contains 144 kbit/s (for 2B+D) plus Monitor and Command/Indication
channels.
Figure 14 Definition of the IOM®-2 Frame in NT Mode
Semiconductor Group3311.97
Page 34
PSB 21911
PSF 21911
IOM®-2 Interface
2.3.2IOM®-2 Command / Indication Channels
The Command/Indication chan nels carry real-time control and status information over
the IOM-2 interface.
C/I Channel 0
C/I channel 0 (C/I0) is available in both operational modes (NT and TE mode). The
channel consists of four bits in each direction. Activation and deactivation of the Utransceiver is always controlled via the C/I0 channel. The C/I codes going to the Utransceiver are called “commands”, those originating from it are referred to as
“indications”. The C/I codes of the U-transceiver are listed and explained in chapter 2.5.8
on page 74.
stand-alone mode
In
3PAC, IPAC or ISAR.
the C/I0 channel is controlled by an external device, e.g. the ICC,
µP mode
In
microprocessor interfa ce. For a description on how to access the C/I0 channel via the
µP-interface please refer to chapter 2.6.3 on page 83.
C/I Channel 1
C/I channel 1 (C/I1) is only available in TE mode (DCL = 1.536 MHz). The channel
consists of six bits in each direction.
stand-alone mode
In
µP mode
In
the C/I0 channel ca n either be controlled by an external device or via the
the C/I1 channel is ignored by the U-transceiver.
it can be accessed via registers CIWI/U and CIRI/U (pag e 83).
Semiconductor Group3411.97
Page 35
PSB 21911
PSF 21911
IOM®-2 Interface
2.3.3IOM®-2 Monitor Channel
The Monitor channel protocol is a handshake protocol used for programming and
monitoring devices in Monitor channel "0" or "1". These can include the on-chip Utransceiver of the IEC-Q TE as well as external devices connected to the IOM-2
interface.
The Monitor channel ope rates on an asynchro nous basis. While data tra nsfers on the
bus take place synchronized to frame sync, the flow of data is controlled by a handshake
procedure. For example: data is placed onto the Monitor channel and the MX bit is
activated. This data will be transmitted repeatedly once per 8-kHz frame until the transfer
is acknowledged via the MR bit.
Monitor Channel 0
Monitor channel 0 is available in both operational modes (NT and TE mod e). The Utransceiver is always controlled and monitored via Monitor channel 0. The Monitor
channel commands and indications of the U-trans ceiver are listed and explained on
page 5 1-61.
stand-alone mode
In
ICC, 3PAC, IPAC or ISAR.
µP mode
In
microprocessor interface. For a descr ipti on on how t o access the Monitor 0 channel via
the µP-interface please refer to chapter 2.6.4 on page 84.
Monitor Channel 1
Monitor channel 1 is only available in TE mode (DCL = 1.536 MHz). The channel
consists of six bits in each direction.
stand-alone mode
In
µP mode
In
external device (e.g. ARCOFI).
the Monitor channel can either be controlled by an external device or via the
it can be accessed via the microprocessor interface (page 83) to control an
the Monitor channel is controlled by an external device, e.g . the
the Monitor 1 channel is ignored by the U-transceiver.
Semiconductor Group3511.97
Page 36
PSB 21911
PSF 21911
IOM®-2 Interface
Monitor Procedure ’Timeout’
The U-transceiver offers an automatic reset (Monitor procedure “Timeout”) for the
Monitor routine. This reset function transfers the Monitor channel into the idle state (MR
and MX set to high) by issuing “EOM” (End of Message) after a timer has elapse d. As
an effect, unacknowledged Monitor messages sent by the U-transceiver are taken away
from the Monitor channel.
The U-transceiver checks for unacknowled ged Monitor messages every 5 m s. In case
the timer expires “EOM” will be issued. The U-transceiver does not repeat the message,
hence it will be lost.
In slow applications e.g. testing or evalution platforms this internal reset function may be
disabled by setting
– Pin MTO in stand-alone mode
– Bit ADF2:MTO in µP mode.
If Monitor Timeout is disabled, no restrictions regarding the time for completing a Monitor
transfe r e xi s ts.
Semiconductor Group3611.97
Page 37
PSB 21911
PSF 21911
IOM®-2 Interface
2.3.4Activation/Deactivation of IOM®-2 Clocks
The IOM-2 clocks may be switched off if the U-transceiver is in state ’Deactivated’. This
reduces power consum ption to a minimum . In this deactivat ed state the cl ock lines are
low and the data lines are high. The power-down condition within the ’Deactivated’ state
will only be entered if no Monitor messages are pending on IOM-2.
For information on how to keep the IOM- 2 clocks a ctive in all states please refer to the
application note ’Providing Clocks in Deacti vate d State’ of 09.97.
deactivation procedure
The
is shown in
figure 15
. After detecting the code DI
(Deactivation Indi cation) the U-transce iver responds by transmitting D C (Deactivation
Confirmation) during s ubsequent frames and stops the timing signals after the fourth
frame.
a)
FSC
DIDIDIDIDIDI
DIN
DCDCDCDCDRDR
DOUT
Detail see Fig.b
b)
DCL
R
IOM -2 Interface
deactivated
R
IOM -2 Interface
deactivated
DIN
Figure 15 Deactivation of the IOM
The IOM-2 clocks a re
DC/ΙΙ/CΙ/CΙ/C
®
-2 Clocks
activated
automatically when the DIN line is pul led low or a line
ITD10292
activation is detected on the U-interface. If a PSB 2186 (ISAC-S TE) or PEB 2070 (I CC)
is connected to the IEC-Q TE via IOM-2, the DIN line of the IEC-Q TE is pulled low by
Semiconductor Group3711.97
Page 38
PSB 21911
PSF 21911
IOM®-2 Interface
setting the SPU bit of the ISAC-S TE or ICC to ’1’. Otherwise, the DU line has to be pulled
to low via an I/O port of the microcontroller
DCL is activated such that its first rising edge occurs with the beginning of the bit
following the C/I0 channel. After the clocks have been enabled this is indicated by the
PU code in the C/I0 channel.
2.3.5Superframe Marker
The start of a new superframe on the U-interface may be indicated with a FSC highphase lasting for one single DCL-period. A FSC high-phase of two DCL-periods is
transmitted for all other IOM-2 frame starts.
The superframe marker is disabled if pin/bit MS2 = 0.
®
2.3.6IOM
-2 Output Driver Selection
In µP mode t he output type o f t he IOM dataline D OUT i s selectable via bit ADF2:DOD.
In stand-alone mode it is configured via pin DOD. Bit/pin DOD set to 0 selects tristate
(reset value) and DOD set to 1 selects open drain outputs.
Ω
In the "open drain" mode pull-up res istors (1 k
–5kΩ) are required on DOUT. FSC
and DCL always are push pull.
Semiconductor Group3811.97
Page 39
PSB 21911
PSF 21911
Microprocessor Interface
2.4Microprocessor Interface
The parallel/serial microprocessor interface can be selected to be either of the
Siemens/Intel non-multiplexed
1.
Motorola
2.
Siemens/Intel multiplexed
3.
signals
Serial mode
4.
The selection is performed via pins ALE/CCLK and SMODE as follows:
The occurrence of an edge on ALE/CCLK, either positive or negative, at any time during
the operation immediately selects interface type 3 or 4. A return to one of the other
interface types is possible only if a hardwar e reset is issued.
2.4.1Micr oproces sor Clock Output
The microprocessor clock is provided in µP mode on the MCLK-output. Four clock rates
are provided by a program mable presc aler. These are 7.68 M Hz, 3.84 MHz, 1.92 MHz
and 0.96 MHz. Switching between the clock rates is realized without spikes. The
oscillator rema ins active all the t ime. The clock is synchr oni zed to the 15.36 M Hz clock
at the XIN pin.
2.4.2Wat chdog Timer
The watchdog is enabled by setting the SWST:WT bit to “1”. The value of SWST:WT
after hardware reset (pin RES
After the microcontroller has enabl ed the watchdog t imer it has to write the bit pa tterns
“10” and “01” in ADF:WTC1 and ADF:WTC2 within a period of 132 ms. If it fails to do so,
a reset signal of 5 ms at pin RST
during this reset.
low and pin TSP low) is "0".
is generated. The cloc k at pin MCLK remains active
Semiconductor Group3911.97
Page 40
PSB 21911
PSF 21911
U-Transceiver
2.5U-Transceiver
The U-interface establishes the direct link between the exchange and the terminal side
over two copper wires. Transmission over the U-interface is performed at a rate of
80 kBaud. The code used is reducing two binary informations to one quaternary symbol
(2B1Q) resulting in a total of 160 kbit/s to be transmitted. 144 kbit/s are user data
(2B + D), 16 kbit/s are used for maintenance and synchronizati on informat ion.
The IEC-Q TE uses two diffe rential outputs (AOUT, BOUT) and two differential inputs
(AIN, BIN) for transmission and recepti on. These differential signals are coupled via a
hybrid and a transformer to the two-wire U-interface.
Figure 16
main blocks:
The System Interface Unit (SIU) provides the connection of the U- and the IOMinterfaces. After scrambling/descram bling and rate adaptation the data channels (2B +
D) are transferred to the appropriate frame. Complete activation and deactivation
procedures are implemented, which are controlled by activation and deactivation
indications from U- or IOM -interfaces. State transition of the procedures depend on the
actual status of the r eceiver (adaptation and synchronization) and timing functions to
watch fault conditions. Two different modes can be selected for maintenance functions:
In the auto-mode all EOC-procedure handli ng and executing as specified by ANSI is
performed. In the transparent mode all bits are transferred t ransparently to the IOM-2
interface without any internal processing.
The Receiver block ( REC) performs the filter algorithmic functions using digital signal
processing techniques. Modules for echo cancellation, pre- and post-equalization,
phase adaptation and frame detection are implemented in a modu lar multi-processor
concept.
shows a block diagram of the U-transceiver which can be subdivided in three
SIUSystem Interface Unit
RECReceiver
LIULine Interface Unit
The Line Interface Unit (LIU) contains the crystal oscillator and all of the analog
functions, such as the A/D-converter and the awake uni t i n the receive path, t he pulseshaping D/A-converter, and the line driver in the transmit path.
Note: Due to the integrated microprocessor interfa ce the IEC-Q TE V5.2 ha s a few µs
more delay from IOM-2 to the U-interface than the IEC-Q V4.4. This may be
relevant in very d elay sensitive appplications like R adio in the Loop (RITL) and
Wireless PBXs.
Semiconductor Group4011.97
Page 41
PSB 21911
PSF 21911
U-Transceiver
Hybrid
D
A
Awake
D
A
EOC
Adapter
Timing
Control
EC
Recovery
CRC
+
AGC
EQ
Adapter
SIURECLIU
R
IOM -2
Framing
Interface
ITB10152
FSC
DCL
DIN
DOUT
Figure 16 U-Transceiver Bloc k Diagram
Semiconductor Group4111.97
Page 42
PSB 21911
PSF 21911
U-Transceiver
2.5.1U-Frame Str ucture
Each basic frame consists of 18 bits for the (inverted) synchronization word; 6 overheads
bits are allocated for system functions, an d 216 bits transfer th e userdata of 2B + D channel (i.e. userdata of 12 IOM-frames is packed into one basic U-frame ).
