5G Toolbox

https://www.mathworks.com/help/5g/
Admin
MATLAB 5G Toolbox provides functions and reference examples for modeling, simulation, and...
Tokens:850,009
Snippets:8,816
Trust Score:9.5
Update:2 months ago
Show doc for...
Tokens:
Raw
### Throughput Output Examples

Source: https://www.mathworks.com/help/5g/ug/nr-ntn-pusch-throughput.html

Example console outputs for throughput calculations.

```text
Throughput(Mbps) for 3 frame(s) at transmit power 18 dBm (modeled SNR per RE -7.9035 dB): 0.0000
```

```text
Throughput(%) for 3 frame(s) at transmit power 18 dBm (modeled SNR per RE -7.9035 dB): 0.0000
```

--------------------------------

### Simulation Output Example

Source: https://www.mathworks.com/help/5g/ug/nr-pdsch-throughput.html

Example output string from the simulation execution.

```text
Simulating transmission scheme 1 (8x2) and SCS=30kHz with CDL-C channel at -5dB SNR for 2 10ms frame(s)
```

--------------------------------

### Example Output Table

Source: https://www.mathworks.com/help/5g/ug/nr-pdsch-throughput.html

Sample output format for the throughput summary table.

```text
    SNRIn    Throughput (%)    Throughput (Mbps)    BLER (%)    Rx Antenna SNR
    _____    ______________    _________________    ________    ______________

     -5            40                24.17             60             5.2     
      0           100                60.43              0            10.2     
      5           100                60.43              0            15.2     
```

--------------------------------

### Example Event Output

Source: https://www.mathworks.com/help/5g/ref/nrgnb.registereventcallback.html

Sample structure of the event data returned by the callback.

```text
         EventName: "AppPacketGenerated"
          NodeName: "Node1"
            NodeID: 1
         Timestamp: 0
    TechnologyType: 2
         EventData: [1×1 struct]

         EventName: "TransmissionStarted"
          NodeName: "Node1"
            NodeID: 1
         Timestamp: 0
    TechnologyType: 2
         EventData: [1×1 struct]
```

--------------------------------

### Display Reference Channel Output

Source: https://www.mathworks.com/help/5g/ug/evm-measurement-of-5g-nr-downlink-waveforms.html

Example output showing the reference channel configuration.

```text
Reference Channel = NR-FR1-TM3.2

```

--------------------------------

### Model Information Output

Source: https://www.mathworks.com/help/5g/ug/identify-5g-and-lte-signals-using-pytorch-and-transfer-learning.html

Example output showing the model architecture details.

```text
Deeplabv3 model with resnet50 backbone has 189 layers and 41.9948 M learnables.
```

--------------------------------

### Signal Datastore Properties Output

Source: https://www.mathworks.com/help/5g/ug/generate-mimo-ofdm-channel-realizations-for-ai-based-systems.html

Example output showing the properties of the generated signal datastore.

```text
data = 
  signalDatastore with properties:

                       Files:{
                             ' .../tpf4c7b9b2/deeplearning_shared-ex43801460/Data/nr_channel_est_1.mat';
                             ' .../tpf4c7b9b2/deeplearning_shared-ex43801460/Data/nr_channel_est_10.mat';
                             ' .../tpf4c7b9b2/deeplearning_shared-ex43801460/Data/nr_channel_est_100.mat'
                              ... and 997 more
                             }
                     Folders: {'/tmp/Bdoc26a_3233028_63455/tpf4c7b9b2/deeplearning_shared-ex43801460/Data'}
    AlternateFileSystemRoots: [0×0 string]
                    ReadSize: 1
              OutputDataType: "same"
           OutputEnvironment: "cpu"
```

--------------------------------

### Setup Channel Parameters

Source: https://www.mathworks.com/help/5g/ug/ai-for-positioning-accuracy-enhancement.html

Configures channel sample rates and calculates maximum channel delay.

```MATLAB
function [channels,maxChDelay] = setupChannels(channels,ofdmInfo)
% Set the sample rate and initial time, then calculate the maximum delay
    
    numBS = size(channels,1);
    maxChDelay = 0;
    
    for bs = 1:numBS
        % Set the channel sample rate
        cdl = channels(bs).SmallScale;
        cdl.SampleRate = ofdmInfo.SampleRate;
      
        % Obtain the maximum delay of the channel
        chInfo = info(cdl);
        channels(bs).chInfo = chInfo;
        maxChDelay = max(maxChDelay,chInfo.MaximumChannelDelay);
    end
end
```

