802.11be Transmitter Measurements

Open Live Script

This example shows how to measure transmitter modulation accuracy, spectral mask, spectral flatness, and in-phase and quadrature (IQ) gain and phase imbalance for IEEE® 802.11be™ (Wi-Fi® 7) waveforms.

Example Overview

In this example, you generate an oversampled IEEE 802.11be extremely high throughput (EHT) multi-user (MU) waveform, as defined in IEEE P802.11be [1]. You can configure the example to generate orthogonal frequency-division multiple-access (OFDMA) waveforms, which transmit data for multiple users over different parts of the band, or non-OFDMA waveforms, which transmit data for a single user over the whole band. Add IQ gain and phase imbalance to the waveform. You can also introduce in-band distortion and spectral regrowth by using a high-power amplifier (HPA) model. You then perform the IQ gain and phase imbalance, transmitter modulation accuracy, required spectral mask, and required spectral flatness measurement on the waveform for the measurement configuration specified in Section 36.3.19 of [1].

For each user, the example decodes the EHT Data field and measures the error vector magnitude (EVM) to determine the modulation accuracy after downsampling the waveform to baseband sampling rate. Additionally, for the non-OFDMA physical layer protocal data unit (PPDU) type, the example measures the spectral flatness of the recovered waveform. This diagram shows the example workflow.

Simulation Setup

Configure the example to generate two EHT MU packets with a 10 microsecond idle period between each packet.

numPackets = 2;
idleTime   = 10;

One approach to improve the EVM is to use data-aided equalization. Data-aided equalization updates the channel estimate using the reconstructed transmitted orthogonal frequency-division multiplexing (OFDM) symbols as a training sequence and exploiting the new available information along with the received symbols. You can optionally equalize the data symbols using a data-aided approach by setting enableDataAidedEqualization to true. This improves the EVM by giving a more accurate channel estimate, at the expense of more complex receiver design. The default value of enableDataAidedEqualization is false.

enableDataAidedEqualization = false;

Set the enableIQImbalanceCorrection setting to true to estimate and correct IQ gain and phase imbalance in the transmitter waveform.The default value of enableIQImbalanceCorrection is false.

enableIQImbalanceCorrection = false;

802.11be Waveform Configuration and Generation

The draft standard defines the EHT MU format for the transmission of non-OFDMA and OFDMA PPDU types. This example supports generation of non-OFDMA and OFDMA EHT MU PPDU types. For more information about the parameterization and generation of IEEE 802.11be EHT MU waveforms, see the 802.11be Waveform Generation example.

savedState = rng(0); % Set random state
ppduType = "OFDMA"; % Set ppduType to OFDMA or Non-OFDMA

Configure transmission parameters of an EHT MU packet by using EHT MU configuration object wlanEHTMUConfig For an OFDMA PPDU type, create an OFDMA configuration for a 20 MHz EHT MU packet with allocation index 47 as defined in Table 36-34 of [1]. This allocation has two 52-tone resource units (RUs) and one 106+26-tone multiple resource unit (MRU). This configuration specifies the transmission of a single user per RU. Set the transmission parameters for each user.

if strcmp(ppduType,"OFDMA")
    allocationIndex = 47;                          % Allocation index 47 specifies three RUs and three users
    mcs = [12 5 4];                                % Modulation and coding scheme (MCS) per user
    spatialMapping = ["Direct" "Direct" "Direct"]; % Set spatial mapping property per RU
    apepLength = [1000 500 400];                   % A-MPDU length pre-EOF padding in bytes per user
    numSTSs = [1 1 1];                             % Number of space-time streams per user
    channelCoding = ["LDPC" "LDPC" "BCC"];         % Set channel coding property per user
    numTx = 1;                                     % Number of transmit antennas

    cfgEHT = wlanEHTMUConfig(allocationIndex);
    chanBW = cfgEHT.ChannelBandwidth;
    numUsers = numel(cfgEHT.User);
    cfgEHT.NumTransmitAntennas = numTx;
    for i = 1:numUsers
        cfgEHT.RU{i}.SpatialMapping = spatialMapping(i);
        cfgEHT.User{i}.APEPLength = apepLength(i);
        cfgEHT.User{i}.MCS = mcs(i);
        cfgEHT.User{i}.NumSpaceTimeStreams = numSTSs(i);
        cfgEHT.User{i}.ChannelCoding = channelCoding(i);
    end
end

