Doc.: IEEE 802.11-05/0020r0 Submission January 2005 Aryan Saed, IceFyre SemiconductorSlide 1 Required Feature Set for Closed Loop BF Notice: This document

January 2005

Aryan Saed, IceFyre Semiconductor

Slide 1

doc.: IEEE 802.11-05/0020r0

Submission

Required Feature Set for Closed Loop BF

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: 2005-01-16

Name Company Address Phone email Aryan Saèd IceFyre

Semiconductor #300 – 411 Legget Drive, Ottawa ON, K2K 3C9 Canada

613-599-3000 x306

[email protected]

Authors:

January 2005


Slide 2

doc.: IEEE 802.11-05/0020r0

Submission

Abstract

• Suggest to make CSI feedback upon request a mandatory aspect of the standard

• Suggest to define a mechanism that allows a lazy transmitter partake in BF

• Suggest to include NF and RSSI to feedback, so that SNR can be inferred from CSI.

• Suggest to separate the issue of header efficiency and header effectiveness for the case of BF

January 2005


Slide 3

doc.: IEEE 802.11-05/0020r0

Submission

Purpose of this presentation

• What do we (TGn) need to define in the standard so that Beam Forming based on closed loop feedback is possible?

• What does a vendor have to implement to enable Beam Forming? We will distinguish between a “lazy” device and a “promoter” device.

January 2005


Slide 4

doc.: IEEE 802.11-05/0020r0

Submission

Wideband OFDM Beam Forming

• For best performance you need recent Channel State Information (the MxN channel) known at the Tx, for each OFDM sub-carrier

• This allows the Tx to form the beams for each OFDM sub-carrier, so that the MIMO Shannon capacity is maximized for the individual channel

January 2005


Slide 5

doc.: IEEE 802.11-05/0020r0

Submission

The role of the Rx and Tx

IFFT

One Matrix per

Subcarrier

IFFT

T11

T22

T21

T12

H11

H22

H21

H12

R11

R22

R21

R12

FFT

FFT

x1

x2

y1

y2

January 2005


Slide 6

doc.: IEEE 802.11-05/0020r0

Submission

Beam Forming

• Tx requires knowledge of the Tx channel

• Tx channel consists of 1. Tx front-end (self)2. Air-Air (shared)3. Rx front-end (the other)

• How to get knowledge of the Tx channel?1. Ask the receiver: a CSI request bit2. Get the channel: a service field containing the CSI

• Calculate the Tx beam forming matrix for the next packet.

January 2005


Slide 7

doc.: IEEE 802.11-05/0020r0

Submission

Calibration or Explicit Feedback

• Air-Air channel changes over time

• Need to update the Tx BF matrix

• Need updated Tx channel information:

a. Ask for it for every packet: an AlwaysCSIRequest bit for explicit feedback

b. Infer it from the received channel estimate, using calibration. Calibration is station specific and removes differences between the up-link and down-link chain.

January 2005


Slide 8

doc.: IEEE 802.11-05/0020r0

Submission

Front-Ends and Air-Air

Hair

GU GD

FU FD

HU HD

Frontend Frontend

Down link

TD TU

January 2005


Slide 9

doc.: IEEE 802.11-05/0020r0

Submission

Explicit CSI (no calibration)

MTS

MTS HD

MTS HU

MTS HD

STA

AP

Calc. HD

Calc. HU Calc. TD

Apply T D

Calc. TU

Apply T U

Calc. HU Calc. TD

January 2005


Slide 10

doc.: IEEE 802.11-05/0020r0

Submission

Explicit CSI (no calibration)

• The CSI request is perpetual and costs overhead in the reverse link

• The CSI request reflects the air channel according to the most recent transmission (not the most recent reception)

January 2005


Slide 11

doc.: IEEE 802.11-05/0020r0

Submission

Interpreted Feedback (calibration)

MTS

MTS HD

MTS HU

MTS

STA

AP

Calc. HD

Calc. HU Calc. TD & calibr.

Apply T D

Calc. TU

Apply T U

Calc. HU

Calc. TD using Calibr.

