doc.: ieee 802.11-06/1490r0 submission september 2006 dr. michael d. foegelle, ets-lindgrenslide 1...

September 2006

Dr. Michael D. Foegelle, ETS-Lindgren

Slide 1

Doc.: IEEE 802.11-06/1490r0

Submission

The Need for TRP and TIS in 802.11.2

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: 2006-9-20

Name Company Address Phone email Dr. Michael D. Foegelle ETS-Lindgren 1301 Arrow Point Drive

Cedar Park, TX 78613 (512) 531-6444 [email protected]

Authors:

mailto:[email protected]

September 2006


Slide 2

Doc.: IEEE 802.11-06/1490r0

Submission

Abstract

This presentation discusses the need for the test capabilities provided by the TRP and TIS test metrics in document 11-06-0906-00-000t, “TRP and TIS Performance Metrics Proposed Text”. It also shows how the measurements possible with the COAT methodology relate to TRP/TIS testing.

September 2006


Slide 3

Doc.: IEEE 802.11-06/1490r0

Submission

• Total Radiated Power (TRP) and Total Isotropic Sensitivity (TIS) are industry standard methods for determining the RF performance of wireless devices.

• They represent traceable quantitative metrics that completely represent the RF performance of a device at the PHY layer, including the antenna(s), the DUT body, and the effect of objects typically found in the near field of the DUT.

• These metrics provide critical link budget information necessary for good network design.

Introduction

September 2006


Slide 4

Doc.: IEEE 802.11-06/1490r0

Submission

• Several industry organizations (CTIA, Wi-Fi Alliance) already use TRP/TIS metrics.

• Wireless service providers use this information to qualify products before they’re allowed on their network.

• The test plans used by these organizations can only be used by members of those organizations.

• We need a public IEEE document that provides the same type of test metrics without the burden of joining an industry organization.

Introduction

September 2006


Slide 5

Doc.: IEEE 802.11-06/1490r0

Submission

• An error of a dB or two is costly to network layout.• In free-space propagation, losing 1 dB of signal in a

link budget increases the number of required APs by 25%!

• Propagation over ground tends to reinforce the direct path, so in that case a 2 dB reduction in the link budget will cause the same 25% increase.

• Indoor LOS/NLOS environments will have more complicated field distributions, but similar network guidelines apply.

Network Design Issues

September 2006


Slide 6

Doc.: IEEE 802.11-06/1490r0

Submission

• An understanding of the RF performance of the APs, as well as that of the target clients, is critical to good network design.

• Once a network is designed, it’s even more costly to add APs to try to fill in gaps caused by poorly performing clients.

• Thus, it’s critical that a certain level of RF performance can be guaranteed.

• We need test techniques that give useful quantitative RF performance metrics.

Network Design Issues

September 2006


Slide 7

Doc.: IEEE 802.11-06/1490r0

Submission

• A common assumption in the wireless industry is that the radiation pattern of a device is defined by the antenna pattern.– i.e. Dipole antenna = dipole radiation pattern– OTA performance = conducted DUT + antenna gain

The Antenna Assumption

+ ?=

September 2006


Slide 8

Doc.: IEEE 802.11-06/1490r0

Submission

• While this may hold for remote antennas or for electrically small devices, electrically large devices (laptops, etc.) generate radiation patterns that rarely match that of the antenna by itself.


=/

September 2006


Slide 9

Doc.: IEEE 802.11-06/1490r0

Submission

• Even if the antenna is tested passively within the body of the DUT so that the effect of the rest of the DUT on the pattern is accounted for, there are other issues:


– Cabled testing of DUT assumes antenna has a 50 impedance. Mismatches between radio circuitry and actual antenna impedance can cause non-linear behavior in radio circuitry.

– Near field coupling to objects typically found near DUT (tables, walls, hands, heads) can change both antenna impedance and radiation pattern.

– Platform noise from electronics can interfere with the receiver through the antenna (and vice-versa).

– Cable effects can still distort pattern.