For the non-OFDMA PPDU type, create a full-band 320 MHz single-user (SU) MIMO configuration and set the transmission parameters of the user.

if strcmp(ppduType,"Non-OFDMA")
    chanBW = "CBW320";  %#ok<UNRCH> % Channel bandwidth
    mcs = 12;           % Modulation and coding scheme
    apepLength = 8000;  % A-MPDU length pre-EOF padding in bytes
    numTx = 1;          % Number of transmit antennas

    cfgEHT = wlanEHTMUConfig(chanBW); % EHT MU configuration object
    numUsers = numel(cfgEHT.User);
    cfgEHT.NumTransmitAntennas = numTx;
    cfgEHT.RU{1}.SpatialMapping = "Direct";
    cfgEHT.User{1}.APEPLength = apepLength;
    cfgEHT.User{1}.MCS = mcs;
    cfgEHT.User{1}.NumSpaceTimeStreams = numTx;
    cfgEHT.User{1}.ChannelCoding = "LDPC";
end

To model the effect of an HPA on the waveform and view the out-of-band spectral emissions, the waveform must be oversampled. Generate the waveform using a larger IFFT than required for the nominal baseband rate, resulting in an oversampled waveform.

osf = 4; % Oversampling factor

Create random bits for all packets.

psduLen = psduLength(cfgEHT).*8;
data = cell(1,numUsers);
for i=1:numUsers
    data{i} = randi([0 1],psduLen(i)*numPackets,1);
end

Generate the EHT MU waveform for the specified bits and configuration by using the wlanWaveformGenerator function, specifying the desired oversampling factor, number of packets, and idle time between each packet.

txWaveform = wlanWaveformGenerator(data,cfgEHT, ...
NumPackets=numPackets,IdleTime=idleTime*1e-6, ...
OversamplingFactor=osf);

Get the baseband sampling rate of the waveform.

fs = wlanSampleRate(chanBW);
disp(['Baseband sampling rate: ' num2str(fs/1e6) ' Msps']);

Baseband sampling rate: 20 Msps

Prepend zeros to the waveform to allow for early timing synchronization.

txWaveform = [zeros(round(idleTime*1e-6*fs),numTx); txWaveform];

Add Impairments

IQ Gain and Phase Imbalance Modeling

IQ gain and phase imbalance is introduced in the radio frequency (RF) section of the transceivers. It occurs because of the gain and phase mismatch between the in-phase and quadrature branches of the modulator or demodulator. This example models IQ gain and phase imbalance in the transmitted waveform based on the modelIQImbalance flag.

modelIQImbalance = false; % Set to true to add IQ gain and phase imbalance
if modelIQImbalance
    iqGaindB = 1;         %#ok<UNRCH> % in dB
    iqPhaseDeg = 2;       % in degrees
    txWaveform = iqimbal(txWaveform,iqGaindB,iqPhaseDeg);
end

HPA Modeling

The HPA introduces nonlinear behavior in the form of in-band distortion and spectral regrowth. This example simulates the power amplifiers by using the Rapp model [2], which introduces AM/AM distortion.

Model the amplifier by using the dsp.FIRInterpolator object and configure reduced distortion by specifying a backoff, hpaBackoff, such that the amplifier operates below its saturation point. You can increase the backoff to reduce EVM for higher MCS values.

pSaturation = 25;                            % Saturation power of a power amplifier in dBm
hpaBackoff = 16;                             % Power amplifier backoff in dB
nonLinearity = comm.MemorylessNonlinearity;
nonLinearity.Method = "Rapp model";
nonLinearity.Smoothness = 3;                 % p parameter
nonLinearity.LinearGain = -hpaBackoff;
nonLinearity.OutputSaturationLevel = db2mag(pSaturation-30);
txWaveform = nonLinearity(txWaveform);