--------------------------------

### Configure BWP Parameters

Source: https://www.mathworks.com/help/5g/ref/nrcsireportconfig.html

Set the bandwidth part size and start index for the CSI report.

```matlab
reportConfig.NSizeBWP = 52;
reportConfig.NStartBWP = 0;
```

--------------------------------

### Example Output for Site 0 Sector 0 Metrics

Source: https://www.mathworks.com/help/5g/ug/nr-interference-modeling-with-toroidal-wrap-around.html

Sample console output showing performance metrics for a specific site and sector.

```text
Metrics for site 0, sector 0 :
```

```text
Peak UL throughput: 51.55 Mbps
Achieved cell UL throughput: 7.20 Mbps
Achieved UL throughput for each UE: [0.87        1.37        0.04        4.92]
Peak UL spectral efficiency: 10.31 bits/s/Hz
Achieved UL spectral efficiency for cell: 1.44 bits/s/Hz 
Block error rate for each UE in the UL direction: [0.017        0.19       0.586       0.138]

Peak DL throughput: 71.10 Mbps
Achieved cell DL throughput: 17.65 Mbps
Achieved DL throughput for each UE: [1.46        6.14        0.39        9.66]
Peak DL spectral efficiency: 14.22 bits/s/Hz
Achieved DL spectral efficiency for cell: 3.53 bits/s/Hz
Block error rate for each UE in the DL direction: [0.034           0       0.022           0]
```

--------------------------------

### Get Example Root Directory

Source: https://www.mathworks.com/help/5g/ug/csi-feedback-with-autoencoders.html

Retrieves the root directory for the example files.

```MATLAB
function rootDir = exRoot()
%exRoot Example root directory
rootDir = fileparts(which("helperCSINetDLNetwork"));
end
```

--------------------------------

### Example Output for Site 6 Sector 0 Metrics

Source: https://www.mathworks.com/help/5g/ug/nr-interference-modeling-with-toroidal-wrap-around.html

Sample console output showing performance metrics for a different site and sector.

```text
Metrics for site 6, sector 0 :
```

```text
Peak UL throughput: 51.55 Mbps
Achieved cell UL throughput: 4.85 Mbps
Achieved UL throughput for each UE: [3.72        0.74        0.14        0.25]
Peak UL spectral efficiency: 10.31 bits/s/Hz
Achieved UL spectral efficiency for cell: 0.97 bits/s/Hz 
Block error rate for each UE in the UL direction: [0       0.241       0.052       0.034]

Peak DL throughput: 71.10 Mbps
Achieved cell DL throughput: 5.02 Mbps
Achieved DL throughput for each UE: [0.52        3.19        0.22        1.08]
Peak DL spectral efficiency: 14.22 bits/s/Hz
Achieved DL spectral efficiency for cell: 1.00 bits/s/Hz
Block error rate for each UE in the DL direction: [0.371       0.056        0.09           0]
```

--------------------------------

### Initialize Simulator and Nodes

Source: https://www.mathworks.com/help/5g/ug/create-and-add-custom-mobility-model-to-ue-nodes-and-visualize-their-movement.html

Setup steps for initializing the wireless network simulator and creating UE nodes with specific initial positions.

```matlab
simulator = wirelessNetworkSimulator.init;
```

```matlab
ue1 = nrUE(Position=[0 0 0]);
ue2 = nrUE(Position=[0 -2 0]);
ue3 = nrUE(Position=[1 -2 0]);
```

```matlab
addNodes(simulator,[ue1 ue2 ue3])
```

--------------------------------

### DCI Search Output Example

Source: https://www.mathworks.com/help/5g/ug/nr-cell-search-and-mib-and-sib1-recovery.html

Example output from the DCI message search process.

```text
 -- Downlink control information message search in PDCCH -- 
 Decoded PDCCH candidate #1 at aggregation level 8
 PDCCH RMS EVM: 12.339%
 PDCCH CRC: 0
```

--------------------------------

### Initialize System and Channel Parameters

Source: https://www.mathworks.com/help/5g/ug/train-pytorch-channel-prediction-models-with-online-training.html

