Rohde&Schwarz FSQ-K70, FSMR-B73 User Manual

Download

Page 1

PAD-T-M: 3574.3259.02/01.00/CI/1/EN

R&S®FSQ-K70/FSMR/FSU-B73 Vector Signal Analysis

Software Manual

Test and Measurement

1161.8073.42 – 13

Page 2

© 2014 Rohde & Schwarz GmbH & Co. KG Muehldorfstr. 15, 81671 Munich, Germany Phone: +49 89 41 29 - 0 Fax: +49 89 41 29 12 164 E-mail: info@rohde-schwarz.com Internet: http://www.rohde-schwarz.com Subject to change – Data without tolerance limits is not binding. R&S® is a registered trademark of Rohde & Schwarz GmbH & Co. KG. Trade names are trademarks of the owners.

The following abbreviations are used throughout this manual: R&S®FSQ-K70/FSMR/FSU-B73 is abbreviated as R&S FSQ-K70/FSMR/FSU-B73

Page 3

R&S FSQ-K70/FSMR/FSU-B73 Contents

Software Manual 1161.8073.42 - 13 3

1 Vector Signal Analysis ....................................................................... 9

1.1 Enabling the Firmware Option .................................................................................... 9

1.2 Test Setup for Measurement on Base Stations and Power Amplifiers ................10

1.2.1 Precautions ..................................................................................................................10

1.2.2 Standard Test Setup ....................................................................................................10

1.3 Calling and Exiting the Option - VSA Softkey .........................................................11

1.3.1 Calling the Option - VSA Softkey .................................................................................11

1.3.2 Exiting the Option - VSA Softkey .................................................................................11

1.3.3 Return to VSA Menu (Home VSA Hotkey) ..................................................................11

1.3.4 Overview ......................................................................................................................12

2 First Measurements - Getting Started ............................................. 13

2.1 Interconnecting Transmitter and Analyzer .............................................................13

2.2 Basic Settings of Test Transmitter ..........................................................................14

2.3 Switching On the R&S FSQ-K70/FSMR-B73/FSU-B73 Option ...............................14

2.4 Basic Analyzer settings for EDGE Measurements .................................................15

2.5 Measurement 1: Demodulation of a Single EDGE Burst .......................................15

2.6 Measurement 2: Selection of a Specific Slot with Trigger Offset .........................19

2.7 Measurement 3: Setting the Burst Search Parameters (LEVEL) ..........................22

2.8 Measurement 4: Suppression of Incorrect Measurements ...................................24

2.9 Measurement 5: Evaluation Lines ............................................................................26

3 Brief Description of Vector Signal Analysis (Function) ................. 29

3.1 Block Diagram of Digital Signal Processing Hardware .........................................29

3.1.1 Description of Block Diagram ......................................................................................29

3.1.2 Bandwidths for Signal Processing ...............................................................................30

3.2 Symbol Mapping ........................................................................................................44

3.3 Demodulation and Algorithms .................................................................................66

3.3.1 Burst Search ................................................................................................................69

3.3.2 Demodulator 1 .............................................................................................................71

3.3.3 Demodulator 2 .............................................................................................................73

3.3.4 Matching ......................................................................................................................74

3.3.5 Pattern Search .............................................................................................................77

3.3.6 Result & Error Calculation, Display ..............................................................................79

Page 4

R&S FSQ-K70/FSMR/FSU-B73 Contents

Software Manual 1161.8073.42 - 13 4

3.3.7 Differences between Modulation Types ......................................................................80

3.4 Vector and Scalar Modulation Errors ......................................................................81

3.4.1 Error Model of Transmitter ...........................................................................................81

3.4.2 Modulation Error (PSK, MSK, QAM, VSB) ..................................................................81

3.4.3 Modulation Error (FSK) ................................................................................................91

4 Operation and Menu Overview ................................ ........................ 93

4.1 Operation ....................................................................................................................93

4.2 Special Features/Differences from the Basic Instrument ......................................93

4.2.1 Display of States Within Softkeys ................................................................................93

4.2.2 Display of Setting Parameters Within Softkeys ...........................................................94

4.2.3 Measurement Window .................................................................................................96

4.3 Menu Overview ...........................................................................................................98

4.3.1 Hotkeys ........................................................................................................................98

4.3.2 Softkeys .......................................................................................................................99

5 Instrument Settings and Measurements ....................................... 104

5.1 Resetting the Option - PRESET VSA Hotkey ........................................................104

5.2 Overview of Current Settings - SETTINGS Hotkey ...............................................104

5.3 Configuration of Measurements - HOME VSA Hotkey .........................................105

5.4 Measurements on Dig. Standards - DIG. STANDARD Softkey ............................106

5.4.1 Predefined Standards and Standard Groups ............................................................106

5.4.2 List of Predefined Standards and Standard Groups ..................................................108

5.4.3 DIGITAL STANDARD Menu ......................................................................................111

5.5 BURST& PATTERN Softkey ....................................................................................116

5.5.1 Burst and Search Parameters ...................................................................................116

5.5.2 Multiple Evaluation of a Captured Data Record (MULTI) ..........................................117

5.5.3 Burst and Search Parameters for Predefined Standards ..........................................122

5.5.4 Pattern and Pattern Lists ...........................................................................................123

5.5.5 BURST& PATTERN Menu ........................................................................................125

5.6 Setting Parameters - MODULATION SETTINGS Softkey .....................................136

5.7 Setting Demodulation - DEMOD SETTINGS Softkey ............................................142

5.7.2 Evaluation Lines / Limiting the Measurement Range ................................................145

5.7.3 Record Buffer, Demodulation Range and Display Range .........................................147

5.8 Display of Measurement Results ...........................................................................149

5.8.1 Spectral Displays .......................................................................................................149

5.8.2 Statistical Displays .....................................................................................................150

5.8.3 MEAS RESULT Softkey ............................................................................................152

Page 5

R&S FSQ-K70/FSMR/FSU-B73 Contents

Software Manual 1161.8073.42 - 13 5

5.8.4 Selection of Displayed Measurement and Reference Signal - MEAS SIGNAL / REF

SIGNAL Softkey .........................................................................................................159

5.8.5 Selection of Error Display - ERROR SIGNAL Softkey ...............................................172

5.8.6 Selection of the Raw Signal - CAPTURE BUFFER Softkey ......................................188

5.8.7 Selection of Adaptive Equalizer Display - EQUALIZER Softkey ...............................198

5.9 Positioning of Display on Screen - FIT TRACE Softkey ......................................206

5.9.1 Scaling of Time Axis in Symbols ................................................................................209

5.9.2 FIT TRACE Menu ......................................................................................................210

5.10 Multiple Evaluation and Section Displays - ZOOM Softkey.................................211

5.11 Setting of Span - RANGE Softkey ..........................................................................212

5.11.1 Automatic Setting of Reference Level - ADJUST LVL Softkey .................................214

5.11.2 Restoring of Factory Settings - FACTORY DEFAULTS Softkey ...............................214

5.11.3 Importing Stand., Mappings, Pattern and Filter - IMPORT Softkey ...........................215

5.11.4 Export of Stand., Mappings, Pattern and Filter - EXPORT Softkey ..........................218

5.12 Overview of Other Menus .......................................................................................221

5.12.1 Default Settings - PRESET Key .................................................................................221

5.12.2 System Error Correction - CAL Key ...........................................................................221

5.12.3 General Instrument Settings - SETUP Key ...............................................................221

5.12.4 Documentation of Results - HCOPY Key ..................................................................223

5.12.5 Frequency Settings - FREQ Key ...............................................................................223

5.12.6 Span ...........................................................................................................................224

5.12.7 Level Settings - AMPT Key ........................................................................................224

5.12.8 Selection of Units for Display - DISPLAY UNIT Key ..................................................224

5.12.9 Setting of Bandwidth for Analog IF Filter - BW Key ...................................................225

5.12.10 Sweep Settings - Sweep Key ....................................................................................225

5.12.11 MEAS Key ..................................................................................................................226

5.12.12 Trigger Settings - TRIGGER Key ...............................................................................226

5.12.13 Trace Functions - TRACE Key ..................................................................................227

5.12.14 Limit Lines Settings - LINES Key ...............................................................................231

5.12.15 Screen Configuration - DISP Key ..............................................................................232

5.12.16 File Management - FILE Key .....................................................................................232

5.12.17 Marker Settings - MARKER Key ................................................................................233

5.12.18 Marker Settings (Marker to) - MKR -> Key ................................................................233

5.12.19 Marker Functions - MKR FCTN Key ..........................................................................233

5.13 Troubleshooting .......................................................................................................239

5.13.1 Different Symbol Rate Setting in Transmitter and Analyzer ......................................240

5.13.2 Different Filter Settings in Transmitter and Analyzer .................................................242

5.13.3 Incorrect Modulation of Analyzer ...............................................................................244

Page 6

R&S FSQ-K70/FSMR/FSU-B73 Contents

Software Manual 1161.8073.42 - 13 6

5.13.4Overdrive Condition of the Analyzer ...............................................................................................246

6 Remote: Control Commands ......................................................... 248

6.1 CALCulate - Subsystem ..........................................................................................248

6.1.1 CALCulate:DDEM - Subsystem .................................................................................249

6.1.2 CALCulate:FEED - Subsystem ..................................................................................250

6.1.3 CALCulate:FORMat - Subsystem ..............................................................................251

6.1.4 CALCulate:ELIN - Subsystem ...................................................................................253

6.1.5 CALCulate:MARKer:FUNCtion Subsystem ...............................................................254

6.1.6 CALCulate:STATistics - Subsystem ..........................................................................282

6.1.7 CALCulate:TRACe - Subsystem ................................................................................283

6.1.8 CALCulate:UNIT - Subsystem ...................................................................................285

6.2 DISPlay - Subsystem ...............................................................................................286

6.3 FORMat -Subsystem ................................................................................................291

6.4 INSTrument - Subsystem ........................................................................................294

6.5 MMEMory - Subsystem ...........................................................................................295

6.6 SENSe - Subsystem .................................................................................................297

6.6.1 SENSe:DDEMod-Subsystem ....................................................................................297

6.6.2 SENSe:FREQuency - Subsystem .............................................................................335

6.7 TRACe – Subsystem ................................................................................................335

6.8 TRIGger - Subsystem ..............................................................................................338

6.9 Table of Softkeys Assigned to IEC/IEEE Bus Commands ..................................339

6.9.1 Hotkey VSA ................................................................................................................339

6.9.2 Hotkeys of Option ......................................................................................................339

6.9.3 FREQ Key ..................................................................................................................352

6.9.4 SPAN Key ..................................................................................................................352

6.9.5 AMPT Key ..................................................................................................................352

6.9.6 MKR Key ....................................................................................................................354

6.9.7 MKR -> Key ...............................................................................................................354

6.9.8 MKR FCTN Key .........................................................................................................355

6.9.9 BW Key ......................................................................................................................356

6.9.10 SWEEP Key ...............................................................................................................356

6.9.11 MEAS Key - not available ..........................................................................................357

6.9.12 TRIG Key ...................................................................................................................357

6.9.13 TRACE Key ................................................................................................................357

6.9.14 LINES Key .................................................................................................................358

6.9.15 DISP Key ...................................................................................................................359

6.9.16 FILE Key ....................................................................................................................360

Page 7

R&S FSQ-K70/FSMR/FSU-B73 Contents

Software Manual 1161.8073.42 - 13 7

6.9.17 CAL Key .....................................................................................................................361

6.9.18 SETUP Key ................................................................................................................362

6.9.19 HCOPY Key ...............................................................................................................364

6.9.20 Hotkey Bar .................................................................................................................364

6.10 STATus-QUEStionable:SYNC Register .................................................................365

6.11 STATus-QUEStionable:POWer Register ...............................................................365

7 Programming Examples ................................................................. 366

7.1 Performing a Measurement with OPC Synchronization ......................................366

7.2 Creating a Search Pattern .......................................................................................367

8 Checking the Rated Specifications ............................................... 369

8.1 Required Test Equipment and Accessories .........................................................369

8.2 Test Sequence ..........................................................................................................369

9 Utilities /External Programs ........................................................... 371

9.1 Mapping Editor (MAPWIZ) .......................................................................................371

9.2 Filter Tool (FILTWIZ) ................................................................................................372

10 Glossary and Formulare ................................................................. 373

10.1 Trace-based Evaluations ........................................................................................373

10.2 Summary - Evaluations ...........................................................................................374

10.3 Statistical Evaluations .............................................................................................376

10.4 Trace Averaging and Marker Functions ................................................................376

10.5 Averaging RMS Quantities ......................................................................................377

10.6 Analytically Calculated Filters ................................................................................377

10.7 Standard-Specific Filters ........................................................................................378

10.8 Analytically Calculated Filters ................................................................................379

10.9 Standard-Specific Filters ........................................................................................380

10.10 Abbreviations Used .................................................................................................381

11 Index ................................................................................................ 383

Page 8

Page 9

R&S FSQ-K70/FSMR/FSU-B73 Vector Signal Analysis

Software Manual 1161.8073.42 - 13 9

1 Vector Signal Analysis

VSA

SPECTRUM

SCREEN B

When equipped with application firmware R&S FSQ-K70 or the VSA Extension R&S FSMR/FSU-B73, the Analyzer R&S FSQ/FSU/FSUP or the Measuring Receiver R&S FSMR performs vector measurements on digitally modulated signals in the time domain. Based on the vector measurements, further evaluations, e.g. statistical evaluations or distortion measurements can be performed.

1.1 Enabling the Firmware Option

Firmware option R&S FSQ-K70/FSMR-B73/FSU-B73 is enabled by entering a keyword in the SETUP  GENERAL SETUP menu. The keyword is supplied with the option. If the option is factory-installed, it is already enabled.

GENERAL SETUP Menu:

OPTIONS

The OPTIONS softkey opens a submenu where the keywords for new firmware options (application firmware modules) can be entered. Available options are listed in a table displayed when the submenu is opened.

INSTALL OPTION

The INSTALL OPTION softkey activates the keyword entry field of a firmware option. One or more keywords can be entered in the entry field. If a valid keyword is entered,

OPTION KEY OK is displayed and the option is added to the FIRMWARE OPTIONS table.

If an invalid keyword is entered, OPTION KEY INVALID is displayed. After installation of the option, VSA ( = vector signal analysis) is displayed in the

hotkey bar of the R&S FSQ/FSMR/FSU. The position of the VSA hotkey may vary depending on the type and number of options installed.

Fig. 1 Hotkey bar of basic unit with option R&S FSQ-K70/FSMR-B73/FSU-B73 installed.

Page 10

R&S FSQ-K70/FSMR/FSU-B73 Vector Signal Analysis

Software Manual 1161.8073.42 - 13 10

1.2 Test Setup for Measurement on Base Stations and Power

Danger of electric shock or from radiation

The relevant safety standards (e.g. EN 60215 and IEC215) must be complied with when operating transmitters and amplifier output stages.

Destruction of the input mixer When transmitters or transmitter output stages with an output power of more than 30

dBm are connected, a suitable power attenuator or power coupler must be used to prevent the analyzer input stages from being damaged.

For R&S FSQ/FSMR/FSU devices with an upper frequency limit of 26.5 GHz or less, the RF input is AC-coupled with switchable AC/DC coupling. For all other R&S FSQ/FSMR/FSU devices (upper frequency limit > 26.5 GHz), the RF input is DCcoupled.

For AC-coupling, a DC input voltage of 50 V must never be exceeded. For DCcoupling, DC voltage must not be applied at the input.

In both cases, noncompliance will destroy the input mixers.

Amplifiers

Special precautions are to be observed when measurements on power amplifiers and mobile radio base stations are performed.

1.2.1 Precautions

1.2.2 Standard Test Setup

Fig. 2 Connection to RF output of a base station (for example R&S FSQ)

Page 11

R&S FSQ-K70/FSMR/FSU-B73 Vector Signal Analysis

Software Manual 1161.8073.42 - 13 11

1.3 Calling and Exiting the Option - VSA Softkey

EXIT VSA

DEFAULTS

SCREEN B

SETTINGS

HOME VSA

SCREEN C

1.3.1 Calling the Option - VSA Softkey

Call the R&S FSQ-K70/FSMR-B73/FSU-B73 option by pressing the VSA hotkey. After activation, the labels in the hotkey bar and the contents of the menus are adapted

to the functions of the VSA option. The menus of the option are described in Chapter 5, "Instrument Settings and Measurements".

Remote: INST:SEL DDEM

Fig. 3 Hotkey bar when option R&S FSQ-K70/FSMR-B73/FSU-B73 is active

1.3.2 Exiting the Option - VSA Softkey

To exit the R&S FSQ-K70/FSMR-B73/FSU-B73 option, press the EXIT VSA hotkey. When the option is closed, the hotkey bar and the menus of the basic unit are restored. When the option is closed, the hotkey bar and the menus of the basic unit are restored.