Data is grouped together into U-superframes of 12 ms. The beginning of a new
superframe is marked w ith an inverted synchronization word (ISW). Each su perframe
consists of eight basic frames (1.5 ms) which begin with a standard synchronization word
(SW) and contain 222 bits of information (
i= information (data / message)
– ACTActivation bitACT = (1) –> Layer 2 ready for communication
– DEADeacti vation bitDEA= (0) –> LT informs NT that it will turn off
– CSOCold Start OnlyCSO= (1) –> NT-ac tivation with cold start only
– UOAU-Only ActivationUOA= (0) –> U-only activated
– SAIS-Activity IndicatorSAI= (0) –> S-interface is deactivated
– FEBE Far-end Block ErrorFEBE = (0) –> Far-end block error occurred
– PS1Power Status Primary SourcePS1= (1) –> Primary power supply ok
– PS2Power Status Secondary SourcePS2= (1) –> Secondary power supply ok
– NTMNT-Test ModeNTM= (0) –> NT busy in test mode
– AIBAlarm Indication BitAIB= (0) –> Interruption (according to ANSI)
– NIBNetwork Indication BitNIB= (1) –> no function
(reserved for network use)
CRC9CRC10
Semiconductor Group4311.97
Page 44
PSB 21911
PSF 21911
U-Transceiver
2.5.1.1Cyclic Redundancy Check
The cyclic redundancy check pro vides a po ssibility to verify the cor rect tran smission of
data. The checksum of a transmitted U-superfra me is calcul ated from the bits in the Dchannel, both B-channels, and the M4 bits according to the CRC polynominal
12
G (u) = u
The check digits (CRC bits CRC1, CRC2, …, CRC12) generated are transmitted at
position M5 and M6 in the U-superframe. At the rec eiving side this value is compared
with the value calculated from the received superframe.
In case these values are not identical a CRC-error will be indicated to bo th sides of the
U-interface. It is indica ted as a NEBE ( Near-end Block Error) on the side where the er ror
is detected and as a FEBE ( Far-end Block Error) on the remote side. The FEBE-bit will
be placed in the next available U-superframe transmitted to the originato r.
+ u11 + u3 + u2 + u + 1
(+ modulo 2 addition)
Far-end or near-end error indicati ons i ncr eme nt the corresponding block error counters
of exchange and terminal side. The IEC-Q TE addi tionally issues a MON-1 message
every time a NEBE or FEBE has occurred (chapter 2.5.3, page 54). The block error
counters can be read via MON-8 commands (refer to chapter 2.5.5, page 59).
It is not possible to directly access the CRC-checksum itself. Hence the user cannot read
or write the checksum values.
Figure 17
Due to the scrambling algorithm described hereafter, a wrong bit decision in the receiver
automatically leads to at least three bit errors. Whether all of these are recorded by a bit
error counter depends on whether all faulty bits are part of the monitored channels
(2B+D, M4) or not.
illustrates the CRC-process.
Semiconductor Group4411.97
Page 45
PSB 21911
PSF 21911
U-Transceiver
R
IOM -2
DD
(MON-1)
NEBE
(MON-8)
DU
NT
(2B + D), M4
G(u)
CRC 1...CRC12
No
=?
Yes
NEBE
Error
Counter
SFR(n)
SFR(n+1)
SFR(n+1.0625)
FEBE= "1"
SFR(4n+1.0625)
FEBE="0"
SFR(n+0.0625)
LTU
G(u)
12CRC...1CRC
FEBE
Error
Counter
(2B + D), M4
R
IOM -2
DD
(MON-8)
DU
(MON-8)
(MON-1)
FEBE
G(u)
G(u)
12CRC...1CRC CRC 1...CRC12
SFR(n+1.0625)
=?
FEBE
Error
Counter
SFR(n+2)
FEBE= "1"
Yes
SFR(n+2)
FEBE= "0"
NEBE
Error
Counter
IEC-Q TEDFE-Q or
IEC-Q
No
(MON-8)
ITD10196
Figure 17 CRC-Proces s
Semiconductor Group4511.97
Page 46
PSB 21911
PSF 21911
U-Transceiver
2.5.1.2Block Error Counters
The U-transceiver provides internal counters for far-end and near-end block errors. This
allows a comfortable surveillance of the transmission quality at the U-interface. In
addition, MON-1 messages indicate the occur rence of near-end errors, far-end errors,
and the simultaneous occurrence of both errors.
A block error is detected ea ch t ime when t he calculated checksum of the recei ved d ata
does not correspond to the control chec ksum transmitted in the f ollowing superframe.
One block error thus indicates that one U-superframe has not been transmitted correctly.
No conclusion with respect to the number of bit errors is therefore possible.
The following two sections describe the operation of near and far-end block error
counters as well as the commands avail able to test them.
Near-End and Far-End Block Error Counter
A near-end block error (NEBE) indicates that the error has been detected in the receive
direction (i.e. NEBE in the NT after an LT --> NT error). This will be indicated with a MON1-message NEBE. Each detected NEBE-error increments the 8-bit NEBE-counter.
When reaching the maximum cou nt, counting is stopped and the cou nter value reads
).
(FF
H
The current value of the NEBE counter is read with the MON-8 command RBEN. The
response comprises two bytes: the first byte always indicates that a MON-8 message is
replied to (80
operation resets the counter to (00
), the second represents the counter value (00H) … (FFH). Each read
H
).
H
A far-end block error identifies errors in transmission direction (i.e. FEBE in the NT = NT
=> LT-error). FEBE errors are processed in the same manner as NEBE errors. A far-end
block error will be indicated with a MON-1 message FEBE. The FEBE counter is read
and reset with the MON-8 command RBEF.
In case both, far-end and near-end block errors occur simultaneously, the MON-1
message FNBE will be issued.
Both counters are also reset in all U-transceiver states except ’Synchr onized’, ’Wait for
Act’, ’Transparent’ and ’Error S/T’.
Semiconductor Group4611.97
Page 47
Testing Block Error Counters
PSB 21911
PSF 21911
U-Transceiver
Figure 18
Transmission errors are simulated with artificially corrupted CRCs. With two commands
the cyclic redundancy checksum can be inverted in the downstream and in the upstream
direction. A third command offers to invert single FEBE-bits.
With EOC command
differences in the functional behavior of the NEBE-counter depending on the EOC mode:
With EOC command
auto mode the IEC-Q TE will react automatically with a permanently inver ted upstream
CRC. In EOC transparent mode this reaction has to be prompted by a MON-8 CCRC
command. Note that MON-8 CCRC is not excecuted if it was not preceeded by the EOC
command RCC.
There are also differences in the functional behavior of the FEBE-counter depending on
the EOC mode:
illustrates how the IEC-Q TE supports testing of the LT’s block error counters.
NCC
the LT notifies the NT of corrupted CRCs. Again, there are
and NEBE-counter disabled
and NEBE-counter enabled
RCC
the LT requests the NT-side to send corrupted CRCs. In EOC
EOC
Auto modeFEBE-detection stopped, no MON-1 FEBE-messages
The EOC command
transparent mode EOC command RTN must be followed by a MON-8 NORM command
to become effective.
With the MON-8-command
command does not provoke permanent FEBE-bit inversion but sets only one FEBE-bit
to (0) per SFB command, it is possible to predict the exact FEBE-counter reading.
RTN
disables all previously sent EOC commands. In EOC
SFB
it is possible to invert single FEBE-bits. Because this
abled
Semiconductor Group4711.97
Page 48
PSB 21911
PSF 21911
U-Transceiver
IOMR-2
(MON-0) NCC
(MON-0) ACK
(MON-1) NEBE
(MON-0) RTN
(MON-0) ACK
(MON-0) RCC
(MON-0) ACK
(MON-8) CCRC
(MON-1) FEBE
EOC
Transparent
ERROR
COUNT
NEBE
ERROR
COUNT
FEBE
EOC
Auto-Mode
STOP
ERROR
DETECT
FREE
ERROR
DETECT
STOP
ERROR
DETECT
EOC:NCC
EOC Acknowledge
Start Inverse
CRC Bits
FEBE = "0"
End Inverse
CRC Bits
EOC:RTN
EOC Acknowledge
EOC:RCC
EOC Acknowledge
Start Inverse
CRC Bits
FEBE ="0"
LTU
ERROR
COUNT
FEBE
ERROR
COUNT
NEBE
R
IOM
-2
(MON-0) NCC
(MON-0) ACK
(MON-8) CCRC
(MON-8) RBEF
(MON-8) NORM
(MON-0) RTN
(MON-0) ACK
(MON-0) RCC
(MON-0) ACK
(MON-8) RBEN
(MON-0) RTN
(MON-0) ACK
:
EOC
RTN
FREE
ERROR
DETECT
IEC-Q TEIEC-Q TEDFE-Q/IEC-Q
EOC Acknowledge
End Inverse
CRC Bits
(MON-0) RTN
ITD10197
Figure 18 Block Error Counter Test
Semiconductor Group4811.97
Page 49
PSB 21911
PSF 21911
U-Transceiver
2.5.1.3Scrambler / Descrambler
The scrambling algorithm as defined by ANSI T1.601 ensures that no sequences of
permanent binary 0s or 1s are transmitted. The algorithms used for scrambling and
descrambling are described in
controlled fully by the IEC-Q TE. Hence, no influence can be taken by the user.
Please refer to page 77 for a descript io n of loop 3.
Transmit Scrambler in normal operation without Loop-Back 3
figure 19
. The scrambling/descrambling process is
Ds
-1
x
-1
x
-1
x
Ds.x
-1
x
-18
Di
Ds = Di
+
-18
x.DsDs.x
-23
+
Transmit Scrambler for Loop-Back 3
Ds
-1
x
-1
x
-1
x
-1
x
-5
.
xDs
Di
Ds = Di
+
.
Ds x
-5
+
.
Ds x
-23
Receive Descrambler in normal operation without Loop-Back 3
Ds
-1
xx
-1
-1
xx
Ds
-1-1
-5
.
x
Do
-5-23
+
+
Do = Ds
x(1
x)
-1
x
-23
x.Ds
-1
x
-23
.
Ds x
x
-23
.
Ds x
Receive Descrambler for Loop-Back 3
Ds
-1
x
-1
xx
-1-1
x
-18
.
xDs
-1
x
-23
.
Ds x
Do
-18
Do = Ds
+
(1)
-23
+
xx
ITD09730
Figure 19 Scrambler / Descrambler Algorithms
Semiconductor Group4911.97
Page 50
PSB 21911
PSF 21911
U-Transceiver
2.5.1.4Embedded Operations Channel (EOC)
EOC-data is inserted into the U-frame at the positions M1, M2 and M3 thereby permitting
the transmission of two complete EOC-messages (2 × 12 bits) within one U-superframe.
The EOC contains an address field, a data/message indicator (d/m) and an eight-bit
information field.
With the
address field
the destination of the transmitted message/data is defined.
Addresses are defined for the NT, 6 repeater stations and broadcasting.
data/message indicator
The
needs to be set to (1) to indicate that the information field
contains a message. If set to (0), numerical data is transferred to the NT. Currently no
numerical data transfer to or from the NT is required.
From the 256 codes possible in the
infor ma tion field
64 are reserved for non-standard
applications, 64 are reserved for inte rnal network use and eight are defined by ANSI/
ETSI for diagnostic and loop-back functions. All remaining 120 free codes ar e availab le
for future standardization.
LBBDClose complete Loop
LB1Close Loop B1
LB2Close Loop B2
RCCRequest Corrupt CRC
NCCNotify of corrup t CRC
RTNReturn to normal
HHold
Page 51
PSB 21911
PSF 21911
U-Transceiver
Table 7EOC-Codes
(cont’d)
EOC
Addressd/mInformation
DirectionMessageFunction
(hex)
a1 a2 a3d/mi1 - i8
1AANT ----> L T
1XXNT ----> L T
UTCU nable to compl y
ACKA ckn ow led ge
2.5.2E OC-Proce ssor and MON-0
An EOC-processor on th e chip is responsibl e for the correct inser tion and ext raction of
EOC-data on the U-interface. The EOC-processor can be programmed to auto mode
(default) or tr anspar ent m ode vi a bit EOCA in the UM O D r egiste r (
table 20
). Access to
the EOC is only possible when a superfram e is transmitted. This is the case in the Utransceiver states ’Synch ronized’, ’Wait for ACT’, ’Transparent’, ’Error S/T’ and ’ Pend.
Deac. U’. In all other states the EOC-bits on the U-interface are clamped to high.