Thermal Noise

Add thermal noise to each transmit antenna by using the comm.ThermalNoise object with a noise figure of 6 dB [3].

thNoise = comm.ThermalNoise(NoiseMethod="Noise Figure",SampleRate=fs*osf,NoiseFigure=6);
txWaveform = thNoise(txWaveform);

EVM and Spectral Flatness Measurements

Downsampling and Filtering

Resample the oversampled waveform down to baseband for physical layer processing and EVM and spectral flatness measurements, applying a low-pass anti-aliasing filter before downsampling. The impact of the low-pass filter is visible in the spectral flatness measurement. Set the parameters for the anti-aliasing filter so that all active subcarriers are within the filter passband.

Design resampling filter.

aStop = 40; % Stopband attenuation
ofdmInfo = wlanEHTOFDMInfo("EHT-Data",cfgEHT,1); % OFDM parameters for the first RU
SCS = fs/ofdmInfo.FFTLength; % Subcarrier spacing
txbw = max(abs(ofdmInfo.ActiveFrequencyIndices))*2*SCS; % Occupied bandwidth
[L,M] = rat(1/osf);
maxLM = max([L M]);
R = (fs-txbw)/fs;
TW = 2*R/maxLM; % Transition width

Resample the waveform to baseband.

firdec = designMultirateFIR(L,M,TW,aStop,SystemObject=true);
rxWaveform = firdec(txWaveform);

Receiver Processing

In this section you detect, synchronize, and extract each packet in rxWaveform, and then measure the EVM and spectral flatness. Perform the spectral flatness measurement for non-OFDMA PPDU type. For each packet, perform these steps.

Detect the start of the packet
Extract the legacy fields
Estimate and correct coarse carrier frequency offset (CFO)
Perform fine symbol timing estimate by using the frequency-corrected legacy fields
Extract the packet from the waveform by using the fine symbol timing offset
Correct the extracted packet with the coarse CFO estimate
Extract the legacy-long training field (L-LTF), then estimate and correct the fine CFO
Estimate and correct IQ gain and phase imbalance

For each packet and each user, perform these steps.

Extract the EHT-LTF and perform channel estimation for each of the transmit streams
Measure the spectral flatness by using the channel estimate for non-OFDMA PPDU type
Extract and OFDM demodulate the EHT Data field
Perform noise estimation by using the demodulated data field pilots and single-stream channel estimate at pilot subcarriers
Phase-correct and equalize the EHT Data field by using the channel and noise estimates
Optionally update the channel estimate by additionally using the demodulated EHT-Data symbols
For each data-carrying subcarrier in each spatial stream, find the closest constellation point and measure the EVM
Recover the PSDU by decoding the equalized symbols

This diagram shows the processing chain.

This example performs two different EVM measurements.

RMS EVM per user per packet, which comprises averaging the EVM over subcarriers, OFDM symbols, and spatial streams.
RMS EVM per subcarrier per spatial stream per user for a packet. Because this configuration maps spatial streams directly to antennas, this measurement can help detect frequency-dependent impairments, which tend to affect individual RF chains differently. This measurement averages the EVM over OFDM symbols only.

Get indices for accessing each field within the time-domain packet.

ind = wlanFieldIndices(cfgEHT);

Define the minimum detectable length of data, in samples.

minPktLen = double(ind.LSTF(2)-ind.LSTF(1))+1;

Detect and process packets within the received waveform by using a while loop, which performs these steps.

