LTE System Toolbox

PDSCH 2x2 Spatial Multiplexing Throughput Conformance Test

This example demonstrates how to measure the Physical Downlink Shared Channel (PDSCH) throughput performance using the LTE System Toolbox™ under conformance test conditions as defined in TS36.101, Section 8.2.1.4.2 [ 1 ]: two codewords, Closed Loop Spatial Multiplexing (CLSM), two transmit and two receive antennas with low correlation and an ETU70 channel.

Introduction

This example implements wideband Precoder Matrix Indication (PMI) feedback to simulate the Closed Loop Spatial Multiplexing (CLSM) MIMO operation. TS36.101, Table 8.2.1.4.2-2 [ 1 ] defines a required throughput of at least 70% of the maximum given an SNR of 14.3 dB. A total of 8 Hybrid Automatic Repeat reQuest (HARQ) processes are used with a maximum of 4 retransmissions permitted.

Working on a subframe by subframe basis when simulating at a particular SNR, a populated resource grid is generated and OFDM modulated to create a transmit waveform. The generated waveform is passed through a noisy Extended Typical Urban (ETU) channel. Channel estimation, equalization, demodulation and decoding are performed at receiver. The throughput performance of the PDSCH is determined using the block CRC result at the output of the channel decoder.

Simulation Configuration

The example is executed for a simulation length of 1 frame at an SNR of 14.3 dB. A large number of NFrames should be used to produce meaningful throughput results. SNRIn can be an array of values or a scalar.

NFrames = 1;     % Number of frames
SNRIn = 14.3;    % SNR range
rng('default');  % Default random number generator seed

eNodeB Configuration

eNodeB settings are specified in a structure enb. This includes a substructure PDSCH to configure the PDSCH according to the conformance test requirements.

In this example a closed loop spatial multiplexing transmission scheme is used. The User Equipment (UE) provides feedback for the entire downlink system bandwidth using wideband Precoder Matrix Indicator (PMI) feedback. These parameters are set in the PDSCH configuration structure PDSCH.

The number of PDSCH codewords is controlled by the parameter ncw. In this example two codewords are used.

A power adjustment factor Rho is required as part of the conformance test. This scales the power of the PDSCH when using multiple transmit antennas to provide the receiver with the same energy as when using a single antenna transmission scheme. In this test two transmit antennas are used so the power is reduced by 3 dB as defined in TS36.101, Table 8.2.1.4.2-1 [ 1 ].

% eNodeB configuration
enb = struct;                        % Create eNodeB config structure
enb.TotSubframes = 1;                % Total subframes to generate
enb.RC = 'R.11';                     % RMC number
ncw = 2;                             % Number of PDSCH codewords

% PDSCH configuration
enb.PDSCH.TxScheme = 'SpatialMux';   % Transmission scheme
enb.PDSCH.RNTI = 1;                  % 16-bit UE specific mask
enb.PDSCH.Rho = -3;                  % PDSCH RE power adjustment factor
enb.PDSCH.CSI = 'On';                % CSI scaling of soft bits
enb.PDSCH.PMIMode = 'Wideband';      % Wideband PMI mode
enb.PDSCH.CodebookSubset = '110000'; % Codebook subset restriction

Set the PMI delay for the closed-loop spatial multiplexing. This is the delay between a PMI being passed from UE to eNodeB as defined in TS36.101, Table 8.2.1.4.2-1 [ 1 ].

pmiDelay = 8;

Propagation Channel Model Configuration

The structure, cfg, contains the channel model configuration parameters.

cfg.Seed = 10;                   % Random channel seed
cfg.NRxAnts = 2;                 % 2 receive antennas
cfg.DelayProfile = 'ETU';        % Delay profile
cfg.DopplerFreq = 70;            % Doppler frequency
cfg.MIMOCorrelation = 'Low';     % Multi-antenna correlation
cfg.NTerms = 16;                 % Oscillators used in fading model
cfg.ModelType = 'GMEDS';         % Rayleigh fading model type
cfg.InitPhase = 'Random';        % Random initial phases
cfg.NormalizePathGains = 'On';   % Normalize delay profile power
cfg.NormalizeTxAnts = 'On';      % Normalize for transmit antennas

Channel Estimator Configuration

The channel estimator is configured with a structure cec. The variable perfectChanEstimator controls channel estimator behavior. Valid values are true or false. When set to true a perfect channel estimator is used otherwise an imperfect estimate of the channel is used, based on the values of received pilot signals.