Transparent
Pin AUTO = 0 or
Bit STCR:AUTO = 0
UIOM-2
EOC
IOM-2
MON-0
Auto mode
Pin AUTO = 1 or
Bit STCR:AUTO = 1
EOCprocessor
Excecute
3x
Echo
U
EOC
EOCEOC
MON-0
MON-0
Figure 20 EOC-Processor: Auto Mode, Transparent Mode
The EOC is controlled and monitored via MON-0 commands and messages in the IOM-2
Monitor channel. The st ructure of a MON-0-mess age is shown bel ow. The structure is
identical in EOC auto and transparent mode.
Semiconductor Group5111.97
Page 52
MON-0 Structure
1. Byte2. Byte
0 0 0 0A A A | Xi1 i2 i3 i4i5 i6 i7 i8
MON-0Addr. | d/mEOC Code
Addr:Address– 0 = NT
– 1 … 6 = Repeater
– 7 = Broadcast
d/m:Data/Message– 0 = Data
– 1 = Message
PSB 21911
PSF 21911
U-Transceiver
i1-i8:EOC Code– 00 … FF
= coded EOC command/indication
H
EOC Auto Mode
Acknowledgment:
immediately without triple-last-look. If an address other than (000
a HOLD message with address 000
All received EOC-frames are echoed back to the exchange
) or (111B) is received,
B
is returned. However, there is an exception: The
B
IEC-Q TE will send a ’UTC’ after three consecutive receptions of d/m = (0) or after an
undefined command.
Latching:
All detected EOC- commands a re latche d, i.e. they are val id as long as t hey
are not disabled with the EOC ’RTN’ command or a deactivation.
Transfer to IOM:
With the triple-last-look criteria fulfilled the new EOC-command will be
passed to IOM-2 with one single MON-0 -message, independent ly of the address used
and the status of the d/m indicator. MON-0-commands from IOM will be ignored.
Execution:
The EOC-commands listed in
PSB 21911 if they were addressed correctl y (000
table 8
will be executed automatically by the
or 111B) and the d/m bit was set to
B
message (1). The execution of a comm and is performed only after the “ triple-la st-lo ok”
criterion is met.
Semiconductor Group5211.97
Page 53
Table 8Executed EOC Commands in Auto Mode
PSB 21911
PSF 21911
U-Transceiver
EOC-code
i1 - i8 (Hex)
DirectionFunction
DU
50LBBD
51LB1
52LB2
53RCC
Close complete loop-back (B1, B2, D).
The U-transceiver
does not close the complete loop-back imm edi ate ly after
receipt of this code. Instead it issues the C/I-command AIL
(in “Transparent” state) or ARL in the states “Error S/T” and
“Synchronized”. This allows the downstream device to
close the loop-back if desired (e.g. S-transceiver or
microc ontroller).
Closes B1 loop-back
in NT. All B1-channel data will be
looped back within the U-transceiver.
Closes B2 loop-back
in NT. All B2-channel data will be
looped back within the U-transceiver.
Request corrupt CRC.
Upon receipt the IEC-Q TE
transmits corrupted (= inverted) CRCs upstream. This
allows to test the near end block error counter on the LTside. The far end block error counter at the NT-side is
disabled and NT-error indications (MON-1) are suppressed.
54NCC
Notify of c orrupt CRC.
Upon receipt of NCC the near end
block error counter is disabled and error indications are
suppressed. This prevents wrong error counts while
corrupted CRCs are sent.
FFRTN
Return t o normal.
With this command all previously sent
EOC-commands will be released. The EOC-process or is
reset to its initial state (FF
).
H
EOC Transparent Mode
In transparent mode no acknowledgment, no triple-last-look and no execution of the
received commands is performed. The received EOC-frame is transmitted directly
downstream via a MO N-0-message. Thus, a M ON-0-message is issu ed on IOM every
6 m s. Ac knowl edgm ent and execution of the received comm and have to be initiated by
the microcontroller. The microcontroller can execute all defined test functions (close/
open loops, send corrupted CRCs) in the NT using MON-8-commands.
In the upstream direction the l ast incoming EOC-code fro m the IOM-2-Moni tor channel
is transmitted to the LT.
Semiconductor Group5311.97
Page 54
PSB 21911
PSF 21911
U-Transceiver
2.5.3Maintenance (MON-1)
This category compr ises c ommands a nd mess ages relating to m aintenance bits of the
U-interface and the self-test according to ANSI T1.601. The co mmands and mes sages
may be mapped to the S/Q channel of the S/ T-interface via the microprocessor. Th is
provides a metho d to exchange U- interface re lated information be tween a terminal on
the S-bus and the NT. Thus, the terminal can be informed about transmission errors that
occurred on the U-interface (NEBE, FEBE, FNBE) or reque st the NT to p erform a selftest (ST).
MON-1 messages ar e two bytes long. The first nibble of the second byte contains S/Qindications, the second nibble contains maintenance bit related commands. The
operation of MON-1-messages is identical in auto- and transpare nt mode.
The following tables give an overview of indications available in the MON-1 category.
The following indications and maintenance bits are defined in MON- 1-messages.
Table 9Mon-1 S/Q-Channel Commands and Indications
S/QD irectionFunction
SSSS (Bin)DUS/Q-Channel
0 0 0 1ST
= coded S/Q-command indication
H
Self-test request
terminal to inquire whether layer 1 is present. No test
is performed within the U-transceiver. Upon reception
the U-transceiver replies with MON-1 “STP”.
. This command is issued by the
0 0 1 0STP
0 1 0 0 FEBE
Semiconductor Group5411.97
Self-test pass
transceiver has received the command “ST” correctly.
Far-end block error
U-interface it is indicated to the NT that transmission
errors occurred in the direction NT –> LT or NT –> LTRP.
. Indicates to the terminal that the U-
. Via the FEBE bit set to (0) on the
Page 55
PSB 21911
PSF 21911
U-Transceiver
Table 9Mon-1 S/Q-Channel Commands and Indications
S/QD irectionFunction
NTM-bit of the U-interface is set active (= 0) in order to
inform the exchange that the NT is involved in testing
and not available for transparent transmission. This
message needs to be sent out by the downstream
device if the terminal requests tests which prevent
transparent transmission (e.g. loop s B1, B2, D).
. Return to normal (idle) state. This command
. After reception of this command the
. Transmission errors occurred
(cont’d)
. Transmission errors
1 1 1 1NORM
Normal
taken.
sets back the NTM-bit to 1. No other action
Semiconductor Group5511.97
Page 56
PSB 21911
PSF 21911
U-Transceiver
2.5.4Overhead Bits (MON-2)
MON-2-indications are used to transfer all overhead bits (M4, M5, M6) except those
representing EOC- and C R C-bits. Star ting wi th the AC T-b it, the or der is id entical to the
position of the bits at the U-interface.
Table 11MON-2 Structure
1. Byte2. Byte
0 0 1 0D11 D10 D9 D8D7 D6 D5 D4 D3 D2 D1 D0
MON-2Overhead BitsOverhead Bits
D0 … 11: Overhead bits
These bit positions in the MON-2-message correspond to the following overhead bits:
– PS1 (Power Prim. Source) = (1) –> PS1 = (1) –> Prim. supply ok
– PS2 (Power Sec. Source)= (1) –> PS2 = (1) –> Sec. supply ok
Control via MON-2
– only the undefined bits marked with binary ’1’
– SAI (S Activity Indicator)= (0) –> S-interface is deactivated;
can be controlled via MON-2 after MON-8 ’PACE’.
Control via other MON-Commands
– NTM (NT-Test Mode)= (0) –> NT busy in test mode (MON-1)
– FEBE (Far-end Block Error);MON-8 message ’SFB’ sets a single FEBE bit to ’0’
For details about the meaning of the overhead bits please refer to ETSI ETR 080 and
ANSI T1.601.
Overhead Bits Upstream Transmission
– The upstream overhead bits ar e controlled by me ans of the U-transceiver due to its
state, pins, MON-2 commands and other MON commands.
– Only the undefined bits market with binary “1” may be controlled directly by a MON-2-
message.
– All overhead bits are set to binary “1” when leaving a power-down state.
Overhead Bits Downstream Reception
– In the receive direction, the overhead bits of the last two U-interface superframes are
compared and a MON-2-message defining all 12 bits is issued if a difference between
both was found on at least one single bit other than the “FEBE” bit. Therefore, a MON2-message is sent not more often than once per superframe (12 ms interval).
Semiconductor Group5711.97
Page 58
PSB 21911
PSF 21911
U-Transceiver
– In order to notify the controller of the initial system status, one MON-2-m essage is
issued immediately after reaching the “Synchronized” state in NTmode.
– The U-transceiver will not issue MON-2- messages w hile CRC-viola tions are detec ted.
Because the CRC-check sum is transmitted one superframe later, a maximum of one
corrupted MON-2-indication can be issued. In this case a MON-2-message indicating
the correct system s tatu s will be issu e d af te r th e transmitte d CRC-checksum is a g ai n
identical to the calculated checksum.
Semiconductor Group5811.97
Page 59
PSB 21911
PSF 21911
U-Transceiver
2.5.5Local Functions (MON-8)
Local functions are controlled via MON-8-commands. The following tables give an
overview of structure and features of commands bel onging to this category.
Format of MON-8-Messages
1. Byte2. Byte
1 0 0 0r | 0 0 0D7 D6 D5 D4 D3 D2 D1 D0
MON-8Register | Addr.Local Command (Message/Data)
r:Register address – 0 = local function register
– 1 = internal register
D0…7 Local command – 00 … FF
– 00 … FF
= local function code
H
= internal register address
H
The following loca l com ma nds are defined. If a response is expected, i t w ill c omprise 4
bytes (2 messages a 2 bytes) if the value of an intern al coefficient is returned, and 2
bytes in all other cases.
In a two-byte response the first byte will indicate that a MON-8 answer is transmitted, the
second byte contains the requested information. This procedure is repeated for a fourbyte transfer (MON-8, Info 1, MON-8, Info 2).
Table 13Mon-8 Local Function Commands
rCodeDirectionFunction
D7-D0 (Bin)DULocal Comm ands
01011 1110PACE
Partial Activation Control External
PACE-command issued, the U-transceiver will
ignore the actual status of the received UOA-bit
and behave as if the UOA-bit was set to (1).
After issuing PACE the UOA/SAI-bits can be
controlled by MON-2-co mmands.
01011 1111PACA
Partial Activation Control Automatic
enables the device to interpret the UOA-bit and
control the SAI-bit automatically. Partial
activation and deactivation is therefore possible.
The U-transceiver is automatically reset into this
mode in the states “Test”, “Receive Reset” and
“Tear Down”.
. With th e
. PACA
Semiconductor Group5911.97
Page 60
PSB 21911
PSF 21911
U-Transceiver
Table 13Mon-8 Local Function Commands
rCodeDirectionFunction
D7-D0 (Bin)DULocal Commands
01111 0000CCRC
01111 0100LB1
01111 0010LB2
Corrup t CRC
recognized if the device is set to EOC
transparent mode. The microcontroller sho uld
issue the command in case the MON-0command RCC was received before. CCRC
then causes corrupt CRCs to be transmitted
upstream.
Loop-back B1
recognized in EOC transparent mode. The
microcontroller should issue the command in
case the MON-0-command LB1 was received
before. LB1 loops back the B1 channel. The
loop is closed near the IOM-2 interface.
Loop-back B2
recognized in EOC transparent mode. The
microcontroller should issue the command in
case the MON-0-command LB2 was received
before. LB2 loops back the B2 channel. The
loop is closed near the IOM-2 interface.
(cont’d)
. This command is only
. The command is only
. The command is only
01111 0001LBBD
01111 1111NORM
01111 1011RBEN
Loop-back B1 + B2 + D
in the EOC transparent and EOC auto-mode.