Sets up antenna sizes, delay profiles, Doppler shifts, and carrier configurations for the channel model.

```matlab
txAntennaSize = 2;
rxAntennaSize = 2;
rmsDelaySpread = 300e-9;     % s
maxDoppler = 37;             % Hz
nSizeGrid = 52;              % Number resource blocks (RB)
subcarrierSpacing = 15;      % 15,30,60,120 kHz
numerology = (subcarrierSpacing/15)-1;

channel = nrTDLChannel;
channel.DelayProfile = 'TDL-A';
channel.DelaySpread = rmsDelaySpread;       % s
channel.MaximumDopplerShift = maxDoppler;   % Hz
channel.RandomStream = "Global stream";
channel.NumTransmitAntennas = txAntennaSize;
channel.NumReceiveAntennas = rxAntennaSize;
channel.ChannelFiltering = false;

% Carrier definition
carrier = nrCarrierConfig;
carrier.NSizeGrid = nSizeGrid;
carrier.SubcarrierSpacing = subcarrierSpacing;

% Channel prediction horizon
horizon = 2;
```

--------------------------------

### Initialize Python Environment with Helper Function

Source: https://www.mathworks.com/help/5g/ug/verify-performance-of-6g-ai-native-receiver-using-pytorch.html

Uses the helperSetupPyenv function to initialize the environment and verify that required libraries are installed.

```matlab
currentPyEnv = helperSetupPyenv(...
    simParameters.PythonPath,...
    simParameters.PythonExecutionMode,...
    simParameters.PythonRequirements);
```

--------------------------------

### Packet Filter Examples

Source: https://www.mathworks.com/help/5g/ref/pcapreader.read.html

Examples of packet filter strings using various protocols and logical operators.

```text
‘pcap.PacketLength == 100’
```

```text
‘eth.SourceAddress == 44FB5A9710AC & eth.Type == 2048’
```

```text
 ‘ecpri.MessageType == IQData || ecpri.MessageType == RemoteReset || ecpri.MessageType == EventIndication’
```

--------------------------------

### Initialize Metrics Visualizer

Source: https://www.mathworks.com/help/5g/ug/nr-cell-performance-evaluation-with-mimo.html

Sets up the visualizer for scheduler and PHY metrics.

```MATLAB
metricsVisualizer = helperNRMetricsVisualizer(gNB,UEs, ...
    RefreshRate=numMetricPlotUpdates,PlotSchedulerMetrics=true, ...
    PlotPhyMetrics=true,PlotCDFMetrics=false);
```

--------------------------------

### Slot Status Output Example

Source: https://www.mathworks.com/help/5g/gs/dl-sch-and-pdsch-transmit-and-receive-processing-chain.html

Example output showing the status of HARQ processes across multiple slots.

```text
Slot 0. HARQ Proc 0: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 1. HARQ Proc 1: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 2. HARQ Proc 2: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 3. HARQ Proc 3: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 4. HARQ Proc 4: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 5. HARQ Proc 5: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 6. HARQ Proc 6: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 7. HARQ Proc 7: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 8. HARQ Proc 8: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 9. HARQ Proc 9: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 10. HARQ Proc 10: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 11. HARQ Proc 11: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 12. HARQ Proc 12: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 13. HARQ Proc 13: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 14. HARQ Proc 14: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 15. HARQ Proc 15: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 16. HARQ Proc 0: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 17. HARQ Proc 1: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 18. HARQ Proc 2: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
Slot 19. HARQ Proc 3: CW0: Initial transmission passed  (TBS=24072,RV=0,CR=0.482212).
```

--------------------------------

### Retrieve Starting Subframes from Configuration Table

Source: https://www.mathworks.com/help/5g/ref/nrprachofdminfo.html

Accesses the configuration table to determine the starting subframe numbers for active preambles.

```matlab
configTable = prach.Tables.ConfigurationsFR1PairedSUL;
subframeNumber = configTable.SubframeNumber{configurationIndex+1,:}
```

```matlab
subframeNumber = _1×2_

     1     6
```

--------------------------------

### Configure Wireless Network Viewer

Source: https://www.mathworks.com/help/5g/ug/create-and-add-custom-mobility-model-to-ue-nodes-and-visualize-their-movement.html

Initialize the visualizer and add the UE nodes to it for tracking movement.

```matlab
visualizer = wirelessNetworkViewer;
```

```matlab
addNodes(visualizer,[ue1 ue2 ue3])
```

--------------------------------

### Open Site Viewer

Source: https://www.mathworks.com/help/5g/ug/analyze-sinr-in-6g-fr3-network-using-digital-twin.html

Initialize the 3-D Site Viewer tool with building data.

```matlab
if exist("viewer","var") && isvalid(viewer) % Viewer handle exists and viewer window is open
    clearMap(viewer);
else
    viewer = siteviewer(Basemap="openstreetmap",Buildings=filename);
end
```