Remote: INST:SEL SAN

1.3.3 Return to VSA Menu (Home VSA Hotkey)

HOME VSA

Pressing HOME VSA in any position of the VSA menu branches to the VSA menu. This function should be used particularly after frequency, level and trigger settings,

because automatic return to the VSA menu is not possible in this case.

Page 12

R&S FSQ-K70/FSMR/FSU-B73 Vector Signal Analysis

Software Manual 1161.8073.42 - 13 12

1.3.4 Overview

Analyzer mode VSA

SPECTRUM

VSA

EXIT VSA

Starting the

application

Exiting the

application

Return to main menu

of R&S FSK-K70

VSA mode

(R&S FSQ-K70/FSMR/FSU-B73)

HOME VSAEXIT VSA

Key

HOME VSA

Operation within

the application

The following functions are shown by the diagram below:

● Starting R&S FSQ-K70/FSMR-B73/FSU-B73 in the spectrum analyzer mode

● Navigation within the application

● Exiting the application

The position of the VSA hotkey may vary depending on the number of activated options.

Fig. 4 Overview: calling and ex iting option FSQ-K70/FSMR/FSU-B73

Page 13

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 13

2 First Measurements - Getting Started

With the aid of a few sample measurements for the digital GSM and EDGE standards, this chapter gives a quick introduction to typical vector analyzer measurements. The individual measurements are in logical order and should familiarize the user gradually with the measurements required of general vector signal analysis. To benefit from this didactics, use the „Continuous – Facing“ view for the display on the screen. The following equipment is required in addition to the Analyzer R&S FSQ/FSU/FSUP/FSG or Measuring Receiver R&S FSMR with option R&S FSQK70/FSMR-B73/FSU-B73:

● 1 test transmitter (GSM-compatible), preferably R&S SMIQ (1125.5555.03)

● 1 ParData Adapter R&S SMIQ-Z5 for R&S SMIQ (1104.8555.02)

● 1 RF cable with 2 male N connectors

● 2 RF cable with 2 male BNC connectors

● 2 power cables

Transmitter operation is only described as far as required for performing the measurements. For more details on the measurements, refer to the test transmitter documentation.

2.1 Interconnecting Transmitter and Analyzer

Fig. 5 Connection to a test transmitter (for example R&S FSQ)

Page 14

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 14

2.2 Basic Settings of Test Transmitter

Parameter

Setting

Level

0 dBm

Frequency

2 GHz

Setting

Operatingsequence SMIQ

Grundeinstellung für GSM / EDGE Softkey Digital Standard

Digital Standard GSM/EDGE State ON

Setting

Measureme nt

EDGE Single Burst

Save/Recall Frame Get predefined Frame EDGE0

EDGE Full Frame

2,3,5,6,7

Save/Recall Frame Get predefined Frame EDGE_ALL

GSM/EDGE Mixed Frame

Save/Recall Frame Get predefined Frame GSM_EDGE

GSM Full Frame

Save/Recall Frame Get predefined Frame GSM_ALL

EDGE Slot Att. (20 dB / slot 1..7)

Slot Attenuation 20 Select Slot Slot 1..7 Slot Level ATTEN

The following frequency and level settings are made on the test transmitter for the measurements below:

Table 1 Basic settings of test transmitter for first measurements

Transmitter settings for the various measurements are listed in the table below:

Table 2 Transmitter settings for various measurements

2.3 Switching On the R&S FSQ-K70/FSMR-B73/FSU-B73 Option

Hotkeys VSA Press the VSA hotkey to call the R&S FSQ-K70/FSMR-B73/FSU-B73 option. After activation, the labels in the hotkey bar and the contents of the menus are adapted

to the functions of the VSA option. The menus of the option are described in Chapter 5, "Instrument Settings and Measurements"

Page 15

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 15

2.4 Basic Analyzer settings for EDGE Measurements

Parameter

Setting

Frequency

2 GHz

Reference level

+6 dBm

Parameter

Setting

Digital standard

EDGE_NB

Sweep

CONTINUOUS

Burst search

Pattern search

Pattern

EDGE_TSC0

Display mode

Screen A: EVM Screen B: Symbols & Modulation Accuracy

In the default setting after PRESET, the R&S FSQ/FSMR/FSU is in the analyzer mode. In this mode the following settings must be made:

Table 3 Basic instrument settings

The following settings of the R&S FSQ-K70/FSMR-B73/FSU-B73 option are only enabled after the vector signal analyzer mode is set and the digital standard EDGE_NB (normal burst) is selected.

Table 4 Basic setting for vector signal analysis measurements

2.5 Measurement 1: Demodulation of a Single EDGE Burst

Objective of the measurement:

● Demodulation of a single EDGE burst and result display

● Switchover of result display to I/Q VECTOR

● Disabling the measurement filter and measuring the raw transmitter signal

Instrument settings:

► Transmitter: GSM default setting

EDGE Single Burst

► Analyzer:: Analyzer: Digital GSM standard  EDGE_NB standard

Adjust Ref Level

Page 16

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 16



Fig. 6 Measurement 1: Frame structure

The burst numbers in the drawing correspond to the timeslots of the GSM frame

(EDGE)

External Frame Trigger

(EDGE)

= Burst, Slot belegt

= Slot nicht belegt

= Burst, abgesenkter Pegel

structure. The transmitter settings cause a single EDGE burst in time slot 0. The time slots 1 to 7

are not assigned.

Page 17

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 17

Measurement:

RMS-EVM:

< 0.5%

Center Frequency Error:

< 2 Hz

Fig. 7 shows a typical result display of the analyzer for the EDGE standard. In the upper half, the magnitude of the vector error is plotted over time; in the lower half

numeric error values in the range of the evaluation lines are listed.

Fig. 7 Measurement 1: Result display of analyzer

For this kind of measurement with adequately set reference level and synchronization of reference oscillators between transmitter and analyzer, the following results should be displayed.

The EDGE measurement must be performed with the measurement filter prescribed by ETSI. If DIGITAL STANDARD EDGE is selected, this filter is automatically switched on.

With the control sequence <SCREEN A>, <MEAS RESULT>, <MEAS SIGNAL>, <I/Q VECTOR>, the associated I/Q trace is displayed (after filtering with the measurement filter, Fig. 8). With the sequence <MEAS RESULT>, <RESULT RAW>, this filter is switched off and the measurement is performed on the raw transmitter signal (before filtering with the measurement filter). The associated display is shown in Fig. 9.

Page 18

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 18

Fig. 8 Measurement 1: I/Q vector

Fig. 9 Measurement 1: RESULT RAW

Switching off the measurement filter may also influence the numeric result display: high-frequency noise components that are to a great extent suppressed by the filter may cause more measurement errors.

Page 19

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 19

2.6 Measurement 2: Selection of a Specific Slot with Trigger



(EDGE)

External Frame Trigger

TRIGGER OFFSET RECORD LENGTH

(EDGE)

Offset

Objective of the measurement:

● Selecting a single EDGE burst by external trigger

● Changing the position of the trace in the display with FIT TRACE

● Reducing the RECORD LENGTH

Instrument settings:

► Transmitter: GSM default setting

EDGE Full Frame

► Analyzer: Digital GSM standard  EDGE_NB standard

1) <MEAS RESULT> <MAG CAP BUFFER>

2) <MEAS RESULT> <RESULT RAW> <MEAS RESULT> <MEAS SIGNAL> <MAGNITUDE ABSOLUTE>

The transmitter settings cause EDGE bursts in time slots 0 to 7.

Page 20

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 20

Measurement:

In the default setting, the TRIGGER OFFSET is set to -100 s and the RECORD LENGTH to 10 times the RESULT LENGTH. The received raw signal is displayed (magnitude capture buffer, Fig. 10).

With this setting the first detected pulse is demodulated. The name of the detected sync pattern that is used for synchronization is displayed (EDGE_TSC0, Fig. 11).

During the measurement, the TRIGGER OFFSET can be varied with the rotary knob until the EDGE_TSC3 sync pattern is displayed. Stable demodulation is achieved with a trigger offset of +1.1 ms.

Fig. 10 Meas. 2: Magnitude capture buffer

Fig. 11 Meas. 2: EDGE_TSC0

Display positioning

When GSM / EDGE is set, FIT PATTERN TO CENTER is selected for the display: the center of the detected sync pattern is represented in the center of the display.

Other possible settings are shown in the figures below:

● FIT TRIGGER TO LEFT: trigger time + trigger offset are displayed at the left screen

edge

Page 21

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 21

● FIT PATTERN TO LEFT: the beginning of the sync pattern is displayed at the left

screen edge

Fig. 12 Meas. 2: FIT TRIGGER TO LEFT

Fig. 13 Meas. 2: FIT PATTERN TO LEFT

Changing the RECORD LENGTH

To speed up the measurement, the data recording time (RECORD LENGTH) can be manually reduced (set RECORD LENGTH = 250 symbols). In some cases, display positioning with FIT TRACE and 'pattern aligned’is no longer possible.

Page 22

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 22

2.7 Measurement 3: Setting the Burst Search Parameters



Fig. 14 Burst-search parameter

(EDGE)

External Frame Trigger

(EDGE)

(LEVEL)

Objective of the measurement:

● Manual setting of burst parameters

● Selective search for sync patterns

Instrument settings:

► Transmitter: GSM default setting

EDGE Full Frame Blank slot 0 and slot 2 Reduce level of slot 1 by 15 dB

► Analyzer: Digital GSM standard  EDGE_NB standard

1) <DISPLAY><SPLIT SCREEN> <DISPLAY><SCREEN B> <MEAS RESULT> <MAG CAP BUFFER> <DISPLAY><SCREEN A> <MEAS RESULT> <MEAS SIGNAL> <MAGNITUDE ABS>

2) <DISPLAY><FULL SCREEN> <MEAS RESULT> <MEAS SIGNAL> <MAGNITUDE ABS>

This basic transmitter setting causes a single burst with reduced level in timeslot 1 and a sequence of bursts in timeslots 3 to 7.

Page 23

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 23

Measurement:

n the previous measurement, a defined burst was selected for the measurement by means of an external trigger signal. If a suitable measurement signal is available, the specific burst can also be selected by manual setting of burst search parameters without external trigger.

The signal consists of a single burst of reduced level and a sequence of bursts of normal level. In automatic burst search, the level threshold depends on the maximum amplitude and slots 3 to 7 are measured. The single burst in slot 2 is not detected. Fig. 15 and Fig. 16 show different untriggered measurements in the AUTO mode.

Fig. 15 Meas. 3: Burst search AUTO, EDGE_TSC4

Fig. 16 Meas. 3: Burst search AUTO, EDGE_TSC3

In the next step, the burst search is set with a level threshold of -30 dB RefLvl (relative to reference level). Because of manual threshold setting, the level-reduced burst in slot 1 is now also detected and demodulated. Fig. 17 shows such a measurement.

Page 24

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 24

Fig. 17 Meas. 3: Burst search, manual level setting

Knowing that slot 1 contains a single burst, the settings for the burst search can be even more selective:

Under <BURST & PATTERN> <EXPERT SETTINGS>, the GAP LENGTH (i.e. the gap between two consecutive bursts) is increased to 50 symbols.

The search algorithm now rejects all bursts in slots 3 to 7 and only identifies the burst in slot 1 as valid because this burst is between two empty timeslots and the only one in the frame to fulfill the burst conditions (see Fig. 17).

2.8 Measurement 4: Suppression of Incorrect Measurements

Objective of the measurement:

● MEAS ONLY ON PATT operating parameter

● Similarity of GSM and EDGE patterns

Instrument settings:

► Transmitter: GSM default setting

GSM Mixed Frame

► Analyzer: Digital GSM standard  EDGE_NB standard

Page 25

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 25



The transmitter settings cause bursts in time slots 1 to 7. GSM and EDGE bursts are

(GSM)

(EDGE)

(GSM)

(EDGE)

(GSM)

(EDGE)

(GSM)

(EDGE)

External Frame Trigger

transmitted alternately.

Measurement:

The signal consists of a fully used frame in which EDGE and GSM bursts are transmitted alternately. In contrast to the standard setting for EDGE_NB, the MEAS ONLY ON PATT parameter is switched off. As a result, the analyzer tries to demodulate each burst that fulfills the burst conditions.

The EDGE demodulation algorithm is optimized for 3pi/8-8PSK modulation. It also synchronizes to GSM signals patterns of identical name, but a great number of error messages are issued in this case.

In the case of untriggered measurements, the following result displays may be obtained.

Fig. 18 Meas. 4: EDGE demodulator, correct demodulation

Page 26

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 26



Fig. 19 Meas. 4: EDGE demodulator, incorrect demodulation of a GSM burst

The incorrect measurements can be avoided when the following settings are made:

● Select appropriate patterns for the EDGE signal (e.g. EDGE_TSC1,

EDGE_TSC3,EDGE_TSC5, EDGE_TSC7)

● Activate MEAS ONLY ON PATT softkeys

The display is only updated after a valid measurement. After a faulty measurement the display remains unchanged and the SEARCHING PATTERN message is displayed.

Despite the similarity of the GSM and EDGE sync patterns, the GSM demodulator is not able to identify EDGE patterns. To suppress invalid measurements (pattern not found), the MEAS ONLY ON PATT softkey must also be activated.

2.9 Measurement 5: Evaluation Lines

Objective of the measurement:

● Use of evaluation lines for determining result ranges

Instrument settings:

► Transmitter: GSM default setting

GSM Full Frame

► Analyzer: Digital GSM standard  GSM_NB standard

1) <SCREEN A> <MEAS RESULT> <MAGNITUDE ABS> <SCREEN B> <MEAS RESULT> <SYM & MODUL ERR>

2) <SCREEN B> <MEAS RESULT> <MAGNITUDE ABS> <SIGNAL STATISTIC>

Page 27

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 27

The transmitter settings cause GSM bursts in time slots 0 to 7.

External Frame Trigger

(GSM)

Measurement:

Evaluation lines delimit the range in which numeric results such as EVM, phase error, magnitude error, RHO are determined. The range is preset and automatically considered when a digital standard is set.

In the first figure below, the EVAL LINES are correctly set; in the second, they are set on the burst edge.

Fig. 20 Meas. 5: Setting the evaluation range: presetting the standard

Fig. 21 Meas. 5: Setting the evaluation range: extension to burst edges

Page 28

R&S FSQ-K70/FSMR/FSU-B73 First Measurements - Getting Started

Software Manual 1161.8073.42 - 13 28

The evaluation lines also affect derived displays such as statistical signal evaluation. Fig. 22 shows the statistical level distribution within the burst. In Fig. 23, the EVAL LINES are extended to ranges outside the burst which is reflected by the level's probability of occurrence.

Fig. 22 Meas. 5: Level distribution within the burst

Fig. 23 Meas. 5: Level distribution within and outside the burst

The displayed measurements were performed in the SINGLE SWEEP mode. The display at the right was obtained solely by varying the EVAL LINE 1 without receiving new data. For this reason the measurement is marked with a red asterisk *. Parameters relating to this measurement (e.g. modulation errors or statistics diagrams) are recalculated, however.

Page 29

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 29

3 Brief Description of Vector Signal Analysis

A D

Decimation

Filters

I Buffer

16 M

Q Buffer

16 M

Signal

Processor

NCO

20.4 MHz

Halfband

Filter

Resampling

Ratio

cos

sin

Sampling Rate =

81.6 MHz... 10 kHz

Trigger

Sampling Rate =

81.6 MHz.

IF=20.4

MHz

Equalizer

Filter

Resampler

Decimation Record Buffer DSP

(Function)

The "Vector Signal Analysis" software option R&S FSQ-K70/FSMR-B73/FSU-B73 performs vector measurements for analyzing modulation errors of RF signals converted to the complex baseband. Carrier envelope and time domain measurements can also be performed but these measurements can be carried out in the basic unit (frequency analyzer) with a considerably wider bandwidth. The same applies to spectral measurements such as adjacent-channel power measurements on mobile radio signals.

The following sections describe the digital signal processing hardware, the interplay of analog and digital filters for bandwidth limiting, system-theoretical modulation and demodulation filters as well as the algorithms used by the measurement demodulator. The implemented modulation modes and the associated predefined symbol mappings are also listed.

The last part of this chapter deals with vector and scalar modulation errors. The required calculation formulae are provided in the Annex to this manual.

3.1 Block Diagram of Digital Signal Processing Hardware

Fig. 24 Block diagram of digital hardware for vector signal analysis

3.1.1 Description of Block Diagram

After having passed several RF, IF and filter stages, the RF input signal is converted to an IF of 20.4 MHz and applied to an A/D converter with a sampling frequency of exactly 81.6 MHz.

The digitized signal is then routed through two ICs for resampling (conversion of sampling rate by a real factor) and for filtering and decimation (reduction of sampling rate by an integral factor). An EQUALIZER FILTER is connected to the RESAMPLER input to compensate for the frequency response of the analog filter stages which would otherwise add to the modulation errors.

