Agilent Technologies E4438C User manual

Agilent
Signal Studio for Bluetooth™
E4438C ESG Vector Signal Generator
Option 406
Application Note 1421
E4438C ESG vector signal
generator main features
• 250 kHz to 1, 2, 3, 4, or 6 GHz
frequency ran e
• Up to 64 me asamples (MSa)
waveform playback memory
• Up to 6 GB non-volatile waveform
stora e
• Sin le-ended and differential
I/Q outputs
• LAN and GPIB connectivity
Signal Studio for Bluetooth
main features:
• Intuitive user interface
• Fully coded Bluetooth packets and
Bluetooth modulated data streams
• Packet types: DH1, DH3, DH5, DM1,
DM3, DM5, AUX1, HV1, HV2, HV3, NULL,
POLL, ID
• Standards-based dirty transmitter
setup for receiver sensitivity tests
• Custom impairments: sinusoidal & linear
frequency drift, frequency offset, symbol
timin error, modulation index, AWGN
• Simplifies BER testin
• Plot BER vs. clock/ ate delay
• Plot the I/Q si nals and spectrum of the
confi ured baseband waveform
Use Signal Studio
to Create Bluetooth
Test Signals
Si nal Studio for Bluetooth software is
a powerful tool for creatin Bluetooth
baseband I/Q waveforms for use with the
A ilent E4438C ESG vector si nal enerator.
Select from thirteen different packet types
to confi ure, download and store the wave-
forms in the ESG, and use the ESG flexible
waveform sequencer to enerate custom
packet sequences.

2
www.agilent.com/find/esg
Software Overview
This application note is a self- uided tutorial describin the test si nals that can be created
with the Si nal Studio for Bluetooth software. This document is not meant to be a
Bluetooth technolo y tutorial. Basic knowled e of the Bluetooth radio specification is
required. For additional information on Bluetooth technolo y, refer to the References
section at the end of this document.
Software Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
Ease of Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
Si nal Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Bluetooth modulation and transmission parameters . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Bluetooth transmission scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Bluetooth link types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
Bluetooth packets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
Connectin to the ESG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Creatin Si nals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Step 1 - Confi ure the Bluetooth si nal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Step 2 - Setup the si nal eneration parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
Step 3 - Confi ure the ESG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
Step 4 - Calculate and download . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
Basic Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Receiver measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
ESG BER analysis capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
BER measurements setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
BER measurements example – clock/ ate delay calibration . . . . . . . . . . . . . . . . . . .38
Si nal Studio for Bluetooth Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42
Acronym List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44
Additional Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46
Recommended Confi uration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47
The Si nal Studio for Bluetooth software is an intuitive Windows®-based tool for confi -
urin Bluetooth packets and Bluetooth modulated data streams. The software calculates
an I/Q waveform file based on the Bluetooth si nal confi uration parameters set by the
user. Most waveforms take only a few seconds to build. The waveform file can then be
downloaded to the ESG internal baseband enerator for playback in arbitrary waveform
mode.
In addition to creatin waveform files, the software provides confi uration menus for key
si nal enerator settin s, includin frequency and amplitude. The instrument settin s,
alon with the I/Q waveform file, are passed to the ESG over the LAN or GPIB interface.
After downloadin the waveform file and instrument settin s to the si nal enerator, it
automatically be ins eneratin the Bluetooth modulated RF si nal. Local control of the
si nal enerator is then re-enabled and instrument settin s, like frequency and amplitude,
can be modified from the ESG front panel. The waveform files cannot be modified once
they have been downloaded to the instrument.
The waveform files can be saved in the ESG non-volatile memory and recalled for
playback at any time. In addition, the Si nal Studio software confi uration can be
saved to the host computer’s local hard drive and recalled at any time to re-calculate
and download the waveform to the si nal enerator for playback.
Table of Contents