September 2006


Slide 10

Doc.: IEEE 802.11-06/1490r0

Submission

• The current COAT methodology is essentially a substitution measurement.– The gain of a reference antenna is used to determine RF

path loss.– DUT is substituted for the reference antenna and path loss

• While the method can produce highly repeatable results, there are inherent assumptions made about the radiation pattern of the DUT.

• COAT methodology requires orienting DUT with antenna in exact position and polarization of reference antenna.

COAT Methodology

September 2006


Slide 11

Doc.: IEEE 802.11-06/1490r0

Submission

• Assumes that orientation/position of antenna is known.

• Even if it is known, that may not be enough to achieve the intended results.

• Testing in only one orientation does not represent the expected behavior of the DUT if the pattern varies as a function of orientation.

• The antenna pattern information is required to be able to determine relationship to actual device performance.

COAT Methodology

September 2006


Slide 12

Doc.: IEEE 802.11-06/1490r0

Submission

• Comparing two NICs in same laptop:

• Testing four different orientations:

COAT Example

+

Internal External

0° 90° 180° 270°

September 2006


Slide 13

Doc.: IEEE 802.11-06/1490r0

Submission

• Evaluating two possible measurement polarizations

COAT Example

Vertically Polarized

Total Field Vector

September 2006


Slide 14

Doc.: IEEE 802.11-06/1490r0

Submission

• Fixing the orientation and/or polarization direction between measurements of DUTs doesn’t help.

• Differences in the deltas between the two NIC configurations are much larger than the expected error from the calibration process.

COAT Example

Vertically Polarized Total Field VectorAngle Internal External Delta

0 3.53 2.07 1.4690 -4.48 5.99 -10.47180 6.13 0.52 5.61270 6.80 -2.04 8.84

Angle Internal External Delta0 4.94 10.33 -5.3990 1.00 14.28 -13.28180 11.32 9.82 1.50270 8.69 16.54 -7.85

September 2006


Slide 15

Doc.: IEEE 802.11-06/1490r0

Submission

• The reason for these issues is simple. The DUTs don’t have simple radiation patterns.

• These patterns definitely aren’t dipole-like!

Analysis

Internal NIC External NIC

Dell Latitude D600, 11 MBPS, 5 Degree Steps, Front Face

Pow

er (

dBm

)

-4

16

-2

0

2

4

6

8

10

12

14

Y X

ZAzimuth = -90.0Elevation = 90.0Roll = 90.0

DWL-AG660 in Latitude D600, 11 MBPS, 5 Degree Steps, Front Face

Pow

er (

dBm

)

-5

25

0

5

10

15

20

Y X


September 2006


Slide 16

Doc.: IEEE 802.11-06/1490r0

Submission

• An ideal dipole produces an omnidirectional pattern.– It radiates symmetrically about one axis of rotation, but not

the same in all directions.

Analysis

-20

0

-16

-12

-8

-4

Y

Z

X

Azimuth = 114.5Elevation = -28.9Roll = -51.4

September 2006


Slide 17

Doc.: IEEE 802.11-06/1490r0

Submission

-20

0

-16

-12

-8

-4

Y

Z

X

Azimuth = 114.5Elevation = -28.9Roll = -51.4

• Assume for the moment that we have a DUT that does have a perfect dipole pattern.

• Wireless networks are commonly considered as a two dimensional problem (floor plan).

• Results in two dimensional analysis of the pattern effects.

Analysis

September 2006


Slide 18

Doc.: IEEE 802.11-06/1490r0

Submission

• If an ideal dipole is vertically polarized, we get the same signal in all directions.

Analysis

Tuned Dipole H-Plane Gain Cut

Gai

n (d

B)

Phi Angle (°)Scale: 0.5/divMin: 0Max: 2.5

0

180

30

210

60

240

90 270

120

300

150

330

• Average Gain = Peak Gain• This is of course the ideal

condition.

September 2006


Slide 19

Doc.: IEEE 802.11-06/1490r0

Submission

• If the antenna were horizontally polarized, the performance becomes a function of angle with respect to the DUT.