Page 30

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 30

During operation, the filters and decimation factors of the instrument are set so that a

.In addition to setting the modulation mode, ACCURATE setting of symbol rate and filter parameters is important for a correct demodulation. Even slight deviations may noticeably impair the measurement result.

Examples are given in the Troubleshooting section.

A D

decimation

filters

I-Memory

16 M

Q-Memory

16 M

Processor

NCO

20.4 MHz

Halfband

Filter

Resampling

ratio

cos

sin

sampling rate =

326.4 MHz... 163.2 MHz

sampling rate =

326.4 MHz...>81.6MHz

Trigger

IF-Filter

120 MHz

sampling rate =

326.4 MHz.

IF=

408 MHz

sampling frequency is obtained at the output of the DECIMATION stage, which exactly corresponds to the following equation:

Sampling rate = Symbol rate * Points/symbol {4,8, or 16}; A higher point/symbol setting automatically results in a corresponding increase of the

I/Q bandwidth. The resulting measurement bandwidths are described in the sections below.

The complex output signal of the DECIMATION stage is stored in the I/Q memory (RECORD BUFFER) and forwarded to a signal processor (DSP) for further processing.

The data recording length and the result length after DSP processing are limited to about 32k samples (irrespective of the set symbol rate or sampling rate).

The received baseband signal is filtered in the subsequent DSP stage as required by the signal, then demodulated without the transmitted data being known (non-dataaided demodulator) and scanned for sync patterns. An ideal transmit signal is reconstructed from the demodulated data, and various modulation and vector errors, which are described in the following sections, are obtained from a comparison of demodulated and ideal I/Q signals.

Supplement to the R&S FSQ-B72 Option

The R&S FSQ-B72 option additionally allows sampling rates from >81.6 MHz to 326.4 MHz. With sampling rates 81.6 MHz, the R&S FSQ-B72 option is not active. The analyzer then behaves in the way described above. Fig. 25 shows the hardware of the analyzer from IF up to the processor for sampling rates above 81.6 MHz. An IF filter of 120 MHz is effective. The A/D converter samples the IF (408 MHz) at a rate of 326.4 MHz. The points/symbol setting parameter is fixed at {4}.

Fig. 25 Block diagram with the signal processing of the R&S FSQ at sampling rates >81.6 MHz

3.1.2 Bandwidths for Signal Processing

Relevant filters for vector signal analysis are shown in the block diagram below.

Page 31

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 31

Halfband Filter

(Baseband)

Decimation Filter

(Baseband)

IF Filter

Measurement Filter

(Baseband)

RBW

(= Resolution

Bandwidth)

IQ-

Bandwidth

Demodulation

Bandwidth

Analog Section Digital Hardware Section DSP Section

Equalizer Filter

(Digital IF)

Fig. 26 Block diagram of bandwidth-relevant filters for vector signal analysis

The total bandwidth is obtained when the shown filter stages are series-connected:

● IF filter (RBW) with selectable nominal bandwidths 120 MHz

50 MHz

**)

, 20 MHz

10 MHz, 5 MHz, 3 MHz, 1 MHz and 300 kHz

● Digital hardware filter (in RESAMPLER and DECIMATION blocks)

● Measurement filter (MEAS FILTER) in the signal processor

Digital filters in the digital hardware section:

● Equalizer filter for compensating amplitude and phase distortions of RBW filters

● Halfband filter for limiting the bandwidth to approx. 40 MHz or 160 MHz (if R&S FSQ-B72 is active)

● Decimation filter for limiting the bandwidth to 0.8 times the output sampling rate. Note: In case of very high sampling rates, this filter is bypassed.

**)

In the DSP section, the demodulation bandwidth can be further reduced by a measurement filter. If this filter is not required for the measurement, measurements are performed with the I/Q bandwidth.

Equalizer filter and halfband filter are only of minor importance for the total bandwidth. The other filters and the filters required for intersymbol-interference-free (ISI-free) demodulation are described in detail in the sections below.

3.1.2.1 Analog RBW Filters

The spectrum of the receive signal is reduced by means of analog prefilters so that the IF stages of the analyzer are optimally driven by the desired signal and undesired mixer products are reduced.

To obtain optimum characteristics for vector signal analysis, the amplitude and phase frequency response within the demodulation bandwidth should be as flat as possible. The permissible IF filters are listed in the table below.

only if R&S FSQ-B72 is active; fixed at 120 MHz

**)

available for R&S FSQ and R&S FSMR only

Page 32

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 32

Filter bandwidths 3 MHz are equalized by means of a built-in calibration procedure

RBW operating parameter

Digitally compensated

Usable bandwidth (effect on filter negligible)

UNCAL display if usable bandwidth is <

300 kHz

1/10*300 kHz = 30 kHz

Symbol rate * Points/symbol

500 kHz

1/10*500 kHz = 50 kHz

Symbol rate * Points/symbol

1 MHz

1/10*1000 kHz = 100 kHz

Symbol rate * Points/symbol

3 MHz

2 MHz

Symbol rate * Points/symbol

5 MHz

3 MHz

Symbol rate * Points/symbol

10 MHz

7 MHz

Symbol rate * Points/symbol

20 MHz*)

17 MHz

Symbol rate * Points/symbol

50 MHz*)

28 MHz

120 MHz**)

120 MHz

and can be used for up to 2/3 of the nominal bandwidth (unless stated otherwise in the table). The maximum equalized IF signal bandwidth that can be used is limited to 28 MHz or 120 MHz (if R&S FSQ-B72 is active).

Filter bandwidths <3 MHz are not equalized and can be used for vector signal analysis up to approx. 1/10 of the nominal bandwidth without noticeably affecting the modulation error. Using the bandwidth above this limit considerably reduces the measurement accuracy.

Unless special measures are required for interference suppression, we recommend using the RBW = AUTO setting.

With RBW = AUTO, the analog RBW filter is set by the analyzer so that the "bandwidth used" (see table below) is wider or equal to the bandwidth of the subsequent digital filter stages.

With RBW = MANUAL, the filter bandwidth specified in the table below may be reduced. If a usable filter bandwidth below the Symbol rate * Points/symbol bandwidth is selected, UNCAL is displayed.

Table 5 RBW filter bandwidths and usable bandwidths

available for R&S FSQ and R&S FSMR only

**) only if R&S FSQ-B72 active; other bandwidths cannot be set

Page 33

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 33

3.1.2.2 I/Q Bandwidth

Sampling rate

sample

[MHz]

RBW bandwidth

Equivalent IF BW (halfband filter)

Equivalent IF BW (decimation filter)

81.6...326.4 MHz*)

120 MHz

approx. 160 MHz

0.68* f_sample

81.6...100 MHz (Interpolation)

Equalized RBW, max. 28**)

approx. 40 MHz

0.35* f_sample

40.8 ... 81.6

Equalized RBW, max. 28**)

approx. 40 MHz

20.4 ... 40.8

Equalized RBW, max. 28**)

approx. 40 MHz

0.68* f_sample

< 20.4

Equalized RBW, max. 28**)

approx. 40 MHz

0.8* f_sample

Parameter POINTS / SYM

IQ baseband-BW (single side)

IQ-IF-BW (double side)

Example: IQ-IF-BW (f_symbol = 100 kHz)

1, 2, 4

(0.8 * F_symbol/2) * 4

(0.8 * F_symbol/2) * 4 *2

360 kHz

4*) (fixed)

(0.68 * F_symbol/2) * 4

(0.68 * F_symbol/2) * 4 *2

(0.68 * F_symbol/2) * 8

(0.8 * F_symbol/2) * 8 *2

720 kHz

(0.8 * F_symbol/2) * 16

(0.8 * F_symbol/2) * 16 *2

1440 kHz

Table 6 specifies the I/Q bandwidth that can be achieved as a function of the sampling rate.

For sampling rates between 40.8 MHz and 81.6 MHz, the bandwidth is limited to approx. 40 MHz by the halfband filter but the RBW of the preceding IF filter (max. 28 MHz, R&S FSU max. 10 MHz) is decisive for the total bandwidth. A decimation filter is not active with this setting.

For lower sampling rates, the bandwidth of the decimation filter is decisive provided no narrower (equalized) RBW is set.

Sampling rates between 81.6 MHz and 100 MHz are achieved by sampling at a fixed rate of 81.6 MHz followed by interpolation. Although a decimation filter is activated again in this mode, the RBW of the IF filter is the determining factor for the total bandwidth. If the R&S FSQ-B72 option is activated, an RBW of 120 MHz, a halfband filter of 160 MHz, as well as a bandwidth of the decimation filter of 0.68 * F_symbol/2 is always active.

Table 6 Maximum I/Q bandwidths of data recording

*) only if R&S FSQ-B72 active

**) or R&S FSU max. 7 MHz The table below shows the effect of the symbol rate and of points/symbol parameters

on the sampling rate.

Table 7 I/Q bandwidth as a function of POINTS/SYM setting

*) only if R&S FSQ-B72 active

For common PSK, QAM and MSK systems, signal sampling with 4 points/symbol fulfills the system-theoretical requirements for a measurement demodulation.

A higher oversampling rate yields a better resolution of displayed traces but it may cause more measurement errors if the extended I/Q bandwidth contains interferences (and the measurement bandwidth corresponds to the I/Q bandwidth). An example is given in the following section.

Page 34

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 34

With FSK systems, oversampling must be set to match the modulation index so that no

Signal

Spectrum

Signal

Spectrum

Signal

Spectrum

Meas_Filter

IQ_Filter

(4 Pts/Symbol)

IQ_Filter

(8 Pts/Symbol)

IQ_Filter

(16 Pts/Symbol)

Interferer

Meas_Filter

fa/2

123

modulation errors are produced by I/Q filtering.

3.1.2.3 Demodulation Bandwidth (Measurement Bandwidth)

The demodulation bandwidth is the part of the spectrum used for demodulation and measurement of the digitally modulated signal. In most cases, the spectrum is routed through a receive filter to obtain intersymbol-interference-free conditions permitting optimum symbol decision. After this receive filter, the modulation error is also measured. For this reason the term MEASUREMENT FILTER (Meas_Filter) is used here. A few modulation systems, especially MSK and FSK, do not use this input filtering. In these cases special care should be taken that no interference or adjacent channels occur within the demodulation bandwidth.

The figure below shows the demodulation bandwidths with different settings of the oversampling rate.

Fig. 27 Selected oversampling rates (I/Q bandwidth, interference)

Fig. 27 shows the spectrum of a digitally modulated signal that was sampled with the oversampling rates 4 ( ), 8 ( ) and 16 ( ).

In addition to the signal spectrum - which is identical in all three cases - different I/Q bandwidths and a single-frequency interfering signal are shown.

If a demodulation or measurement filter is used, the interferer is suppressed in all three examples and the measurement bandwidth corresponds to that of the measurement filter.

If no filter is used for the measurement, the interfering carrier is suppressed by the I/Q filter only in example 1; in example 2 it is partly suppressed and in example 3 not at all.

The same effect occurs if the measurement filter is switched off for special measurements on unfiltered PSK and QAM signals (RESULT = RAW setting).

Page 35

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 35

Typical PSK systems prescribe special receive or measurement filters (e.g. root-raised

Filter

ISI

Filter

MEAS

Filter

Signal

Pro-

cessing

Demod

IQ Receive

Signal

Transmitter

Analyzer

(ISI Demodulation)

Correction Parameter

Analyzer

(Meas Demodulation)

REF

Filter

Symbols

Ideal IQ

Reference Signal

IQ Measurement

Signal

REF_FILTER =

TX_FILTER *

MEAS_FILTER

IQ Bandwidth

cosine receive filter or EDGE measurement filter). If no such filtering is performed, care should be taken that neither interfering signals nor adjacent channels fall within the demodulation bandwidth.

3.1.2.4 System-Theoretical Modulation and Demodulation Filters

Sampling points are required for demodulation in the analyzer, where only information of the current symbol and none of neighbouring symbols is present (symbol points). These points are also called ISI-free points (ISI = intersymbol interference). If the transmitter does not provide an ISI-free signal after the transmit filter, this condition can be fulfilled by signal-specific filtering of the analyzer input signal (ISI filter). If an RRC (root-raised cosine) filter is used in the transmitter, an RRC filter is also required in the analyzer to obtain ISI-free points.

n many PSK systems, RRC filters are used as transmit, ISI and measurement filters. To determine the I/Q measurement error, the measurement signal must be compared with the I/Q trace of an ideal signal. For this purpose a REFERENCE FILTER is required which is calculated by the analyzer from the coefficient convolution of the transmit filter (TX FILTER) and the MEAS FILTER (see Fig. 28, RESULT = FILT).

If unfiltered signals have to be measured as well (e.g. to determine nonlinear signal distortions), no measurement filter is switched into the signal path and the REFERENCE FILTER is identical with the Tx filter (see Fig. 29, RESULT = RAW)

In the baseband block diagrams below, the system-theoretical transmitter and analyzer filters are shown for PSK, QAM and VSB demodulation. For the sake of clearness, RF stages, RBW filters and the filter stages of the digital hardware section are not shown.

Fig. 28 Block diagram of filters in the PSK mode (RESULT = FILT setting)

Page 36

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 36

Mod. type

Modulation filter (transmit filter)

Demodulation filter = receive filter (analyzer)

Measurement filter (analyzer)

Remarks

PSK, QAM, VSB

RC (Raised Cosine)

- - ISI system

PSK, QAM, VSB

RRC (Root Raised Cosine)

RRC

ISI system

FSK

Gauss

- - Near ISI system

MSK

Gauss

- - Near ISI system

EDGE

GAUSS_LINARIZED

EDGE_ISI

EDGE_MEAS

Standard specific filters NO ISI system!

Cdma2k

CDMA2k_1X_TX

CDMA2k_1X_ISI

Standard specific filters, but ISI-system

Filter

ISI

Filter

Signal

Pro-

cessing

Demod

IQ Measurement

Signal

IQ Receive

Signal

Transmitter

Analyzer

(ISI Demodulation)

Correction Parameter

Analyzer

(Meas Demodulation)

REF

Filter

Symbols

Ideal IQ

Reference Signal

REF_FILTER

TX_FILTER

Fig. 29 Block diagram of filters in the PSK mode (RESULT = RAW setting)

For a correct demodulation, 3 filters have to be accurately specified for the analyzer:

● transmit filter (TX filter): filter characteristic of transmitter

● receive filter (ISI filter): filter characteristic of a receive filter producing intersymbol-interference-free points from the Tx-filtered signal

● MEAS filter: filter used for measurements. In many applications, this filter is identical with the ISI filter.

The REFERENCE filter synthesizes the ideal transmit signal (after MEAS filtering). It is calculated by the analyzer from the above filters (convolution operation TX_FILTER * MEAS_FILTER).

Table 8 Typical combinations of TX, ISI and MEAS filters

Typical combinations of TX, ISI and MEAS filters are shown in the table above; they can be set in the analyzer as a FILTER SET. If RC (raised cosine), RRC (root-raised cosine) and GAUSSIAN filters are used, the ALFA (RC, RRC filters) and BT (GAUSSIAN filters) parameters must be set in addition to the filter characteristic (rolloff factor).

Page 37

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 37

Fig. 30 Block diagram of filter stages in the MSK and FSK modes

Stage

1010

Signal

Pro-

cessing

Demod

Measurement

Signal

IQ Receive

Signal

Transmitter

Analyzer

( Demodulation)

Correction

Parameter

Analyzer

(Meas Demodulation)

REF

Stage

Symbols

Ideal Reference

Signal

REF_FILTER

TX_FILTER

No further band limiting is performed in FSK and MSK systems by MEAS or ISI filters in the signal path. Some parts of signal generation in the transmitter and generation of the reference signal in the analyzer are much more involved. The next section contains detailed block diagrams for signal generation and describes requirements caused by customized filters in the instrument.

3.1.2.5 Design and Use of Customized Filters

The analytical filter types RC (raised cosine), RRC (root-raised cosine) and GAUSSIAN as well as the most important standard-specific filters are already integrated in the basic unit. The requirements described in this chapter should be observed when customized filters are designed.

Customized filters may be useful for the following purposes:

● Development of new networks and modulation methods for which no filters are defined yet.

● Measurements of transmitter characteristics with slightly modified (e.g. shortened) transmitter filters.

Page 38

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 38

Filter for PSK, QAM, USER-QAM and VSB

Mapper

Symbols

1010

Zero

Stuffing

Filter

IQ IQ

0 1 2 3 4 5 6 7 8 9

-2

-1

Inphase

0 1 2 3 4 5 6 7 8 9

-2

-1

Quadrature

time [Symbols]

0 1 2 3 4 5 6 7 8 9

-2

-1

Inphase

0 1 2 3 4 5 6 7 8 9

-2

-1

Quadrature

time [Symbols]

0 1 2 3 4 5 6 7 8 9

-2

-1

Inphase

0 1 2 3 4 5 6 7 8 9

-2

-1

Quadrature

time [Symbols]

Frequ

Mapper

Symbols

1010

Zero Stuffing + RECT

Filter

f f

INTEG exp(j*..)



