Analysis of Multi Path Channels - Matlab...Evaluating IEEE 802.11ax (Wi-Fi 6) without Commercial WLAN Hardware Matlab models allow evaluation of WLAN standards before commercial hardware

Alper Akbilek, Dr. Florian Pfeiffer, Manzar Hussain

perisens GmbH

Analysis of Multi-Path Channels Using

the WLAN Packet Preamble

Analysis of Multi-Path Channels Using the WLAN Packet Preamble

Agenda

1. About perisens

2. Motivation

3. Background

4. Implementation and Results

5. Conclusion

24.05.2019 Analysis of Multi-Path Channels Using the WLAN Packet Preamble 2

Agenda

1. About perisens

2. Motivation

3. Background


5. Conclusion


About perisens…

perisens GmbH

Founded in 2009 as Spin-Off from the Technical University of Munich(TUM) with ongoing cooperation

24.05.2019 4

Services

• Technical Consulting / Studies

• RF measurements & Simulations (up to 90 GHz)

• Development and Evaluation of Wireless

Communication Systems

• Signal Processing

• Development of RF Prototypes

• Solutions in Automotive Radar Sensors

Products

• In-House Development, Production and Sale of

Radar Target Simulators (RTS)


Agenda

1. About perisens

2. Motivation

3. Background


5. Conclusion


Motivation

24.05.2019 6

▪ Wifi is everywhere

▪ Higher data rates enabled by new amendments

▪ High data rates require very favorable channel

conditions

▪ Evaluating the wireless channel is required to

determine if implementing new wireless standards

(e.g IEEE 802.11ax) would bring improvement


Goal:

Evaluating IEEE 802.11ax (Wi-Fi 6) without Commercial WLAN Hardware

▪ Matlab models allow evaluation of WLAN standards

before commercial hardware hits the market

▪ We aim to evaluate the standard for specific

environment (e.g. in-vehicle environments)

▪ Simulations and over-the-air testing with Matlab

➢ IEEE 802.11ax models are available in MATLAB WLAN

Toolbox since 2018

Timeline


May

2014

Start

TGax

March

2016

Draft

0.1

Nov.

2016

Draft

1.0

Sept.

2017

Draft

2.0

May

2018

Draft

3.0

Jan.

2019

Draft

4.0

Jan.

2020

Final

Approval

Jun.

2020

Publication

MATLAB R2018a

802.11ax Models in WLAN Toolbox

Agenda

1. About perisens

2. Motivation

3. Background


5. Conclusion


OFDM

▪ Orthogonal frequency-division multiplexing

▪ Data is transmitted over independent sub-carriers with some redundancy (channel coding)

▪ Able to cope with severe channel conditions (frequency-selective fading, narrowband interference…)


WLAN sub-carrier distance:

IEEE 802.11a/g/n/ac: 312.5 kHz

IEEE 802.11ax : 78.125 kHz

BW: 20, 40, 80 or 160 MHz

Enabling High Throughput

Higher Modulation and Coding Schemes

24.05.2019 10

6 Bits/Symbol/Carrier

MCS 5,6,7

IEEE 802.11a/g/n


MCS 8,9

IEEE 802.11ac


MCS 10,11

IEEE 802.11ax


Received Signal Strength

▪ Total received signal power in the channel

▪ Available at the receiver

▪ Conventional way to evaluate the link

▪ High data rates require high signal power at the receiver

▪ Signal strength is NOT the only factor determining the link speed

24.05.2019 11Analysis of Multi-Path Channels Using the WLAN Packet Preamble

Multi-Path Channel

▪ Several echoes of the signal is detected at receiver

▪ Signal power and SNR are not equally distributed in the channel

▪ Causes frequency-selective fading which increases the required SNR

▪ Sensitive to moving objects on the signal path

▪ Channel models available for IEEE 802.11 simulations

24.05.2019 12

Transmitter Receiver

time

frequency

h(t

)H

(f)

Noise floor

High SNR

Low SNR


IEEE 802.11 Packet Preamble

Channel Estimation by HE-LTF for Equalization

24.05.2019 13

Wireless

Channel

HE-LTF at TX HE-LTF at RX

Channel State

Information

(CSI)


Agenda

1. About perisens

2. Motivation

3. Background


5. Conclusion


WLAN PHY Simulations in MATLAB

▪ Simulation loop is already included in MATLAB

WLAN Toolbox

▪ IEEE channel models available for example

scenarios (small office, conference room etc.)