3
www.agilent.com/find/esg
Si nal Studio simplifies creatin Bluetooth test si nals for use with ESG. Rather than
spendin valuable time codin Bluetooth packets at the bit level, use the software
to create fully coded, standard-based Bluetooth packets and Bluetooth modulated
data streams.
ntuitive user interface
Si nals can be rapidly confi ured in an easy-to-use raphical software environment.
Thirteen fully coded Bluetooth packet types
The software provides 13 different Bluetooth packet types to confi ure, includin both
voice and data packets. Because these packets are fully coded, demodulation and decod-
in capabilities of the DUT can be tested.
Bluetooth modulated data streams
Continuous, non-bursted, non-packetized Bluetooth modulated data streams can be con-
fi ured usin the software. These si nals are ideal for eneratin Bluetooth interference
si nals and basic test si nals that do not require decodin to perform BER analysis.
Add signal impairments
Create standards-based dirty transmitter or custom impaired test si nals. Simulate a
realistic Bluetooth device si nal by addin frequency, modulation, and noise impairments
to the si nal when performin receiver tests.
Waveform sequencing
Take advanta e of the ESG waveform playback memory (8 or 64 MSa) to playback lon er
waveforms or create custom packet sequences. The Si nal Studio for Bluetooth software
and the ESG built-in waveform sequencin capability combine to provide flexible eneration
of custom Bluetooth packet sequences.
Baseband and RF signals
Use the ESG sin le-ended and differential I/Q outputs to inject baseband si nals in the
transmit or receive paths of the device under test. Or drive the DUT usin the ESG RF port
tuned to the IF or carrier frequency of interest.
Simplified BER analysis
To facilitate receiver BER measurements on Bluetooth devices, an automated clock/ ate
delay calibration utility is provided in the software. Usin the utility, the data, clock, and
ate si nal timin ali nment at the input of the ESG internal BER analyzer can be easily
determined and modified to attain accurate results.
Ease of Use

4
www.agilent.com/find/esg
This section provides a quick review of key Bluetooth si nal characteristics. Additional
information about the Bluetooth si nal structure can be found in the Bluetooth
System Specification.
Bluetooth modulation and transmission parameters
Modulation parameters
Modulation type FSK – frequency shift keyin
Baseband filterin Gaussian, BT = 0.5
Maximum frequency deviation 140 – 175 kHz
Modulation index 0.28 – 0.35
Symbol rate 1 Msymbol/second
Transmission parameters
Frequency band 2.4 GHz ISM band
Transmission scheme TDD – time division duplex
Spreadin type FHSS – frequency hoppin spread spectrum
Number of channels 79
Channel spacin 1 MHz [(f = 2.402 + k) GHz, k = 0,1,2,…78]
Hop rate 1600 hops/second
Bluetooth transmission scheme
Time-division duplex
In Bluetooth systems, master and slave devices exchan e information by alternatin
packet transmissions in time-division duplex, Fi ure 1. The master starts its transmission
in even-numbered time slots only, and the slave starts its transmission in odd-numbered
timeslots only.
Figure 1. Bluetooth time-division duplex transmission scheme.
Frequency hopping
In normal operation, Bluetooth devices typically hop to a new carrier frequency each
timeslot. However, when performin the majority of transmitter and receiver tests as
indicated in the Bluetooth RF Test Specification, frequency hoppin is disabled.
Signal Structure

5
www.agilent.com/find/esg
Bluetooth link types
Two types of links can be established between Bluetooth devices:
Asynchronous Connection-Less ACL) link
The ACL link is a point-to-multipoint link between the master and all slaves participatin
on the piconet. It is considered a packet-switched connection primarily intended for
transmittin non-time sensitive information, like data. The fundamental packet len th in
Bluetooth systems occupies a sin le timeslot, as shown in Fi ure 1. Packets that occupy
multiple timeslots are used to achieve increased data rates on the ACL link because they
have lar er payload fields.
Synchronous Connection-Oriented SCO) link
The SCO link is a point-to-point link between the master and a sin le slave. It is
considered a circuit-switched connection primarily intended for transmittin time
sensitive information, like voice. All SCO link packets occupy a sin le timeslot; however,
the packet transmission rate varies dependin upon how much voice information is
carried in the packet payload. The objective is to maintain a 64 kb/s link for voice
transmission. However, because of the overhead associated with FEC codin , packets
with FEC on the speech data must be transmitted at a faster rate than packets with
no FEC to achieve an equivalent speech data rate (64 kb/s).
Bluetooth packets
Packet types
There are many different types of packets used to exchan e information on a Bluetooth
link. They can be enerally cate orized into three roups:
1. ACL link packets
2. SCO link packets
3. Control packets
Packet structure
All Bluetooth packet types are constructed based upon a standard packet structure,
Fi ure 2. The standard packet structure consists of three fields: access code, header,
and payload. In eneral, the access code and header fields carry overhead information
necessary to communicate over a Bluetooth link and the payload field carries the user data
to be exchan ed. The amount of information contained within the payload field and de ree
of forward error correction (FEC) chan es si nificantly dependin upon the packet type.
Figure 2. Standard Bluetooth packet structure.
Si nal Studio software supports 13 different packet types, includin both synchronous and
asynchronous link type packets. The Creating signals section of this document provides
additional detail about the structure of these packets and explains how to confi ure them
usin Si nal Studio.
Signal Structure
Access code
72 bits
Payload
0 to 2744 bits
Header
54 bits