0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

Fig. 31 Generation of baseband transmit signal (PSK, QAM, USER-QAM and VSB)

Fig. 31 illustrates generation of a QPSK signal in the complex baseband. In an I/Q mapper, logic symbols are mapped onto complex symbols in the I/Q plane. In

the ZERO STUFFING stage, zeros are inserted between the symbols, and this oversampled signal is then filtered in the TX filter stage. For the sake of clearness, the signals in the figures are oversampled with 4 points/symbol.

Filter for FSK / MSK

Fig. 32 Generation of transmit signals (FSK, MSK)

Fig. 32 illustrates the generation of a 2-level FSK signal. An I/Q mapper maps logic symbols onto real Dirac pulses in the frequency-versus-time

plane. In the ZERO STUFFING + RECT stage, each Dirac pulse is replaced by a square pulse of one symbol length. This oversampled signal is then filtered in the TX filter stage.

Page 39

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 39

The INTEGRATOR and EXP stages have nothing to do with filtering; they only convert

Filter

Analog

filter

MEAS

Filter

DUT

Analyzer

(Meas Demodulation)

IQ Measurement

Signal

Error of transfer function E

(f)

Filter

Analog

filter

MEAS

Filter

DUT

Analyzer

(Meas Demodulation)

IQ Measurement

Signal

Adaptive

equalizer

Error of transfer function E(f) Compensation function E-1(f)

the signal to the I/Q plane. As in the previous example, the signals are oversampled by the factor 4.

The following requirement must be met by all customized filters:

● Oversampling rate (f

sample

/ f

● The filter must feature purely real coefficients

● The number of coefficients must be uneven

● The filter must be symmetrical to the central filter coefficient.

3.1.2.6 Adaptive Equalizer Filter

A possible source of high modulation errors of the DUT with PSK and QAM signals is a non-flat frequency response or ripple in frequency response within the modulation bandwidth.

This could be caused by the DUT’s:

● Analog filter sections

● Digital filter sections, if a shortened filter length is used

● Digital arithmetic sections, if a shortened bit-length is used

Fig. 33 Base band schematic of the modulation- and demodulation stages

) of 32 in the time domain

symbol

In the case of low linear distortions an equalizer filter (with reverse frequency response characteristic) is able to compensate the distorted frequency response in order to improve the modulation analysis results (see figure below).

Fig. 34 Base band schematic: compensation of the transfer function’s error by inserting an adaptive equalizer in the receive path

The measurement demodulator’s signal path -including the adaptive equalizer filter- is shown in following figure . In front of the demodulation chain the adaptive filter is arranged. The filter coefficients are adapted in such a way that the mean square value of the error vector magnitude (EVM) is minimized. By comparing the demodulated measuring signal and the ideal signal (generated from the demodulated symbols) a control signal for the equalizer is extracted. When analyzing the filter coefficients (trained equalizer state) with a FFT the compensating transfer function can be gained and from it the error function E(f) can be gathered.

Page 40

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 40

ISI

Filter

MEAS

Filter

Signal

Pro-

cessing

Demod

IQ Receive

Signal

Analyzer

(ISI Demodulation)

Correction Parameter

Analyzer

(Meas Demodulation)

REF Filter

Symbols

Ideal IQ

Reference Signal

IQ Measurement

Signal

Adapt.

Equalizer

control

Sampled

IQ baseband signal

Equalizer

Fig. 35 Base band schematic: compensation of the transfer function’s error by inserting an adaptive equalizer in the receive path

Another range of application is the analysis of an unknown or approximately known transmitter filter. The adaptive filter algorithm delivers a matched receiver filters for an intersymbol-interference-free demodulation when the following filter setting is set.

● Transmit-Filter = raised cosine

● Receive-Filter = none

● Measurement-Filter = none

The algorithm is limited to PSK and QAM modulation schemes, because of the optimization criterion of the algorithm is based on minimizing the mean square error vector magnitude. So it cannot be used for MSK, FSK and VSB schemes.

3.1.2.7 Training process of the equalizer

During operation of the equalizer we have to distinguish between two states: TRAIN The equalizer is trained; the filter coefficients are continual adjusted by using

the current demodulation results in order to minimize the RMS EVM. This process needs a lot of calculation so that the measurement update rate of the instrument decreases distinctly.

FREEZE The current filter coefficients are frozen, that means they no longer adapted.

The display update rate increases distinctly again

Training phase of the adaptive equalizer starts

The screen plot (upper diagram) shows a broad distribution of the constellation points (dots) around the ideal decision points (cross hairs)

The magnitude of the filter coefficients is shown in the lower part of the diagram in logarithmic scaling. The equalizer has not been trained yet, so a neutral filter is arranged in the signal path (all filter coefficients are zero, only the middle filter tap has the value ‘one’)

Page 41

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 41

Ref

0 dBm

4 MHz

CF 1 GHz

Meas Signal

ConstDiag

200

mU/

CLRWR

Const

Att

25 dB

Ref

0 dBm

4 MHz

CF 1 GHz

Eq Train

Magnitude

CLRWR

Magn

-1 U

1 U

Symbols

dB/

16QAM

FILT

16QAM

SGL

-2.908505 U

2.908505 U

581.701 mU/

-15 sym

15 sym

3 sym/

-40

-35

-30

-25

-20

-15

-10

-5

Ref

0 dBm

4 MHz

CF 1 GHz

Meas Signal

ConstDiag

200

mU/

CLRWR

Const

Att

25 dB

Ref

0 dBm

4 MHz

CF 1 GHz

Eq Train

Magnitude

Symbols

dB/

CLRWR

Magn

-1 U

1 U

16QAM

FILT

16QAM

SGL

-2.908505 U

2.908505 U

581.701 mU/

-15 sym

15 sym

3 sym/

-40

-35

-30

-25

-20

-15

-10

-5

During the training phase

The screenshot (upper diagram) indicates a distinct improvement because of the variance of constellation points distribution has decreased observably. On either side of the adaptive filter’s middle filter tap more non-zero coefficients are coming up (lower diagram). The logarithmic scaling makes the diagram very sensitive to.

Slight variations of the filter coefficients are easy to observe due to the logarithmic scaling of the diagram.

End of the training phase

The screenshot (upper diagram) indicates a nearly perfect constellation diagram. All constellation points are located close to their ideal positions in the cross hairs. The variance of the constellation distribution cannot be observed anymore. The accuracy of equalizer’s coefficients has further improved and the number of non-zero coefficients has slightly increased.

Please note that there are still some zero coefficients, so the filter length could be a little reduced for the shown measurement problem (saves calculation time during the equalizer’s training phase).

Page 42

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 42

Ref

0 dBm

4 MHz

CF 1 GHz

Meas Signal

ConstDiag

581.701 mU/

200

mU/

CLRWR

Const

Att

25 dB

Ref

0 dBm

4 MHz

CF 1 GHz

Eq Train

Magnitude

Symbols

3 sym/

dB/

CLRWR

Magn

-1 U

1 U

-2.908505 U

2.908505 U

-15 sym

15 sym

16QAM

FILT

16QAM

-40

-35

-30

-25

-20

-15

-10

-5

Optimisation Range of

adaptive equalizer filter (with 4 pts/symbol and a RRC 0.22 transmit filter)

Operating range of the Equalizer

The total frequency response can be flattened by the equalizer filter only in the pass-band of the transmitter- and receiver filter respectively. Because of the ideal reference signal

doesn’t generate any signal power outside of the pass-band, the equalizer eliminates most of the measurement signal’s out of band power if necessary. The equalizer’s out-of- band

characteristic is mainly influenced by the existence or not-existence of any interfering signal power (e.g. noise, spurious signals, interfering signals). If there are any interfering out-ofband signals, the equalizer algorithm is going to suppress by its transfer characteristic (high out of band attenuation).

If there are no interfering signals, there is no need for the equalizer to suppress out of band signals (flat but poor out-of-band attenuation). The user has to consider this

behavior when interpreting the filter’s frequency characteristic.

The following figure exemplifies the equalizer’s frequency response for a linear

distorted measurement signal (raise cosine filter, alpha = 0.22). The optimization range is enhanced by red lines. An estimate of the pass-band with the pre-known signal

parameters gives a good approximation to the equalizer’s optimization range as

demonstrated in the figure (signal has a very good signal to noise ratio, therefore the out-of band response is flat):

Filter-bandwidth = symbol-rate*(1+alpha) = 4MHz*1.22=4.88 MHz

Fig. 36 Optimization range of the adaptive equalizer filter

Page 43

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 43

The adaptive equalizer’s out-of-band transfer function is mainly influenced by the

Ref

0 dBm

3.84 MHz

CF 1 GHz

Freq Resp

-7.68 MHz

7.68 MHz

1.536 MHz/

dB/

CLRWR

FreqR

Att

25 dB

Ref

0 dBm

3.84 MHz

CF 1 GHz

CLRWR

QPSK

Eq Freeze

FILT

QPSK

Sym&Mod Acc

-40

-35

-30

-25

-20

-15

-10

-5

SYMBOL TABLE (Hexadecimal)

00000 0 2 3 2 3 1 3 0 2 0 3 0 2 1 1 2 1 1

00018 1 2 2 3 2 2 3 3 2 3 0 3 3 0 0 3 0 1

00036 1 0 1 0 1 1 3 3 0 3 0 3 3 1 0 1 1 1

00054 1 0 2 0 3 3 3 3 0 2 1 1 0 0 0 1 2 2

00072 3 0 0 0 3 2 3 0 1 0 0 1 3 0 1 3 0 0

00090 3 1 1 3 1 1 0 1 2 3 1 2 3 0 3 2 0 2

00108 2 0 1 1 3 1 2 3 1 3 3 1 2 2 0 2 1 0

00126 2 2 3 1 2 2 0 2 3 0 2 2 3 1 2 0 1 2

00144 3 0 2 2 1 3 2 0 1 2 3 2 1 3 1 3 2 0

00162 3 2 1 2 1 3 1 1 3 2 2 2 1 2 3 1 3 3

00180 3 2 2 0 2 1 0 0 3 3 1 2 2 0 0 1 0 2

00198 2 3 1 0 3 0 2 3 0 2 0 1 1 2 0 1 2 1

00216 3 3 2 1 3 2 2 1 0 3 2 1 3 3 2 0 1 3

MODULATION ACCURACY

Result Peak atSym Unit

EVM 6.872 18.457 734 %

Magnitude Err 4.880 18.389 734 %

Phase Error 2.78 -9.60 13 deg

CarrierFreq Err 15.63 Hz

Ampt Droop 0.05 dB

Origin Offset -44.62 dB

Gain Imbalance -0.03 dB

Quadrature Err 0.19 deg

RHO 0.995278

Mean Power -40.14 -34.13 557 dBm

SNR (MER) 23.26 dB

Ref

0 dBm

3.84 MHz

CF 1 GHz

Freq Resp

-7.68 MHz

7.68 MHz

1.536 MHz/

dB/

CLRWR

FreqR

Att

25 dB

Ref

0 dBm

3.84 MHz

CF 1 GHz

CLRWR

QPSK

Eq Freeze

FILT

QPSK

Sym&Mod Acc

-40

-35

-30

-25

-20

-15

-10

-5

SYMBOL TABLE (Hexadecimal)

00000 3 0 0 0 3 2 2 2 2 3 3 2 0 0 0 0 3 3

00018 0 2 3 3 3 2 3 2 1 3 3 3 0 3 0 1 2 3

00036 3 2 2 2 0 1 3 3 0 0 0 2 0 2 3 2 3 3

00054 1 2 1 3 0 3 3 1 1 1 2 2 2 3 1 0 0 1

00072 0 0 3 1 3 3 0 3 3 2 1 2 3 2 2 3 0 1

00090 1 3 0 0 3 2 0 0 2 2 3 2 0 2 3 1 0 3

00108 0 2 1 3 1 3 2 2 1 1 2 1 2 0 0 1 0 1

00126 1 1 2 3 0 3 0 0 0 1 3 2 2 2 3 3 0 2

00144 0 0 0 3 3 2 1 2 3 3 2 3 0 1 1 3 3 0

00162 3 2 0 0 2 3 2 2 0 2 3 1 3 0 0 2 1 3

00180 1 2 2 3 1 1 2 1 1 0 3 1 0 1 1 0 3 2

00198 1 3 0 0 3 2 0 1 2 2 3 2 0 2 0 1 0 3

00216 0 2 1 2 0 3 2 2 1 1 1 2 2 0 0 1 0 0

MODULATION ACCURACY

Result Peak atSym Unit

EVM 1.646 3.575 420 %

Magnitude Err 1.154 3.467 420 %

Phase Error 0.67 1.93 239 deg

CarrierFreq Err 40.08 Hz

Ampt Droop -0.01 dB

Origin Offset -49.24 dB

Gain Imbalance -0.01 dB

Quadrature Err -0.17 deg

RHO 0.999729

Mean Power -1.19 4.22 239 dBm

SNR (MER) 35.67 dB

signal to noise ratio and interfering signals, as mentioned before. The algorithm tries to suppress any interfering signals in order to improve the RMS EVM value. Hence the out-of-band transfer function does not represent an inverse frequency response of the DUT or the channel.

The equalizer’s frequency response to an input signal providing with poor SNR is shown in Fig. 37 whereas the response to a signal with good SNR is demonstrated in Fig. 38. The left diagram (bad SNR) indicates a good suppression of interfering signals.

Fig. 37 Upper diagram: frequency response of a trained equalizer filter (bad SNR at the instrument’s input)

Fig. 38 Upper diagram: frequency response of a trained equalizer filter (good SNR)

Page 44

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 44

3.2 Symbol Mapping

12Q

Mapping or symbol mapping means that logic symbols or symbol numbers are assigned to points or transitions in the I/Q (e.g. PSK and QAM) or frequency plane (e.g. FSK).

Mapping in the analyzer serves for decoding the transmitted symbols from the sampled I/Q or frequency/time data records.

The mappings for all standards used in the analyzer and for all employed modulation modes are described in the following. Unless characterized otherwise, symbol numbers are specified in hexadecimal form (MSB at the left).

If logical symbol mapping does not exactly correspond to the display on the screen, the corresponding physical constellation diagram is shown in addition to mapping.

3.2.1.1 Phase Shift Keying (PSK)

With this type of modulation, the information is represented by the absolute phase position of the receive signal at the decision points. All transitions in the I/Q diagram are permissible for modulation types using static mapping. The complex constellation diagram is shown. The symbol numbers are entered in the diagram according to the mapping rule. The diagram displayed on the analyzer corresponds to symbol mapping.

BPSK (NATURAL)

Fig. 39 Symbol mapping – BPSK / NATURAL

QPSK (WCDMA)

Fig. 40 Symbol mapping – QPSK / WCDMA

Page 45

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 45

QPSK (NATURAL)

Fig. 41 Symbol mapping – QPSK / NATURAL

QPSK (CDMA2K_FWD)

Fig. 42 Symbol mapping – QPSK / CDMA2K_FWD

8PSK (NATURAL)

Fig. 43 Symbol mapping – 8PSK / NATURAL

Page 46

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 46

3.2.1.2 Phase Offset PSK

0 1 2 3 4 5 6 7

-2

-1

Inphase

0 1 2 3 4 5 6 7

-2

-1

Quadrature

time [Symbols]

0 1 2 3 4 5 6 7

-2

-1

Inphase

0 1 2 3 4 5 6 7

-2

-1

Quadrature

time [Symbols]

With this type of modulation, the digital information is represented by the absolute position in the constellation diagram, a phase offset of (n*phi_offset) (n = symbol number) being taken into account for each I/Q symbol. This offset has the same effect as a rotation of the basic system of coordinates by the offset angle after each symbol.

This phase offset is automatically considered when the symbols are decoded and displayed.

The method is highly important in practical applications because it prevents signal transitions through the zeros in the I/Q plane. This reduces the dynamic range of the modulated signal and the linearity requirements for the amplifier.

In practice, the method is used for 3pi/8-8PSK and (in conjunction with phasedifferential coding) for pi/4-DQPSK.

The logical constellation diagram for 3pi/8-8PSK comprises 8 points that correspond to the modulation level. A counter-clockwise offset (rotation) of 3pi/8 is inserted after each symbol transition.

Fig. 44 I/Q symbol stream before 3pi/8 rotation

Fig. 45 IQ-I/Q symbol stream after 3pi/8 rotation

Page 47

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 47

Fig. 44 and Fig. 45 illustrate the influence of the 3pi/8 rotation. Fig. 44 shows the I/Q

'7'

3pi/8

'7' +3pi/8

'7' +6pi/8

'7' +9pi/8

'7' +12pi/8

'7' +15pi/8

symbol stream in the transmitter before rotation (corresponding to an 8PSK modulation), Fig. 45 after rotation (3pi/8 PSK). 1+j*0 was constantly assumed as the modulating symbol.