Detect a packet by indexing into rxWaveform with the sample offset, searchOffset
Detect and process the first packet within rxWaveform
Detect and process the next packet by incrementing the sample index offset
Repeat until no further packets are detected

rxWaveformLength = size(rxWaveform,1);
pktLength = double(ind.EHTData(2));
rmsEVM = zeros(numPackets,numUsers);
eqSym = cell(1,numUsers);
evmPerSC = cell(1,numUsers);
decodeSuccess = false(numPackets,numUsers);
passSF = false(numPackets,1);
pktOffsetStore = zeros(numPackets,1);
iqImbalEst = [0 0];

rng(savedState); % Restore random state
pktNum = 0;
searchOffset = 0; % Start at first sample (no offset)
while (searchOffset+minPktLen)<=rxWaveformLength
    
    % Detect packet and determine coarse packet offset
    pktOffset = wlanPacketDetect(rxWaveform,cfgEHT.ChannelBandwidth,searchOffset);
    % Packet offset from start of the waveform
    pktOffset = searchOffset+pktOffset; 
    % Skip packet if legacy-short training field (L-STF) is empty
    if isempty(pktOffset) || (pktOffset<0) || ...
            ((pktOffset+ind.LSIG(2))>rxWaveformLength)
        break;
    end
    
    % Extract L-STF and perform coarse frequency offset correction
    nonht = rxWaveform(pktOffset+(ind.LSTF(1):ind.LSIG(2)),:);  
    coarsefreqOff = wlanCoarseCFOEstimate(nonht,cfgEHT.ChannelBandwidth);
    nonht = frequencyOffset(nonht,fs,-coarsefreqOff);
    
    % Extract the legacy fields and determine fine packet offset
    lltfOffset = wlanSymbolTimingEstimate(nonht,cfgEHT.ChannelBandwidth);
    pktOffset = pktOffset+lltfOffset; % Determine packet offset

    % If offset is outside the bounds of the waveform, then skip samples
    % and continue searching within remainder of the waveform
    if (pktOffset<0) || ((pktOffset+pktLength)>rxWaveformLength)
        searchOffset = pktOffset+double(ind.LSTF(2))+1;
        continue;
    end
    
    % Timing synchronization complete; extract the detected packet
    rxPacket = rxWaveform(pktOffset+(1:pktLength),:);
    pktNum = pktNum+1;
    
    % Apply coarse frequency correction to the extracted packet
    rxPacket = frequencyOffset(rxPacket,fs,-coarsefreqOff);
    
    % Perform fine frequency offset correction on the extracted packet
    lltf = rxPacket(ind.LLTF(1):ind.LLTF(2),:); % Extract L-LTF
    fineFreqOff = wlanFineCFOEstimate(lltf,cfgEHT.ChannelBandwidth);
    rxPacket = frequencyOffset(rxPacket,fs,-fineFreqOff);

    % Perform IQ gain and phase imbalance estimation and correction
    if enableIQImbalanceCorrection
        % Estimate and correct residual CFO
        resCFO = helperResidualCFOEstimate(rxPacket,fs,cfgEHT); %#ok<UNRCH>
        rxPacket = frequencyOffset(rxPacket,fs,-resCFO);

        % Estimate and correct IQ gain and phase imbalance
        [rxPacket, iqGainEst, iqPhaseEst] = helperIQImbalanceCorrect(rxPacket,fs);
        iqImbalEst = [iqGainEst iqPhaseEst];
    end

    % Extract EHT-LTF samples, demodulate, and perform channel estimation
    ehtLTF = rxPacket(ind.EHTLTF(1):ind.EHTLTF(2),:);
    for i = 1:numUsers
        ehtLTFDemod = wlanEHTDemodulate(ehtLTF,"EHT-LTF",cfgEHT,i);

        % Estimate channel
        [chanEst,pilotEst] = wlanEHTLTFChannelEstimate(ehtLTFDemod,cfgEHT,i);

        % Measure spectral flatness for non-OFDMA PPDU type
        if strcmp(ppduType,"Non-OFDMA")
            [passSF(pktNum),deviation,testsc] = wlanSpectralFlatness(chanEst,"EHT",cfgEHT.ChannelBandwidth);  %#ok<UNRCH>
            % Plot deviation against limits
            ehtPlotTxSpectralFlatness(deviation,testsc,pktNum);
        end

        % Data demodulate
        rxData = rxPacket(ind.EHTData(1):ind.EHTData(2),:);
        demodSym = wlanEHTDemodulate(rxData,"EHT-Data",cfgEHT,i);