6
www.agilent.com/find/esg
The Si nal Studio software must be installed in a host computer prior to creatin
Bluetooth test si nals. Before attemptin to download waveforms to the instrument,
verify that the host PC is communicatin with the ESG. For additional information on
connectin to the si nal enerator, refer to the E4438C Signal Studio Installation Guide,
available on the website: www.agilent.com/find/signalstudio.
This section describes how to confi ure and download Bluetooth si nals to the ESG
usin the Si nal Studio software. There are four basic steps:
Step 1 - Confi ure the Bluetooth si nal
Step 2 - Set up the si nal eneration parameters
Step 3 - Confi ure the ESG
Step 4 - Calculate and download
Step 1 Configure the Bluetooth signal
Figure 3. Signal Configuration and Payload Setup menus.
The Si nal Confi uration and Payload Setup menus provide an intuitive interface for
constructin Bluetooth modulated data streams and fully coded Bluetooth packets,
Fi ure 3. To distin uish the type of si nal select a mode from the Operation Mode
pull-down menu, provided in the Si nal Confi uration menu, Fi ure 4. Two modes
of operation are available to choose from: Data Stream and Packet Generator.
Figure 4. Operation Mode pull-down menu.
Connecting to the ESG
Creating Signals

7
www.agilent.com/find/esg
Data stream mode
When operatin in data stream mode, a waveform comprised of a continuous non-bursted
stream of data can be confi ured. This type of waveform is typically used as an inter-
ference si nal or a basic test si nal that doesn’t require decodin to perform BER
analysis. The si nal confi uration and Payload Setup menu parameters used to confi ure
Bluetooth packet structures (link type, packet type, device address, etc.) are disabled in
data stream operatin mode. Only fields necessary to confi ure a non-bursted raw data
stream can be modified.
To confi ure the data stream, select the desired data pattern in the Payload Setup menu.
The len th of the data stream can be defined by either settin the number of bits in the
len th field or settin the number of times the pattern is repeated in Repetitions field.
Figure 5. Setup for continuous PN15 Bluetooth modulated interference signal.
For example, to confi ure a continuous PN15 Bluetooth modulated interference si nal,
select PN15 in the data pattern pull-down menu and enter 1 in the pattern repetition
field, Fi ure 5. Alternatively, 32767 could have been entered into the len th field. When
the repetitions field is modified, the len th field is automatically updated, and vice versa.
When downloaded to the ESG, the baseband enerator will continuously repeat this data
pattern durin waveform playback.
Creating Signals

8
www.agilent.com/find/esg
Creating Signals User file format
Both modes of operation, Data Stream and Packet Generator, allow users to select
User File as the Data Pattern. This data must be in binary form. The most si nificant
bit (MSB) of the first byte in the user data file will be the first bit of the user payload.
The MSB of the second byte in the user file will be the ninth bit of the user payload,
and so on. An example MATLAB®pro ram to enerate and save user data is included
in the appendix section of this document.
User File
If User File is selected as the data pattern, the number of bits in the user’s data file will
equal the value displayed in the Pattern Len th field, when the Pattern Reps field is set
to one. The number of times the pattern is repeated can by selected by enterin the
desired repetition value in the Pattern Reps field. The Pattern Reps and Pattern Len th
fields are coupled. If one of these fields is chan ed, the other field will automatically be
updated.
Packet generator mode
When operatin in Packet Generator mode, the software creates fully coded, standard-
based Bluetooth packets. Table 1 provides a summary of the different packets that can be
confi ured usin the software.
Figure 6. Link Type pull-down menu.

