doc.: ieee 802.11-05/0181r0 submission march 2005 c. young, et alslide 1 legacy device testing with...

March 2005

C. Young, et al

Slide 1

doc.: IEEE 802.11-05/0181r0

Submission

Legacy Device Testing with Mixed Mode Preambles

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 <http:// ieee802.org/guides/bylaws/sb-bylaws.pdf>, 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 <[email protected]> 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 <[email protected]>.

Date: 2005-03-11

Name Company Address Phone email Christopher Young

Broadcom 190 Mathilda Place, Sunnyvale, CA 94086

+1 408 543 3334

[email protected]

Tushar Moorti Broadcom 190 Mathilda Place, Sunnyvale, CA 94086

+1 408 543 3352

[email protected]

Christopher Hansen

Broadcom 190 Mathilda Place, Sunnyvale, CA 94086

+1 408 543 3378

[email protected]

Authors:

March 2005

C. Young, et al

Slide 2

doc.: IEEE 802.11-05/0181r0

Submission

Abstract

• Measurement results of proposed .11n mixed mode preambles with legacy devices are presented.

March 2005

C. Young, et al

Slide 3

doc.: IEEE 802.11-05/0181r0

Submission

Motivation

• Previous tests had indicated that using the WWiSE preamble can cause high error flooring for some legacy devices (see 802.11-05/0006r1 from January 2005 meeting in Monterey)

• Our testing indicates that this assertion is true for one device.

• A new mixed mode preamble is needed.

• We note that as long as the cyclic shift is the same for both the STS and LTS, there is no error floor.

• The WWiSE mixed mode preamble has been updated accordingly.

March 2005

C. Young, et al

Slide 4

doc.: IEEE 802.11-05/0181r0

Submission

Motivation

• Tested legacy devices using these proposed preambles– 2 TX antenna preambles:

• WWiSE ( +/-400ns STS, -100ns LTS shifts)

• TGnsync ( +50ns STS, +50ns LTS shifts)

• Compromise 1 ( -400ns STS, -400ns LTS shifts)

• Compromise 2 ( +400ns STS, +400ns LTS shifts)

• Compromise 3 ( -200ns STS, -200ns LTS shifts)

• Compromise 4 ( +200ns STS, +200ns LTS shifts)

– “-” is cyclic advance, “+” is cyclic delay

• 3 and 4 TX testing is pending

March 2005

C. Young, et al

Slide 5

doc.: IEEE 802.11-05/0181r0

Submission

Motivation

• Testing methodology same as before (802.11-05/1590r0)– Controlled environment to determine sensitivity and error floors

• Legacy devices tested– Atheros (Netgear 108 Mbps Wireless PC card WG511T, driver version 2.4.0.7181

6/20/03)

– Broadcom (internal reference card, representative of publicly available devices)

– Conexant (Intersil PRISM LAN 802.11a/g Adapter, driver version 1.0.20.0 7/9/03)

– Intel (Intel PRO/Wireless 2200BG Adapter, driver version 9.0.1.9 10/29/04)

– Texas Instruments (US Robotics 802.11g Wireless Turbo Adapter, driver version 4.0.40.3 7/21/03)

– Marvell (Netgear WG511 v2, driver version 3.1.0.19 9/17/04)

• Two new devices (Intel and Marvell) added since testing originally done last year

March 2005

C. Young, et al

Slide 6

doc.: IEEE 802.11-05/0181r0

Submission

Test Setup

AWG-50 dBmMAC addr = AAASSID = ‘test’

2:1 Splitter

Attenuator 2:1 Splitter

DUT

Sniffer (Brcm device)

2:1 Splitter

50dB pad

Driver runningMAC addr = BBBSSID = ‘test’

B

CAWG = Arbitrary Waveform Generator (Agilent E8267C)DUT = Device Under Test

No driver runningMAC addr = AAA (same as packets from AWG)SSID = ‘test’

