S-38.3310 Tietoverkkotekniikan diplomity

1
S-38.3310 Tietoverkkotekniikan diplomityöseminaari
09.05.2006

• Title Automatic Bluetooth Radio Frequency
  Measurement System
• Author Jussi Savolainen
• Supervisor Prof. Sven-Gustav Häggman
• Instructor Petteri Yrjälä, M.Sc.
• Work carried out at Elcoteq Design Center Oy,
  Salo

2
Presentation outline

• Background
• Research problem
• Research methods
• Bluetooth essentials
• Bluetooth RF measurements
• System setup
• Software implementation
• Conclusions and further work

3
Background

• Bluetooth RF conformance testing by far the most
  expensive (time and cost) of the conformance
  tests, because
• off-the-shelf test systems gt 1 M (cost)
• response time using external test houses too long
  (time)
• Projects need a heap of test data in a short
  amount of time for statistical analysis, etc.
• Failing in type approval tests is expensive

4
Research problem

• How to automatically perform the Bluetooth RF
  conformance tests according to Bluetooth Test
  Specification version 2.0EDR in prequalification
  phase of the design
• How to perform the tests simultaneously for a
  number of devices
• How to support different Bluetooth products and
  product types

5
Research methods

• Literary study (Bluetooth RF specification,
  Bluetooth test specification, measurement device
  vendor application notes, etc.)
• Vendor audits/interviews
• HW Designer interviews
• UI usability study
• Data storage study

6
Bluetooth essentials

• Bluetooth connects portable or fixed equipment,
  and it is a cable replacing technology, which is
  cheap and robust with low power consumption
• Technology initially developed at Ericsson in
  1994
• Bluetooth SIG established in 1998, first version
  of specification out same year
• Most recent version 2.0EDR came out in 2004 with
  support to higher data rates than previously
• Operates on unlicensed (but not unregulated!)
  2.4GHz ISM band

ps
S
S
S
sb
M/S
piconet 2
M
S
ps
piconet 1
Scatternet with two piconets
7
Bluetooth essentials protocol stack

• Bluetooth protocol stack is a fuzzy structure
• Controller and Host can be separated using HCI,
  chipset and applications implemented separately
• Thesis discusses mainly radio layer

8
Bluetooth essentials packet structure

• Basic rate 1Mb/s
• Enhanced Data Rate 2Mb/s or 3Mb/s
• in EDR packets modulation is changed during the
  burst from 2FSK to PSK (backwards compatible)
• EDR mostly affects radio layer

General basic rate packet format
General enhanced data rate packet format
9
Bluetooth RF measurements - basics

• Device Under Test (DUT) can be placed to test
  mode, which allows configuration and control of
  the DUT
• Some tests must be performed under extreme
  temperature and voltage conditions
• Tests divided in transmitter (TX), receiver (RX)
  and transceiver (TRC) measurements
• TX and RX tests according to Bluetooth RF test
  specification
• TRC tests according to ETSI EN 300 328

10
Bluetooth RF measurements TX tests

• TX tests either power measurement as function of
  frequency or frequency as function of time

Measurement Description
Output power Maximum peak and average power of the burst
Power density Maximum RF-output power density
Power control Verification of power control
Frequency range Emissions inside the operating frequency within limits
20dB bandwidth Emissions inside the operating frequency within limits
Adjacent channel power Emissions inside the operating frequency within limits
Modulation characteristics Modulation index
Initial carrier frequency tolerance Transmitter carrier frequency accuracy
Carrier frequency drift Transmitter centre frequency drift within a packet
EDR relative transmit power Difference in power during 2FSK and PSK portions of packet
EDR carrier frequency stability and modulation accuracy Carrier frequency stability and modulation accuracy
EDR differential phase encoding Differential phase encoded data correctness
EDR in-band spurious emissions Unwanted signal levels on adjacent channels
11
Bluetooth RF measurements RX tests

• RX tests are BER measurements performed under
  different stimulus conditions

Measurement Description
Sensitivity Single slot packets Required sensitivity with non-ideal signal
Sensitivity Multi slot packets Required sensitivity with non-ideal signal
C/I performance Sensitivity in presence of co-/adjacent channel interference
Blocking performance Sensitivity in presence of out-of-band interference
Intermodulation performance Verification of receivers intermodulation characteristics
Maximum input level Receiver performance in presence of high power signal
EDR sensitivity Required sensitivity with non-ideal signal
EDR BER floor performance Receiver performance for 10-5 BER
EDR C/I performance Sensitivity in presence of co-/adjacent channel interference
EDR maximum input level Receiver performance in presence of high power signal
12
Bluetooth RF measurements TRC tests

• Out-of-band spurious emissions measured according
  to ETSI EN 300 328
• Does not belong to Bluetooth conformance tests
• Measures harmonics of the carrier signal, cabinet
  radiation and modulation products

13
System setup

• All Bluetooth RF conformance tests and
  out-of-band spurious emission measurement must be
  performed automatically with one setup
• 12 DUTs measured simultaneously
• Capable of measuring any Bluetooth device which
  implements at minimum the controller

14
System setup - equipment
Device Description
BT tester Bluetooth signaling tester capable of performing EDR measurements
RF shield Attenuates interfering RF signals
Temperature chamber Used in extreme temperature conditions (typ. -20ºC and 55 ºC)
Relay box Used to filter unwanted frequencies e.g. in out-of-band spurious measurement
IQ Signal generator capable of producing Bluetooth modulated interfering signals
CW Signal generator capable of producing continuous wave signals
Spectr an Spectrum analyzer capable of gating
GSM signaling GSM 3G signaling tester for coexistence testing

• Altogether 200 k
• Calibration of the transmission paths from 30MHz
  to 13GHz with 1MHz steps

15
Software implementation - UI
16
Software implementation data storage

• System creates 50Mb/day of data ? text files or
  excel sheets are not an option
• Writer (test system/test engineer) and reader
  (customer project/RF designer) are separated
• Graphs and reports must be easily extractable
  from database
• Short response time for new graphs and reports

17
Conclusions and further work

• This type of system is never finished as
  interface to a new DUT must be built every time
  and data storage is updated
• Error handling updates
• Calibration
• Measurement equipment SW updates

18
Questions?

• DANKE!

19
Worlds leading EMS providerto communications
technology customers