LBBD loops back both B-channels and the Dchannel. The loops are closed near the IOM-2
interface. In transparent mode the loop is closed
unconditionally. In auto-mode the loop is closed
only if LBBD was received in the EOC-channel
before.
Return to Normal
resets the device into its default mode, i.e. loops
are resolved and corrupted CRCs are stopped.
It is only used in transparent mode.
Read Near-End Block Error Counter
value of the near-end block error counter is
returned and the counter is reset to zero. The
maximum value is FF
. The NORM-command
. The command is used
. The
.
H
Semiconductor Group6011.97
Page 61
PSB 21911
PSF 21911
U-Transceiver
Table 13Mon-8 Local Function Commands
rCodeDirectionFunction
D7-D0 (Bin)DULocal Comm ands
01111 1010RBEF
Read Far-End Block Error Counter
of the far-end block error counter is returned and
the counter is reset to zero. The maximum value
.
H
0r r r r r r r rABEC
is FF
Answer Block Error Counter
requested block error counter is returned (8 bit).
00000 0000RID
Read Identification
identification.
0r r r r r r r rAID
01111 1001SFB
1cccc ccccRCOEF
DCOEF
11bbbb bbbb
bbbb bbbb
Answer identification
Set FEBE Bit
Read Coeffiecient
Data Coefficients
Data bits D0 … D 7, 1. byte
Data bits D8 … D15, 2. byte
(cont’d)
to 0
. The value of the
. Request for device
. Reply to a n RID is ’03
, 2 bytes.
. The value
H
’.
Notes:
b … binternal coefficient value
c … cinternal coefficient address
r … rresult from block error counter
Semiconductor Group6111.97
Page 62
PSB 21911
PSF 21911
U-Transceiver
2.5.6State Machine Notation Rules
The state machine in cludes all information necessary for the user to understand and
predict the activation/deactivation status of the U-transceiver. The information contained
in a state bubble is:
State
name,
indication
U-signal
transmitted,
Single Bits
transmitted on the C/I-channel,
Signal Transmitted
to U-Interface
(general)
State Name
Indication Transmitted on C/I-Channel
(Overhead bits) transmitted,
Transition criteria
IN
Single Bit
Transmitted
to U-Interface
(DD)
ITD04257
OUT
and
Timers
C/I-
.
Figure 21 State Diagram Notation U-Transceiver
The following example explains the use of the state diagram by an extract of the NTstate diagram. The state explained is the “EC-Training” state.
Example:
The state may be entered by either of two methods:
–from state “Alerting” after time T11 has expired.
–from state “EC-Training 1” after the C/I command “DI” has been received.
The following informantion is transmitted:
–SN1 is sent on the U-interface.
–No overhead bits are sent
–C/I message “DC” is issued on the IOM-2 interface.
The state is be left at occurrence of one of the following events:
–Leave for state “EQ-Training” after LSEC has been detected.
–Leave for state “EQ-Training” after timer T12 has expired.
Semiconductor Group6211.97
Page 63
PSB 21911
PSF 21911
U-Transceiver
Combinations of transition crit eria ar e possibl e. Logi cal “AN D” is ind icated b y “&” (TN &
DC), logical “OR” is written “or” and for a negation “/” is used. The start of a timer is
indicated with “TxS” (“x” being equivalent to the timer number). Timers are always
started when entering the new state. The act ion resulting after a timer h as expired is
indicated by the path labelled “TxE”.
2.5.7State Machine
This chapter describes the activation and deactivation behavior of the IEC-Q TE. It
applies for both NT and TE mode.
2.5.7.1Cold and Warm Starts
Two types of start-up procedures are supported by the U-transceiver : cold starts and
warm starts.
Cold starts are performed after a reset and require all echo and equalizer coefficients to
be recalculated. This procedure typically is completed after 1-7 seconds depe nding on
the line characteristics. Cold starts are recommended for activations where the line
characteristics have cha nged consi der abl y sin ce the last dea ctivati on.
A warm start procedure uses the coef ficient set saved during the last deactivati on. It is
therefore completed much faster (maximum 300 ms). Warm starts are however
restricted to activations where the line characteristics do not change significantly
between two activations.
Regarding the path in the transition diagram, cold starts have in particular that the Utransceiver has entere d the state ’Test’ (e.g. due to a rese t) prior to an activation. The
activation procedure itself is then identical in both cases. Therefore, the following
sections apply to both warm and cold starts .
Semiconductor Group6311.97
Page 64
2.5.7.2State Diagram
PSB 21911
PSF 21911
U-Transceiver
T14 S
Any State
Pin-SSP or
Pin-RES or
SSP or RES
T7E
& DI
SN0
Pending Timing
SN0/SP
ARL
1SN
EC-Training AL
3SN
Wait for SF AL
SN3T
Analog Loop Back
Any State
Pin-DT or DT
SN0
Pend. Receive Res.
LSU or (/LOF
& T13E)
0SN
Receive Reset
DC
T14 S
DI
Test
DR
T12S
DC
LSEC or T12E
act = 0
DC
BBD1 & SFD
act = 0
AR
EI1
T7S
DR
.
.
.
T14 E
T14 S
TL
DI
T1S,
T11S
LSUE
or T1E
T1E
LOF
LOF
LOF
EI1
LOF
EI1
act = 0
0SN
Deactivated
SN0
IOM
TN
DC
T11E
SN1
EC-Training
SN0
EQ-Training
SN2
Wait for SF
SN3/SN3T
Synchronized 1
SN3/SN3T
Synchronized 2
SN3/SN3T
Wait for Act
SN3T
Transparent
SN3T
.
T13S
T7S
.
TL
.
DC
TIM or DIN = 0
.
R
Awaked
PU
AR or TL
.
Alerting
PU
T12S
.
DC
LSEC or T12E
.
DC
BBD0 & FD
.
DC
BBD0 & SFD
act = 0
DC
uoa = 1
act = 0
AR/ARL
AI
act = 1
AR/ARL
act = 1act = 0
act = 1
AI/AIL
act = 1 & AI
act = 0
Error S/T
AR/ARL
LOF
AR or TL
T1S,
T11S
dea = 0
LSUE
uoa = 0
dea = 0
LSUE
uoa = 0
dea = 0
LSUE
uoa = 0
dea = 0
LSUE
dea = 0
uoa = 0
LSUE
DI
LOF
DI
DI
TN
1
Alerting
DR
T11E
T12S
1SN
EC-Training 1
DR
(LSEC or T12E)
& DI
act = 1/0
3SN
Pend. Deact. S/T
DR
LSUE
SN3T
Pend. Deact. U
dea = 0
uoa = 1
?
dea = 1
act = 1
DC
LSU
.
.
Yes
T1S,
T11S
No
ITD09705
Figure 22 State Transition Diagram
Semiconductor Group6411.97
Page 65
2.5.7.3Inputs to the U-Transceiver
C/I-Commands
AIActivati on In dic a tion
The S-transceiver issues this indication to announce that the S-receiver is
synchronized. The U-transceiver informs the LT side by setting the “ACT” bit to
“1”.
ARActivation Request
INFO1 has been received b y the S-transceiver or the Intelligent NT wan ts to
activate the U-interface. The U-transceiver is requested to start the activation
process by sending the wake-up signal TN.
ARLActivation Request Local Loop-back
The U-transceiver is requested to operate an analog loop-back (close to the Uinterface) and to begin the start-up sequence by sending SN1 (without starting
timer T1). This command may be issued only after the U-transceiver has been
reset by making use of the C/I-channel code RES or a hardware reset. This
assures that the EC- and EQ-coefficient updating algorithms converge
correctly. The ARL-command has to be i ssued continuously as long as the
loop-back is required.
PSB 21911
PSF 21911
U-Transceiver
DIDe activati on Indicati on
This indication is used during a deactivation procedure to inform the Utransceiver that timing signals are needed no longer and that the U-transceiver
may enter the deactivated (power-down) state. Th e DI-indication has to be
issued until the U-transceiver has answered with the DC-code.
DIN = 0Binary “0” polarity on DIN
This asynchronous signal r equests the U- transceiver to pr ovide IO M clocks.
Hereafter, binary “0s” in the C/I-channel (code TIM “0000” or any other code
different from DI “1111”) keep the IOM-2 interface active.
DTData Through
This unconditional command is use d for tes t purposes only and forces the Utransceiver into a state equivalent to the “Transparent” state. The far-end
transceiver is assumed to be in the same condition.
EI1Error Indication 1
The S-transceiver indicates an error condition on its receive side (loss of frame
alignment or loss of incoming signal). The U-transceiver informs the LT-side
by setting the ACT-bit to “0” thus indicating that transparency has been lost.
RESReset
Unconditional command which res ets the whole chip; especi ally the EC- and
EQ-coefficients are set to zero.
Semiconductor Group6511.97
Page 66
SSPSend Single Pulses
Unconditional com mand whi ch requests the tr ansmission of single pul ses on
the U-interface. The pulses are issued at 1.5 ms intervals and have a duration
of 12.5 µs.
The chip is in the “Test” state, the receiver will not be reset.
TIMTiming
In the NT-mode the U-transcei ver is requested to continue pr oviding timing
signals and not to leave the “Power-up” state.
Pins
Pin-ResPin-Reset
Corresponds to a low-level at pin RES
pin is the same as of the C/I-code RES. C/I-message DR w ill be issued. The
duration of the reset pulse must be longer than 10 ns.
PSB 21911
PSF 21911
U-Transceiver
or a power-on reset. The function of this
Pin-SSP Pin-Send Single Pulses
Corresponds to a high-level at pin TSP in stand-alon e mode. The funct ion of
this pin is the same as o f the C/I-code SSP. C/I-message DR will be issued.
The high-level must be applied continuously for single pulses.
Pin-DTPin-Data Thro ugh
This function is activated when both pins RES
alone mode (RES
of the C/I-code DT. C/I-message DR will be issued.
U-Interface Events
The signals SLx and TL received on the U-interface are defined in
ACTACT-bit received from LT-side.
–ACT = 1 requests the U-transceiver to transmit transparently in both
directions. As transparency in receive direction (U-interface to IOM) is
already performed when the receiver is synchronized, the receipt of
ACT = 1 establishes transparency in transmit direction (IOM to Uinterface), too. In the case of loop-backs, howe ver, transparency in both
directions of transmission is established when the receiver is
synchronized.
= ’0’ and TSP = ’1’). The function of this pin is the same as
and TSP are active in stand-
table 22
page 100
on
–AC T = 0 indicates t hat the LT-side has l ost trans pare n cy.
Semiconductor Group6611.97
Page 67
DEADEA-bit received from the LT-side
–DEA = 0 informs the U-transceiver that a deactivation procedure has been
started by the LT-side.
–DEA = 1 reflects the case when DEA = 0 was detected by faults due to e.g.
transmission errors and allows the U-transceiver to recover from this
situation (see state ’Pend. Deact. U’).
UOAUOA-bit received from network side
–UOA = 0 informs the U-transceiver that only the U-interface is to be
activated. The S/T-interface must remain deactivated.
–UOA = 1 enables the S/T-interface to activat e.
LOFLoss of Framing on the U-interface
This condition is fulfilled if framing is lost for 576 ms. 576 ms are the upper
limit. If the correlation between synchronization word and the input signal is not
optimal, LOF may be issued earlier.
PSB 21911
PSF 21911
U-Transceiver
LSECLoss of Signal level behind the Echo Canceler
Internal signal which indicates that the echo cancel er has converged.
LSULoss of Signal level on the U-interface
This signal indicates that a loss of signal level for a duration of 3 ms has been
detected on the U-int erface. This short respo nse time is relevant in all cases
where the NT waits for a response (no signal level) from the LT-side, i.e. after
a deactivation has been announced (receipt of DEA = 0), after the NT has lost
framing, and after timer T1 has elapsed.