% Configure channel estimator
cec.PilotAverage = 'UserDefined';   % Type of pilot symbol averaging
cec.FreqWindow = 1;                 % Frequency window size in REs
cec.TimeWindow = 31;                % Time window size in REs
cec.InterpType = 'Cubic';           % 2D interpolation type
cec.InterpWindow = 'Centered';       % Interpolation window type
cec.InterpWinSize = 1;              % Interpolation window size

% Channel estimator behaviour
perfectChanEstimator = false;

RMC-DL Configuration

A downlink Reference Measurement Channel (RMC) configuration structure, rmc, is created using lteRMCDLlteRMCDL for test model R.11. The RMC configuration contained in rmc is used by lteRMCDLToollteRMCDLTool to generate an OFDM modulated waveform.

% Generate RMC configuration structure for R.11
rmc = lteRMCDL(enb, ncw);

% Redundancy version sequence for each codeword and transport block sizes
% for each subframe within a frame are taken from this structure
rvSequence = rmc.PDSCH.RVSeq;
trBlkSizes = rmc.PDSCH.TrBlkSizes;

Setup HARQ Processes

This example models the implementation of HARQ retransmission scheme. The HARQ table stores the index of each HARQ processes to use. This is generated for 8-HARQ processes using the helper function hHARQTable.mhHARQTable.m.

harqTable = hHARQTable();

Set Propagation Channel Model Sampling Rate

The sampling rate for the channel model is set using the value returned from lteOFDMInfolteOFDMInfo.

ofdmInfo = lteOFDMInfo(rmc);
cfg.SamplingRate = ofdmInfo.SamplingRate;

Processing Loop

The conformance test may be carried out over a number of SNR points. To determine the throughput at each SNR point, the PDSCH data is analyzed on a subframe by subframe basis using the following steps:

  • Update Current HARQ Process. The HARQ process either carries new transport data or a retransmission of previously sent transport data depending upon the Acknowledgment (ACK) or Negative Acknowledgment (NACK) based on CRC results. All this is handled by the HARQ scheduler, hHARQScheduling.mhHARQScheduling.m. The PDSCH data is updated based on the HARQ state.

  • Set PMI. A PMI is taken sequentially from a set of PMIs, txPMIs, each subframe and used by the eNodeB to select a precoding matrix. When a PMI is used by the eNodeB for a transmission it is replaced with a PMI selected by the UE. This PMI is then used to select a precoding matrix after pmiDelay subframes. Initially a set of pmiDelay random PMIs is used.

  • Create Transmit Waveform. The data generated by the HARQ process is passed to lteRMCDLToollteRMCDLTool which produces an OFDM modulated waveform, containing the physical channels and signals.

  • Noisy Channel Modeling. The waveform is passed through a fading channel and AWGN noise added. The noise power is normalized to take account of the sampling rate.

  • Perform Synchronization and OFDM Demodulation. The received symbols are offset to account for a combination of implementation delay and channel delay spread. The symbols are then OFDM demodulated.

  • Perform Channel Estimation. The channel response and noise are estimated using either a perfect or imperfect channel estimator. These estimates are used to aid the soft decoding of the PDSCH.

  • Decode the PDSCH. The recovered PDSCH symbols for all transmit and receive antenna pairs, along with a noise estimate, are demodulated and descrambled by ltePDSCHDecodeltePDSCHDecode to obtain an estimate of the received codeword.