9
www.agilent.com/find/esg
To confi ure a packet, first set the Link Type pull-down menu to ACL or SCO, Fi ure 6.
Then select the desired packet in the packet type pull-down menu. A different set of
packets is available to choose from in the packet type pull-down menu dependin on
which link type is selected, see Table 1. Control packets are always available because
they are common to both link types.
Table 1. Summary of packet types that can be configured using Signal Studio for Bluetooth.
ACL and SCO packets
The Si nal Studio software performs the all the necessary codin to create the selected
packet type; includin forward error correction (FEC) and cyclic redundancy check (CRC)
for the header and payload fields when required. As a result, the process for confi urin
the different ACL and SCO packets is very similar. Only a few basic parameters need to
be set so that the access code, header, and payload fields of the Bluetooth packet can
be created.
Access code
The access code field is created based on the user-defined Bluetooth Device Address
(BD ADDR). The BD ADDR is a unique 48-bit device address allocated to each Bluetooth
transceiver. To confi ure the BD ADDR, enter a 48-bit device address in hexadecimal
notation (most si nificant bit to least si nificant bit).
Header
The header field is created usin the type code of the selected packet type and the
active member address (AM ADDR) of the Bluetooth device for which the packet is
intended. The AM ADDR is a 3-bit address used to distin uish between the active
participatin slaves on a piconet. To confi ure the AM ADDR, enter the 3-bit address
in octal notation (most si nificant bit to least si nificant bit).
Payload
The structure of the payload field varies dependin on which packet type is selected.
Table 2 provides a summary of the payload field structure for all packet types supported
by the software. Additional details re ardin the packet structure can be found in the
Bluetooth System Specification.
Creating Signals
Packet
Type
Description Len th
ACL SCO Control Code (Slot)
(b
3
b
2
b
1
b
0
)
DH1
•
0100 Hi h data rate packet 1
DH3
•
1011 Hi h data rate packet 3
DH5
•
1111 Hi h data rate packet 5
DM1
••
0011 Medium data rate packet 1
DM3
•
1010 Medium data rate packet 3
DM5
•
1110 Medium data rate packet 5
AUX1
•
1001 Auxiliary packet 1
HV1
•
0101 Hi h quality voice packet 1
HV2
•
0110 Hi h quality voice packet 1
HV3
•
0111 Hi h quality voice packet 1
NULL
•
0000 Null packet 1
POLL
•
0001 Poll packet 1
ID
•
n/a Identification packet 1

10
www.agilent.com/find/esg
Table 2. Basic structure of the packets supported by Signal Studio for Bluetooth.
The Payload Setup menu offers a variety of custom payload data confi urations for
sin le and multi packet sequences. First select whether the data pattern is to be
distributed into a sin le packet or over a multiple packet sequence usin the Data
Distribution pull-down menu.
Configuring a single packet
When Sin le Packet is selected, Fi ure 7, the software creates a waveform that consists
of one Bluetooth packet. When downloaded to the ESG, the waveform is repeatedly
played back resultin in a packet sequence comprised of identical packets.
Figure 7. Example single packet setup.
The payload field, for the selected packet type, is filled to capacity with the data type
selected in the data pattern pull-down menu. If PN data patterns are selected, the
PN sequence is truncated after the maximum amount of user data has been placed
in the packet.
User File
If User File is selected as the data pattern, the number of bits, shown in the Pattern
Len th field, equals the number of bits in the user data file, up to the number of
bits allowed for that packet type. Any remainin bits in the user data file beyond the
maximum amount will be truncated.
The Pattern Len th and Pattern Reps fields are automatically updated as different packet
types and data sequences are selected. However, these fields cannot be modified when
Sin le Packet has been selected from the Data Distribution pull-down menu.
Configuring a multiple packet sequence
When Multi Packet is selected, Fi ure 8, the software creates a waveform that consists
of a multiple packet sequence. A data pattern is distributed into the payload portion of
each packet in the sequence. Once downloaded to the instrument, the si nal enerator
repeatedly plays back the entire packet sequence. The net result is the ability to confi ure
lon er data sequences for analysis.
Creating Signals
Packet Access Header Payload Payload summary
code field field field Header User data FEC CRC
(bytes) (bytes)
DH1
•••
1 0 to 27 no yes
DH3
•••
2 0 to 183 no yes
DH5
•••
2 0 to 339 no yes
DM1
•••
1 0 to 17 2/3 yes
DM3
•••
2 0 to 121 2/3 yes
DM5
•••
2 0 to 224 2/3 yes
AUX1
•••
1 0 to 29 no no
HV1
•••
n/a 10 1/3 no
HV2
•••
n/a 20 2/3 no
HV3
•••
n/a 30 no no
NULL
••
POLL
••
ID
•