LSUELoss of Signal level on the U-interface
After a loss of signal has been noticed, a 588 ms timer is started. When it has
elapsed, the LSUE-criterion is fulfilled. This long response time (see also LSU)
is valid in all cases where the NT is not prepared to lose signal level i.e. the LT
has stopped transmission because of loss of framing, an unsuccessful
activation, or the transmission line is interrupted.
Note that 588 ms represe nt a minim um value; the actual loss o f signa l might
have occurred earlier, e.g. when a long loop is cut at the NT-side and the echo
coefficients need to be readjusted to the new parameters. Only after the
adjusted coefficients cancel the echo completely, the loss of signal is detected
and the timer can be started (if the long loop is cut at the remote end, the
coefficients are still correct and loss of signal will be detected immediately).
SFDSuperframe (ISW) Detected on U-inte rface
FDFrame (SW) Detected on U-inte rface
Semiconductor Group6711.97
Page 68
TLWake-up signal received from the LT
The U-transceiver is requested to start an activation procedure. The TL-
criterion is fulfilled when 12 consecutive periods of the 10-kHz wake-up tone
were detected.
When in the “Pending Timing” state and automatic activation after reset is
selected (NT-AUTO mode), a recognition of TL is assumed every time the
“Pending Timing” state has been entered from the “Test” state (caused by C/I
code DI). This behavior allows the U-transceiver to initiate one single
activation attempt after having been reseted.
BBD0/1Binary “0s” or “1s” detected in the B- and D-channels
This internal signal indicates that for 6-12 ms, a continuous stream of binary
“0s” or “1s” has been detected. It is used as a criterion that the r eceiver has
acquired frame synchronization and both its EC- and EQ-coefficients have
converged. BBD0 corresponds to the signal SL2 in the case of normal
activation and BBD1 corresponds to the internally received signal SN3 in case
of an analog loop back in the NT-mode.
PSB 21911
PSF 21911
U-Transceiver
Timers
The start of timers is indicated by TxS, the expiry by TxE. The following
which timers are used by the U-transceiver:
Signals and indications are issued on IOM-2 (C/I-indications) and on the U-interface
(predefined U-signals).
C/I Indications
AIActivati on In dic a tion
The U-transceiver has established transparency of transmission in the
direction IOM to U-interface. In an NT1, the S-transceiver is requested to send
INFO4 and to achieve transparency of transmission in the direction IOM to S/
T-interface.
AILActivation Indication Loop-back
The U-transceiver has detected ACT = 1 while loop-back 2 is still established.
In an NT1, the S-transceiver is requested to send INFO4 (if a transparent loop-
back 2 is to be implemented) and to keep loop-back 2 active.
ARActivation Request
The U-receiver has synchronized on the incom ing signal. In an NT1, the S-
transceiver is requested to start the acti vat ion procedure on the S/T-interface
by sending INFO2.
ARLActivation Request Loop-back
The U-transceiver has det ect ed a loop-back 2 command in th e EO C-cha nnel
and has established transparency of transmission in the direction IOM to U-
interface. In an NT1, the S-transceiver is requested to send INFO2 (if a
transparent loop-back 2 is to be implemented) and to operate loop-back 2.
DCDeactivation Co nfirmat io n
Idle code on the IOM-2 interface. The U-transceiver stays in the power-down
mode unless an activation procedure has been star ted from the LT-side. The
U-interface may be activated but the S/T-interface has to remain deactivated.
DRDeactivati on Re quest
The U-transceiver has detected a deactivation request command from the LT-
side for a complete deactivation or a S/T only deactivation. In an NT1, the S-
transceiver is requested to start the deactivation procedure on the S/T-
interface by sending INFO0.
EI1Error Indication 1
The U-transceiver has entered a failure condition caused by loss of framing on
the U-interface or expiry of timer T1.
Semiconductor Group6911.97
Page 70
INTInterrupt (Stand-alone mode only)
A level change on input pin INT triggers the transmission of this C/I code in four
successive IOM-2 frames. Please refer to page 96 for details.
PUPower Up
The U-transceiver provides IOM-2 clocks.
Signals on U-Interface
PSB 21911
PSF 21911
U-Transceiver
The signals SNx, TN and SP transmitted on the U -int erface a re d efined in
page 100
The polarity of the transmitted ACT-bit is as follows:
The polarity of the issued SAI-bit depends on the received C/I-channel code: DI and TIM
leads to SAI = 0, any other C/I-code sets the SAI-bit to 1 indicating any activ ity on the
S/T-interface.
2.5.7.5States
The following states are used:
Alerting
The wake-up signal TN is transmitted for a period of T11 either in response to a received
wake-up signal TL or to start an activation procedure on the LT-side.
Alerting 1
“Alerting 1” state is entered when a wake -up tone wa s received in t he “Recei ve Reset”
state and the deactivation procedure on the NT-side was not yet finished. The
transmission of wake-up tone TN is started.
.
a = 0/1 corresponds to ACT-bit set to binary “0/1”
Table 22 on
Analog Loop-Back
Upon detection of bi nar y “ 1s” fo r a per io d of 6–12 ms and of the super f rame i ndi cati on,
the “Analog loop-back” state is entered and transparency is achieved i n both directions
of transmission. This st ate can be left by making use of any unconditional command.
Only the C/I-channel code RES should be used, however. This assures that the EC- and
EQ-coefficients are set to zero and that for a subsequent normal activation procedure
the receiver updating algorithms converge correctly.
Deactivated
The ’Deactivated’ state is a power-down state. If there are no pending Monitor channel
messages from the U-transceiver, i.e. all Monitor channel messages have been
Semiconductor Group7011.97
Page 71
PSB 21911
PSF 21911
U-Transceiver
acknowledged, the IOM-clocks are turned off. No signal is sent on the U-interface .
The U-transceiver w aits for a wa ke-up signa l TL fro m the LT- side to start an activation
procedure. To enter state ’IOM Awake’ a wake-up signal (DIN = 0) is required if the IOMclocks are disabled. The wake-u p signal is pr ovided v ia the IOM-2 interface (p in DIN =
0). If the IOM-clocks were active in sta te ’Deactivated’ C/I-code TIM is sufficient for a
transition to state ’IOM Awake’.
EC Training
The signal SN1 i s transmitted on the U-interface to allow the NT- rec eiver to update the
EC-coefficients. The automatic gain control (AGC), the timing recovery and the EQ
updating algorithm are disabled. The “EC-training” state is left when the EC has
converged (LSEC) or when timer T12 has elapsed.
EC-Training 1
The “EC-Training 1” state is entered if transmission of signal SN1 has to be started and
the deactivation procedure on the NT-side is not yet finished.
EC-Training AL
This state is entered in the case of an analog loop-back . The signal SN1 is transm itted
on the U-interface to allow the NT-receiver to update the EC-coefficients. The automatic
gain control (AGC), the timing recovery and the EQ updating algorithm are disabled. The
“EC-training” state is left when the EC has conv erged (LSEC) or when timer T12 has
elapsed.
EQ-Training
The receiver waits f or signal SL1 or SL2 to be able to update the AG C, to recover the
timing phase, to detect the synch-word (SW), and to update the EQ-coefficients. The
“EQ-training” state is left upon detection of binary “0s” in the B- and D-channels for a
period of 6–12 ms corresponding to the detection of SL2.
Error S/T
Loss of framing or loss of incoming signal has been detected on the S/T-interface (EI1).
The LT-side is informed by setting the ACT-bit to “0” (loss of transpa rency on the NTside). The following codes are issued on the C/I-channel:
– Normal activation or single-channel loop-back:AR
– Loop-back 2:ARL
®
IOM
Timing signals are delivered on the IOM-2 interface. The U-transceiver enters the
“Deactivated” state again upon detection of the C/I-channel code DI (idle code).
Semiconductor Group7111.97
-2 Awaked
Page 72
PSB 21911
PSF 21911
U-Transceiver
Pending Deactivation of S/T
The U-transceiver has received the UOA-bit at zero after a complete activation of the S/
T-interface. The U-transceive r deactivates the S/T-interface by issuing DR in the C/Ichannel. The value of the ACT-bit depends on its value in the previous state.
Pending Deactivation of U-Interface
The U-transceiver waits for the receive signal level to be turned off (LSU) to enter the
“Receiver Reset” state and start the deactivation procedure.
Pending Receive Reset
The “Pending Recei ve Reset” st ate is enter ed upon det ection of l oss of fram ing on the
U-interface or expiry of timer T1. This failure condition is signalled to the LT-side by
turning off the transmit level (SN0). The U-transceiver then waits for a response (no
signal level LSU) from the LT-side to enter the “Receive Reset” state.
Pending Timing
The pending timing state assures that the C/I-channel code DC is issued four times
before the timing signals on the IOM-2 interface are turned off.
In case the NT-auto m ode (Pin AUA=1) is selected the recogni tion of the LT wake-up
tone TL is assumed ev erytime the “Pending Timing” state ha s been entered from the
“Test” state. This function guarantees that the NT (in NT-auto mode) starts one single
activation attempt after h aving been resetted. After the a ssumed TL recog nition in this
state the activation will proceed norma lly.
Receive Reset
The “Receive Reset” state is entered upon detection of a deactivation request f rom the
LT-side, after a failure condition on the U-interface (loss of signal level LSUE), or
following the “Pending Reset” state upon expiry of timer T1 or loss of framing. No signal
is transmitted on the U-interface, especially no wake-up signal TN, and the S-transceiver
or microcontroller is requested to start the deactivation procedure on the NT-side (D R).
Timer T7 assures that no activation procedure is started from the NT-side for a minimum
period of T7. This gives the LT a chance to activate the NT.
The state is left only after completion of the deactivation procedure on the NT-side
(receipt of the C/I-channel code DI), unless a wake-up tone is received from the LT-side.
Semiconductor Group7211.97
Page 73
PSB 21911
PSF 21911
U-Transceiver
Synchronized 1
When reaching this state the U-transceiver informs the LT-side by sending the
superframe indication (inverted synch.-word). The loop-back commands decoded by the
EOC-processor control the output of the transmit signals:
– Normal ACT and UOA = 0:SN3
– Any loop-back and UOA = 0 (no loop-back):SN3T
The value of the issued SAI-bit depends on the received C/I-chan nel code: DI and TIM
lead to SAI = 0, any other C/I-code sets the SAI-bit to 1 indicating activity on the S/Tinterface. The U-transceiver waits for the receipt of UOA = 1 to enter the “Synchronized
2” state.
Synchronized 2
In this state the U-transceiver has recei ved UOA = 1. This is a request to activate the
S/T-reference point. The l oop-back commands d etected by the EO C-processo r control
the output of indications and transmit signals:
– Normal activation and UOA = (1):SN3 and AR
– Single channel loop-back and UOA = (1):SN3T and AR
– Loop-back 2 (LBBD):SN3T and ARL
The value of the issued SAI-bit depends on the received C/I-chan nel code: DI and TIM
lead to SAI = 0, any other C/I-code sets the SAI-bit to 1 indicating activity on the S/Tinterface. The U-transceiver waits for the receipt of the C/I-channel code AI to enter the
“Wait for ACT” state.
Test
The “Test” mode is entered when the unconditional co mmands RES, SSP, Pin-RES or
Pin-SSP are used. It is left when normal U-transceiver operation is selected via pins RES
and TSP and the C/I-channel codes DI or ARL are received.
The following signals are transmitted on the U-interface:
– No signal level (SN0) when the C/I-channel code RES is applied or a hardware reset
is activated.
– Single pulses (SP) when the C/I-channel code SSP is applied or pin TSP =1.
Transparent
This state is entered upon the detection of ACT = 1 received from the LT-side and
corresponds to the fully active state. In the case of a normal activation in both directions
of transmission the the following codes are output:
– Normal activation or single-channel loop-back:AI
– Loop-back 2:AIL
Semiconductor Group7311.97
Page 74
PSB 21911
PSF 21911
U-Transceiver
Wait for ACT
Upon the receipt of AI, the ACT-bit is set to “1” and the NT waits for a response (ACT =
1) from the LT-side. The output of indi cations and transmit signals i s as defined for the
“Synchronized” state.