        % Perform pilot phase tracking
        demodSym = wlanEHTTrackPilotError(demodSym,chanEst,cfgEHT,"EHT-Data",i);

        % Estimate noise power in EHT fields
        ofdmInfo = wlanEHTOFDMInfo("EHT-Data",cfgEHT,i); % OFDM parameters
        nVarEst = wlanEHTDataNoiseEstimate(demodSym(ofdmInfo.PilotIndices,:,:),pilotEst,cfgEHT,i);

        % Extract data subcarriers from demodulated symbols and channel
        % estimate
        demodDataSym = demodSym(ofdmInfo.DataIndices,:,:);
        chanEstData = chanEst(ofdmInfo.DataIndices,:,:);

        if enableDataAidedEqualization
            % Data-aided channel estimation using demodulated EHT-LTF and EHT-Data symbols
            demodLTFSym = ehtLTFDemod(ofdmInfo.DataIndices,:,:); %#ok<UNRCH> % Demodulated EHT-LTF symbols of data subcarriers
            chanEstData = ehtDataAidedChannelEstimate(demodLTFSym,demodDataSym,chanEstData,nVarEst,cfgEHT,i);
        end

        % Equalize
        [eqSym{i},csi] = wlanEHTEqualize(demodDataSym,chanEstData,nVarEst,cfgEHT,"EHT-Data",i);
        
        % Set up EVM measurements
        [EVMPerPkt,EVMPerSC] = evmSetup(cfgEHT,i);

        % Compute RMS EVM over all spatial streams for the packet
        rmsEVM(pktNum,i) = EVMPerPkt(eqSym{i});

        % Compute RMS EVM per subcarrier and spatial stream for the packet
        evmPerSC{i} = EVMPerSC(eqSym{i}); % Nst-by-1-by-Nss

        % Recover data field bits
        rxPSDU = wlanEHTDataBitRecover(eqSym{i},nVarEst,csi,cfgEHT,i);

        if isequal(rxPSDU,data{i}((1:psduLen(i))+(pktNum-1)*psduLen(i)))
            decodeSuccess(pktNum,i) = true;
        end
    end

    % 1st Plot: equalized constellation per packet per spatial stream per user
    % 2nd Plot: RMS EVM per subcarrier per packet per spatial stream per user
    ehtTxEVMConstellationPlots(eqSym,evmPerSC,cfgEHT,pktNum);

    % Store the offset of each packet within the waveform
    pktOffsetStore(pktNum) = pktOffset;
    
    % Increment waveform offset and search remaining waveform for a packet
    searchOffset = pktOffset+pktLength+minPktLen;
end

Set the unit of EVM to decibel or percentage.

evmUnit = "Decibel";

Display tables for decode status and measurement summary. The Packet EVM column of the Measurement Summary table displays the average EVM for all users in a packet.

ehtMeasurementSummary(cfgEHT,rmsEVM,decodeSuccess,pktOffsetStore,passSF,evmUnit,enableIQImbalanceCorrection,iqImbalEst);

Decode Status

    Packet Number    Start Index     User 1       User 2       User 3  
    _____________    ___________    _________    _________    _________

          1               74        "Success"    "Success"    "Success"
          2             9074        "Success"    "Success"    "Success"

Measurement Summary

    Packet Number    User 1 EVM (dB)    User 2 EVM (dB)    User 3 EVM (dB)    Packet EVM (dB)
    _____________    _______________    _______________    _______________    _______________

          1              -49.438            -49.408            -49.607            -49.484    
          2               -48.78            -49.123            -49.103            -49.001    

Average EVM for 3 users:
  User 1: -49.10dB
  User 2: -49.26dB
  User 3: -49.35dB
  All users: -49.24dB

Spectral Mask Measurement

In this section, you measure the spectral mask of the filtered and impaired waveform after HPA modeling. The transmitter spectral mask test [4] uses a time-gated spectral measurement of the EHT Data field. The example extracts the EHT Data field of each packet from the oversampled waveform by using the start indices of each packet within the baseband waveform. Any delay introduced in the baseband processing chain used to determine the packet indices must be accounted for when gating the EHT Data field within txWaveform. Concatenate the extracted EHT Data fields in preparation for measurement.