Fig. 46 and Fig. 47 show the corresponding display in the I/Q plane. The logical constellation diagram (Fig. 46) comprises 8 points corresponding to the

modulation level. When looking at the decision points of an ISI-free receive signal, a physical constellation diagram (Fig. 47) with 16 possible points is obtained.

Eingezeichnet sind Examplehaft 5 Symbolübergänge ‚Symbol 7‘->‘Symbol 7’in der Five symbol transitions are shown in the 'symbol 7’ 'symbol 7’diagram in Fig. 47.

3pi/8-8PSK (EDGE)

Fig. 46 Logical symbol mapping – 3pi/8-8PSK / EDGE

3pi/8-8PSK (display)

Fig. 47 Physical constellation diagram with ISI-free demodulation (taking into account the 3pi/8 phase offset)

Fig. 48shows the TX filter prescribed for the EDGE standard. Fig. 49 shows the vector diagram of a transmitted EDGE signal and the reduced dynamic range of the signal in the case of phase offset modulation (eye aperture in the center of the diagram). The displayed signal is not filtered at the receiver end so that the ISI-free points cannot be seen in the diagram.

Page 48

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 48

Fig. 48 EDGE TX filter

0 2 4 6 8 10 12 14 16 18 20

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Edge TX Filter (impulse response)

Filtertaps

Fig. 49 Vector diagram: transmitted EDGE signal

Page 49

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 49

3.2.1.3 Differential PSK (DPSK)

pi/2

3pi/2

With differential PSK, the information is represented by the phase shift between two consecutive decision points. The absolute position of the complex sampling value at the decision point does not carry information.

In the logical mapping diagram, all permissible symbol transitions (phase transitions) are represented by points in the I/Q plane. The phase position of a point corresponds to the phase difference of the symbol transition. The arrow in the diagram highlights the phase shift and indicates the corresponding symbol number.

In the physical constellation diagram, the constellation points at the symbol decision points obtained after ISI-free demodulation are shown (as with common PSK methods). This diagram corresponds to the display on the analyzer. The position of the constellation points is standard-specific. For example, some QPSK standards define the constellation points on the diagonals, while other standards define the coordinate axes.

The symbol transitions at any constellation point in the diagram are indicated by arrows and labelled according to the mapping.

The indicated QPSK (ISAT) mapping corresponds to simple QPKS with phasedifferential coding. Other types of modulation using this coding method are described in the section 'Mixed PSK modulation'.

DQPSK (INMARSAT)

Fig. 50 Logical symbol mapping – DQPSK / INMARSAT

Fig. 51 Physical constellation diagram – DQPSK / INMARSAT

Page 50

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 50

D8PSK (NATURAL)

6pi/4

7pi/4

5pi/4

4pi/4

3pi/4

2pi/4

1pi/4

Fig. 52 Logical symbol mapping – D8PSK / NATURAL

Fig. 53 Physical constellation diagram – D8PSK / NATURAL

Page 51

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 51

3.2.1.4 Mixed PSK Modulation

pi/2

3pi/2

Phase-differential modulation is frequently combined with an additional phase offset (e.g. pi/4 DQPSK = pi/4 phase offset modulation + differential modulated 4PSK).

The logical mapping diagram corresponds to the diagram for DPSK. In the physical constellation diagram, the constellation points at the symbol decision

points obtained after ISI-free demodulation are shown. This diagram corresponds to the display on the analyzer and, in the case of pi/4-QPSK modulation, the displayed constellation points are doubled.

Pi/4 DQPSK (NADC, PDC, PHS, TETRA)

Fig. 54 Logical mapping – (NADC, PDC, PHS, TETRA)

Fig. 55 Physical constellation diagram –pi4-DQPSK (NADC, PDC, PHS, TETRA); the pi/4 phase offset is taken into account

Page 52

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 52

Pi/4 DQPSK (TFTS)

pi/2

3pi/2

Fig. 56 Logical mapping – pi/4 DQPSK (TFTS)

Fig. 57 Physical constellation diagram – pi/4DQPSK (TFTS); the pi/4 phase offset is taken into account

Page 53

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 53

3.2.1.5 Offset QPSK

0 1 2 3 4 5 6 7 8 9

-2

-1

Inphase

0 1 2 3 4 5 6 7 8 9

-2

-1

Quadrature

time [Symbols]

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-2

-1.5

-1

-0.5

0.5

1.5

Inphase

Quadrature

Vector Diagram QPSK

With this method, the Q component is delayed by half a symbol period against the I component in the time domain. This method is used with QPSK and illustrated by the diagrams below.

Derivation of OQPSK

QPSK

Fig. 58 PSK vector diagram with alpha = 0.35

Page 54

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 54

OQPSK (delayed Q component)

0 1 2 3 4 5 6 7 8 9

-2

-1

Inphase

0 1 2 3 4 5 6 7 8 9

-2

-1

Quadrature

time [Symbols]

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-2

-1.5

-1

-0.5

0.5

1.5

Inphase

Quadrature

Vector Diagram OQPSK

Fig. 59 OQPSK vector diagram with alpha = 0.35

This method (as phase offset PSK) reduces the dynamic range of the modulated signal and the demands on amplifier linearity by avoiding the zero crossing.

A distinction is made in the analyzer display:

● In the I/Q diagram (I/Q VECTOR), the time delay is not compensated for. The display corresponds to the physical diagram shown in Fig. 59.

● In the constellation diagram (I/Q CONSTELLATION), the time delay is compensated for. The display corresponds to the logical mapping (Fig. 60)

OQPSK (CDMA2K_REV)

Fig. 60 Logical symbol mapping – OQPSK / CDMA2K_REV

Page 55

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 55

3.2.1.6 Frequency Shift Keying (FSK)



















0)(0

0)(1

)(

tffür

Symbol-

nummern

f in

[FSK REF

DEVIATION]

-1

'1'

'0'































ffür

tffür

)(02

0)(

)(0

)(

In the case of FSK demodulation, a frequency/time diagram is displayed instead of the constellation and vector diagrams. The symbol decision is based on the signal frequency at the decision points. To illustrate the symbol decision thresholds, the symbol numbers are marked in the logical mapping diagram versus the instantaneous frequency fi . The 0 frequency in the baseband corresponds to the input frequency of the analyzer.

2-FSK (NATURAL) With 2FSK, the symbol decision is made by a simple frequency discriminator with

reference to the 0 frequency in the baseband:

for all symbol decision points t = n*Ts, fi = instantaneous frequency normalized to FSK REF DEVIATION

Fig. 61 Symbol mapping – 2FSK / NATURAL

4-FSK

With 4FSK, the symbol decision is made by a frequency discriminator with 3 decision thresholds (-2/3; 0; +2/3) normalized to the FSK REF DEVIATION parameter.

for all symbol decision points t = n*Ts, fi = normalized instantaneous frequency

Page 56

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 56

Fig. 62 Symbol mapping – 4-FSK / NATURAL

Symbol-

nummern

f in

[FSK REF

DEVIATION]

1/3

-1/3

'1'

'3'

'0'

'2'

-1

pi/2

-pi/2

3.2.1.7 Minimum Shift Keying (MSK)

MSK modulation is a special case of 2FSK with FSK REF DEVIATION = ¼ * symbol rate. This special characteristic causes modulation-dependent phase shifts of +/- 90° which can be shown in an I/Q constellation diagram. As with PSK, demodulation is performed by evaluation of the phase positions.

MSK (NATURAL)

Fig. 63 Logical symbol mapping – MSK / NATURAL

Fig. 64 Physical constellation diagram – MSK

Similar to PSK, differential coding can also be used with MSK. In this case, too, the information is represented by the transition of two consecutive symbols. The block diagram of the coder is shown below.

Page 57

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 57

Fig. 65 DMSK: differential encoder in the transmitter

input symbol {0;1} of differential encoder

input symbol delayed by the symbol period Ts

output symbol {0;1} of differential encoder

To ensure reliable demodulation, the statistical distribution of the available symbol quantity should be as even as possible.

For instance, if only

- single symbols

- single amplitude ranges or

- single quadrants are used, demodulation errors may occur. As a rule of thumb, the RESULT LENGTH

should correspond to at least 8 times the modulation level. For example, with 64 QAM a RESULT LENGTH of at least 4*64 = 256 symbols should be used.

XOR

i-1

id1idi

During demodulation and symbol decision in the analyzer, the original symbols are restored by a differential decoder and displayed.

This modulation method used for the digital GSM standard in conjunction with a GAUSSIAN transmitter filter is called GMSK.

Signal mapping with the differential encoder is called MSK / GSM.

3.2.1.8 Quadrature Amplitude Modulation (QAM)

In the case of QAM the information is represented by the signal amplitude and phase. The symbols are arranged in a square constellation (16, 64, 256QAM) or as cross-

shaped structures (21, 128QAM) in the I/Q plane. The differential mappings below meet ETSI EN 300429 V1.2.1 (DVB-C).

Page 58

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 58

Statistical QAM Mappings

00.xx10.xx

11.xx 01.xx

pi/2

.00

.10

.01

.11

.00

.10

.01

.11

00.xx10.xx

11.xx 01.xx

.00

.10

.01

.11

.10 .00

.01

.11

.00.01

.10.11

.00

.01 .11

.10

EhFh

Dh Ch 4h

5h 7h

9h 2h0h3h

1h8h

00.00

00.10

00.01

00.11

00.xx10.xx

11.xx 01.xx

10.10

10.11 10.01

10.00

01.11

01.1001.00

11.10

11.0011.01

11.11 01.01

02h06h

04h

00h

05h

01h

07h

03h

13h17h

12h

16h

15h

14h 10h

11h

18h19h1Bh

1Fh 1Dh 1Ch

1Eh1Ah

08h 0Ch 0Eh

09h

0Bh

0Dh 0Ah

0Fh

00.01000.110

00.100

00.000

00.101

00.001

00.111

00.011

10.01110.111

10.010

10.110

10.101

10.100 10.000

10.001

11.00011.00111.011

11.111 11.101 11.100

11.11011.010

01.000 01.100 01110

01.001

01.011

01.101 01.010

01.111

The following QAM mappings are obtained from the mapping of the 1st quadrant, which is always rotated by pi/2 for the subsequent quadrants and supplemented by a (GRAY-coded) prefix for each quadrant.

Derivation of QAM mappings

Fig. 66 Rotation of 1st quadrant

In the following diagrams, the symbol mappings are indicated in hexadecimal and binary form.

16 QAM (DVB-C)

Fig. 67 Symbol mapping – 16QAM / DVB-C

32 QAM (DVB-C)

Fig. 68 Symbol mapping – 32QAM / DVB-C

Page 59

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 59

64 QAM (DVB-C)

00h 01h

02h 03h

04h05h

06h07h

08h 09h

0Ch 0Bh 0Eh

0Ch0Dh

0Fh

2Eh

2Dh 27h 25h2Fh

2Bh

2Ah

2Ch

29h

28h

26h

23h

22h

24h

21h

20h

3Fh

3Dh

37h

35h

3Eh 3Bh 3Ah

3Ch 39h 38h

36h 33h 32h

34h 31h 30h

1Fh

1Eh

1Bh

1Ah

1Dh

1Ch

19h

18h

17h

16h

13h

12h

15h

14h

11h

10h

00.0000 00.0001

00.0010 00.0011

00.010000.0101

00.011000.0111

00.1000 00.1001

00.1010 00.1011 00.1110

00.110000.1101

00.1111

10.1110

10.1101 10.0111 10.010110.1111

10.1011

10.1010

10.1100

10.1001

10.1000

10.0110

10.0011

10.0010

10.0100

10.0001

10.0000

11.1111

11.1101

11.0111

11.0101

11.1110 11.1011 11.1010

11.1100 11.1001 11.1000

11.0110 11.0011 11.0010

11.0100 11.0001 11.0000

01.1111

01.1110

01.1011

01.1010

01.1101

01.1100

01.1001

01.1000

01.0111

01.0110

01.0011

01.0010

01.0101

01.0100

01.0001

01.0000

00.0010000.0010100.0000100.00000 00.01100 00.01101

00.0011000.0011100.0001100.00010 00.01110 00.01111

00.1011000.1011100.1001100.10010 00.11110 00.11111

00.1010000.1010100.1000100.10000 00.11100 00.11101

00.0100000.0100100.1100100.11000

00.0101000.0101100.1101100.11010

01.1000001.1001001.0001001.00000 01.11000 01.11010

01.1000101.1001101.0001101.00001 01.11001 01.11011

01.1010001.1011001.0011001.00100 01.01000 01.01010

01.1110001.1111001.0111001.01100

01.1110101.1111101.0111101.0110111.1101011.1101111.0101111.01010

11.1100011.1100111.0100111.01000

11.1010011.1110011.11101 11.10101 11.10001 11.10000

11.0011011.0111011.01111 11.00111 11.00011 11.00010

11.0010011.0110011.01101 11.00101 11.00001 11.00000

10.1000010.1100010.11010 10.10010 10.00010 10.0000

10.1000110.1100110.11011 10.10011 10.00011 10.00001

10.1010110.0100110.01011 10.10111 10.00111 10.00101

10.1010010.0100010.01010 10.10110 10.00110 10.00100

10.11100 10.11110 10.01110 10.01100

10.11101 10.11111 10.01111 10.01101

5Dh 5Fh 4Fh 4Dh

5Ch 5Eh 4Eh 4Ch

54h48h4Ah 56h 46h 44h

55h49h4Bh 57h 47h 45h

51h59h5Bh 53h 43h 41h

50h58h5Ah 52h 42h 40h

64h6Ch6Dh 65h 61h 60h

66h6Eh6Fh 67h 63h 62h

74h7Ch7Dh 75h 71h 70h

68h 69h 79h 78h

6Ah 6Bh 7Bh 7Ah 2Dh 2Fh 3Fh 3Dh

2Ch 2Eh 3Eh 3Ch

24h 26h 36h 34h 28h 2Ah

21h 23h 33h 31h 39h 3Bh

20h 22h 32h 30h 38h 3Ah

00h 01h 05h 04h 0Ch 0Dh

02h 03h 07h 06h 0Eh 0Fh

12h 13h 17h 16h 1Eh 1Fh

10h 11h 15h 14h 1Ch 1Dh

18h 19h 09h 08h

1Ah 1Bh 0Bh 0Ah

11.1011011.1111011.11111 11.10111 11.10011 11.10010

76h7Eh7Fh 77h 73h 72h

01.1010101.1011101.0011101.00101 01.01001 01.01011

25h 27h 37h 35h 29h 2Bh

Fig. 69 Symbol mapping – 64QAM / DVB-C

128 QAM (DVB-C)

Fig. 70 Symbol mapping 128 QAM / DVB-C

Page 60

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 60

256 QAM (DVB-C)

7Bh

00.001100

00.001110

00.001101

00.001111

00.000110

00.000100

00.000111

00.000101

00.001001

00.001011

00.001000

00.001010

00.000010 00.000011

00.00000100.000000 00.01000000.01000100.01010100.010100

00.01001000.01001100.01011100.010110

00.01101000.01101100.01111100.011110

00.01100000.01100100.01110100.011100

00.10110000.10110100.10100100.101000 00.11100000.11100100.11110100.111100

00.10111000.10111100.10101100.101010 00.11101000.11101100.11111100.111110

00.11001000.11001100.11011100.110110

00.11000000.11000100.11010100.110100

00.10011000.10011100.10001100.100010

00.10010000.10010100.10000100.100000

10.000000

10.000001

10.000101

10.000100

10.010000

10.010001

10.010101

10.010100

10.000110

10.000111

10.000010

10.000011

10.010010

10.010011

10.010111

10.010110

10.001110

10.001111

10.001011

10.001010

10.011010

10.011011

10.011111

10.011110

10.001100

10.001101

10.001001

10.001000

10.011000

10.011001

10.011101

10.011100

10.101100

10.101101

10.101001

10.101000

10.111000

10.111001

10.111101

10.111100

10.101110

10.101111

10.101011

10.101010

10.111010

10.111011

10.111111

10.111110

10.110010

10.110011

10.110111

10.110110

10.100110

10.100111

10.100011

10.100010

10.110000

10.110001

10.110101

10.110100

10.100100

10.100101

10.100001

10.100000

11.000000

11.000010

11.001010

11.001000

11.101000

11.101010

11.100010

11.100000

11.000001

11.000011

11.001011

11.001001

11.101001

11.101011

11.100011

11.100001

11.000101

11.000111

11.001111

11.001101

11.101101

11.101111

11.100111

11.100101

11.000100

11.000110

11.001110

11.001100

11.101100

11.101110

11.100110

11.100100

11.010100

11.010110

11.011110

11.011100

11.111100

11.111110

11.110110

11.11010011.110101

11.110111

11.111111

11.011101

11.111101

11.011111

11.110111

11.010101

11.110001

11.110011

11.111011

11.111001

11.011001

11.011011

11.010011

11.01000111.010000

11.010010

11.011010

11.011000

11.111000

11.111010

11.110010

11.110000 01.010000 01.010010 01.011010 01.011000 01.111000 01.111010 01.110010 01.110000