11
www.agilent.com/find/esg
Figure 8. Example multi packet sequence setup.
The data type selected in the data pattern pull-down menu is spread over the payload
portion of the packet sequence, Fi ure 9. The len th of the data stream can be defined
by either settin the number of bits in the len th field or the number of times the pattern
is repeated in repetitions field. When either of these fields is modified, the other is
automatically updated. The number of packets required to accommodate the len th
and repetition field settin is automatically updated in the Payload Summary section of
the menu.
Figure 9. Data Pattern pull-down menu.
The payload of each packet in the sequence is filled to capacity with the selected data
pattern if possible. When the len th of the data pattern is selected such that it is not
divisible by an inte ral number of packet payloads, a packet with a partial payload is
enerated to accommodate the remainin portion of the data pattern. This packet is
then appended to the sequence of full packets.
Note: The payload portion of a Bluetooth packet must be filled with an inte ral number of data bytes
(1 byte = 8 bits). If the len th of the data pattern is not evenly divisible by ei ht bits, then zeros are appended
to the selected data pattern until this criteria is met.
Two packet sequence presets, continuous PN9 and continuous PN15, are provided in
the data pattern pull-down menu when multi packet is selected, Fi ure 9. These presets
are intended to simplify the setup process when confi urin a packet sequence with
continuous PN data distributed over the payload portion of each packet payload. When
continuous PN9 or PN15 is selected in the Data Pattern pull-down menu, the software
automatically confi ures the Len th and Repetitions fields, Fi ure 10. The number of
packets required to accommodate the selected continuous PN pattern is also
automatically determined and updated in the Payload Summary section.
Creating Signals

12
www.agilent.com/find/esg
Figure 10. Payload Setup for a multi packet sequence with continuous PN9 user data.
Payload summary section
Full packets – Number of packets with payload filled to capacity.
Partial packets – Because the payload portion of each packet is filled to capacity if
possible, the number of partial packets appended to the end of a packet sequence will
always be 0 or 1.
Pad packets – When the software creates a multiple packet sequence, the sequence
bit in the header field is to led after each packet transmission in the sequence. This
indicates to the Bluetooth device that the incomin packet is not a re-transmitted packet.
To maintain an alternatin sequence bit when the waveform is repeatedly played back, an
even number of packets must be enerated. If the data pattern len th is set such that an
odd number of packets are required to send the data, then a pad packet is appended to
the end of the packet sequence. The pad packet has no payload and is only enerated in
this circumstance to accommodate the alternatin sequence bit.
User File
If User File is selected as the Data Pattern, the bits in the user data file will be distributed
across multiple packets. The payload of each packet in the sequence is filled to capacity
with the user’s data file, up to a maximum of 80,000 bits. The packets will contain the
maximum number of bits for each packet type as lon as bits are available from the
user’s data file. The same conditions apply to the User File as the other, multipacket
Data Patterns mentioned above, except the Pattern Reps settin is fixed at 1 repetition.
Creating Signals

13
www.agilent.com/find/esg
Calculating a long multiple packet sequence
If a PN15, Continuous PN15, or a User File is chosen as the data pattern, a calculation
warnin will appear. See Fi ure 11. The amount of time it takes your PC to calculate
the si nal depends on the PC’s amount of RAM and the speed of the processor. As the
amount of samples shown in the Projected Len th field increases, so does the amount
of time it takes the PC to calculate the si nal. If the calculation takes lon er than
20 minutes, it is likely that it has failed, even if an error messa e is not visible.
Figure 11. Calculation Warning message for multi packet generation using
PN15 sequences and user files
Control packets
As indicated in Table 2, the ID, NULL, and POLL packets do not contain a payload field. As
a result, the payload setup menu is disabled when one of these packet types is selected in
the packet type pull-down menu.
NULL and POLL packets are composed of an access code and header field (126-bit fixed
len th). When confi urin NULL and POLL packets, only the BD ADDR and AM ADDR
need to be set.
The ID packet is comprised of only an access code (68-bit fixed len th). Because the
ID packet does not contain header and payload fields, only the BD ADDR needs to
be confi ured.
Bursting
In Packet Generator mode, burstin can be enabled or disabled in the Si nal
Confi uration menu, Fi ure 12.
Figure 12. Disable Bursting or set the Power Ramp and Settling times.
When enabled, the packet transmission timin of the test si nal adheres to the
Bluetooth TDD slot structure for the selected packet type. A summary of the
transmission rates for each the packet types supported by the software is provided
in Fi ure 13.

