This example shows how to measure the channel quality indicator (CQI) block error rate (BLER), which indicates the probability of incorrectly decoding CQI information sent using physical uplink control channel (PUCCH) format 2. The CQI BLER performance requirements are defined in TS 36.104 Section 22.214.171.124.
This example uses a simulation length of 10 subframes. This value has been chosen to speed up the simulation. A larger value should be chosen to obtain more accurate results. The CQI BLER is calculated for a number of SNR points. The target defined in TS 36.104 Section 126.96.36.199 [ 1 ] for 1.4 MHz bandwidth (6 RBs) and a single transmit antenna is a CQI BLER of 1% (i.e. probability of erroneous block detection P = 0.01) at an SNR of -3.9 dB. The test is defined for 1 transmit antenna.
numSubframes = 10; % Number of subframes SNRdB = [-9.9 -7.9 -5.9 -3.9 -1.9]; % SNR range NTxAnts = 1; % Number of transmit antennas
ue = struct; % UE config structure ue.NULRB = 6; % 6 resource blocks ue.CyclicPrefixUL = 'Normal'; % Normal cyclic prefix ue.Hopping = 'Off'; % No frequency hopping ue.NCellID = 9; ue.RNTI = 1; % Radio network temporary id ue.NTxAnts = NTxAnts;
% Empty hybrid ACK vector is used for Physical Uplink Control Channel % (PUCCH) 2 ACK = ; pucch = struct; % PUCCH config structure % Vector of PUCCH resource indices, one per transmission antenna. This is % the n2pucch parameter pucch.ResourceIdx = 0:ue.NTxAnts-1; % Set the size of resources allocated to PUCCH format 2 pucch.ResourceSize = 0; % Number of cyclic shifts used for PUCCH format 1 in resource blocks with a % mixture of formats 1 and 2. This is the N1cs parameter pucch.CyclicShifts = 0;
Configure the channel model with the parameters specified in the tests described in TS 36.104 Section 188.8.131.52 [ 1 ].
channel = struct; % Channel config structure channel.NRxAnts = 2; % Number of receive antennas channel.DelayProfile = 'ETU'; % Channel delay profile channel.DopplerFreq = 70.0; % Doppler frequency in Hz channel.MIMOCorrelation = 'Low'; % Low MIMO correlation channel.NTerms = 16; % Oscillators used in fading model channel.ModelType = 'GMEDS'; % Rayleigh fading model type channel.Seed = 3; % Channel seed channel.InitPhase = 'Random'; % Random initial phases channel.NormalizePathGains = 'On'; % Normalize delay profile power channel.NormalizeTxAnts = 'On'; % Normalize for transmit antennas % SC-FDMA modulation information: required to get the sampling rate info = lteSCFDMAInfo(ue); channel.SamplingRate = info.SamplingRate; % Channel sampling rate
The channel estimator is configured using a structure
cec. Here cubic interpolation will be used with an averaging window of 12-by-1 Resource Elements (REs). This configures the channel estimator to use a special mode which ensures the ability to despread and orthogonalize the different overlapping PUCCH transmissions.
cec = struct; % Channel estimation config structure cec.PilotAverage = 'UserDefined'; % Type of pilot averaging cec.FreqWindow = 12; % Frequency averaging window in REs (special mode) cec.TimeWindow = 1; % Time averaging window in REs (Special mode) cec.InterpType = 'cubic'; % Cubic interpolation
For each SNR point the loop below calculates the CQI BLER using information obtained from
NSubframes consecutive subframes. The following operations are performed for each subframe and SNR values:
Create an empty resource grid
Generate and map PUCCH 2 and its Demodulation Reference Signal (DRS) to the resource grid
Send the modulated signal through the channel
Minimum Mean Squared Error (MMSE) equalization
PUCCH 2 demodulation/decoding
Record decoding failures
PUCCH 2 DRS decoding. This is not required as part of this test but is included to illustrate the steps involved