--------------------------------

### Performance Indicator Output Example

Source: https://www.mathworks.com/help/5g/ug/evaluate-performance-of-cell-free-mmimo-networks.html

Example output showing peak and achieved throughput, spectral efficiency, and block error rates.

```text
Peak UL throughput: 54.64 Mbps
Achieved cell UL throughput: 45.48 Mbps
Achieved UL throughput for each UE: [11.31        11.38        11.42        11.36]
Peak UL spectral efficiency: 2.73 bits/s/Hz
Achieved UL spectral efficiency for cell: 2.27 bits/s/Hz 
Block error rate for each UE in the UL direction: [0  0  0  0]

Peak DL throughput: 75.37 Mbps
Achieved cell DL throughput: 66.25 Mbps
Achieved DL throughput for each UE: [16.37        16.73        16.54        16.61]
Peak DL spectral efficiency: 3.77 bits/s/Hz
Achieved DL spectral efficiency for cell: 3.31 bits/s/Hz
Block error rate for each UE in the DL direction: [0  0  0  0]
```

--------------------------------

### Open 5G NR Interference Modeling Example

Source: https://www.mathworks.com/help/5g/ug/run-5g-system-level-simulations-on-aws-cloud.html

Command to open the NR Interference Modeling with Toroidal Wrap-Around example in the MATLAB environment.

```matlab
%openExample("5g/NRInterferenceModelingWithToroidalWrapAroundExample");
```

--------------------------------

### nrDLSCHInfo

Source: https://www.mathworks.com/help/5g/referencelist.html?addon=&blocktype=all&capability=&category=downlink-channels&endrelease=&listtype=cat&s_tid=CRUX_lftnav&startrelease=&type=function

Get downlink shared channel (DL-SCH) information.

```APIDOC
## nrDLSCHInfo

### Description
Retrieves information regarding the downlink shared channel (DL-SCH).
```

--------------------------------

### Initialize Metrics Visualizer

Source: https://www.mathworks.com/help/5g/ug/nr-fdd-scheduling-performance-evaluation.html

Sets up the visualizer for scheduler and RLC metrics using the specified refresh rate.

```MATLAB
showSchedulerMetrics = true;
showRLCMetrics = true;
metricsVisualizer = helperNRMetricsVisualizer(gNB,UEs, ...
    RefreshRate=numMetricPlotUpdates, ...
    PlotSchedulerMetrics=showSchedulerMetrics, ...
    PlotRLCMetrics=showRLCMetrics);
```

--------------------------------

### Peak EVM output

Source: https://www.mathworks.com/help/5g/ug/evm-measurement-of-5g-nr-pusch-waveforms.html

Example output for peak EVM measurement.

```text
Peak EVM : 8.0145%
```

--------------------------------

### Display Worker Count Output

Source: https://www.mathworks.com/help/5g/ug/accelerate-link-level-simulations-with-parallel-processing.html

Example output for the number of workers.

```MATLAB
Number of workers: 12
```

--------------------------------

### Initialize Wireless Network Simulator

Source: https://www.mathworks.com/help/5g/ug/nr-cell-performance-evaluation-with-beam-management.html

Sets up the wireless network simulator with the specified gNB and UE nodes.

```MATLAB
% Initialize wireless network simulator
nrNodes = [{gNB}; UEs];
networkSimulator = hWirelessNetworkSimulator(nrNodes);
```

--------------------------------

### Display Measurement Structure

Source: https://www.mathworks.com/help/5g/ug/5g-nr-csi-rs-measurements.html

Example output structure for CSI-RS measurements.

```MATLAB
meas = _struct with fields:_
    RSRPPerAntenna: [-94.1599 -91.3258]
    RSSIPerAntenna: [-65.6220 -64.9823]
    RSRQPerAntenna: [-11.3779 -9.1834]
```

--------------------------------

### Configure Indoor Factory Scenario

Source: https://www.mathworks.com/help/5g/ug/ai-for-positioning-accuracy-enhancement.html

