1. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 1 Conducted Power and Sensitivity Measurements Notice: This document has been prepared to assist IEEE 802.11. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.11. Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE 802.11 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at.http:// ieee802.org/guides/bylaws/sb-bylaws.pdfstuart.kerry@philips.compatcom@ieee.org Date: 2005-9-20 Authors:

2. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 2 Abstract This presentation demonstrates techniques for measuring the transmit power and receiver sensitivity of individual DUTs using off-the-shelf test equipment with traceable calibrations. The results of measurements made using these techniques are compared to conducted rate vs. range measurements made using a variable attenuator inserted between to DUTs.

3. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 3 Overview Introduction Transmit Power Receiver Sensitivity Individual DUT tests compared to variable attenuator rate vs. range tests. –RF Link Calculations –Rate vs. Range measurement –Comparisons –Problem Diagnostics Conclusions

4. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 4 Introduction The biggest limitation to most of the existing techniques proposed for measuring the RF performance of 802.11 devices is that these techniques rely on system level testing. –Devices are tested as a pair or more. Testing the transmit power and sensitivity of each individual DUT allows prediction of the resulting system performance for any combination of DUTs.

5. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 5 Introduction There are a number of important issues these tests must be able to address to obtain desired results. –Measure DUT in native operational mode (ensure testing can be performed by independent parties and discourage use of test modes that may not mirror normal operation). –Control the data rate being tested. –Keep the DUT at full power. –Generate necessary traffic, ideally without modifying DUT. Possible to meet these criteria with most DUTs.

6. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 6 Transmit Power Conceptually simple procedure: –Generate traffic from DUT at full power with desired data rate and packet size. –Measure average power of one or more packets. Generating traffic from the DUT: –A simple technique that will work for most 802.11 devices is to use the ICMP Echo request (ping the DUT with the desired packet size). Doesn’t require client software, etc. –Operate DUT in typical use case to test packets from normal traffic.

7. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 7 Transmit Power Controlling Data Rate –Interface to traffic generator with dot11SupportedDataRatesRx table set to desired data rate. –Send Echo request at desired data rate (doesn’t work in all cases) –Configure DUT to transmit desired data rate (requires some level of DUT control) –Selectively acknowledge packets to force DUT to desired data rate. Measuring power of packets: –Use calibrated wide bandwidth device such as vector signal analyzer or wideband power meter. –Detect pulse from DUT and calculate average power.

8. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 8 Transmit Power Sample Echo traffic at 11 MBPS

9. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 9 Transmit Power Anatomy of an 11 MBPS 802.11 PPDU Packet

10. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 10 Transmit Power PLCP Preamble and Header not necessarily at same power level as PSDU. –Differences in modulation schemes and data content can result in different levels. Power level of PSDU (more complex modulation) determines relationship with sensitivity of receiver. Ideally only measure power of PSDU. Also need to stay away from edges (rise/fall) of packet. –Typically use center 85% of packet for wireless protocols (GSM & TDMA).

11. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 11 Transmit Power Effect of modulation on transmit power. –Largest amplitude (peak power) determines limits on output power. –Constant envelope modulations can use total power of output amplifier (average power = peak power). –Non-constant modulation schemes such as QAM result in average power less than maximum output power of amplifier for same peak signal.

12. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 12 Transmit Power Effect of modulation on transmit power.

13. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 13 Transmit Power Conducted Transmit (Output) Power Measurement Configuration.

14. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 14 Transmit Power Conducted Transmit (Output) Power Measurement Configuration. –Cabled connection to DUT required. –Traffic generator (PC with wireless NIC) pings DUT If possible, DUT can also ping TG. Simplifies pulse triggering slightly. –Variable attenuator keeps path loss between TG and DUT at a point where input stages are not overloaded. If necessary, can also be used to help ensure full power and/or control TX data rate of DUT. –Directional coupler allows DUT signal to be routed to measurement receiver/power meter at higher level than that from TG. –Calibrated measurement path from DUT to receiver allows correction for path loss introduced by cabling and directional coupler. –Optional shield around DUT eliminates unwanted extraneous signals and/or inter-device coupling.

15. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 15 Transmit Power Component 1: PCI NIC.

16. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 16 Transmit Power Component 2: “Single Antenna” AP.

17. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 17 Transmit Power Varying levels of signal stability. –Different packet content can result in different power levels for non-constant envelope modulations. –Other internal oddities of a given design Points to the need to average multiple packets to get a valid reading of transmit power. For DUTs with marked max/min ratio (i.e. 1 dB or more) it may be desirable to use the minimum value for link budget calculations. –At a minimum, deviation should be reported. Other interesting effects observed. –20 dB drop in TX signal level at low path loss (explanation later)

18. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 18 Receiver Sensitivity Again, conceptually simple procedure: –Perform Packet Error Rate (PER) test by: Sending packets to DUT at a fixed level with desired data rate and packet size. Counting ACKs received from DUT. –Repeat PER test at continually lower packet power levels until desired sensitivity level is detected (typically 10% PER). Does not require a continuous search curve. Can be optimized starting with larger steps and narrowing down on target.

19. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 19 Receiver Sensitivity PER tester requires some special capabilities: –Packet source must be separate from receiver used to count ACKs. Must be able to reduce the PER tester output power across a broad dynamic range without affecting the receiver. Receiver needs its own attenuator (and possibly amplifier) to keep received ACKs in optimum range of receiver. Ensures that missed ACKs are due to the DUT failing to decode the packet and not due to the PER tester missing the response. Can be same radio with separate transmit/receive paths. –Packet source needs to offer calibrated output power levels. Build around stable calibrated source (VSG, Golden Radio). Use power meter to measure output packet power real-time.

20. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 20 Receiver Sensitivity Possible conducted PER setup:

21. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 21 Receiver Sensitivity Shielded enclosure around DUT is a MUST. –Most DUTs have insufficient shielding to ensure that only signal reaching DUT is conducted on cable. –Need to isolate DUT to ensure that external interference/collisions don’t affect measured PER. Packet generator and detector must have a high level of shielding as well. –Use of non-specialized test equipment (off-the-shelf 802.11 components) requires additional shielding. Several off-the-shelf and system solutions available. –Anritsu MT8860A used here.

22. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 22 Receiver Sensitivity Component 1: PCI NIC.

23. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 23 Receiver Sensitivity Component 2: “Single Antenna” AP.

24. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 24 RF Link Calculations. Combining the Transmit Power of Device 1 with the Sensitivity of Device 2 gives the NIC to AP link performance: 16.7 dBm – (-87.7 dBm) = 104.4 dB Combining the Transmit Power of Device 2 with the Sensitivity of Device 1 gives the AP to NIC link performance: 15.3 dBm – (-86.2 dBm) = 101.5 dB

25. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 25 Rate vs. Range Throughput from Device 1 to Device 2.

26. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 26 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation 101.5 dB < 103 dB < 104.4 dB

27. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 27 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation ~1.5 dB uncertainty in result not unexpected. –Attenuation step size is 1 dB for thoughput test and sensitivity test. –Power measurement uncertainty >0.5 dB It should be expected that the NIC to AP link will be the primary contribution given that the traffic is from NIC to AP. Interesting to look at this link vs. data rates…

28. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 28 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation PER can be measured for each data rate.

29. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 29 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Scaling PER by data rate gives throughput.

30. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 30 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Adding TX Power gives Throughput vs. Path Loss.

31. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 31 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Comparing prediction to measured values…

32. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 32 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Throughput from Device 1 to Device 2.

33. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 33 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Component 2: “Single Antenna” AP.

34. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 34 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Component 2: “Single Antenna” AP.

35. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 35 Comparison of Conducted Power/Sensitivity with Throughput vs. Attenuation Component 2: “Single Antenna” AP has a sensitivity problem and apparently a transmit signal level problem! Turns out the internal circuitry is based on a diversity reference design. At high RX power, the diversity channel (20 dB down from the test channel as seen through the diversity switch) leaks through and begins trying to communicate, resulting in packet errors. Above a certain level, the leakage from the unused diversity channel takes over (TX Power -20 dB from max!) and is used for all communication. Points to a possible problem using this AP in close proximity to the client.

36. Doc.: IEEE 802.11-05/0943r0 Submission September 2005 Dr. Michael D. Foegelle, ETS-LindgrenSlide 36 Conclusions This presentation shows clearly that it is possible to measure transmit power and sensitivity of individual DUTs using off-the-shelf components and accurately reproduce the results of rate vs. range type measurements performed on pairs of devices. The ability to qualify individual devices provides desired performance criteria for a specific product by itself. This testing also provides needed diagnostic information. This measurement method can now be extended for OTA testing.