Wait for SF
Upon detection of SL2, the sig nal SN2 is sent on th e U -interface and the receiver waits
for detection of the superframe indication. Timer T1 is then stopped and the
“Synchronized” state is entered.
Wait for SF AL
This state is entered in the case of an analog loop-back and allows the receiver to update
the AGC, to recover the timing phase, and to updat e the EQ- coeffici ent s. Signal SN 3 is
sent instead of signal SN2 in the “Wait-for-SF” state.
2.5.8C/I Codes
Both commands and indications depend on the data direction.
defined C/I codes. A ne w com man d or indication will be recognized as valid after it has
been detected in two successive IOM frames (Double last-look criterion). Indications are
strictly state orientated. Refer to the state diagrams in the previous sections for
commands and indications applicable in various st ates.
AIActivation IndicationEI1Error Indication 1
ARActivation RequestINTInterrupt
ARLActivation Request Local LoopPUPower-Up
DCDeactivation ConfirmationRESReset
DIDea ctivation Indicati onSSPSend-Single-Pulses test mode
DRDeactivation RequestTIMTimi ng reque st
DTData- T hrou gh test mode
Semiconductor Group7511.97
Page 76
PSB 21911
PSF 21911
U-Transceiver
2.5.9Layer 1 Loop-Back s
Test loop-backs are specified by the national PTTs in order to facilitate the location of
defect systems. Four different loop-backs are defined. The position of each loop-back is
S-Bus
figure 23
Loop 22Loop
.
IOM
R
UU
illustrated in
SBCXIEC-Q TE
NT
R
IOM
Loop 2
ICC
ICC
PBX or TE
IEC-Q TE
3Loop
IEC-Q TE
1aLoop
Repeater (optional)
IOM
R
R
IEC-QIEC-Q
Loop 1
IEC-Q
Exchange
IOM
ITD10198
Figure 23 Test Loop-Backs Supported by the IEC-Q TE
Loop-backs #1, #1A and #2 are controlled by the exchange. Loop-backs #3 is controlled
by the terminal. All loop-ba ck types are transparent. This means all bi ts tha t are l ooped
back will also be passed onwards in the normal manner.
2.5.9.1Loop-Back (No. 2)
Normally loop-back #2 is controlled f rom the exc hange via the EOC command s LBBD,
LB1 and LB2. In EOC aut o mode the EOC commands are recogn ized and execu ted
automatically (see page 51). The single channel loop-backs (LB1, LB2) are closed in the
U-transceiver itself whereas the complete loop-back (LBBD) is closed in a connected Stransceiver.
All loop-back functions are latched. This allows channel B1 and channel B2 to be looped
back simultaneousl y. All loop-back s are op ened wh en the EOC c ommand RTN is s ent
by the LT.
Semiconductor Group7611.97
Page 77
PSB 21911
PSF 21911
U-Transceiver
Complete loop-back
The complete loop-back comprises both B-channels and the D-channel. It may be closed
either in the U-transceiver itself , in the S-transceiver or in an external device.
When receiving the EO C-command LBBD in EOC auto m ode, the U-transceiver does
not close the loop-back immediately. Because the intention of this loop-back is to test the
complete NT, the U-transceiver passes the complete loop-back request on to the IOM-2
interface. This is achieved by issuing the C/I-code AIL in the “Transparent” state or C/I =
ARL in states different than “Transparent”. In applications that include a microcontroller,
the software decides where to close the loop, whereas in an NT1 the loop is closed
automatically in the S-transceiver (e.g. SBCX).
Single-channel loop-back (B1/B2)
Single-channel loop-backs are always performed directly in the U-transceiver if EOC
auto mode is selected. No difference between the B1-channel and the B2-channel loopback control procedure exists. They are therefore discussed together.
The B1-channel is closed with the EOC-comm and LB1. LB2 causes the c hannel B2 to
loop-back. Because these functions are latched, both channels may be looped back
simultaneously by sending first the command to close one channel followed by the
command for the remaining channel.
2.5.9.2Analog Loop-Back (No. 3)
Loop-back #3 is invoked by sending C/I command ARL to the U-transceiver. The loop is
closed by th e U-transcei ver as near to t he U -interface a s possi ble. For this reaso n it is
also called analog loop-back. All analog signals will still be passed on to the U-interface.
As a result the opposite station (LT) is activated as well.
In order to open an analog loop-back correctly, res et the U-transceiver into the TEST
state with the C/I-command RES. This e nsure s tha t the echo coef ficients and equalizer
coefficients will converge correctly when activating the following time.
Semiconductor Group7711.97
Page 78
PSB 21911
PSF 21911
U-Transceiver
2.5.10Analog Line Port
The analog part of the IEC-Q TE consists of three main building blocks:
– The analog-to-digital converter in the receive path
– The digital-to-analog converter in the transmit path
– The output buffer in the transmit path
Furthermore it contains some special functions. Thes e are:
– Analog test loop-back
– Level detect function
Analog-to-Digital Converter
The ADC is a sigma-delta modulator of second order using a clock rate of 15.36-MHz.
The peak input signal measured between AIN and BIN must be below 4 Vpp. In case the
signal input is too low (long range), the received signa l is amplified internally by 6 dB.
The maximum signal to noise ratio is achieved with 1.3 Vpp (long range) and 2.6 Vpp
(short range) input signal voltage.
Digital-to-Analog Converter
The output pulse is shaped by a special DAC. The DAC was optimized for excellent
matching between positive and negative pulses and high linearity. It uses a fully
differential capacitor appr oach. The staircase-like output signal of the DAC drives the
output buffers. The shape of a DAC-output signal is shown below, the peak amplitude is
normalized to one. This signal is fed to an RC-lowpass of first order with a corner
frequency of 1 MHz ± 50%.
The duration of each pulse is 24 steps, with
the pulse rate is 8 0-kHz or one pulse per 16 ste ps. Thus, the subsequent pulses are
overlapping for a duration of 8 steps.
= 0.78 µs per step. On t he other ha nd,
T0
Semiconductor Group7811.97
Page 79
PSB 21911
PSF 21911
U-Transceiver
Figure 24 DAC-Output for a Single Pulse
Output Stage
The output stage consists of two identical buffers , operated in a differ ential mode. This
concept allows an output-voltage sw ing of 6.4 Vpp at the output pins of the IEC-Q TE.
The buffers are optimized for:
– High output swing
– High linearity
– Low quiescent current to minimize power consumption
The output jitter produced by the transmitter ( with j itter-free i nput si gna ls) i s below 0.02
UIpp (Unit intervall = 12.5 µs, peak-peak) meas ured with a high-p ass filter of 30-Hz
cutoff frequency. Withou t the filter the cutoff frequ ency is below 0. 1 UIpp.
Analog Loop-Back Function
The loop-back C/I command ARL activates an internal, analog loop-back. This loop-back
is closed near the U-interface. All signals received on AIN / BIN will neither be evaluated
nor recognized after reaching the “Synchroni zed” state in NT-mode.
Level Detect
The level detect circuit evaluates the differential signal between AIN and BIN. The
differential threshold level is betwe en 4 mV an d 28 mV . The DC-level (common mode
level) may be between 0 V and 3 V. Level detect is not effected by the range setting.
Semiconductor Group7911.97
Page 80
PSB 21911
PSF 21911
U-Transceiver
Pulse Shape
The pulse mask for a single positive pulse measured between AOUT and BOUT at a load
of 98 Ω is given in the following figure.
Figure 25 Pulse Mask for a Single Positive Pulse
Hybrid
The hybrid circuit for the IEC-Q TE is shown on page 110.
Semiconductor Group8011.97
Page 81
2.6Access to IOM-2 Channels
PSB 21911
PSF 21911
Access to IOM-2 Channels
Important
: This chapter applies only in µP mode
In µP mode the microcontroller has access to the IOM-2 channels via the processor
interface (PI) and registers.
FSC
R
2
IOM -2
PI
UU
ITS10193
Figure 26 Access to IOM-2 Channels (µP mode)
The processor interface can be unde rstood as an intelligent switch between IOM-2 and
the U-transceiver. It handles D, B1, B2, C/I and Monitor-channel data. The data can
either be transferred directly bet we en IOM-2 and the U-transceiver, or be controlled via
the PI. The PI acts as an additional participant to the Monitor channel.
Switching directions are selected by setting the register SWST as indicat ed below:
SWST-Register
WTB1B 2DCIMONBSSGL
• Setting one of the 5 bits B1, B2, D, CI, or MON of SWST to "1" enables the µP access
to the corresponding data.
• Setting the bits listed above to "0" directly passes the corresponding data from IOM-2
to the U-transceiver and vice versa.
For a description of the bits WT, BS and SGL pl ease refer to
page 127
. The default value
after hardware reset is "0" at all 8 positions.
Semiconductor Group8111.97
Page 82
PSB 21911
PSF 21911
Access to IOM-2 Channels
2.6.1B -Channe l Access
Setting SWST:B1 (B2) to "1" enables the microprocessor to access B1 (B2)-channel
data between IOM-2 and the U-transceiver.
Eight registers (see
µP to IOM-2, from the µP to U and from U to the µP:
Table 16B1/B2-Channel Data Registers
RegisterFunction
WB1Uwrite B1-channel data to U-interface
RB1Uread B1-channel data from U-interface
WB1Iwrite B1-channel data to IOM-2
RB1Iread B1-channel data from IOM-2
WB2Uwrite B2-channel data to U-interface
RB2Uread B2-channel data from U-interface
WB2Iwrite B2-channel data to IOM-2
RB2Iread B2-channel data from IOM-2
Every time B-channel bytes arrive, an interrupt ISTA:B1 or ISTA:B2 respectively is
created. It is cleared after the corresponding registers have been read. ISTA:B1 is
cleared after RB1U and RB1I have been read. ISTA:B2 is cleared after RB2I and RB2U
have been read. After an interrupt the data in RB1U and RB1I is stable for 125µs.
table 16
) handle the transfer of data from IOM-2 to the µP, from the
2.6.2D -Channe l Access
Setting SWST:D to " 1" enables the micropr ocessor to access D-c hannel da ta between
the IOM-2 and the U-interface.
Four registers (see
µP to IOM-2, from the µP to U and from U to the µP.
Table 17D-channel data registers
RegisterFunction
DWUwrite D-channel data to U-interface
DRUread D-channel data from U-interface
DWIwrite D-channel data to IOM-2
DRIread D-channel data from IOM -2
Two 2-bit FIFOs of length 4 collect the incoming D-channel packets from IOM and U.
Every fourth IOM-frame they are full, an interrupt ISTA:D is generated and the cont ent s
table 17
) handle the transfer of data from IOM-2 to the µP, from the
Semiconductor Group8211.97
Page 83
PSB 21911
PSF 21911
Access to IOM-2 Channels
of the FIFOs are para llely shif ted to DRU and DRI re spectivel y. DRU and DRI have to be
read before the ne xt interrupt I S TA:D c an occur, otherwise 8 bits will be l ost . D WU and
DWI have to be loaded with data for 4 IOM-frames. Data in DWU and DWI is assumed
to be valid at the time ISTA:D occurs. The register contents are shifted parallely into two
2-bit FIFOs of length four, from where the data is put to IOM-2 and U respectively during
the following 4 IOM-frames. During this time, new data can be placed on DWU and DWI.
DWU and DWI are not cl eared a fter the data w as pass ed to the FI FOs. That i s, a byte
may be put into DWU or DWI once and continously passed to IOM or U, respectively.
Figure 27
illustrates this procedure:
Figure 27 Procedure for the D-Channel Processing
Note:
2.6.3C/I Channel Access
Setting SWST:CI to "1" enables the microprocessor to access C/I-commands and
indications between IOM-2 and the U-transceiver.