Sets up the simulation environment parameters including scenario type, hall dimensions, clutter properties, and UE/BS deployment settings.

```MATLAB
simParameters.Mode            = "Simulate"; % Simulate with the pretrained network or train the network
simParameters.TrainRatio      = 0.95; % Define the ratio of training data relative to validation and test data

simParameters.Scenario        = "InF-DH"; % "InF-SL", "InF-DL", "InF-SH", "InF-DH", "InF-HH"
simParameters.HallSize        = [120 60 10]; % Dimensions of hall for InF scenarios: [length width height] in meters
simParameters.ClutterSize     = 2; % Clutter size (m)
simParameters.ClutterHeight   = 6; % Clutter height (m)
simParameters.ClutterDensity  = 0.6; % Clutter density, between 0 and 1

simParameters.BSSeparation    = [20 20]; 
simParameters.BSAntennaHeight = 8;
simParameters.BSNoiseFigure   = 5;

simParameters.UEDrop          = "random"; % UE drop type can be either 'random' or 'grid'
simParameters.NumUEs          = 15; % Number of UE positions; applicable if ueDropType is 'random'
simParameters.UESeparation    = [30 30]; % UE separation in x and y axes in meters; applicable if ueDropType is 'grid'
simParameters.UEAntennaHeight = 1.5; % UE antenna height on the z axis in meters

% Set a random number seed for repeatability
simParameters.Seed = 23;
rng(simParameters.Seed,"twister");

% Configure and visualize the scenario
simParameters = h5GIndoorFactoryScenario(simParameters);
```

--------------------------------

### Time-Domain Noise Power Output

Source: https://www.mathworks.com/help/5g/ug/snr-definition-used-in-link-simulations.html

Example output for time-domain noise power.

```text
Time-domain channel modeling noise power: 4.9932
```

--------------------------------

### Frequency-Domain Noise Power Output

Source: https://www.mathworks.com/help/5g/ug/snr-definition-used-in-link-simulations.html

Example output for frequency-domain noise power.

```text
Frequency-domain channel modeling noise power: 4.9115
```

--------------------------------

### Configure and Execute Network Training

Source: https://www.mathworks.com/help/5g/ug/csi-feedback-transformer-autoencoder.html

Sets up training options and executes the training process or loads a pretrained network based on the trainNow flag.

```MATLAB
trainNow = false;
saveNetwork = false;

options = trainingOptions("adam", ...
    InitialLearnRate=initLearnRate, ...
    LearnRateSchedule=schedule, ...
    MaxEpochs=epochs, ...
    MiniBatchSize=batchSize, ...
    Shuffle="every-epoch", ...
    Verbose=false, ...
    VerboseFrequency=100, ...
    ValidationData={HValReal,HValReal}, ...
    ValidationFrequency=200, ...
    OutputNetwork="best-validation-loss", ...
    Metrics=trainingMetric, ...
    Plots="training-progress");

if trainNow
    [trainedNet, trainInfo] = trainnet(HTrainReal,HTrainReal,CSIFormerNet,@(X,T) mse(X,T),options);%#ok

    if saveNetwork
        save("CSIFormerTrainedNetwork_" ...
            + string(datetime("now","Format","dd_MM_HH_mm")), 'trainedNet')
    end

else
    trainedNetName = "trainedNetEnc" + num2str(compressionFactor);
    load("CSIFormerTrainedNets.mat",trainedNetName)
end
```

--------------------------------

### Initialize DQN Agent

Source: https://www.mathworks.com/help/5g/ug/train-dqn-agent-for-beam-selection.html

Create a DQN agent using observation and action information from the environment and configure the experience buffer.

```MATLAB
obsInfo = getObservationInfo(envTrain);
actInfo = getActionInfo(envTrain);
agent = rlDQNAgent(obsInfo,actInfo);

agent.ExperienceBuffer = rlPrioritizedReplayMemory(obsInfo,actInfo);
```

--------------------------------

### Display Transmission Parameters Output

Source: https://www.mathworks.com/help/5g/ug/nr-sidelink-vector-waveform-generation.html

Example output of the SLTransmission configuration properties.

```text
    SubchannelAllocation: [0 2]
               NumLayers: 1
              Modulation: '16QAM'
        NumDMRSPositions: 3
    NumDMRSPositionsList: [2 3]
          TargetCodeRate: 0.3691
          BetaSCI2Offset: 5
                   OSCI2: 35
                    NXID: []
       RedundancyVersion: 0
           HARQProcessID: 0
           PSFCHOverhead: 0
                   OSCI1: 26
```

--------------------------------

### Configure and execute ray tracing

Source: https://www.mathworks.com/help/5g/ug/analyze-sinr-in-6g-fr3-network-using-digital-twin.html

Sets up the propagation model and calculates paths between gNB and UE sites.

```MATLAB
pm = propagationModel("raytracing",MaxNumReflections=maxNumReflections,MaxNumDiffractions=maxNumDiffractions);
```

```MATLAB
rays = raytrace(tx,rx,pm);
```