startIdx = osf*(ind.EHTData(1)-1)+1; % Upsampled start of EHT Data
endIdx = osf*ind.EHTData(2);         % Upsampled end of EHT Data
delay = grpdelay(firdec,1);          % Group delay of downsampling filter
numPackets = pktNum;
idx = zeros(endIdx-startIdx+1,numPackets);
for pktIdx = 1:numPackets
    % Start of packet in txWaveform
    pktOffset = round(osf*pktOffsetStore(pktIdx))-delay;
    % Indices of EHT Data in txWaveform
    idx(:,pktIdx) = (pktOffset+(startIdx:endIdx));
end
gatedEHTTData = txWaveform(idx(:),:);
if numPackets>0
    ehtSpectralMaskTest(gatedEHTTData,fs,osf);
end

   Spectral mask passed

Conclusion and Further Exploration

This example shows how to measure and plot these properties of an IEEE 802.11be waveform.

Spectral flatness for non-OFDMA PPDU type
RMS EVM per subcarrier
Equalized constellation
Spectral mask
IQ gain and phase imbalance

The HPA model introduces significant in-band distortion and spectral regrowth, which is visible in the EVM results, noisy constellation, and out-of-band emissions in the spectral mask plot. Try increasing the HPA backoff and observe the improved EVM, constellation, and lower out-of-band emissions. The downsampling (to bring the waveform to baseband for processing) stage includes filtering. The filter response affects the spectral flatness measurement. The ripple in the spectral flatness measurement is due to downsampling to baseband. Try using a different filter or changing the stop-band attenuation and observe the impact on the spectral flatness. For meaningful EVM results generate at least 20 EHT MU packets as specified in Section 36.3.19.4.4 of [1].

This example also optionally equalizes data symbols by using data-aided equalization to improve EVM performance. Below is a comparison of EVM with and without data-aided equalization. Data-aided equalization improved the EVM by approximately 1 dB with the default configurations in this example. Try using different APEP lengths, MCSs, numbers of space-time streams or hardware impairment models and observe the impact on the EVM.

Average EVM (dB)	Without data-aided equalization	With data-aided equalization
User 1	-49.10	-50.10
User 2	-49.26	-50.20
User 3	-49.35	-50.46
All users	-49.24	-50.25

The example also models IQ gain and phase imbalance by setting the modelIQImbalance flag to true. This introduces default gain imbalance of 1 dB and phase imbalance of 2 degrees in the waveform. The table below compares the EVM values with and without IQ imbalance correction by enabling and disabling the enableIQImbalanceCorrection setting. It shows that IQ imbalance correction significantly improves the EVM. Try using different IQ gain and phase imbalance values, MCSs, noise figures, and observe the impact on the EVM.

Average EVM (dB)	Without IQ imbalance correction	With IQ imbalance correction
User 1	-32.81	-47.21
User 2	-21.59	-47.52
User 3	-21.79	-48.46
All users	-24.08	-47.72

References

[1] IEEE Draft Standard for Information Technology–Telecommunications and Information Exchange between Systems Local and Metropolitan Area Networks–Specific Requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications Amendment: Enhancements for Extremely High Throughput (EHT).” IEEE P802.11be/D5.0, November 2023, Jan. 2024, pp. 1–1045. IEEE Xplore, https://ieeexplore.ieee.org/document/10381585.

[2] Loc and Cheong. IEEE P802.11 Wireless LANs. TGac Functional Requirements and Evaluation Methodology Rev. 16. 2011-01-19.

[3] Perahia, Eldad, and Robert Stacey. Next Generation Wireless LANs: 802.11n, 802.11ac, and Wi-Fi Direct. Second edition, Cambridge University Press, 2013.

[4] Archambault, Jerry, and Shravan Surineni. IEEE 802.11 spectral measurements using vector signal analyzers. RF Design 27.6 (2004): 38-49.