14
www.agilent.com/find/esg
Figure 13. Bursted transmission timing relationships for different packet types.
At the be innin and end of a packet transmission, the ESG ramps power followin a
user-defined burst profile. The burst profile is defined by settin the Power Ramp and
Ramp Settlin fields with the desired time duration (in microseconds). See Fi ure 13.
The ramp profile is the duration of time prior to the first symbol transmission, over
which the carrier frequency is ramped from idle power to transmit power. The power
ramp is shaped with a cosine function, and len th of the ramp can be adjusted. The
Ramp Settlin field sets the len th of time allowed for the power level to stabilize prior
to the first symbol transmission of the first bit of the preamble, tpo. The settlin time
includes the power ramp time as indicated in the Bluetooth specification, see section
6.2.2 of the SIG’s RF Test Specification under the “Reference Si nal Definition” section.
The default settin s of a 6 µs power ramp and ramp settlin time produce the ramp
profile shown in Fi ure 14.
Figure 14. Burst profile of a single slot packet transmission with 6 µs power ramp and ramp settling.
A reference burst profile for si nal enerator test si nals is indicated in the Bluetooth
RF test specification. This burst profile is recommended when performin the dirty
transmitter test. The reference burst power ramp is 2 µs and the ramp settlin time
is 4 µs. This ramp profile is shown in Fi ure 15.
Figure 15. Burst profile of a single slot packet transmission with 2 µs power ramp and 4 µs ramp settling.
Creating Signals
Time
Power
Tx Timeslot period = 625 µs
Access
Code Header Payload
6µs
power ramp
down
power settlin
6µs
6µs
power ramp
up
t
po
Time
Power
Tx Timeslot period = 625 s
Access
Code Header Payload
power settlin
4s
4 s
power ramp
up
2s
t
po
4s
power ramp
down

15
www.agilent.com/find/esg
When modifyin the burst profile, keep in mind that as the rise/fall time of the bursted
si nal is decreased, the amount of spectral splatter is increased due to burstin .
When burstin is disabled, the ESG transmits concatenated packets with no power
rampin . The effect of disablin bursted transmission is shown in Fi ure 16. Notice that
there is no off time between packets when bursted packet transmission is disabled.
This concept can be extended to all the packet types shown in Fi ure 13. This non-bursted
transmission structure is similar to data stream mode; however, in data stream mode
there is no access code or header associated with the data transmission.
Figure 16. Bursted and non-bursted single slot packet transmission.
Step 2 Set up the signal generation parameters
Figure 17. Signal Generation Setup and Signal nformation menus.
The Si nal Generation Setup menu, boxed in Fi ure 17, provides a simple interface for
definin the oversamplin ratio, mirrorin the spectrum, delayin clock and ate si nals,
and addin si nal impairments.
Oversampling ratio
The oversamplin ratio defines the number of samples calculated per I/Q symbol.
Increasin the oversamplin ratio of the si nal increases the separation of the samplin
ima es from the desired si nal. This allows for better ima e rejection by the baseband
reconstruction filter. However, the improved ima e rejection comes with a price. Increasin
the oversamplin ratio increases the waveform calculation time and file size. Notice that
the projected file len th in the si nal information section is updated as the oversamplin
ratio settin is increased. The default settin is sufficient for most applications. Additional
benefits of modifyin the oversamplin ratio are discussed in the Clock/gate delay
section.
Creating Signals