Analysis

Tuned Dipole E-Plane Gain Cut

Gai

n (d

B)

Theta Angle (°)Scale: 5/divMin: -25Max: 5

0

180

30

210

60

240

90 270

120

300

150

330• One common solution is to take the average performance as a function of angle.• Peak Gain = 2.15 dBi• Average Gain = -1.35 dBi

September 2006


Slide 20

Doc.: IEEE 802.11-06/1490r0

Submission

Theta = 90° Cut, Total Power

EIRP

(dB

m)

Phi Angle (°)Scale: 5/divMin: -5Max: 20

0

180

30

210

60

240

90 270

120

300

150

330

Built-in NIC PCMCIA NIC

• Comparing cuts through the two patterns shows that on average the external NIC performs better.

Analysis

• Comparing the average power gives:• Internal NIC = 8.85 dBm• External NIC = 13.76 dBm• Difference = 4.91 dB.

September 2006


Slide 21

Doc.: IEEE 802.11-06/1490r0

Submission

Theta = 70° Cut, Total Power

EIRP

(dB

m)

Phi Angle (°)Scale: 2/divMin: -2Max: 20

0

180

30

210

60

240

90 270

120

300

150

330

Built-in NIC PCMCIA NIC

• AP and Client unlikely to be at the same height.– Performance at angles other than horizontal is required.

Analysis

• Comparing the average power for this cut gives:• Internal NIC = 8.32 dBm• External NIC = 13.24 dBm• Difference = 4.92 dB.

• Considerably different pattern, but similar results.

September 2006


Slide 22

Doc.: IEEE 802.11-06/1490r0

Submission

• TRP is nothing more than the average transmit performance over an entire spherical surface (i.e. in all directions) rather than just a single cut.

Analysis

• Comparing the TRPs shows results similar to that for the individual cuts:• Internal NIC = 8.44 dBm• External NIC = 12.22 dBm• Difference = 3.78 dB.

• Principal cuts tend to overestimate actual performance.

Dell Latitude D600, 11 MBPS, 5 Degree Steps, Front Face

Pow

er (

dBm

)

-4

16

-2

0

2

4

6

8

10

12

14

Y X


September 2006


Slide 23

Doc.: IEEE 802.11-06/1490r0

Submission

• The proposed TRP and TIS methods are simply methods for performing radiation pattern measurements on active devices.

• While the basic methodology for TRP and TIS involves measuring the entire transmit power and sensitivity radiation patterns, the relationship between these quantities can be used to simplify the total amount of testing required.

• Once the “real” radiation pattern information is known, it is possible to use single point COAT type tests to determine equivalent total pattern quantities.

Analysis

September 2006


Slide 24

Doc.: IEEE 802.11-06/1490r0

Submission

AnalysisEI

RP (

dBm

)

Phi Angle (°)Scale: 2/divMin: 2Max: 18

0

180

30

210

60

240

90 270

120

300

150

330

Quantity 1 Quantity 2

Point measured as part of pattern measurement.

Same point measured for a different quantity and normalized to available pattern information.

September 2006


Slide 25

Doc.: IEEE 802.11-06/1490r0

Submission

Conclusion

• There is a real need for TRP/TIS testing for 802.11 devices.

• Effective network design and operation utilizes link budgets that rely on TRP/TIS values.

• Without this type of measurement, too many assumptions are made about the DUT RF performance.

• TRP/TIS complements the COAT concept and provides ways to determine/verify those quantities that are assumed.

September 2006


Slide 26

Doc.: IEEE 802.11-06/1490r0

Submission

Conclusion

• Failure to incorporate these methods into the TGT draft will be doing a disservice to users of the Recommended Practice who do not have access to industry organization test plans.

• By adding these metrics, the TGT draft becomes harmonized with industry test requirements.

September 2006


Slide 27

Doc.: IEEE 802.11-06/1490r0

Submission

References

1. IEEE 802.11-06/0906r0, “TRP and TIS Performance Metrics Proposed Text,” M.D. Foegelle.

2. IEEE 802.11-06/0760r0, “Calibrated Over the Air Test Environment Proposed Draft Text ,” M.D. Foegelle.

doc.: ieee 802.11-06/1490r0 submission september 2006 dr. michael d. foegelle, ets-lindgrenslide 1...

Documents