A change in tw o consecutive frames (d ouble last look) in the C/I-channel on IOM-2 is
indicated by an interrupt ISTA:CICI. The received C/I-command can be read from
register CIRI. A change in the C/I-channel coming from the U-transceiver is indicated by
an interrupt ISTA:CICU. The new C/I-indication can be read from regist er CIRU .
Default of DWU, DWI, DRU and DRI aft er res et is "FF
".
H
Note: The term C/I-indication always refers to a C/I-code coming from the U-transceiver,
whereas the term C/I-command refers to a C/I-code going into the U-transc eiver.
A C/I-code going to the U-transceiver has t o be written into the CIWU-register. A C/Icode to IOM-2 h as to be written into the CIWI-register. The contents of both registers
(CIWU and CIWI) will be transferred at the next available IOM-2 frame. The registers are
not cleared after the transfer. Therefore, it is possible to continously send C/I codes to
IOM-2 or the U-transceiver by only writing the code into the register once.
Semiconductor Group8311.97
Page 84
PSB 21911
PSF 21911
Access to IOM-2 Channels
C/I-commands to the U-transceiver have to be applied at least for two IOM-2 frames (250
µs) to be considered as valid.
In TE mode (i.e. 1.53 6 MHz DCL), the ADF2 :TE1 bit is used to direct the C/I-channel
access either to IOM -2 channel 0 (ADF2:TE1 = 0, de fault) or to IOM-2 channel 1 of the
IOM-2 terminal structure (ADF2:TE1 = 1),
terminal devices such as t he AR C O FI via the proc essor in terfa ce of the IEC- Q TE. The
C/I code going to IOM-2 is 4 bits long if it is written to IOM-2 channel 0 (ADF2:TE1 = 0).
If written to IOM-2 channel 1 this C/I code is 6 ibts long (ADF2:TE1 = 1). If the ADF2:TE1
bit is 1, the C/I channel on IOM-2 ch annel 0 is passed transparently from the IOM-2
interface to the IEC-Q TE itself.
figure 28 on page 84
. This allows to program
R
C/I 0 (4 Bit)
IOM -2
C/I access to IOM -2 channel 0
R
(4 Bit)C/I
CIWUCIRUCIWICIRI
IEC-Q Core
SWST : CI = 1
ADF2 : TE1 = 0
ITB10295
R
C/I 1 (6 Bit)
IOM -2
CIWUCIRUCIWICIRI
C/I access to IOM -2 channel 1
R
IEC-Q Core
SWST : CI = 1
ADF2 : TE1 = 1
ITB10296
Figure 28 C/I Channel Access
2.6.4Monitor Channel Access
Setting SWST:MON to "1" enables the microprocessor to access Monitor-channel
messages at IOM-2 interface and the U-transceiver. Mon itor-channel access ca n be
performed in three different IOM-2 channels (see
Setting SWST:MO N t o ’ 0’ disables the contro ller access to the M onitor channel (
29
upper left part).
figure
Setting SWST:MON to ’1’ enables three different ways of controller access to the
Monitor channel. ADF2:TE1 set to ’0’ allows to either access the U-transc eiver core of
the IEC-Q TE (see
the IEC-Q TE (
figure 29
figure 29
upper right part, ADF2:MIN = ’1’) or the IOM-2 interface of
lower left part, ADF2:MIN = ’0’).
Setting ADF2:TE1 to ’1’ in TE mode gives access to IOM-2 channel 1 rather than IOM-2
channel 0 directed out of the IEC-Q TE. This allows to program devices linked to IOM-2
channel 1 (e.g. ARCOFI) via the processor interface of the IEC-Q TE.
Semiconductor Group8511.97
Page 86
PSB 21911
PSF 21911
Access to IOM-2 Channels
2.6.4.1Monitor Channel Protocol
The PI allows to program the IEC-Q TE Monitor channel in the way known from the PEB
2070 ICC.
The Monitor channel oper ates on an
IOM-bus occur synchronized to frame sync FSC, the flow of data is controlled by a
handshake procedur e using the Monitor Channel Receive (MR) and Monitor Chann el
Transmit (MX) bits. For example: data is placed onto the Monitor channel and the MX bit
is activated. This data will be transmitted repeatedly once per 8-kHz frame until the
transfer is acknowledged via the MR bit.
The microprocessor may either enf orce a "1" (idle) in M R, MX by se tting the cont rol bit
MOCR:MRC or MOCR:MXC to "0", or enable the control of these bi ts internally by the
IEC-Q TE according to the Monitor channel protocol. Thus, before a data exchange can
begin, the control bits MRC or MXC should be set to "1" by the microprocessor.
The Monitor channel protocol is illustrated in
bits for transmission and reception are:
Monitor Transmit Bits
RegisterBitcontrol / statusFunct ion
Before starting a transmission, the micr oprocessor should verify that the transm itter is
inactive, i.e. that a possible previous transmission has been terminated. This is indicated
by a "0" in MOSR:MAC, the Monitor Channel Active status bit.
To enable interrupts for the transmitter the MOCR:MXE bit must be set to “1“ (For details
section 4.1.1 on page 115
see
register, the microprocessor sets the Monitor Transmit Control bit MXC to "1". This
enables the MX bit to go active (“0“), indicating the presence of valid Monitor data
(contents of M OX) in the corr esponding f rame. As a r esult, the receiving device stores
the Monitor byte in its Monitor Receive (MOR) register and generates an MDR interrupt
status.
Alerted by the MDR interrupt, the microprocessor reads the Monitor Receive (MOR)
Semiconductor Group8711.97
). After having written the Monitor Data Transmit (MOX)
Page 88
PSB 21911
PSF 21911
Access to IOM-2 Channels
register. When it is r eady to accept data (e.g. based on the value in M OR, which in a
point-to-multipoint application might be the address of the destination device), it sets the
MR control bit MRC to "1" to enable the recei ver to store succeeding Monitor channel
bytes and ackn owledge them a ccording to the Monitor c hannel protocol. I n addition, it
enables other Monitor channel interrupts by setting Monitor receive Interrupt Enable
(MRE) to "1".
As a result, the first Monitor byte is acknowledged by the receiving device setting the MR
bit to "0". This causes a Monitor Data Acknowledge (MDA) interrupt status at the
transmitter.
A new Monitor data byte can now be written by the microprocessor in MOX. The MX bit
is still in the active (“0“) state. The transmitter indicates a new byte in the Monitor channel
by returning the MX bit act ive af ter sending it once in the inactive sta te. As a result, the
receiver stores the M onitor byte in MOR and g enerates a new MDR interrupt status.
When the microproces sor has read the MOR register, the receiver acknowledges the
data by returning the MR bit active after sending it once in the inactive state. This in turn
causes the transmitter to generate an MDA interrupt status.
This "MDA interrupt – write data – MDR interrupt – read data – MDA interrupt"
handshake is repeated as long as the transmitter has data to send. Note that the Monitor
channel protocol imposes no maximum reaction times to the microprocessor.
When the last byte has been ac knowledged by the rec eiver (MDA interrupt status ), the
microprocessor sets the Monitor Transmit Control bit (MXC) to "0". This enforces an
inactive ("1") state in the MX bit. Two frames of MX inactive signifies the end of a
message. Thus, a Monitor Channel End of Reception (MER) interrupt status is
generated by the receiver when the MX bit is received in the inactive state in two
consecutive frames. As a result, the microprocessor sets the MR control bit MRC to “0“,
which in turn enforces an inactive state in the MR bit. This marks the end of the
transmission, making the Monitor Channel Active (MAC ) bit return to "0".
During a transmission process, it is possible for the receiver to ask for a transmission to
be aborted by sending an inactive MR bit value in two consecutive frames. This is
effected by the microprocessor writing the MR control bit MRC to "0". An aborted
transmission is indicated by a Monitor Channel Data Abort (MAB) interrupt status at the
transmitter.
In TE mode, the ADF2:TE1 bit is used to direct the Monitor access either to IOM-channel
0 (ADF2:TE1 = "0" , default) or to I OM-channel 1 of the IOM-Terminal structure. This
allows to program terminal devices such as the ARCO FI via the processor interface of
the IEC-Q TE. If the ADF2:TE1 bit is "1", the Monitor channel on IOM-channel 0 is
passed transparently from the IOM-2 interface to the IEC-Q TE itself.
Semiconductor Group8811.97
Page 89
2.7S/G Bit and BAC Bit in TE Mode
PSB 21911
PSF 21911
S/G Bit and BAC Bit in TE Mode
Important
: This chapter applies only in µP mode and if DCL = 1.536 MHz (TE mode).
If DCL = 1.536 MHz the IOM-2 interface consists of three IOM-2 channels. The last octet
of an IOM-2 f rame i ncludes the S/G and the BAC bit (chapter 2.3.1.1, page 31). Either
or both bits can be used in various applications including
• the D-channel arbitration in a PBX via an ELIC on the linecard
• the synchronization o f a base station in radio in t he local loop (RLL) or w ireless PBX
applications
The S/G is always written and never read by the IEC-Q TE. Its value depends on the last
received EOC-command and on the status of the BAC bit.The proc essi ng m ode f or the
S/G bit is selected via bits SWST:BS, SWST:SGL and ADF:CBAC according t o
A detailed description of the S/G bit in all modes is provided in
Appendix B
table 18
.
Table 18S/G Processing Mode
SWST:ADF:Description (X is don’t care)Application
BSSGLCBAC
00xS/G bit always "0" (default)
.
010S/G bit always "1" S/G and BAC are
handled by other
devices than the
IEC-Q T E
0 11S/G bit set to "1" continously with EOC
received, reset to "0" with EOC 27H
25
H
received
BAC bit controlls S/G-bit, upstream Dchannel not affected
10xS/G bit set to "1" for 4 IOM-frames with
EOC 25
receivced, automatically reset
H
to "0" after that
ELIC on linecard,
Interframe fill of
terminals
contains zeroes
(e.g. ’01111110’)
Synchronizaiton
of base station,
e.g. IBMC or
MBMC
110S/G bit set to "1" continously with EOC
receivced, reset to "0" with EOC 27H
25
H
received
111S/G bit set to "1" continously with EOC
receivced, reset to "0" with EOC 27H
25
H
received
BAC bit controlls S/G bit and upstream Dchannel according to
table 19
.
ELIC on linecard,
Interframe fill of
terminals are
’ones’
Semiconductor Group8911.97
Page 90
PSB 21911
PSF 21911
S/G Bit and BAC Bit in TE Mode
2.7.1Applications w ith ELIC on the Linecard (PBX)
The S/G bit on DOUT (downstream) and the BAC bit on DIN (upstream) can be used to
allow D-channel arbitration similar to the operation of the Upn interface realised with the
OCTAT-P and the ISAC-P TE. The basic function is as follows:
The PBX linecard using the ELIC assigns one HDLC controller to a number of terminals.
T
As soon as one terminal
terminals receive a message indicating the D-channel to be blocked for them. The
request is done with the BAC bit. At terminal
transfers the need for the D-channel to the LT-side. There, the HDLC-controller is
assigned to the appropriate IOM-channel. Once this is done and indicated to the terminal
by means of the S/G bit, the terminal begins to send D-channel messages.
Note that this procedure is somewhat different from the opera tion of the OCTAT/ISACP TE. There, the beginn of the upstream D-channel data transfer itself indicates the need
for the HDLC-controller. This implies that any other terminal, that incidentally sent a
HDLC-message the same time , can be stopped bef ore the mess age is lost in case the
HDLC-controller is not available. The U-interface featured by the IEC-Q TE is not able to
transfer the available/blocked information often enough to ensure this. Hence, it is
necessary to indicate a D-channel access by the terminal in advance. "In advance"
actually means about 14 ms.
requests the D-channel, e.g. for signalisation, all other
T
the BAC bit is set and the IEC-Q TE
Giving MON-0 25
access at the NT-side to be on " STOP". As one EOC-message is tran smitted via the
EOC-channel once every 6 ms, the S/G bit on IOM-2 can be set in 6 ms intervals.