01.010001 01.010011 01.011011 01.011001 01.111001 01.111011 01.110011 01.110001

01.010111 01.011111 01.011101 01.111101 01.111111 01.110111 01.11010101.010101

01.010100 01.010110 01.011110 01.011100 01.111100 01.111110 01.110110 01.110100

01.000100 01.000110 01.001110 01.001100 01.101100 01.101110 01.100110 01.100100

01.000101 01.000111 01.001111 01.001101 01.101101 01.101111 01.100111 01.100101

01.000001 01.000011 01.001011 01.001001 01.101001 01.101011 01.100011 01.100001

01.000000 01.000010 01.001010 01.001000 01.101000 01.101010 01.100010 01.100000

00h 01h 05h 04h 14h 15h 11h 10h

02h 03h 07h 06h 16h 17h 13h 12h

0Ah 0Bh 0Fh 0Eh 1Eh 1Fh 1Bh 1Ah

08h 09h 0Dh 0Ch 1Ch 1Dh 19h 18h

28h 29 2Dh 2Ch 3Ch 3Dh 39h 38h

2Ah 2Bh 2Fh 2Eh 3Eh 3Fh 3Bh 3Ah

22h 23h 27h 26h 36h 37h 33h 32h

20h 21h 25h 24h 34h 35h 31h 30h

A0h A2h AAh A8h 88h 8Ah 82h 80h

A1h A3h ABh A9h 89h 8Bh 83h 81h

A5h A7h AFh ADh 8Dh 8Fh 87h 85h

A4h A6h AEh ACh 8Ch 8Eh 86h 84h

B4h B6h BEh BCh 9Ch 9Eh 96h 94h

B5h B7h BFh BDh 9Dh 9Fh 97h 95h

B1h B3h BBh B9h 99h 9Bh 93h 91h

B0h B2h BAh B8h 98h 9Ah 92h 90h

D0h D1h D5h D4h C4h C5h C1h C0h

D2h D3h D7h D6h C6h C7h C3h C2h

DAh DBh DFh DEh CEh CFh CBh CAh

D8h D9h DDh DCh CCh CDh C9h C8h

F8h F9h FDh FCh ECh EDh E9h E8h

FAh FBh FFh FEh EEh EFh EBh EAh

F2h F3h F7h F6h E6h E7h E3h E2h 51h 53h 5Bh 59h 79h 73h 71h

55h 57h 5Fh 5Dh 7Dh 7Fh 77h 75h

54h 56h 5Eh 5Ch 7Ch 7Eh 76h 74h

44h 46h 4Eh 4Ch 6Ch 6Eh 66h 64h

45h 47h 4Fh 4Dh 6Dh 6Fh 67h 65h

41h 43h 4Bh 49h 69h 6Bh 63h 61h

40h 42h 4Ah 48h 68h 6Ah 62h 60h

50h 52h 5Ah 58h 78h 7Ah 72h 70hF0h F1h F5h F4h E4h E5h E1h E0h

3.2.1.9 Differential QAM Mappings

Fig. 71 Symbol mapping 256 QAM / DVB-C

The following differential QAM mappings show the mapping in a quadrant (1st quadrant) and differential mapping. In the case of differential mapping, the quadrant transitions are coded (as with DQPSK).

Differential 16 QAM (DVB-C)

Page 61

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 61

00.xx10.xx

11.xx 01.xx

yy.00

yy.10

yy.01

yy.11

00.xx10.xx

11.xx 01.xx

yy.010yy.110

yy.100

yy.000

yy.101

yy.001

yy.111

yy.011

00.xx10.xx

11.xx 01.xx

yy.0000 yy.0001

yy.0010 yy.0011

yy.0100yy.0101

yy.0110yy.0111

yy.1000 yy.1001

yy.1010 yy.1011 yy.1110

yy.1100yy.1101

yy.1111

Fig. 72 Symbol mapping D16 QAM / DVB-C

Differential 32 QAM (DVB-C)

Fig. 73 Symbol mapping D32 QAM / DVB-C

Differential 64 QAM (DVB-C)

Fig. 74 Symbol mapping D64 QAM / DVB-C

Page 62

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 62

Differential 128 QAM (DVB-C)

yy.00100yy.00101yy.00001yy.00000 yy.01100 yy.01101

yy.00110yy.00111yy.00011yy.00010 yy.01110 yy.01111

yy.10110yy.10111yy.10011yy.10010 yy.11110 yy.11111

yy.10100yy.10101yy.10001yy.10000 yy.11100 yy.11101

yy.01000yy.01001yy.11001yy.11000

yy.01010yy.01011yy.11011yy.11010

00h 01h 05h 04h 0Ch 0Dh

02h 03h 07h 06h 0Eh 0Fh

12h 13h 17h 16h 1Eh 1Fh

10h 11h 15h 14h 1Ch 1Dh

18h 19h 09h 08h

1Ah 1Bh 0Bh 0Ah

00.xx10.xx

11.xx 01.xx

yy.001100

yy.001110

yy.001101

yy.001111

yy.000110

yy.000100

yy.000111

yy.000101

yy.001001

yy.001011

yy.001000

yy.001010

yy.000010 yy.000011

yy.000001yy.000000 yy.010000yy.010001yy.010101yy.010100

yy.010010yy.010011yy.010111yy.010110

yy.011010yy.011011yy.011111yy.011110

yy.011000yy.011001yy.011101yy.011100

yy.101100yy.101101yy.101001yy.101000 yy.111000yy.111001yy.111101yy.111100

yy.101110yy.101111yy.101011yy.101010 yy.111010yy.111011yy.111111yy.111110

yy.110010yy.110011yy.110111yy.110110

yy.110000yy.110001yy.110101yy.110100

yy.100110yy.100111yy.100011yy.100010

yy.100100yy.100101yy.100001yy.100000

00h 01h 05h 04h 14h 15h 11h 10h

02h 03h 07h 06h 16h 17h 13h 12h

0Ah 0Bh 0Fh 0Eh 1Eh 1Fh 1Bh 1Ah

08h 09h 0Dh 0Ch 1Ch 1Dh 19h 18h

28h 29 2Dh 2Ch 3Ch 3Dh 39h 38h

2Ah 2Bh 2Fh 2Eh 3Eh 3Fh 3Bh 3Ah

22h 23h 27h 26h 36h 37h 33h 32h

20h 21h 25h 24h 34h 35h 31h 30h

00.xx10.xx

11.xx 01.xx

Fig. 75 Symbol mapping D128 QAM / DVB-C

Differential 256 QAM (DVB-C)

Fig. 76 Symbol mapping D256 QAM / DVB-C

Page 63

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 63

3.2.1.10 User Defined Constellations (USER-QAM)

Customized constellations (including symbol mappings) can be defined with the external utility MAPWIZ (PC Windows environment).

For a description of this tool see chapter 8, Utilities /External Programs The example in the following figure shows the constellation diagram of the 16-level

USER-QUAM that has the minimum probability of symbol errors in the case of AWGN (Source: "Optimization of Two-Dimensional Signal Constellations in the Presence of Gaussian Noise", G. J. Foschini et al., IEEE Transactions on Communications, Vol. COM-22, 01/1974, pp. 28).

Fig. 77 Demodulation of a 16ary USER-QAM

Page 64

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 64

3.2.1.11 Vestigial Sideband Modulation (VSB)

Like BPSK, digital vestigial sideband modulation (VSB) transfers the information in the real component, in which case different amplitude stages must additionally be used. Owing to the real baseband signal, transmitting a single sideband is sufficient, e.g. VSB signals have half the bandwidth of BPSK signals. Rather than completely suppressing one of the two sidebands, a vestige of the sideband to be suppressed is permitted, thus reducing the effort for implementing filters. However, halving the bandwidth produces intersymbol interference (ISI), which is indicated by vertical lines in the constellation diagram (see Fig. 78).

Fig. 78 8VSB constellation diagram

A further and primary difference compared to PSK methods is that VSB signals additionally contain a pilot carrier. The pilot carrier is removed from the signals for all measurements (except capture buffer). To make it possible to analyze VSB signals with the vector signal analyzer, the center frequency and the frequency position (normal position or inverted position) must be adjusted in such a manner that a spectrum that is symmetrical about the center frequency is present at the analyzer input. In this case, the pilot carrier must be located to the left of the center frequency (see Fig. 79). Compared with the true VSB spectrum that has been freed from the pilot carrier (see Fig. 80), the spectrum must be shifted to the left by symbol rate/4.

Page 65

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 65

0 7

321

Fig. 79 8VSB spectrum at the input of the analyzer (pilot carrier visible to the left)

Fig. 80 Spectrum of measurement signal 8VSB (pilot carrier always removed) 8VSB (ATSC)

Fig. 81 Symbol mapping 8VSB (ATSC)

Page 66

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 66

3.3 Demodulation and Algorithms

DEMODULATOR

(non data

aided)

MATCHING

BURST

PATTERN

ERROR CALC &

FIT TRACE

I/Q

Burst Position

Symbols

Pattern

Position

I/Q (Ref) I/Q (Meas) Symbols Burst & Pattern Pos.

Symbols I/Q (Ref) I/Q (Meas) Burst Pos.

RESULT DISPLAY

(Numerical Results)

Coarse Estimation of:

-Center Freq

-Timing

Fine estimation:

-Center Freq

-Timing Calculation of:

-Origin Offset

-Amplitude Droop

-IQ Imbalance

RESULT DISPLAY

(Signal & Error

Traces)

I/Q (Ref) I/Q (Meas) Symbols Burst & Pattern Pos.

Calculation of:

-RMS EVM

-RMS MAG ERR

-RMS Phase Error

-RHO

-Trigger to PatternPos

IQ Record Buffer

Burst Parameter, Modulation, Result Len, Fit Trace,

SettingsPSK Demodulator

Modulation, Filter

Modulation, Pattern Parameter

Result Type Result Len, Fit Trace

Fig. 82 Digital demodulation of a PSK demodulator

Fig. 82 gives an overview of the demodulation stages of the vector signal analysis option, using PSK demodulation as an example. Differences to other types of modulation will be dealt with at the end of this chapter.

The function blocks for demodulation are shown at the left, settings for the function blocks at the right.

After data recording in the RECORD BUFFER, the I/Q data is forwarded to the BURST SEARCH In this stage, the RECORD BUFFER is searched through for burst structures. The first

burst found is forwarded together with its environment to the next processing stage.

Page 67

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 67

The length of the transferred data record normally corresponds to the RESULT LENGTH. The internal length may be automatically extended because of the delays required by the demodulation filters to settle and for trace positioning in the display (FIT TRACE).

If the burst search is switched off, a data record from the beginning of the RECORD BUFFER is transmitted.

DEMODULATOR

This stage performs demodulation down to symbol level. Correction values for timing, frequency and phase position are determined during demodulation and applied to the data record so that a correct symbol decision is possible. Network-specific synchronization aids such as sync patterns are not used in this case so that the measurement demodulator operates without knowing the transmitted data contents (NDA (non-data-aided) demodulator). A reference signal corresponding to an ideal, error-free transmission signal is regenerated from the various symbols and forwarded to the MATCHING stage together with the corrected measurement signal.

PATTERN SEARCH

The symbol data record is searched through for one or more user-defined sync patterns. The measurement results (TRACES) can be positioned with the aid of the patterns found. The pattern search is optional.

MATCHING

In this stage the reference and measurement signals are correlated. The matching algorithm determines accurate correction values for signal amplitude and signal timing as well as for frequency errors and phase position of the measurement signal with the aid of the optimization criterion in order to minimize the RMS vector error, and then corrects the measurement data record.

First numeric measurement results such as center frequency error, origin offset and I/Q imbalance are obtained at this stage.

ERROR CALC & FIT TRACE

At this processing stage, further modulation errors are calculated which are either displayed as results or used for further result calculation. Results are available in numeric form (e.g. RMS EVM), display versus time (EVM trace) or as a statistical evaluation of error parameters (e.g. 95:th percentile).

RESULT DISPLAY

The selected measurement results are positioned in the display and scaled according to user settings. Special points in time or ranges of the measurement signal (e.g. sync pattern or symbol decision points) can be highlighted in the display.

A detailed description of the function blocks follows on the next pages.

Page 68

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 68

Power

(Averaged)

Find Rising &

Falling Edge

(Level)

Burst Search

Calculate

(Auto)Search

Parameters

Check

Burst Length

Burst Level

Level &

Length ok ?

End of Record Buffer

Runin &

Runout

Range

Prefit

Burst/Result

I/Q

Search Again

No Burst

Burst Search

on/OFF

Burst Search

ON/OFF

Calculate

Fit & Result

Range

Burst Parameter, Modulation, Result Len, Fit Trace,

Settings

Fig. 83 Burst Search

Page 69

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 69

3.3.1 Burst Search

B1 B2 B3 B4

RECORD LEN

Min Burstlen Max Burstlen

With the burst search switched on, the magnitude of the sample in the record buffer is calculated and then averaged with a square filter to reduce modulation-responsive signal amplitude variations and to suppress short noise peaks.

In the AUTO mode, the global minimum and maximum of this data record are determined and two level threshold values are calculated by taking into account a modulation-responsive factor.

With the aid of these thresholds, the magnitude data record is searched through for rising and falling burst edges. Brief level drops are ignored.

When the first burst is found that fulfills the requirements regarding minimum and maximum length, the burst search is terminated and the part of the record buffer containing the burst is forwarded to the subsequent processing stage.

The minimum and maximum lengths that can be detected, the calculation of threshold values and the sensitivity for short level drops can be varied in the AUTO mode by selecting a digital standard.

In the MAN (manual) mode, these parameters can be set by the user. However, the MAN mode is only recommended under difficult receive conditions.

If the burst search is switched off, a block with a length required for result display from the beginning of the record buffer is forwarded to the next processing stage.

Fig. 84 Record buffer containing several bursts

Fig. 84 shows the contents of a record buffer with several bursts. The upper (UT) and the lower burst threshold (LT) and bursts of different levels are

shown. All bursts fulfill the level requirements, i.e. the burst edges cross both burst thresholds;

burst B1 also has the required length, B2 is too short, B3 is too long and B4 has no falling burst edge.

B1 is the first burst to fulfill all requirements and therefore forwarded to the subsequent processing stages.

Page 70

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 70

Timing

Recovery

Phase &

Frequency

Recovery

Timing Corrected

IQ_raw

Tming, Phase & Freq. Corrected

IQ_raw

Demodulator 1

Timing

Numerical

Results

ISI-Filter, Modulation

Settings

Center Freq

Error

Fig. 85 IQ-Demodulator: Timing, Phase, Frequency Recovery

Page 71

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 71

3.3.2 Demodulator 1

;







0 0.5 1 1.5 2 2.5 3

-1

Sampled Signal

Symbol Timing Detection & Correction

0 0.5 1 1.5 2 2.5 3

Symbol Time

0 0.5 1 1.5 2 2.5

-1

Corrected Signal

time [Symbols]

The first part of the demodulator comprises the following function groups:

● Timing recovery

● Phase & frequency recovery

3.3.2.1 Phase & Frequency Recovery

If a burst structure is found, the burst (without edges) is used as the demodulator estimation range although the determined correction parameters are applied to the full demodulation range.

For reasons of algorithm, the signals are filtered in these function blocks to obtain ISIfree points. However, the output signals are timing-, frequency- and phase-corrected raw signals (as shown in the drawings) so that subsequent distortion measurements can be performed or customized measurement filters used.

At the input of this stage, the I/Q data record in the complex baseband contains

● a time offset

● a center frequency error and a phase error of .

3.3.2.2 Timing Recovery

This function group determines the ideal symbol decision points in the signal. The I/Q data record must then be corrected so that the samples occur exactly at the symbol decision points (resampling).

Fig. 86 Symbol timing detection & correction

Fig. 86 illustrates the correction using the sampled input signal, the ideal symbol decision points and the corrected data record (time axis adapted).

A calculated timebase correction also affects numeric results (e.g. trigger to sync measurement).

SIGNAL 2(Fig. 85) corresponds to the I/Q data record before timing correction,

Page 72

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 72

SIGNAL 2a to the record after timing correction. Since the frequency error is not yet





I/Q Symbol

Decision

Reference

Filter

Timing, Phase

& Frequency

Corrected IQ_raw

IQ_Symbols:

e.g.:

(+1+j),(+1-j);

(-1+j);(-1-j);

Symbols

Measure-

ment Filter

e.g. 'none'

ISI Filter

Demodulator 2

IQ_ref

Settings

ISI-Filter

IQ & Symbol Mapping

TX / Meas Filter

3b3a

Corrected &

Filtered IQ_raw

eliminated, the symbol points in the constellation diagram are shown as a circular band.