  • Decode the DownLink Shared Channel (DL-SCH) and Store the Block CRC Error for a HARQ Process. The vector of decoded soft bits is passed to lteDLSCHDecodelteDLSCHDecode; this decodes the codeword and returns the block CRC error used to determine the throughput of the system. The contents of the new soft buffer, harqProc(harqIdx).decState, is available at the output of this function to be used for the next subframe.

  • Update PMI. A PMI is selected and feedback to the eNodeB for future use.

% The number of transmit antennas P is obtained from the resource grid
% dimensions. Dims is M-by-N-by-P where M is the number of subcarriers, N
% is the number of symbols and P is the number of transmit antennas.
Dims = lteDLResourceGridSize(rmc);
P = Dims(3);

% Initialize variables used in the simulation and analysis
resultIndex = 1;   % Initialize SNR test point counter index
offsets = 0;       % Initialize frame offset value

% Store results for each SNR point and each subframe containing data for
% the whole simulation
nDataTBS = sum(trBlkSizes(:)~=0)*NFrames;
totalBLKCRC = zeros(numel(SNRIn), nDataTBS);
bitThroughput = zeros(numel(SNRIn), nDataTBS);

for SNRdB = SNRIn
    fprintf('\nSimulating at %gdB SNR for %d Frame(s)\n' ,SNRdB, NFrames);

    % Initialize result store for SNR point tested
    blkCRC = zeros(nDataTBS/ncw, ncw);  % Intermediate block CRC
    bitTput = zeros(nDataTBS/ncw, ncw); % Intermediate throughput
    dataSubframeIndex = 1;

    % Initialize state of all HARQ processes
    harqprocess = hDownlinkSpatialMuxNewHARQProcess(rmc);
    harqProcesses = repmat(harqprocess, 1, max(harqTable));

    % Use random PMIs for the first 'pmiDelay' subframes until feedback is
    % available from the UE
    pmidims = ltePMIInfo(rmc,rmc.PDSCH);
    txPMIs = randi([0 pmidims.MaxPMI], pmidims.NSubbands, pmiDelay);

    for subframeNo = 0:(NFrames*10-1)

        % Update subframe number
        rmc.NSubframe = subframeNo;

        % Get HARQ index for given subframe from HARQ index table
        harqIdx = harqTable(mod(subframeNo, length(harqTable))+1);

        % Update current HARQ process
        harqProcesses(harqIdx) = hHARQScheduling( ...
            harqProcesses(harqIdx), subframeNo, rvSequence);

        % Update the PDSCH transmission config with HARQ process state
        rmc.PDSCH = harqProcesses(harqIdx).txConfig;
        data = harqProcesses(harqIdx).data;

        % Set the PMI to the appropriate value in the delay queue
        pmiIdx = mod(subframeNo, pmiDelay);     % PMI index in delay queue
        rmc.PDSCH.PMISet = txPMIs(:, pmiIdx+1); % Set PMI

        % Create transmit waveform and add 25 sample padding
        txWaveform = [lteRMCDLTool(rmc, data); zeros(25, P)];

        % The initialization time for channel modeling is set each subframe
        % to simulate a continuously varying channel
        cfg.InitTime = subframeNo/1000;

        % Pass data through channel model
        rxWaveform = lteFadingChannel(cfg, txWaveform);

        % Calculate noise gain including compensation for downlink power
        % allocation
        SNR = 10^((SNRdB-rmc.PDSCH.Rho)/20);

        % Normalize noise power to take account of sampling rate, which is
        % a function of the IFFT size used in OFDM modulation, and the
        % number of antennas
        N0 = 1/(sqrt(2.0*rmc.CellRefP*double(ofdmInfo.Nfft))*SNR);

        % Create additive white Gaussian noise
        noise = N0*complex(randn(size(rxWaveform)), ...
                            randn(size(rxWaveform)));