--------------------------------

### Display Pool Configuration Output

Source: https://www.mathworks.com/help/5g/ug/nr-sidelink-vector-waveform-generation.html

Example output of the NRSidelinkResourcePool configuration properties.

```text
  NRSidelinkResourcePool with properties:

                   SCSCarrier: [1×1 nrCarrierConfig]
                    NStartBWP: []
                     NSizeBWP: []
           sl_StartSymbol_r16: 0
         sl_LengthSymbols_r16: 14
        sl_SubchannelSize_r16: 10
    sl_StartRB_Subchannel_r16: 0
         sl_NumSubchannel_r16: 5
            sl_X_Overhead_r16: 0
               sl_Scaling_r16: 1
     sl_TimeResourcePSCCH_r16: 2
     sl_FreqResourcePSCCH_r16: 10
       sl_DMRS_ScrambleID_r16: 0
          sl_PSFCH_Period_r16: 4
            MaxNumSubchannels: 5
                  IsPSFCHSlot: 0
```

--------------------------------

### Initialize Metric Visualizer

Source: https://www.mathworks.com/help/5g/ug/nr-cell-performance-evaluation-with-physical-layer-integration.html

Set up the metrics visualizer with specified refresh rates and enabled plot categories.

```MATLAB
metricsVisualizer = helperNRMetricsVisualizer(gNB,UEs, ...
    RefreshRate=numMetricPlotUpdates,PlotSchedulerMetrics=true, ...
    PlotPhyMetrics=true);
```

--------------------------------

### Performance Metrics Table Output

Source: https://www.mathworks.com/help/5g/ug/matlab-pytorch-coexecution-for-csi-feedback-compression-offline-training.html

Example output of the performance metrics table.

```text
     Model     Compression Factor    NMSE(dB)      Rho      InferenceTime    NumberOfLearnables
    _______    __________________    ________    _______    _____________    __________________

    "CLNet"            4             -27.691     0.99958      0.035358           1.0289e+05    
```

--------------------------------

### Configure and Execute Online Training Loop

Source: https://www.mathworks.com/help/5g/ug/train-pytorch-channel-prediction-models-with-online-training.html

Sets up training parameters, initializes the trainer via a Python wrapper, and executes the training loop with periodic validation.

```matlab
trainNow = false;
if trainNow
  dltAvailable = checkDeepLearningToolbox();
  numIterations            = 50000;
  trainingBatchSize        = 512;
  validationBatchSize      = 1000;
  initialLearningRate      = 1e-3;
  learningRateUpdatePeriod = 175;
  validationFrequency      = 1000;
  progressPlot             = false;    % Set to true if you want progress plot
  verbose                  = true;

  if dltAvailable && progressPlot
    monitor = trainingProgressMonitor( ...
      Metrics=["TrainingLoss","ValidationLoss"], ...
      Info=["LearningRate","Iteration"], ...
      XLabel="Iteration");
    groupSubPlot(monitor,"Loss",["TrainingLoss","ValidationLoss"]);
    yscale(monitor,"Loss","log")
    monitor.Status = "Running";
  end

  tStartLoop = tic;
  dataGen.BatchSize = trainingBatchSize;

  % Set up the trainer for online training using the Python wrapper.
  trainer = py.nr_channel_predictor_wrapper.setup_trainer( ...
    Ntx, ...                 % number of transmit antennas
    gruHiddenSize, ...       % hidden size
    gruNumLayers, ...        % number of layers
    initialLearningRate, ... % learning rate
    trainingBatchSize, ...   % batch size
    verbose);

  % Training loop
  numOutputs = floor(numIterations/validationFrequency);
  trainingLoss = nan(1,numIterations);
  validationLoss = nan(1,numOutputs);
  iterations = nan(1,numOutputs);
  outCount = 1;
  tStart = tic;
  fprintf('Starting online training...\n');
  for iteration = 1:numIterations
    % Get the next batch from the data generator
    [inputData,targetData] = dataGen();

    % Perform one training update
    trainLoss = py.nr_channel_predictor_wrapper.train_one_iteration(trainer,inputData,targetData, ...
      ~mod(iteration,learningRateUpdatePeriod));
    trainingLoss(iteration) = trainLoss;

    % Every validationFrequency iterations, perform validation
    if mod(iteration,validationFrequency) == 0
      % Access data from Python
      learningRate = trainer.get_current_learning_rate();

      % Generate validation batch
      dataGen.BatchSize = validationBatchSize;
      [inputData,targetData] = dataGen();
      dataGen.BatchSize = trainingBatchSize;

      % Perform validation
      valLoss = py.nr_channel_predictor_wrapper.validate(trainer,inputData,targetData);

      % Report metrics
      valLoss = double(valLoss);
      if verbose
        et = seconds(toc(tStartLoop)); et.Format = "hh:mm:ss.SSS";
        fprintf('%s : [%d/%d] Training Loss: %1.3f, Validation Loss: %1.3f, Learning Rate: %1.3e\n',...
          et,iteration,numIterations,trainLoss,valLoss,learningRate);
      end
      if dltAvailable && progressPlot
        recordMetrics(monitor, ...
          iteration, ...
          TrainingLoss=trainLoss, ...
          ValidationLoss=valLoss);
        updateInfo(monitor, ...
          LearningRate=learningRate, ...
          Iteration=sprintf("%d of %d",iteration,numIterations));
        monitor.Progress = 100*iteration/numIterations;
        if monitor.Stop
          monitor.Status = "Stopped by user";
          break
        end
      else
        trainingLoss(outCount) = trainLoss;
        validationLoss(outCount) = valLoss;
        iterations(outCount) = iteration;
        outCount = outCount+1;
      end
    end
  end
  if dltAvailable && progressPlot
    monitor.Status = "Training complete";
    trainingLoss = monitor.MetricData.TrainingLoss(:,2);
    validationLoss = monitor.MetricData.ValidationLoss(:,2);
    iterationNum = monitor.MetricData.ValidationLoss(:,1);
    iterations = str2double(extractBefore(monitor.InfoData.Iteration," of"));
  end
  et = toc(tStart); et = seconds(et); et.Format = "hh:mm:ss.SSS";
  fprintf('%s : Training complete.\n',et);
end
```

