Spatial techniques in WiFi 802.11ac

1. Overview on basic MIMO techniques

2. Single User Beamforming

3. Multi User MIMO

Mohamed Seif

STBC

WiFi 802.11ac uses STBC (Alamouti Scheme).

STBC is a transmit diversity scheme that comes with a

robust performance achieved with low cost.

Figure: Alamouti’s Scheme for 2*1

STBC (cont’d)

STBC (cont’d)

In 802.11ac, four STBC modes were defined (2x1, 4x2,

6x3 or 8x4).

Figure: Alamouti’s Scheme for 4*2.

Spatial Multiplexing

Figure: 2*2 MIMO system.

Motivation

Transmit beamforming is used to enhance the reception

of signals.

Figure: Depicts the difference between two

schemes.

Motivation (cont’d)

In broader sense, higher SINR.

In Transmit beamforming, parallel channel paths are

established such that each symbol is transmitted.

Figure: Conventional MIMO

system equivalent channel.

Figure: TX beamforming

equivalent channel (Ideal).

SVD approach

Where X is either U or V

SU-Beamforming Pre-coding

Figure: Equivalent channel model after pre-coding.

U

SU-Beamforming Pre-coding

Receiver design

The received signal is then

ZF or MMSE technique is applied to estimate the

symbols.

Figure: Tx Beamforming gain over conventional MIMO

in terms of un coded BER.

Results

Channel Correlation and Interference

Figure: Channel correlation can degrade the

beamforming gain.

Figure: Effect of channel correlation.

Results

Channel Feedback

The Access Point (AP) does not have any Channel State

Information about the channel (CSI).

Transmit Beamforming requires CSI in order to apply the

weighting matrix V.

Channel feedback techniques are used.

Channel Feedback (cont’d)

Communication system is a closed loop system.

TDD is commonly used.

Compressed feedback technique is used to reduce

overhead.

Figure: Closed loop system.

Compressed Channel Feedback

Compressed Channel Feedback (cont’d)

Compressed Channel Feedback (cont’d)

Histogram and Optimization

Variable feedback rate may be investigated using

source coding.

Ref. W.H. Chin, C. Yuen, Design of differential quantization for low bitrate channel state

information feedback in MIMO-OFDM systems. IEEE VTC 2008 Spring.

Results

Multi-User MIMO (MU-MIMO)

Capacity increasing.

Interference Problem.

Figure: MU-MIMO scheme.

MU-MIMO modeling

MU-MIMO Pre-coding

ZF Pre-coding (channel inversion)

Regularizing the inversion

Figure: Block diagram for MU-MIMO pre-coding process.

Ref. An Introduction to the Multi-User MIMO Downlink, IEEE comsoc.

Receiver design

Receiver design (cont’d)

The estimated symbol is then

Where,

Figure: Comparison between the two different

pre-coding techniques for MU-MIMO

Performance of the two schemes will converge at higher

SNR.

Results

Figure: Effect of feedback delay in MU-MIMO.

Results

THANK YOU!