16
www.agilent.com/find/esg
Mirror spectrum
Enablin the mirror spectrum feature inverts the Q channel, resultin in a mirrored
spectrum. As a si nal normally propa ates throu h the different functional blocks of a
receiver, for example the mixer block, the si nal spectrum may be reversed. Usin this
feature facilitates realistic testin of receiver functional blocks that would normally be
presented with a mirrored spectrum si nal.
Clock and gate signals
The Si nal Studio software makes use of the ESG markers to enerate clock and ate
si nals alon with the confi ured waveform. These si nals are necessary to perform BER
analysis on Bluetooth packets and data streams usin the ESG internal BER analyzer.
A symbol clock is enerated on the ESG event one port for all waveforms created usin
the Si nal Studio for Bluetooth software. Because Bluetooth is 2-level FSK modulated
(1 bit/symbol), the symbol clock can be used to indicate the bit rate of the incomin
data sequence to the ESG internal BER analyzer. When operatin in packet enerator
mode, a payload data ate si nal is also provided on the ESG event two port. This ate
si nal is used to recover continuous PN9 payload data from an incomin Bluetooth
packet sequence for BER analysis. Refer to the Basic measurements section for a
detailed example of how these si nals are used to perform BER analysis.
Clock/gate delay
When clock and ate si nals enerated by the ESG are used to perform BER analysis, it is
important to realize that the test si nal transmitted by the ESG experiences a propa ation
delay throu h the device under test. As a result, the demodulated loopback si nal must
be reali ned in time with the clock and ate si nals at the input of the BER analyzer,
Fi ure 18. Delay control over the clock and ate si nals is provided by the software to
enable reali nment with the test si nal at the input of the BER analyzer.
Figure 18. To perform BER analysis using this measurement setup, the clock/gate delay setting
(∆) must be equal to the propagation delay the test signal experiences through the Bluetooth
device (a).
If the propa ation delay characteristics for the device under test are known, enter the
delay value in the clock/ ate delay field durin waveform confi uration. The clock and
ate si nals associated with the waveform will be delayed by the indicated amount durin
waveform playback. The resolution of the clock/ ate delay parameter is directly coupled
to the oversamplin ratio settin . It can be determined by dividin the symbol period
(1 µs) by the oversamplin ratio. To increase the incremental delay resolution, increase
the oversamplin ratio of the confi ured waveform. When doin so, remember that
increasin the oversamplin ratio also increases the projected len th of the waveform.
Creating Signals
E4438C ESG vector si nal enerator
• Option 602 Baseband enerator with 64 MSa
• Option 406 Si nal Studio for Bluetooth
• Option UN7 Internal BER analyzer
BER
ate in
Event 2BER
data in
BER
clock in
Event 1
Delay = (t + a)
Delay = (t + ∆)
Delay = (t + ∆)
Bluetooth
DUT
Propa ation
delay = a
RF test si nal
Demodulated test si nal
Rear panel

17
www.agilent.com/find/esg
Clock/gate delay calibration
If the delay parameter is unknown, leave the default value (0 µs) in the clock/ ate delay
field and finish confi urin the waveform. After the waveform has been calculated, the
clock/ ate delay calibration utility can be used to determine the delay characteristics of
the device under test.
Figure 19. Signal Generation Setup menu.
After the waveform has been downloaded to the ESG, select the calibrate button in the
Si nal Generation Setup menu, Fi ure 19, to brin up the clock/ ate delay calibration
utility, Fi ure 20. Once confi ured, the calibration utility enerates a plot of BER versus
clock/ ate delay. From the plot, the correct clock ate delay settin to reali n the
si nals at the input of the BER analyzer can be determined.
Figure 20. Clock/Gate Delay Calibration utility.
The Clock/Gate Sweep Setup menu is used to confi ure the calibration utility, Fi ure 21.
Enter the Clock/Gate Delay ran e in the Start and End fields. Typical propa ation delay
throu h a Bluetooth device ran es from one to 10 µs; however, dependin on the test
setup, lon er delays are possible.
Figure 21. Clock/Gate Sweep Setup menu.
Creating Signals

18
www.agilent.com/find/esg
The incremental delay step at which BER tests are performed is set in the step size field.
The resolution of this field is also directly coupled to the oversamplin ratio settin as
described previously. The Repeat field is used to set the number of BER test iterations for
each clock/ ate delay increment. When this field is set to two or reater, the BER results
are plotted on the raph for each iteration of the measurement. By increasin the number
of BER test repetitions, it can easily be determined if repeatable results are achievable.
The ESG internal BER analyzer settin s required to perform the clock/ ate delay
calibration are also confi ured remotely usin the Si nal Studio software, Fi ure 22. The
total number of data bits to be analyzed at each delay increment is set in the total bits
field. The ate si nal is automatically enabled in packet enerator mode and disabled
in data stream mode. The default polarity of the clock, data, and ate si nals is set to
positive. This is the correct settin for the basic measurement setup and should not be
modified unless there is a need to do so in a particular application.
Figure 22. ESG internal BER Analyzer setup menu.
Once confi ured, initiate the BER test sequence by pressin the Start button. A plot of
BER vs. clock/ ate delay will be incrementally filled in as each BER test in the sweep
is completed. With no impairments added to the si nal, a delay that results in 0 % BER
should be achievable.
After the plot has been enerated, a marker is automatically placed at the clock/ ate
delay settin with the minimum BER test results. The results section directly below
the raph is updated to reflect the marker position. The arrow buttons can be used to
incrementally move the marker to other points on the curve. The clock/ ate delay settin
and correspondin BER are continuously updated in the results section as the marker
is moved across the raph. If the Repeat field is set to 2 or reater, the minimum,
maximum, and avera e BER is provided in the results section for each clock/ ate delay
increment.
An example plot is provided in Fi ure 23. In this case, the repetitions field was set to
three. The dots indicate the individual results for each clock/ ate delay increment
(note that some results overlap). The trian les are placed at the avera e BER for each
clock/ ate delay increment.
Creating Signals