% Preallocate memory for vector of BLERs versus SNR BLER = zeros(size(SNRdB)); for nSNR = 1:length(SNRdB) % Detection failures counter failCount = 0; % Noise configuration SNR = 10^(SNRdB(nSNR)/20); % Convert dB to linear % The noise added before SC-FDMA demodulation will be amplified by the % IFFT. The amplification is the square root of the size of the IFFT. % To achieve the desired SNR after demodulation the noise power is % normalized by this value. In addition, because real and imaginary % parts of the noise are created separately before being combined into % complex additive white Gaussian noise, the noise amplitude must be % scaled by 1/sqrt(2*ue.NTxAnts) so the generated noise power is 1. N = 1/(SNR*sqrt(double(info.Nfft)))/sqrt(2.0*ue.NTxAnts); % Set the type of random number generator and its seed to the default % value rng('default'); % Loop for subframes offsetused = 0; for nsf = 1:numSubframes % Create resource grid ue.NSubframe = mod(nsf-1, 10); % Subframe number reGrid = lteULResourceGrid(ue); % Resource grid % Create PUCCH 2 and its DRS CQI = randi([0 1], 4, 1); % Generate 4 CQI bits to send % Encode CQI bits to produce 20 bits coded = lteUCIEncode(CQI); pucch2Sym = ltePUCCH2(ue, pucch, coded); % PUCCH 2 modulation pucch2DRSSym = ltePUCCH2DRS(ue, pucch, ACK); % PUCCH 2 DRS creation % Generate indices for PUCCH 2 and its DRS pucch2Indices = ltePUCCH2Indices(ue, pucch); pucch2DRSIndices = ltePUCCH2DRSIndices(ue, pucch); % Map PUCCH 2 and its DRS to the resource grid reGrid(pucch2Indices) = pucch2Sym; reGrid(pucch2DRSIndices) = pucch2DRSSym; % SC-FDMA modulation txwave = lteSCFDMAModulate(ue, reGrid); % Channel state information: set the init time to the correct value % to guarantee continuity of the fading waveform channel.InitTime = (nsf-1)/1000; % Channel modeling % The additional 25 samples added to the end of the waveform are to % cover the range of delays expected from the channel modeling (a % combination of implementation delay and channel delay spread) rxwave = lteFadingChannel(channel, [txwave;zeros(25, ue.NTxAnts)]); % Add noise at receiver noise = N*complex(randn(size(rxwave)), randn(size(rxwave))); rxwave = rxwave + noise; % Receiver % Synchronization % An offset within the range of delays expected from the channel % modeling (a combination of implementation delay and channel % delay spread) indicates success [offset, rxACK] = lteULFrameOffsetPUCCH2( ... ue, pucch, rxwave, length(ACK)); if (offset<25) offsetused = offset; end % SC-FDMA demodulation rxgrid = lteSCFDMADemodulate(ue, rxwave(1+offsetused:end, :)); % Channel estimation [H, n0] = lteULChannelEstimatePUCCH2(ue, pucch, cec, rxgrid, rxACK); % Extract REs corresponding to the PUCCH 2 from the given subframe % across all receive antennas and channel estimates [pucch2Rx, pucch2H] = lteExtractResources(pucch2Indices, rxgrid, H); % MMSE Equalization eqgrid = lteULResourceGrid(ue); eqgrid(pucch2Indices) = lteEqualizeMMSE(pucch2Rx, pucch2H, n0); % PUCCH 2 demodulation rxBits = ltePUCCH2Decode(ue, pucch, eqgrid(pucch2Indices)); % PUCCH 2 decoding decoded = lteUCIDecode(rxBits, length(CQI)); % Record any decoding failures if (sum(decoded~=CQI)~=0) failCount = failCount + 1; end % Perform PUCCH 2 DRS decoding. This is not required as part of % this test, but illustrates the steps involved. % Extract REs corresponding to the PUCCH 2 DRS from the given % subframe across all receive antennas and channel estimates [drsRx, drsH] = lteExtractResources(pucch2DRSIndices, rxgrid, H); % PUCCH 2 DRS Equalization eqgrid(pucch2DRSIndices) = lteEqualizeMMSE(drsRx, drsH, n0); % PUCCH 2 DRS decoding rxACK = ltePUCCH2DRSDecode( ... ue, pucch, length(ACK), eqgrid(pucch2DRSIndices)); end % Probability of erroneous block detection BLER(nSNR) = (failCount/numSubframes); end
plot(SNRdB, BLER, 'b-o', 'LineWidth', 2, 'MarkerSize', 7); hold on; plot(-3.9, 0.01, 'rx', 'LineWidth', 2, 'MarkerSize', 7); xlabel('SNR (dB)'); ylabel('CQI BLER'); title('CQI BLER test (TS 36.104 Section 184.108.40.206)'); axis([SNRdB(1)-0.1 SNRdB(end)+0.1 -0.05 0.4]); legend('simulated performance', 'target');
3GPP TS 36.104 "Base Station (BS) radio transmission and reception"