January 2005


Slide 12

doc.: IEEE 802.11-05/0020r0

Submission

Interpreted Feedback (calibration)

• Overhead is negligible because it occurs infrequently, usually during a link set-up

• Will cover variations in air-air channel, but not variations in Rx and Tx front ends

• May require bounds on front-end variation (due to temperature)

• The CSI request reflects the air channel according to the most recent reception (not the most recent transmission)

January 2005


Slide 13

doc.: IEEE 802.11-05/0020r0

Submission

Requirements for either

• Need a MIMO Training Sequence (MTS) that ensures ability to perform beam forming based on a wide spectrum of methods, e.g. phased array, SVD, other matrix operations. Must meet spectral requirements (interpolation / smoothing) and time requirements (AGC, PAPR)

• Need a quality metric for the CSI, like Tx EVM in 11a. Could make CSI MSE a function of SNR, e.g. always 6dB better.

• Define CSI as H or as raw FFT output

January 2005


Slide 14

doc.: IEEE 802.11-05/0020r0

Submission

Required in the Standard

• Receiver must respond to a CSI request(mandatory)

Respond with H (or unprocessed FFT) without revealing any IP for smoothing, decoding, interpolation, etc…

Suggest including RSSI and NF in response so that SNR per carrier can be inferred from the CSI. This aids in rate adaptation.

January 2005


Slide 15

doc.: IEEE 802.11-05/0020r0

Submission

Preferred in the Standard

• Rate is defined per stream for all sub-carriers, not just for entire packet

Opens up broadest possibilities for maximizing the rate, e.g. using SVD

(A different method for BF matrix calculation has been shown to allow one rate for entire packet, see 04/882 by IceFyre)

January 2005


Slide 16

doc.: IEEE 802.11-05/0020r0

Submission

Promoter AP with Lazy STA

MTS

MTS HD

MTS TU

MTS

STA

AP

Calc. HD (for EQ)

Calc. HU (for EQ)

Calc. TD & calib. HD/HU

Apply T D

Calc. TU

Apply T U

Calc. HU

Update: T 'D and T 'U using calib.

January 2005


Slide 17

doc.: IEEE 802.11-05/0020r0

Submission

Lazy participant

• Benefit from BF in either direction of the link, regardless which side supports the matrix calculation

• For frequency domain DSP (e.g. SVD), need a mechanism in the standard to insert provided coefficients at Tx if requested to do so by Rx.

• For RF level phase adjustment, need mechanism to apply phase array.

• Discussion: make the DSP for Tx coefficient insertion mandatory (must insert coefficients if provided)

January 2005


Slide 18

doc.: IEEE 802.11-05/0020r0

Submission

Pre-amble

• Discussion this week:

Is BF impeded by the MTS due to risk of rank reversal? (TGnSync concerns with Wwise header)

Is overhead from a full spectrum MTS warranted if BF is an optional mode? (Wwise concern with TGnSync header)

January 2005


Slide 19

doc.: IEEE 802.11-05/0020r0

Submission

Pre-amble

• Comparison and feasibility tests are still needed to address the concerns from both sides

• The solution is not necessarily one or the other.

a. CSI MSE errors can be overcome with longer headers: full spectrum plus multi-symbol averaging

b. Overhead concerns can be overcome by applying a longer header only when SVD is intended

January 2005


Slide 20

doc.: IEEE 802.11-05/0020r0

Submission

Conclusions

• Suggest to make CSI feedback upon request a mandatory aspect of the standard

• Suggest to define a mechanism that allows a lazy transmitter partake in BF

• Suggest to include NF and RSSI to feedback, so that SNR can be inferred from CSI.

• Suggest to separate the issue of header efficiency and header effectiveness for the case of BF

January 2005


Slide 21

doc.: IEEE 802.11-05/0020r0

Submission

References

• 11-05-1645-02-000n-preambles-beamforming-wwise-proposal

• 11-05-1635-01-000n-wwise-preambles-and-mimo-tx-beamforming

• 11-04-0882-02-000n-optimized-mimo

Documents

Doc.: IEEE 802.11-05/0020r0 Submission January 2005 Aryan Saed, IceFyre SemiconductorSlide 1 Required Feature Set for Closed Loop BF Notice: This document