3.3.2.3 Phase & Frequency Recovery

This function group determines and corrects the frequency and phase offset. With the aid of a robust, maximum-likelihood frequency and phase estimator, the stage determines the optimum estimation value for the data record after timing correction (center frequency error ).

After correction of these quantities, a 'non-rotating constellation diagram’(for an unfiltered raw signal) is obtained (see SIGNAL 2b).

Fig. 87 Demodulator 2

Page 73

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 73

3.3.3 Demodulator 2

arg(x(n)) = phase of I/Q input sample at the decision point s(n) = decided I/Q symbol

s(n)=

(1-j) / sqrt(2)

s(n)=

(-1+j) / sqrt(2)

s(n)=

(1+j) / sqrt(2)

s(n)=

(-1-j) / sqrt(2)

 





























jnsnx

)(

4))(arg(

)(

3))(arg(

)(

2))(arg(

)(

))(arg(0





The timing-, frequency- and phase-corrected data record (signal 2b) is forwarded to an ISI FILTER to eliminate the ISI of adjacent symbols (see section „System-Theoretical

Modulation and Demodulation Filters")

I/Q symbols (signal 3c) and - if symbol mappings are taken into account - logical symbols are then produced in the I/Q SYMBOL DECISION block (for PSK).

In the case of QPSK, the segment decider is a simple quadrant decider which only affects the input signal phase.

Fig. 88 QPSK segment decider

The I/Q REF data record (signal 3b) is generated from the data record of the decided I/Q symbols after null stuffing (to attain the required oversampling rate) and filtering with the REFERENCE FILTER. After filtering with the MEASUREMENT FILTER, the measurement data record is forwarded as signal 3a to the subsequent processing stages. When MEASUREMENT FILTER = NONE is set, the data record is forwarded unchanged.

Phase ambiguity of demodulator

Up to now, the demodulator operated without knowing the transmitted signal. Since phase shifts may occur on the transmission path, the result of demodulation is ambiguous with respect to the phase position (because of the rotation symmetry in the PSK constellation). In the case of QPSK with static symbol mapping, this means that the I/Q measurement and I/Q reference signals as well as the decided symbols may have a constant phase offset of {0, pi/2, pi, or 3pi/2}. This offset can only be detected and eliminated in all 3 data records after sync pattern search in the data record.

If modulation types without static mapping are used, e.g. differential PSK or MSK, the information represented by the phase transition is encrypted so that static symbol mapping and the ambiguity of the starting phase are no longer a problem.

Page 74

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 74

Timing

Phase-Freq

I/Q-Offset

Correction

IQ_Meas

Matching

Calculation

Numerical Results:

Center Frequency Error

Origin-Offset

Amplitude Droop

I/Q-Imbalance

Matching

Range

Matching Parameters

Numerical

Results

3a 3b 4

Match IQ Settings

Result Settings

Fig. 89 Matching

3.3.4 Matching

The measurement signal was processed in the previous modulation stages so that error-free demodulation, symbol decision and reference signal generation could be performed.

In the MATCHING group, the error parameters (e.g. RMS EVM in case of PSK) are minimized.

Page 75

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 75

With the aid of the following equation a transmit signal Y(t) in the time domain can be

 

;0)()(1)( WCtERRtREFCtY

txtx



;

tjt









)(



;

)(







nREF

nEV

EVMRMS

;0)()()( CnREFnMEASnEV 

;





obtained in the baseband (all parameters used are complex):

● REFtx is the ideal transmit signal,

● ERRtx the error signal of the transmitter (linear and nonlinear distortions),

● C0 the I/Q offset (origin offset) and

● C1 a complex constant (phase and amplitude of transmitter)

is a complex factor which represents the amplitude variations in the burst ( ) and a

center frequency offset .

The parameter to be minimized (valid for EDGE, for formulae of other modulation types see chapter 10 is defined by

containing the error vector: where

● EV is the error vector after the prescribed measurement filtering,

● MEAS is the measured transmit signal (Y8t) after measurement filtering in the

analyzer,

● REF is the reference signal and

● (n) the symbol points in the useful part (length N) of the demodulator range.

The RMS_EVM is minimized by means of a maximum likelihood function in the MATCHING block and the associated parameters (C0, C1, ) are determined.

During minimizing, a residual time offset is also determined to compensate for the estimation uncertainty of the non-data-aided demodulator.

Page 76

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 76

== ?

Pattern Search

Pattern Index Pattern Position Pattern Phase

Pattern_1 Pattern_2

Rotate

Symbol-

Mapping

2 3 0 0 0 3 2 2 3 1 1 1 3 2

Data (Example):

1 0 3 2

0 2 1 3

2 3 0 1

3 1 2 0

2 3 0 0 0 3 2* * * * * * * * * * * * * *

Pattern_1a Pattern_2a

2 3 0 0 0 3 2 2 3 1 1 1 3 2

Pattern_1b Pattern_2b

3 1 2 2 2 1 3 3 1 0 0 0 1 3

Pattern_1c Pattern_2c

1 0 3 3 3 0 1 1 0 2 2 2 0 1

Pattern_1d Pattern_2d

0 2 1 1 1 2 0 0 2 3 3 3 2 0

1 0 3 2

0 2 1 3

2 3 0 1

3 1 2 0

Rotated Pattern &

Rotation Phase

Symbol Data

Pattern (Example):

QPSK (GRAY)

Symbolmapping

Settings

-> Pattern_1 .._n

Fig. 90 Pattern search

Page 77

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 77

3.3.5 Pattern Search

Original

Hypotheses

Mapping

Temporary Pattern

If pattern is found

Hypothesis a) (phase = 0pi/2)

- I/Q data records are unchanged

- Symbol data record are unchanged

Hypothesis b) (phase = pi/2)

- I/Q data records are rotated clockwise by pi/2

- The symbol data record is remapped (2->0, 0->1, 1->3, 3->2)

Hypothesis c) (phase = 2pi/2)

- I/Q data records are rotated clockwise by 2pi/2

- The symbol data record is remapped>1, 0->3, 1->2)

Hypothesis c) (phase = 3pi/2)

- I/Q data records are rotated clockwise by 3pi/2

- he symbol data record is remapped (1->0, 3->1, 2->3, 0->2)

1 0 3 2

Pattern_1a Pattern_2a

2 3 0 0 0 3 2 2 3 1 1 1 3 2

1 0 3 2

Pattern_1a Pattern_2a

2 3 0 0 0 3 2 2 3 1 1 1 3 2

0 2 1 3

Pattern_1b Pattern_2b

3 1 2 2 2 1 3 3 1 0 0 0 1 3

2 3 0 1

Pattern_1c Pattern_2c

1 0 3 3 3 0 1 1 0 2 2 2 0 1

3 1 2 0

Pattern_1d Pattern_2d

0 2 1 1 1 2 0 0 2 3 3 3 2 0

Many digital standards use constant symbol sequences (here called patterns) at defined positions in the burst, which are used in the mobile network for estimating transmission channel characteristics.

In the analyzer, the pattern position defined by the standard is used for scaling and for determining the standard-specific measurement range.

In pattern search, a distinction is made between static mappings and differential mappings:

With static mappings, the symbol information is represented by the absolute position of the symbol in the I/Q plane. Examples are QPSK, 8PSK and regular QAM constellations (see section "Symbol Mapping"). Because of the rotation symmetry of these mappings, an unambiguous symbol decision is only possible after a pattern search.

When the pattern is found, the absolute phase position of the signal is also identified, the I/Q measurement data record and the I/Q reference data record are appropriately rotated and the symbol data record is corrected.

Fig. 91 illustrates the function principle for QPSK (GRAY mapping). The user predefines 2 possible sync patterns (Pattern_1 and Pattern_2). With QPSK, 4

symmetry states (mapping a to d) are possible, which correspond to a rotation of coordinates by 0, pi/2, pi, 3pi/2, respectively.

Fig. 91 Pattern search for static QPSK mapping

The algorithm internally converts the predefined pattern by taking the symmetry states

Page 78

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 78

into account (pattern 1a to d and pattern 2a to d) and searches in the symbol data

3a,

Result & Error Calculation

Display

Settings

RESULT Settings FIT Settings EVAL Lines

ERROR

VECTOR

MAGNITUDE

PHASE

RHO

EVM

MAGNITUDE

ERROR

PHASE

ERROR

FREQUENCY

ERROR

MAGNITUDE

CAP BUFFER

STATISTIC

TRACE

AM &PM

Conversion

SYMBOLS

13c

MODULATION

ACCURACY

Numerical

Results

TRACE Functions (Peak, Average...)

AVERAGE Function (Modulation Accuracy)

DISPLAY Scaling

TRACE Settings (Clear/Write, Peak, Average) RANGE Settings

95 pctl

record for this "rotating" search pattern. If the patterns exactly coincide, the search is successfully terminated and, if required, the I/Q data records and the symbol data record are corrected according to the hypothesis found. With differential mappings, only a single-stage procedure is required because the symbol information is represented by the phase difference of two consecutive decision points. Correction of data records is therefore not required.

Fig. 92 Result & Error Calculation

Page 79

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 79

3.3.6 Result & Error Calculation, Display

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

-1

Trace 1

Averaging of Time Based Traces

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

-1

Trace 2

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

-1

Trace 3

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

-1

Average

);3..1(

)3(

)2(

)1(

EVMRMS

TRACEEVM



The result displays selected by the user are calculated and scaled in the two last processing stages.

Extreme values and average values over several measurements can be calculated for result display. This function can be switched on and off in the Trace menu.

The calculation formulae can be found in the description of the specific display modes and at the end of this manual (chapter 10").

Fig. 93 Trace averaging

In the case of trace display, average and extreme values are calculated for each trace point derived from the measured value samples.

Fig. 93 illustrates this process of linear averaging over three measurements. The smoothed measurement trace (average) is also displayed.

Fig. 94 Averaging of scalar parameters

For numeric (scalar) result display, the results of all single measurements are considered. Square averaging of the scalar EVM parameter is shown as an example. The linear average and the standard deviation are calculated for these measurement parameters in addition to the square average value.

Average and extreme value functions are not available for display in the I/Q plane. Fig. 95 shows the different result displays that can be calculated from the I/Q

measurement and I/Q reference data records (PSK, MSK, QAM).

Page 80

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 80

QAM

Processing is very similar to that of PSK, but evaluation of amplitude statistics and signal scaling are performed in the first processing stages. As with PSK, the optimization criterion for the MATCHING stage is the minimization of RSM EVM.

VSB

Processing is very similar to that of PSK, but evaluation of amplitude statistics and signal scaling are performed in the first processing stages (as with QAM). In addition the pilot carrier typical for VSB are removed from the signals. As with PSK, the optimization criterion for the MATCHING stage is the minimization of RSM EVM.

MSK

Demodulation and matching are based on I/Q data records; the optimization criterion for the MATCHING stage is the minimization of RMS phase errors. All available samples are used, not only the decision points.

FSK

Output data of the demodulator stage (and therefore the basis for all subsequent stages) comprises real data records with instantaneous frequencies. Optimization criterion for the MATCHING stage is the minimization of the RMS frequency error between reference and measurement signal.

Trace

(I/Q Plane)

Trace

(t, Symb)

Numerical

Results

Statistic

Trace

Average

RMS, Mean,

StdDev,

Total Peak

Trace

Average

Error Calc

I/Q

(REF&

MEAS)

Num.Results

(Matching)

Signal & Error Trace

(Single Measurement)

Signal & Error Trace

(Averaged Measurement)

I/Q Data & Numerical Results

(from Demodulator & Matching)

Trace

Average

Fig. 95 Result display

3.3.7 Differences between Modulation Types

There are slight differences between the function blocks for QAM, VSB, MSK and FSK.

Page 81

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 81

3.4 Vector and Scalar Modulation Errors

Carrier

Modelling Modulation Errors

cos(2pi*f*t)

-sin(2pi*f*t+phi_q)

c1_i

c1_q

cos(2pi*f*t+phi_arb)

Amplitude

Inbalance

Quadrature

Offset

I/Q Offset

Baseband

Inputs

Modulated

RF Signal

Nonlin.

Distor-

sion

n(t)

Distorsion Noise

0 0.2 0.4 0.6 0.8 1 1.2

0.2

0.4

0.6

0.8

1.2

Ref Vector, Meas Vector, Error Vector

Real

Imaginary

Error Vector

Ref Vector

Meas Vector

3.4.1 Error Model of Transmitter

The following error model is used for the examples below:

Fig. 96 The following error model is used for the examples below:

3.4.2 Modulation Error (PSK, MSK, QAM, VSB)

3.4.2.1 Error vector (EV)

Fig. 97 Modulation error: error vector

Definition of error vector (EV):: The error vector is the difference between the measurement signal vector (Meas

vector) and the reference signal vector (Ref vector).

Page 82

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 82

3.4.2.2 Error vector Magnitude (EVM)

0 0.2 0.4 0.6 0.8 1 1.2

0.2

0.4

0.6

0.8

1.2

Phase Error, Magnitude Error, Phase Error, EVM

Real

Imaginary

Mag Error

EVM

Phase Error

Fig. 98 Modulation error: EVM, magnitude error

The error vector in the diagram is specified as error vector magnitude (EVM). The difference between the reference vector magnitude and the measurement vector magnitude is referred to as magnitude error.

In some modern networks, the basic EVM definition is modified so that the calculation is weighted with half the average signal power in the observed period. This is sometimes referred to as modulation error ratio (MER). In the case of ISI-free demodulation and measurements, the two definitions are identical.

Page 83

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 83

3.4.2.3 Phase Error

0 0.2 0.4 0.6 0.8 1 1.2

0.2

0.4

0.6

0.8

1.2

Phase Error, ErrorVectorPhase

Real

Imaginary

ErrorVectorPhase

Phase Error

 

;arg

REFMEAS

err





);arg(EV





Fig. 99 Modulation error: EVM, magnitude error

Fig. 99 illustrates the definition of the phase error: The phase error is the phase difference between the measurement vector and the

reference vector.

This measurement parameter is of great importance for MSK modulation measurements.

In contrast, the error vector phase is defined as:

The effects of the different modulation errors in the transmitter on the result display of the analyzer are described on the next pages. All diagrams show the equivalent, complex baseband signal. Errors for FSK are shown in the frequency/time diagram.

Page 84

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 84

3.4.2.4 IQ-Offset (Origin Offset)

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

IQ-Offset (Transmitter)

Inphase

Quadrature

Origin Offset

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

IQ-Offset (Analyzer)

Inphase

Quadrature

Fig. 100 Modulation error: origin offset (I/Q offset)

Fig. 101 Modulation error: compensation of origin offset

Fig. 100 and Fig. 101 show the effect of an I/Q offset or origin offset in the transmitter and in the analyzer after demodulation and error compensation.

The residual carrier of the amplitude C0 and any phase is superimposed on the ideal transmit signal. The result is a noise vector in the complex baseband that shifts the constellation diagram out of its complex 0 position. Fig. 100 shows an ideal constellation diagram and a diagram shifted by the I/Q offset.

This error parameter is determined during demodulation and deducted from the complex measurement data record.

The result after error compensation is shown in Fig. 101. The ideal constellation diagram is restored after demodulation. The unit circle around the constellation points remains unchanged.

Page 85

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 85

3.4.2.5 Gain Imbalance

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

gain inbalance (Transmitter)

Inphase

Quadrature

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

gain inbalance (Analyzer)

Inphase

Quadrature

Fig. 102 Modulation error: gain imbalance (transmitter)

Fig. 103 Modulation error: gain imbalance (analyzer)

The gain difference in the I and Q channels during signal generation in the transmitter is referred to as gain imbalance. The effect of this error on the constellation diagram and the unit circle are shown in Fig. 102. In the example, the gain in the I channel is slightly reduced which causes a distortion of coordinates in the I direction. The unit circle of the ideal constellation points has an elliptic shape.

This distortion is not corrected in the analyzer. It increases the EVM and is part of the displayed I/Q imbalance error. Fig. 103 shows that the analyzer chooses linear scaling for the measurement signal to minimize the RMS EVM. The elliptic shape of the unit circle remains unchanged.

Page 86

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 86

3.4.2.6 Quadrature Imbalance

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

quadrature inbalance (Transmitter)

Inphase

Quadrature

-1.5 -1 -0.5 0 0.5 1 1.5

-1.5

-1

-0.5

0.5

1.5

quadrature inbalance (Analyzer)

Inphase

Quadrature

Fig. 104 Modulation error: quadrature imbalance (transmitter)

Fig. 105 Modulation error: quadrature imbalance (analyzer)

Quadrature imbalance is another modulation error which is shown in Fig. 104 and Fig.