--------------------------------

### Initialize Simulation Environment

Source: https://www.mathworks.com/help/5g/ug/create-and-add-custom-traffic-in-5g-simulations.html

Set the random seed and initialize the wireless network simulator.

```MATLAB
rng(1,"combRecursive");
```

```MATLAB
networkSimulator = wirelessNetworkSimulator.init;
```

--------------------------------

### Generation Progress Output

Source: https://www.mathworks.com/help/5g/ug/generate-mimo-ofdm-channel-realizations-for-ai-based-systems.html

Example console output during the channel generation process.

```text
Removing invalid data directory: /tmp/Bdoc26a_3233028_63455/tpf4c7b9b2/deeplearning_shared-ex43801460/Data
Starting channel realization generation
1 worker(s) running
00:00:02 - 10% Completed
00:00:04 - 20% Completed
00:00:06 - 30% Completed
00:00:07 - 40% Completed
00:00:09 - 50% Completed
00:00:11 - 60% Completed
00:00:13 - 70% Completed
00:00:15 - 80% Completed
00:00:17 - 90% Completed
00:00:19 - 100% Completed
00:00:19 - 100% Completed
```

--------------------------------

### Display PDSCH Symbol Output

Source: https://www.mathworks.com/help/5g/ref/nrpdschindices.html

Example output for generated PDSCH symbols.

```text
sym = _3900×1 single column vector_

  -0.9487 + 0.9487i
  -0.9487 - 0.9487i
  -0.3162 - 0.9487i
   0.9487 - 0.3162i
  -0.9487 + 0.3162i
   0.3162 + 0.9487i
   0.3162 + 0.9487i
  -0.3162 + 0.3162i
   0.3162 + 0.3162i
   0.9487 - 0.3162i
   0.3162 + 0.3162i
  -0.3162 + 0.3162i
  -0.9487 + 0.3162i
   0.3162 - 0.3162i
   0.3162 + 0.3162i
      ⋮
```

--------------------------------

### Configure the C/C++ compiler

Source: https://www.mathworks.com/help/5g/ug/what-is-c-code-generation-from-matlab.html

Run the mex -setup command at the MATLAB prompt to configure the environment for code generation.

```matlab
mex -setup
```

--------------------------------

### Initialize Metric Visualizers

Source: https://www.mathworks.com/help/5g/ug/nr-interference-modeling-with-toroidal-wrap-around.html

Create visualization objects for MAC and PHY metrics for the selected cells.

```matlab
metricsVisualizer = cell(numCellsOfInterest,1);

for cellIdx = 1:numCellsOfInterest

    % Obtain the gNB and UEs for the cell of interest from the scenario object
    [gNB, UEs] = getNodesForCell(scenario, siteOfInterest, sectorOfInterest, cellIdx);
    
    % Create visualization object for the MAC and PHY metrics
    metricsVisualizer{cellIdx} = helperNRMetricsVisualizer(gNB, UEs, ...
        RefreshRate=numMetricPlotUpdates, PlotSchedulerMetrics=true, ...
        PlotPhyMetrics=true, CellOfInterest=siteOfInterest(cellIdx));

end
```