▪ Output: SNR requirement for a definite WLAN

packet format

▪ We use simulations for link budget calculations for

specific scenarios in which the path loss is known

24.05.2019 15Analysis of Multi-Path Channels Using the WLAN Packet Preamble

IEEE802.11

Baseband

Waveform Creation

TX bits

Channel Model

Synchronization,

Equalization and

Demodulation

RX bits

Baseband IQ

(TX)

Baseband IQ

(RX)

Packet Error

Rate (PER)

Opening the Simulation Loop

Measurements Over the Air

24.05.2019 16

▪ Sending WLAN packets over the RF channel

▪ Evaluate the real wireless channel

▪ Requires RF instruments or Software-Defined-Radios

(SDRs) as TX and RX

▪ Can measure the PER for a definite packet type

▪ Drawback: Long measurement time


IEEE802.11

Baseband

Waveform Creation

TX bits

Channel Model

Synchronization,

Equalization and

Demodulation

Baseband IQ

(TX)

Baseband IQ

(RX)

RX bits

Packet Error

Rate (PER)

IEEE802.11

Baseband

Waveform Creation

TX bits

Synchronization,

Equalization and

Demodulation

RX bits

RF Signal

over the air

Channel Sounding

Extracting the Channel State Information (CSI)

24.05.2019 17

▪ Instead of making PER measurements, we only

extract the Channel State Information (CSI) and signal

power

▪ We collect several CSI samples to use in simulations

▪ Packet type & modulation are varied in simulations to

determine the highest achievable data rates


Baseband IQ

(TX)

IEEE802.11

Baseband

Waveform Creation

Synchronization,

Equalization and

Demodulation

Channel State Information (CSI)

RF Signal

over the air

Channel Sounding Method

Using the CSI in Simulations

24.05.2019 18

Channel State Information

Extraction


IEEE802.11

Baseband

Waveform Creation

Baseband IQ

(TX)

RF Signal

over the air

Synchronization,

Equalization and

Demodulation

IEEE802.11

Baseband

Waveform Creation

Multiplication

in

frequency domain

AWGN

channel

simulation

Synchronization,

Equalization and

Demodulation

RX bitsTX bits

Packet Error

Rate (PER)

Baseband IQ

(RX)

Channel State Information

Channel Frequency Response & Error Vector Magnitude (EVM)


Simulation using

the measured

channel response

EVM threshold

for 1024-QAM:

-35 dB

Channel Sounding with WLAN Packets

Measurement Setup

▪ Measurement in a closed room

▪ IEEE 802.11ax waveforms generated in MATLAB

▪ First transmission with line of sight channel

➢ Measured path loss: 53 dB

▪ Second transmission with non line of sight channel



TX

RX

Line of sight

distance:

4.3m

BLOCK


Measurement Setup

▪ Measurement in a closed room

▪ IEEE 802.11ax waveforms generated in MATLAB

▪ First transmission with line of sight channel


▪ Second transmission with non line of sight channel



TX

RX

BLOCK


Results


Amplitude variation in the

line of sight (LOS) channel:

6 dB

Amplitude variation in the

non line of sight (NLOS) channel:

28 dB

Simulation Using the LoS and NLoS Channel Responses

Results


~8 dB

Agenda

1. About perisens

2. Motivation

3. Background


5. Conclusion


Conclusion

Summary

▪ New wireless standards can be evaluated with MATLAB before the hardware hits the market

▪ Wireless channels can be analyzed using the WLAN signals

▪ Energy is not evenly distributed in a non line of sight (NLOS) channel

▪ Receiver needs higher signal power to decode the signal in NLOS scenarios

▪ The channel state information (CSI) changes if there are moving objects in the room

Future Work

▪ Other applications using the channel state information (CSI) of WLAN signals:

➢ Human presece detection

➢ Vital signal monitoring (breathing & heart rate)


Documents

Analysis of Multi Path Channels - Matlab...Evaluating IEEE 802.11ax (Wi-Fi 6) without Commercial WLAN Hardware Matlab models allow evaluation of WLAN standards before commercial hardware