A

March 2005

C. Young, et al

Slide 7

doc.: IEEE 802.11-05/0181r0

Submission

Transmit Signal Details

• Packets created in MATLAB. Each packet is a 6 Mbps signal, with the DUT’s destination address/BSSID (unicast), data frame, 5 data bytes long

– Short packet because we just want to check that PLCP was properly decoded for legacy devices (I.e., legacy devices will properly detect mixed mode MIMO transmission and hold off its own transmissions)

• For candidate preambles: – TX1 antenna has legacy preamble. 500 packets created, each packet convolved

with randomly generated channel h1(t)– TX2 antenna has candidate cyclically shifted STS/LTS/SIG/DATA. 500

packets created, each packet convolved with randomly generated channel h2(t) (Note: h1(t) does not equal h2(t))

– Sum the two, 500 packets (with 100 usec gaps) saved for playback with the AWG

March 2005

C. Young, et al

Slide 8

doc.: IEEE 802.11-05/0181r0

Submission

Received signal at the DUT

• Received power level carefully calibrated at point B.

• If the DUT was able to receive and correctly decode the unicast packet, then an acknowledgement frame (ACK) will be sent back after a SIFS time interval.

• Driver is running at the DUT– Only have access to drivers for legacy devices

– Means there will be some amount of scanning/beacon sending/etc. that can not be controlled. This means there potentially is a small percentage of data and/or ACKs lost due to this problem.

March 2005

C. Young, et al

Slide 9

doc.: IEEE 802.11-05/0181r0

Submission

Received signal at the DUT (cont.)

• Each legacy device will have a different sensitivity point

• Thus, report data for each legacy device with different reference received power seen at point B, then show data in 1 dB received power increments

March 2005

C. Young, et al

Slide 10

doc.: IEEE 802.11-05/0181r0

Submission

Sniffer• Sniffer is a Brcm device without driver running

– No beacons, no scanning, etc.

• Sniffer has same MAC address (AAA) as the Source address of the packets sent from the AWG

• Thus, sniffer can count with accuracy:

– Data frames sent from the AWG to the DUT (from MAC address AAA to MAC address BBB. No other possible data frames exist)

– ACKs sent from the DUT back to the AWG (BBB to AAA. No other possible ACKs exist)

• Power level seen at the sniffer for both transmitted data packets from AWG and received ACKs from DUT at point C is large/reasonable/will have no issues detecting either.

March 2005

C. Young, et al

Slide 11

doc.: IEEE 802.11-05/0181r0

Submission

Results• Results reported across received signal powers, for 25

and 50 nsec rms channels, for the candidate preambles

• For each test case, the 500 packets were sent 20 times, thus 10,000 packets sent. Ran each point 10 times, throwing out the high and low values and averaging.

• Results reported as ACK failure rate– A failure rate of 0% would mean ‘the legacy station received and

acknowledged as many packets using the candidate preamble as were sent’. Ideally, the ratio across all received powers would be 0%.

March 2005

C. Young, et al

Slide 12

doc.: IEEE 802.11-05/0181r0

Submission

Sensitivity Results for Candidate 2TX Preambles (Atheros)

• Legacy device #1 for 25ns and 50ns RMS delay spreads– Showing WWiSE, TGnsync, -200 and -400 (cyclic advance) preamble results– Does not do well with the TGnsync preamble

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate

Power Relative to Reference (dB)

Legacy device #1 over 25 nsec rms channels 2TX case

-400ns-200nsTgnsync

WWiSE

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200nsTgnsync

WWiSE

March 2005

C. Young, et al

Slide 13

doc.: IEEE 802.11-05/0181r0

Submission

Sensitivity Results for Candidate 2TX Preambles (Atheros)

• Legacy device #1 for 25ns and 50ns RMS delay spreads– Showing -200, + 200, -400, +400 ns STS/LTS preamble results– -200 and -400 (cyclic advance) is preferred