105. In this diagram, the I and Q components of the modulated carrier are of identical

amplitude but the phase between the two components deviates from 90°. This error also distorts the coordinates. In the example in Fig. 104 the Q axis is shifted. During demodulation in the analyzer, the phase is shifted in addition to linear amplitude

scaling to minimize the RMS EVM. The elliptic shape of the unit circle remains unchanged.

Page 87

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 87

Fig. 106 Modulation error: I/Q imbalance

;

Q"I"

jQ""

InbalanceIQ







The effect of quadrature imbalance and gain imbalance are combined to form the error parameter I/Q imbalance.

Fig. 106 shows this measurement parameter for the quadrature imbalance.

3.4.2.7 Gain Distortion

Fig. 107 Nonlinear distortions: amplitude distortion (transmitter)

Fig. 108 Amplitude distortion (analyzer)

Page 88

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 88

Fig. 107 illustrates the effect of nonlinear amplitude distortions on a 64QAM signal

-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0

-1

-0.8

-0.6

-0.4

-0.2

0.2

0.4

0.6

0.8

Amplitude Transfer Function (Transmitter)

Input Power (log)

Output Power / Input Power (log)

-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0

-1

-0.8

-0.6

-0.4

-0.2

0.2

0.4

0.6

0.8

Amplitude Transfer Function (Analyzer)

Input Power (log)

Output Power / Input Power (log)

(only the 1st quadrant is shown). The transfer function is level-dependent: the highest effects occur at high input levels while low signal levels are hardly affected. The signal is scaled in the analyzer so that the average square magnitude of the error vector is minimized. Fig. 108 shows the signal after scaling.

Fig. 109 Amplitude transfer function (transmitter)

Fig. 110 Amplitude transfer function (analyzer)

Fig. 109 and Fig. 110 show a logarithmic display of the amplitude transfer functions. The analyzer trace is shifted against the transmitter trace by this scale factor.

Page 89

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 89

3.4.2.8 Phase Distortion

-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0

-0.2

-0.15

-0.1

-0.05

0.05

0.1

0.15

0.2

Phase Transfer Function (Transmitter)

Input Power (log)

Phase Error (lin)

Fig. 111 Nonlinear distortions: phase distortion (transmitter)

Fig. 112 Phase distortion (analyzer)

Fig. 111 illustrates the effect of nonlinear phase distortions on a 64QAM signal (only the 1st quadrant is shown). The transfer function is level-dependent: the highest effects occur at high input levels while low signal levels are hardly affected. These effects are caused, for instance, by saturation in the transmitter output stages. The signal is scaled in the analyzer so that the average square magnitude of the error vector is minimized. Fig. 112 shows the signal after scaling.

Fig. 113 Nonlinear distortions: phase distortion (transmitter)

Page 90

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 90

Fig. 114 Phase distortions (analyzer)

-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0

-0.2

-0.15

-0.1

-0.05

0.05

0.1

0.15

0.2

Phase Transfer Function (Analyzer)

Input Power (log)

Phase Error (lin)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Noise (Analyzer)

Real

Imaginary

Fig. 113 and Fig. 114 show a logarithmic display of the phase transfer functions. The analyzer trace is shifted by the phase described above as against the transmitter trace.

3.4.2.9 Noise

Fig. 115 Additive noise

Fig. 115 shows a 64QAM signal (only the 1st quadrant is shown) with additive noise. The symbol decision thresholds are also shown.

The noise signal forms a "cloud" around the ideal symbol point in the constellation diagram. Exceeding the symbol decision boundaries leads to wrong symbol decisions and increases the bit error rate.

Similar displays are obtained in case of incorrect filter settings (transmitter filter or corresponding receive filter in the analyzer). When an incorrect filter is selected, crosstalk occurs between neighbouring symbol decision points instead of the ISI-free points. The effect increases the more the filtering deviates from actual requirements.

The two effects described cannot be distinguished in the I/Q constellation diagram but in statistical and spectral analyses of the error signal.

Page 91

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 91

3.4.3 Modulation Error (FSK)

0123456789-2

--1-

00.1

Fr time

0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

Freq

time [Symbols]

Fig. 116 Modulation error: reference signal (REFDEVCOMP = OFF) and measurement signal

Fig. 117 Modulation error: frequency error, reference signal not normalized

Fig. 116 shows the instantaneous frequency characteristic of the MEAS signal and the REF signal characteristic.

The FSK demodulator demodulates the signal down to symbol level and generates the REF signal using the transmitter filter and the reference deviation set.

A center frequency error is automatically compensated for during demodulation (as with PSK, MSK and QAM) and has no effect on subsequent error calculations.

The following error parameters are calculated by correlation or simply by forming the difference:

Deviation error = numeric value for the entire measurement range Frequency error = deviation from the instantaneous frequency of the two signals Fig. 117 shows the frequency error calculated from the MEAS and REF signals in Fig.

116. A striking feature is the modulation-dependent error signal variations.

Page 92

R&S FSQ-K70/FSMR/FSU-B73 Brief Description of Vector Signal Analysis (Function)

Software Manual 1161.8073.42 - 13 92

0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

Freq

time [Symbols]

0 1 2 3 4 5 6 7 8 9

-2

-1.5

-1

-0.5

0.5

1.5

Freq

time [Symbols]

Fig. 118 Modulation error: reference signal normalized

Fig. 119 Modulation error: frequency error, reference signal normalized

With FERDEVCOMP ON, the reference signal is scaled so that the RMS error between the scaled REF signal and the MEAS signal is minimized.

Fig. 118 shows the same MEAS signal as Fig. 119 and a REF signal with rescaled reference deviation.

The error plot (Fig. 119) no longer shows modulation-dependent variations; the errors are statistically distributed around the 0 frequency.

Page 93

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 93

4 Operation and Menu Overview

For softkeys that offer more than one setting, the softkey labelling indicates the current setting. For example, the following settings are possible for the measurement evaluation IQ Error:

● IQ ERROR VECTOR Display of I/Q error in the vector diagram

● IQ ERROR CONST. Display of I/Q error in the constellation diagram

The state of the softkey is indicated by its color: The measurement is switched off: The softkey is grey

The measurement is switched on with the display mode VECTOR. The softkey is highlighted in green, the setting VECTOR is indicated in brackets.

The measurement is switched on with the display mode CONSTELLATION The softkey is highlighted in green, the setting CONST is indicated in brackets.

Pressing the inactive softkey re-activates the measurement set last and the softkey colour changes from grey to green.

Error Signal

ERROR SIGNAL

IQ ERROR (VECTOR)

IQ ERROR

(CONST)

IQ ERROR

(VECTOR)

IQ ERROR

(CONST)

IQ ERROR

(VECT)

4.1 Operation

The R&S FSQ-K70/FSMR-B73/FSU-B73 option is menu-guided using keys, hotkeys and softkeys.

4.2 Special Features/Differences from the Basic Instrument

The standard unit is symbols. In some cases (e.g. RECORD LENGTH), time can be selected as the basic unit. If so, the values are automatically rounded up to the next integer that expresses the number of symbols.

4.2.1 Display of States Within Softkeys

Page 94

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 94

Pressing the active softkey open the window for selecting the softkey setting.

4.2.2 Display of Setting Parameters Within Softkeys

The current set value of some numeric entry parameters is displayed in the softkey labelling.

Examples: RECORD LENGTH LENGTH (with unit) RESULT LENGTH (without unit; SYMBOLS is used as the standard unit

here) The current set value can thus be immediately read off without opening

the associated softkey menu. The selected unit is also displayed in the labelling of softkeys that enable parameters to be entered with different basic units (e.g. TIME or SYMBOLS).

The ERROR STATISTIC and ERROR SPECTRUM softkeys offer additional evaluation modes:

When the ERROR STATISTIC softkey is selected, not the error parameter itself but its statistical distribution is output in the selected display mode (e.g. EVM).

When the ERROR SPECTRUM softkey is selected, a fast Fourier transform (FFT) for determining the spectrum is carried out for the selected type of display (e.g. EVM).

The basic display mode is restored by again pressing (switching off) the ERROR STATISTIC or the ERROR SPECTRUM softkey.

When a new display mode is activated (e.g. MAGNITUDE ERROR, PHASE ERROR), the ERROR STATISTIC and ERROR SPECTRUM softkeys are automatically switched off.

Suitable evaluation modes are available for the record buffer and the measurement and reference signal (see section 5.8).

IQ ERROR

(VECT)

IQ ERROR

(CONST)

RESULT LEN

(800)

DEMOD

SETTINGS

RESULT LEN

(800)

RESULT LEN

(1200)

ENTER

ERROR

SPECTRUM

ERROR SIGNAL

ERROR

STATISTIC

PHASE

ERROR

. . .

EVM

MAGNITUDE

ERROR

Page 95

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 95

Ref

-20 dBm

3.84 MHz

CF 1 GHz

3G_WCDMA_FWD

Error Signal

EVM

0 sym

799.75 sym

Symbols

79.975 sym/

CLRWR

EVM

FILT

Att

5 dB

Ref

-20 dBm

3.84 MHz

CF 1 GHz

3G_WCDMA_FWD

Error Signal

Spect(EVM)

0 Hz

7.68 MHz

768 kHz/

CLRWR

EVM

PRN

0.5

1.5

2.5

3.5

4.5

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

Fig. 120 Result display split screen EVM (upper diagram) ERROR STATISTIC + EVM (lower diagram)

Fig. 121 Result display split screen EVM (upper diagram) ERROR SPECTRUM + EVM (lower diagram)

STATISTIC: The unit and the scaling of the y-axis of the basic diagram is also used

for the x-axis of the statistic diagram.

SPECTRUM: The unit and the scaling of the y-axis of the basic diagram is also used

for the y-axis of the spectrum diagram. The scaling of the x-axis depends on the I/Q bandwidth.

Page 96

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 96

4.2.3 Measurement Window

The measurement window configuration is only slightly different from that of the basic instrument. Information on vector signal analysis has replaced the displays that are typical for the spectrum analyzer mode such as filter settings and sweep time (RBW, VBW, SWT). For displays of the measurement window that are not described here, refer to the documentation for the basic instrument.

The new fields above the measurement curve are provided to display the following:

● Digital standard or modulation mode

● Symbol rate

● Designation of the result display

The following status information is displayed in the curve: Warnings and status information on the current measurement (e.g. BURST NOT

FOUND) Consecutive number and number of measurements for averaging measurements Additional information on the type of filtering in signal processing is provided to the left

of the curve: RAW or FILT for measurements on non-filtered or measurement-filtered signals

Fig. 122 Measurement window of the R&S FSQ-K70/FSMR-B73/FSU-B73 option

Page 97

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 97

4.2.3.1 Warnings and Messages of Signal Processing Stages

Priority

Warning

Cause

Message suppressed in the presence of a warning with a higher priority

Very High

NO VALID SIGNAL

Demodulation not possible

High

END OF BUFFER

End of the recorded data set reached

Medium

BURST NOT FOUND

No burst in the signal, but BURST SRCH ON

Low

PATTERN NOT FOUND

No pattern in the signal, but PAT SRCH ON

BURST NOT FOUND

Depending on the type of input signal, various errors may occur during demodulation.

BURST NOT FOUND

The analyzer was parameterized with BURST SRCH ON (search for bursts = ON) but no burst was found in the signal.

PATTERN NOT FOUND

The analyzer was parameterized with PAT SRCH ON (search for patterns = ON) but no set synchronization pattern was found.

END OF BUFFER

The analyzer has reached the end of the captured data record. No more data for demodulation and measurement is present. This message occurs only if multiple evaluation mode (MULTI) as well as SINGLE SWEEP are active and no new data is captured automatically (AUTO CAPTURE = OFF).

NO VALID SIGNAL

The analyzer cannot demodulate the input signal. This message may occur if noise, an unmodulated carrier, or a signal with noncompliant modulation parameters is present at the input.

In the signal and modulation error traces, such measurements are marked with a warning on the function panel. If several warnings occur at the same time, only the warning with the highest priority is displayed on this panel and further ones are suppressed.

Table 9 Warnings displayed in the order of priority

With an error-free measurement, the name of the pattern found (e.g. GSM_TSC0) is displayed on this function panel. If a pattern search is not active, the panel remains blank.

4.2.3.2 Discarding a Measurement

With MEAS ONLY ON BURST and MEAS ONLY ON PATT, the analyzer only performs and displays measurements with a valid burst signal or pattern. Otherwise, both measurement is suppressed and status Message SEARCHING BURST or SEARCHING PATTER is indicated on the display. For averaged measurements with the setting BURST SRCH=ON, MEAS ONLY ON BURST should also be activated so that erroneous measurements do not affect the result of averaging. The same applies to pattern searches.

Page 98

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 98

4.3 Menu Overview

VSA

SPECTRUM

SCREEN B

EXIT VSA

SCREEN B

SETTINGS

HOME VSA

SCREEN C

PRESET VSA

4.3.1 Hotkeys

4.3.1.1 Assignment of the Hotkey Bar of the Basic Instrument

The position of the VSA hotkey varies depending on the type and number of installed options.

Fig. 123 Hotkey bar of the basic instrument with the R&S FSQ-K70/FSMR-B73/FSU-B73 option installed

4.3.1.2 Assignment of the Hotkey Bar of the Option

Fig. 124 Hotkey bar with the R&S FSQ-K70/FSMR-B73/FSU-B73 option switched on

Page 99

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 99

4.3.2 Softkeys

ZOOM

MODULATION

SETTINGS

MEAS

RESULT

FIT TRACE

DEMOD

SETTINGS

DIGITAL

STANDARD

ADJUST

REF LVL

RANGE

BURST &

PATTERN

BASEBAND

DIGITAL

RF PATH

BASEBAND

ANALOG

FACTORY

DEFAULTS

EX-IQ-BOX

DIGITAL BB

INFO

ALL

STANDARDS

GROUPS

PATTERN

FILTER

HOME VSA

IMPORT

EXPORT

PATH

STANDARDS

PATTERNS

MAPPINGS

EQUALIZERS

FILTERS

see option FSQ-B71

see option FSQ-B17

GENERIC

LIST

STANDARD

LIST

SAVE AS

STANDARD

DEFAULTS

DELETE

STANDARD

NEW

GENERIC..

EDIT

GENERIC..

DELETE

GENERIC STD

SAVE

EDIT

GENERIC STD

SHOW ALL

STANDARDS

INSERT

STANDARD

CANCEL

REMOVE

STANDARD

Digital

Standard

Page 100

R&S FSQ-K70/FSMR/FSU-B73 Operation and Menu Overview

Software Manual 1161.8073.42 - 13 100

POINTS/SYM

(4)

RESULT LEN

(800 Sym)

RESULT

FILT

ERROR SIGNAL

MEAS

SIGNAL

CAPTURE

BUFFER

SYMBOLS &

MOD ACC

MAGER CALC

(|MAX|)

RAW

EVM CALC

(|MAX|)

EQUALIZER

MAGER ABS

ON OFF

REFDEVCOMP

ON OFF

Meas Result

REAL / IMAG

SPECTRUM

SIGNAL

STATISTIC

PHASE ERR

MAGNITUDE

ERROR

SPECTRUM

AM & PM

CONVERSION

ERROR

STATISTIC

EVM

REAL / IMAG

FREQUENCY

PHASE

MAGNITUDE

(VECT)

(RELATIVE)

(WRAP)

EYE

FREQ ERR

(VECT)

(RELATIVE)

(I)

REF

SIGNAL

MAG CAP

BUFFER

SPECTRUM

SIGNAL

STATISTIC

REAL / IMAG

ZOOM

FREQUENCY

PHASE (WRAP)

MAGNITUDE

(LIN)

FREQ RESP

(LIN)

CHAN RESP

(LIN)

GROUP DELAY

REAL / IMAG

PHASE RESP

(WRAP)

OFFSET EMV

ON OFF

HIGHLIGHT

NORMALIZE

ON OFF

EQALIZER

TRAIN

EQUALIZER

ON OFF

EQALIZER

SAVE

EQALIZER

LANGTH

EQALIZER

DELETE

EQALIZER

STEP

EQALIZER

RESET

EQALIZER

LOAD

EQALIZER

FREEZE

FIT AUTO

FIT

BURST

SET SYMB #

(58SYM)

FIT ALIGN

(20%)

FIT ALIGN

CENTER

FIT ALIGN

LEFT

FIT

PATTERN

FIT ALIGN

RIGHT

FIT

TRIGGER

RECORD LEN

AUTO

RECORD LEN

(1.579 us)

FIT OFFSET

(-10SYM)

RESULT LEN

(456SYM)

PAT POS

(123SYM)

RECORD LEN

(800 SYM)

Fit Trace

ZOOM

START

DEMOD

NEXT RIGHT

DEMOD @

ZOOM START

DEMOD

RESTART

ZOOM

LENGTH

ZOOM

CAPTURE

AUTO OFF

MULTI

ON OFF