        % Add AWGN to the received time domain waveform
        rxWaveform = rxWaveform + noise;

        % Once every frame, on subframe 0, calculate a new synchronization
        % offset
        if (mod(subframeNo,10) == 0)
            offset = lteDLFrameOffset(rmc, rxWaveform);
            if (offset > 25)
                offset = offsets(end);
            end
            offsets = [offsets offset]; %#ok
        end

        % Synchronize the received waveform
        rxWaveform = rxWaveform(1+offset:end, :);

        % Perform OFDM demodulation on the received data to recreate the
        % resource grid
        rxSubframe = lteOFDMDemodulate(rmc, rxWaveform);

        % Channel estimation
        if(perfectChanEstimator)
            estChannelGrid = lteDLPerfectChannelEstimate(rmc, cfg, offset); %#ok
            n = lteOFDMDemodulate(rmc, noise(1+offset:end ,:));
            noiseEst = var(n(:));
        else
            [estChannelGrid, noiseEst] = lteDLChannelEstimate( ...
                rmc, cec, rxSubframe);
        end

        % Decode the PDSCH. The received subframe is scaled to counteract
        % any attenuation applied at the transmitter.
        rxEncodedBits = ltePDSCHDecode(rmc, rmc.PDSCH,...
              rxSubframe*(10^(-rmc.PDSCH.Rho/20)), ...
              estChannelGrid, noiseEst);

        % Extract the current subframe transport block size(s)
        TBSs = trBlkSizes(:, mod(subframeNo, 10)+1).';

        % Decode the DL-SCH
        [decbits, harqProcesses(harqIdx).blkerr, ...
            harqProcesses(harqIdx).decState] = lteDLSCHDecode( ...
            rmc, rmc.PDSCH, TBSs, rxEncodedBits, ...
            harqProcesses(harqIdx).decState);

        % Store block CRC and throughput results for subframes containing
        % transport data, ignore subframes containing no data
        if(any(TBSs ~= 0))
            blkCRC(dataSubframeIndex,:) = harqProcesses(harqIdx).blkerr;
            bitTput(dataSubframeIndex,:) = ...
                TBSs.*(1-harqProcesses(harqIdx).blkerr);
            dataSubframeIndex = dataSubframeIndex + 1;
        end

        % Provide PMI feedback to the eNodeB
        PMI = ltePMISelect(rmc, rmc.PDSCH, estChannelGrid, noiseEst);
        txPMIs(:, pmiIdx+1) = PMI;

    end

    % Record the block CRC error and bit throughput for the total number of
    % frames simulated at an SNR point
    totalBLKCRC(resultIndex,:) = blkCRC(:);
    bitThroughput(resultIndex,:) = bitTput(:);
    resultIndex = resultIndex + 1;

    % Display the results dynamically in the command window
    fprintf([['Throughput(Mbps) for ', num2str(NFrames) ' Frame(s) '],...
              '= %.4f\n'], mean(bitTput(:))* (nDataTBS/(NFrames*10))/1e3);
    fprintf(['Throughput(%%) for ', num2str(NFrames) ' Frame(s) = %.4f\n'],...
             100*(1-mean(blkCRC(:))));
end
Simulating at 14.3dB SNR for 1 Frame(s)
Throughput(Mbps) for 1 Frame(s) = 20.7360
Throughput(%) for 1 Frame(s) = 88.8889

Throughput Results

The throughput results for the simulation are displayed in the MATLAB® command window after each SNR point is completed, and are also captured in totalBLKCRC and bitThroughput. totalBLKCRC is a matrix where each row contains the results of decoding the CRC for a particular SNR. Each column contains the CRC result for a transport block containing PDSCH data at an SNR. bitThroughput is a matrix where each row contains the bit throughput per subframe for a particular SNR. Each column contains the throughput result for a transport block containing PDSCH data at an SNR.

Appendix

This example uses the helper functions:

Selected Bibliography

  1. 3GPP TS 36.101