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

March 2005

C. Young, et al

Slide 14

doc.: IEEE 802.11-05/0181r0

Submission

Sensitivity Results for Candidate 2TX Preambles

• Legacy device #2 for 25ns and 50ns RMS delay spreads– Showing WWiSE, TGnsync, 200 and 400 cyclic advance preamble results

– Does not prefer the TGnsync preamble

0 1 2 3 4 5 6 7 80

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200ns

TgnsyncWWiSE

0 1 2 3 4 5 6 7 80

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200ns

TgnsyncWWiSE

March 2005

C. Young, et al

Slide 15

doc.: IEEE 802.11-05/0181r0

Submission


• Legacy device #2 for 25ns and 50ns RMS delay spreads– Showing -200, + 200, -400, +400 ns STS/LTS preamble results

– All compromise preambles are acceptable

0 1 2 3 4 5 6 7 80

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

0 1 2 3 4 5 6 7 80

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

March 2005

C. Young, et al

Slide 16

doc.: IEEE 802.11-05/0181r0

Submission


• Legacy device #3 for 25ns and 50ns RMS delay spreads– All preambles comparable

– +/- 200 nsec preambles not tested

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



WWISETgnsync

-400ns Both+400ns Both

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



WWISETgnsync

-400ns Both+400ns Both

March 2005

C. Young, et al

Slide 17

doc.: IEEE 802.11-05/0181r0

Submission


• Legacy device for 25ns and 50ns RMS delay spreads– Showing old WWiSE, TGnsync, 200 and 400 cyclic advance preamble results

– Does not prefer the old WWiSE preamble, does like the TGnsync preamble

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200nsTgnsyncWWiSE

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate




March 2005

C. Young, et al

Slide 18

doc.: IEEE 802.11-05/0181r0

Submission



– Prefers TGnsync, and -400 nsec, then +200 nsec (error floor consideration)

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

0 2 4 6 8 10 120

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

March 2005

C. Young, et al

Slide 19

doc.: IEEE 802.11-05/0181r0

Submission


• Legacy device #5 for 25ns and 50ns RMS delay spreads– Showing old WWiSE, TGnsync, 200 and 400 cyclic advance preamble results

– Does not prefer TGnsync

0 1 2 3 4 5 6 7 8 9 100

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200ns

TgnsyncWWiSE

0 1 2 3 4 5 6 7 8 9 100

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate




March 2005

C. Young, et al

Slide 20

doc.: IEEE 802.11-05/0181r0

Submission



– Prefers -400 ns, then -200 ns

0 1 2 3 4 5 6 7 8 9 100

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

0 1 2 3 4 5 6 7 8 9 100

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

March 2005

C. Young, et al

Slide 21

doc.: IEEE 802.11-05/0181r0

Submission


• Legacy device #6 for 25ns and 50ns RMS delay spreads– Showing WWiSE, TGnsync, 200 and 400 cyclic advance preamble results

– Prefers WWiSE

0 2 4 6 8 10 12 140

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200ns

TgnsyncWWiSE

0 2 4 6 8 10 12 140

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns-200ns

TgnsyncWWiSE

March 2005

C. Young, et al

Slide 22

doc.: IEEE 802.11-05/0181r0

Submission



– Prefers +200 ns, then -200 ns

0 2 4 6 8 10 12 140

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

0 2 4 6 8 10 12 140

0.05

0.1

0.15

0.2

0.25

AC

K F

ailu

re R

ate



-400ns+400ns-200ns+200ns

March 2005

C. Young, et al

Slide 23

doc.: IEEE 802.11-05/0181r0

Submission

Conclusions

• Overall, cyclic shift of -200 ns or -400 ns are best.

• WWiSE has updated its proposal to reflect this data.

March 2005

C. Young, et al

Slide 24

doc.: IEEE 802.11-05/0181r0

Submission

References

doc.: ieee 802.11-05/0181r0 submission march 2005 c. young, et alslide 1 legacy device testing with...

Documents