If the 4 channel LT chip set PEB 24911 (DFE-Q) is used in the LT, a PEB 20550 (ELIC)
can arbitrate the D-channel via the C/I command as known from the OCTAT-P and
QUAT-S devices. Please refer to the PEB 24911 Data Sheet for detailed information on
this.
The BAC bit together with the EOC-message s rece ive d from the LT control the S/G bit
and the upstream D-channel according to the
Table 19Control Structure of the S/G Bit and of the D-Channel
BAC bit of
last
IOM-frame
0reflects last received EOC
at the LT during transparent operation will cause the D-channel
H
table 19.
S/G bit
1 = stop
0 = go
message after falling edge after
delay TD1 (1.5 ms and two EOCframes)
D-channel upstream
tied to "0"
11set transparent with first "0" in D-
channel
Semiconductor Group9011.97
Page 91
D-Channel Request by the Terminal
PSB 21911
PSF 21911
S/G Bit and BAC Bit in TE Mode
Figure 31
is BAC = 1 at DIN after TD1 has expired. That causes the S/G bit to be set to the stop
position.
BAC = 1 received on DIN sets the S/G bit on DOUT to the stop position ("1") at the next
IOM-frame. When the terminal requests access to the HDLC-controller in the ELIC it sets
the BAC-bit at DIN of it’s IEC-Q TE to "0". That causes the D -ch annel data upst ream to
be tied to "0" and the S/G-bit to be set to "1". The ELIC receives the zeros and reacts by
assigning the HDLC-controller to this very terminal. This is indicated via the change of C/
I code downstream at the LT side resulting in the S/G bit to be set to "0" (’go’) after delay
TD1 (see below for the explanation of TD1 and TD2).
The IEC-Q TE will continue to send "0" upstream in the D-channel until the actual HDLC
data arrives at DIN.The HDLC -frame i tself, mar ked by t he first "0" in the D-c hannel w ill
reset the D-channel back to transparent. This allows to have arbitrary delay s between
the S/G bit going to "0" and the D-channel being used without the risk of loosing the
HDLC-controller by sending an abort request con sisting of all "1".
At the end of the HDLC-frame the BAC bit is reset to "1" again by the layer-2 controller
(e.g . SMARTL INK; ICC). Thi s c a u s e s the S/G b it to be set to "1" i n th e next IOM frame
which stops a possible second HD LC-frame that could not be processed in th e ELIC
anymo re.
illustrates the request for the HDLC-controller by the terminal. The start state
TD1 and TD2
The delays TD1 and TD2 (see
the 6ms interval in which an EOC message can be transmitted on the Uko interface. As
an EOC-message can start once every 6 ms and w ill take 6 ms to be transm itted, TD2
will be 12 ms in the worst case.
TD1 is at minimum 7.5 ms depending on the location of the superframe at th e time the
HDLC-controller is requested by the terminal. This delay is necessary becaus e instead
of receiving an EOC-message "go" as req uest ed, the termin al cou ld as well recei ve the
EOC message "stop" because the HDLC-controller was assigned to an other subscriber
just before .
Flags as interframe Fill
The influence to the upstr eam D-c hannel can b e disab led whi le the c ontrol of the S/Gbit via EOC-mess ages and via th e BAC bit still is given as describ ed above by se tting
SWST:BS to "0", SWST:SGL to "1" and ADF:CBAC to " 1". This i s usefull when having a
controlling device in the terminal, that is able to send the interframe timefill "flags".
figure 31
) have the following reasons: TD2 is caused by
Semiconductor Group9111.97
Page 92
PSB 21911
PSF 21911
S/G Bit and BAC Bit in TE Mode
ELIC
C/I =
R
xxxx
R
IOM -2
IEC-Q
AFE/DFE
HDLC occupied by other Terminal:
1100=C/I
Delay TD2
"00" on D-Channel
EOC 25
HDLC assigned:
C/I = 1000
Delay TD2
27EOC
U
k0
IEC-Q TE
R
IOM -2
TE Mode
1=BAC
S/G = 1
D-Channel Transparent
BAC = 0
D-Channel fied to "0"
H
Delay TD1
H
Transparent;=S/G
HDLC-Controller available
HDLC-Frame on D-Channel
HDLC ready:
1100=C/I
Delay TD2
EOC 25
H
Figure 31 D-Channel Request by the Terminal
D-Channel
Transparent
HDLC
Frame
BAC = 1
"stop"
1;=S/G
D-Channel Transparent
ITD10200
Semiconductor Group9211.97
Page 93
2.8Reset
PSB 21911
PSF 21911
Reset
Important
Several resets are provided in the IEC-Q TE. Their effects are summarized in
Table 20Reset
ResetConditionEffectPin RST
Power-onPower-onResets the state machine and all
Hardware
Reset
WatchdogWatchdog
Software
Reset
The IOM-2 clocks DCL and FSC as well as MCLk are delivered during reset (except for
power-on).
: This chapter applies only in µP mode.
registers
Pin RES
expired
C/I = 0001Resets the state machine and does
= 0Resets the state machine and all
registers exept for STCR register
Resets no register and does not
affect the state machine
not affect the registers
yes
no
yes
no
tabl e 20
active
.
The IEC-Q TE provides a low active reset output (pin RST
power-on reset and the watchdog timer. The watchdog is enabled by setting the
SWST:WT bit to “1”. Default after hardware reset of SWST:WT is "0". Please refer to
page 3 9 for information on how to use the watchdo g timer.
Figure 32
Figure 32 Reset Sources
illustrates the reset sources that have an impact on pin RST
Power-on
) which is c ontrolled by a
.
Semiconductor Group9311.97
Page 94
2.9Power Controller Interface
PSB 21911
PSF 21911
Power Controller Interface
Important
A power controller inte rface is implemented in the PSB 21 911 to provide comfortable
access to peripheral circuits which are not connected directly to the microprocessor .
Because this interface was specifically designed to support the ISDN Excha nge Power
Controller IEPC (PEB 2025) it is referred to as “Power Controller Interf ace”. Despite t his
dedication to the IEPC, the controller interface is just as suited for other general-purpose
applications.
Interface Bits (PCD, PCA, PCRD, PCWR, INT)
The interface structure is adapted to the register structure of the IEPC. It consists of three
data bits PCD0 ... 2, two address bits PCA0,1, read and write signals PCRD and PCWR
respectively as well as an interrupt facility INT.
The address bits are latched, they may therefore in genera l interface applications be
used as output lines. For general interface inputs each of the three data bits is suitable.
Read and write operations are performed via MON-8 com m ands. Thr ee i nput s and two
outputs are thus available to connect external circuitry.
The interrupt pin is edge sensitive. Each change of level at the pin INT will initiate a C/Icode “INT” (0110
resulting actions need to be performed by the control unit.
: This chapter applies only in stand-alone mode.
) lasting for four IOM-frames. Interpretation of the interrupt cause and
B
Table 21
interface.
Table 21MON-8 and C/I-Commands
ChannelCodeFunction
MON-8WCIWrite to interface. Address and data is contained in the MON-
MON-8RCIRead from interface at specified address. Address is latched
MON-8ACIAnswer from interface. After a RCI-request the value of three
C/IINTInterrupt. After a change of level has been observed, the C/I-
lists all MON-8- and C/I-commands that are relevant to the power controller
command. The address is latched, data is not latched.
and the current value of the data port is read. The result is
returned to the user with MON-8 “ACI”.
data bits at the specified address is returned.
code “INT” is issued for 4 IOM-frames. Note the special timing
of the interrupt signals described on page 96.
Semiconductor Group9411.97
Page 95
PSB 21911
PSF 21911
Power Controller Interface
Communication with the power controller interface is established with local Monitor
messages (MON-8) on IOM-2. The following two-byte messages are matched to the
IEPC-power controller status register read and write operations but can be used in
general, too.
MON-8WCIWrite Controller Interface
10000000011D0D1D2A0A1
MON-8RCIRead Controller Interface
10000000010–––A0A1
MON-8ACIAnswer Controller Interface
10000000D0D1D2–––––
After the receipt of a M O N -8-com ma nd t he I EC -Q TE will set the address/data bi t s and
generate a read or write pulse.
The address bits are latched, and th e output is stable until it is overwritten by a new
dedicated MON-8-command. All data lines are connected to an internal pull-up resistor.
The initial value on the address lines after a soft or hardware reset is (11
).
B
Semiconductor Group9511.97
Page 96
PSB 21911
PSF 21911
Power Controller Interface
Interrupt
For every change at the input pin “INT”, the IEC-Q TE will transmit a C/I-channel code
(0110
sampled every 4 IOM-2-frames. An interrup t indi catio n m ust th erefore be applied to pin
“INT” for at least 4 IOM-2-frames.
), INT, in 4 successive IOM-2-frames. The input condition of the “INT” pin is
B
125 µs
R
IOM -2 Frames
1ms
INT
ms0.5
Figure 33 Sampling of Interrupts
C/I Code INT
INT
C/I Code INT
Example AExample B
ITD10294
Semiconductor Group9611.97
Page 97
3Operational Description
3.1C/I Channel Programming
PSB 21911
PSF 21911
C/I Channel Programming
Important
CIWU = C7
ISTA = 02
CIRU = 03
: This chapter applies only in µP mode.
µ
P
CICU
Interrupt
C/I Channel Handler
Transmit C/ICommand "RES"
New C/I-Code
received from U
Read New C/I-Code.
=C/I = 0001b
C/I = 0000b
U-Transceiver
Transfer to "TEST"
State
Send C/I-Indication
"DR"
ITD10291
Figure 34 Example: C/I-Channel Use (all data values hexadecimal)
Semiconductor Group9711.97
Page 98
3.2Monitor Channel Programming
PSB 21911
PSF 21911
Monitor Channel Programming
Important
The example on
number (ID). It consists of the transmission of a two-byte message f rom the control unit
to the transceiver in IOM channel 0. The transceiver acknowledges the receipt by
returning a two-byte l ong message in the moni t or channel. The procedure is absol ut ely
identical for Monitor channel 1.
The µP starts the transfer procedure after having confirmed that the monitor channel is
inactive. The first byte of monitor data is loaded into the transmit register MOX. Via the
Monitor Control Register MOCR monitor interrupts are enabled and control of the MX-bit
is handed over to the IEC-Q TE. Then transmission of the first byte begins. The Utransceiver reacts to a low level of the MX-bit by reading and acknowledging the monitor
channel byte automatically. On detection of the confirmation, the IEC-Q TE issues a
monitor interrupt to inform the µP that the next byte may be sent. Loading the second
byte into the transmit register results in an immediate transmissi on (timing is controlled
by IEC-Q TE). The U-transceiver receives the second byte in the same manner as
before. When transmission is completed, the U-transceiver sends “End of Message”
(MX-bit high).
: This chapter applies only in µP mode.
page 99
illustrates the read-out of the transceiver’s identification
It is assumed that a monitor command was sent that needs to be answered by the Utransceiver (e.g. read-out of a register). Therefore, the U-transceiver commences to
issue a two-byte confirmation after an End-of-Message indication from the IEC-Q TE has
been detected. The IEC-Q TE notifie s the µP via interrupt when new monitor data has
been received. The processor may then read and acknowledge the byte at a convenient
instant. When confirmati on has been completed, the U-transceive r sends “EOM”. This
generates a c orresponding interrupt in the I EC -Q TE. By setting the MR- bit t o high, the
monitor channel is inactive, the transmission is finished.
Semiconductor Group9811.97
Page 99
PSB 21911
PSF 21911
Monitor Channel Programming
Example: Monitor Channel Transmission and Reception
Basic Configuration, IOM-2 Clo cks must be active
wSTCR =0x15// TE Mode, EOC Auto Mode
wSWST =0x06// Access to C/I and Monitor channel
wADF2 =0x48// Monitor access to U-transceiver