19
www.agilent.com/find/esg
Figure 23. Plot of BER vs. Clock/Gate Delay with repetitions field set to three.
The statistics displayed in the box in fi ure 23 are the current marker value and not for
the entire plot. All the repetitions are summed internally. The maximum, minimum, and
avera e for each delay point are plotted on the raph.
When the Set Delay Button is selected, see fi ure 20, the delay settin of the current
marker location is returned to the Clock/Gate Delay field in the Si nal Generation Setup
menu. As a result, it is important to remember to return the marker to the delay settin
that yields the minimum BER test results prior to pressin the Set Delay button.
After pressin the Set Delay button, select the OK button to close the Clock/Gate Delay
calibration utility and return to the main user interface. The BER calibration can be
aborted at any time by pressin the Cancel button.
An example is provided in the Basic Measurements section of this document that details
the measurement setup and procedure for performin Clock/Gate Delay calibration.
Marker rotate SCPI command
To adjust the clock/ ate si nal delay, the Si nal Studio software sends a marker rotate
SCPI command to the ESG baseband enerator. The marker rotate feature currently only
exists in the Si nal Studio for Bluetooth software. There is no equivalent soft key available
in the ESG dual arb menu. The marker rotate SCPI command can also be used when settin
up custom SCPI based test routines. The command and description are provided below
for reference.
Marker rotate
[:SOURce]:RADio:ARB:MARKer:ROTate “filename”, <rotate_count>.
The marker rotate SCPI command shifts the I/Q waveform markers by the indicated
number of samples in the rotate_count field. The shift is based on the current position of
the markers. The rotate_count cannot exceed +/- (number of sample points - 1).
Creating Signals Max BER
Marker
Avera e BER
Min BER

20
www.agilent.com/find/esg
Signal impairments
The Si nal Studio software provides a strai htforward raphical menu for addin
impairments to the Bluetooth si nal. Custom- and standards-based dirty transmitter
impairments can be added to the Bluetooth si nal. See Fi ure 24.
In eneral, the impairments are derived from the frequency and modulation tolerances
specified for Bluetooth transmitters. Addin these impairments enables the evaluation of
receiver performance with si nals realistically anticipated from Bluetooth transmitters.
All impairments required by the Bluetooth RF Test Specification to perform sin le- and
multi-slot sensitivity tests are provided in Si nal Studio for Bluetooth.
The full complement of impairment conditions can be used for testin a iven frequency,
or a study can be done to determine the reatest contributors to error in symbol timin
recovery. Many of the specified impairment conditions are not stressin enou h to
access the desi n limitations of the devices. These studies would allow the desi ner
to optimize the receiver al orithms for symbol recovery, and extend the sensitivity
levels achieved.
Dirty transmitter impairments
Clickin the Dirty Transmitter mpairments Setup button in the Si nal Generation Setup
menu brin s up the Dirty Transmitter Test confi uration menu. See Fi ure 25. This menu
is completely independent of the main Si nal Studio for Bluetooth menu. The settin s in
the Dirty Transmitter Test confi uration menu are not coupled with the main Si nal Studio
for Bluetooth menu, and vice versa. While the Dirty Transmitter Test confi uration menu
is active, settin s in the main Si nal Studio for Bluetooth menu are disabled and cannot
be modified until the Dirty Transmitter Test menu is closed.
Creating Signals
Figure 24. Add either custom- or standards-based dirty transmitter impairments.
Figure 25. Dirty Transmitter Test menu