--------------------------------

### MIB Structure Output

Source: https://www.mathworks.com/help/5g/ug/nr-cell-search-and-mib-and-sib1-recovery.html

Example output showing the contents of the parsed MIB structure.

```text
 BCH/MIB Content:
                     NFrame: 0
    SubcarrierSpacingCommon: 15
                      k_SSB: 0
          DMRSTypeAPosition: 3
            PDCCHConfigSIB1: [1×1 struct]
                 CellBarred: 0
       IntraFreqReselection: 0
```

--------------------------------

### Create UAV Tracking Scenario

Source: https://www.mathworks.com/help/5g/ug/vehicle-tracking-for-urban-air-mobility-using-5g-prs.html

Initializes a uavScenario object with three UAVs and six base stations, setting up trajectories and ground station positions.

```matlab
function scene = createUAVTrackingScenario()
    % scene = createUAVTrackingScenario() creates the uavScenario object with three UAVs
    % and six base stations.

    minHeight = 1.6925e+03;
    latlonCenter = [39.9786 -105.2882 minHeight];
    
    % Set the update rate as integer multiples of 1 ms, so that it
    % corresponds to the start of a slot when the instantaneous time is converted to
    % slot number in the 5G processing
    scene = uavScenario(UpdateRate=10,...
        StopTime=50,...
        ReferenceLocation=latlonCenter,...
        MaxNumFrames=200,...
        HistoryBufferSize=1000);
    
    % Package delivery drone
    waypointsA = [1895 90 20; 1995 180 35; 1820 125 20];
    timeA = [0 25 50];
    trajA = waypointTrajectory(waypointsA, TimeOfArrival=timeA,...
        ReferenceFrame="ENU",...
        AutoBank=true);

    uavA = uavPlatform("UAV1", scene, ...
        Trajectory=trajA,...
        ReferenceFrame="ENU");
    updateMesh(uavA, "quadrotor", {5}, 0*[1 1 1], eye(4));
    
    % Air Taxi
    waypointsB = [1940 120 50; 1240 -230 50];
    timeB = [0 50];
    trajB = waypointTrajectory(waypointsB, TimeOfArrival=timeB,...
        ReferenceFrame="ENU",...
        AutoBank=true);
    uavB = uavPlatform("UAV2", scene, ...
        Trajectory=trajB,...
        ReferenceFrame="ENU");
    updateMesh(uavB, "fixedwing", {10},  0*[1 1 1], eye(4));
    
    % Grid line maintenance 
    waypointsC = [1950 60 35; 1700 60 35; 1900 160 35];
    timeC = linspace(0,50,size(waypointsC,1));
    trajC = waypointTrajectory(waypointsC, TimeOfArrival=timeC,...
        ReferenceFrame="ENU",...
        AutoBank=true);
    uavC = uavPlatform("UAV3", scene, ...
        Trajectory=trajC,...
        ReferenceFrame="ENU");
    updateMesh(uavC, "quadrotor", {5}, 0*[1 1 1], eye(4));
    
    % Add six 5G ground stations in a circular region
    xC = [1850;250;0];
    r = 500;
    n = 6;
    dT = 360/n;
    theta = dT:dT:360;
    z = [50 75 100 240 10 50];
    pos = xC + [r.*cosd(theta);r.*sind(theta);z];
    for i = 1:size(pos,2)
        gs = uavPlatform(strcat("GS",num2str(i)),...
            scene,ReferenceFrame="ENU",...
            InitialPosition=pos(:,i)');
        updateMesh(gs,"cuboid",{[50,50,pos(3,i)]},0.3*[1 1 1],[0 0 -pos(3,i)/2],[1 0 0 0]);
    end
    
    % Set up the scene
    setup(scene);
end
```

--------------------------------

### Initialize Resource Grid Prototype

Source: https://www.mathworks.com/help/5g/ug/nr-pdsch-throughput.html

Creates a resource grid prototype based on the data type and GPU usage settings.

```MATLAB
if simParameters.UseGPU == "on" || (simParameters.UseGPU == "auto" && canUseGPU)
    device = gpuDevice().Name;
    fprintf('\nUsing GPU device %s for computation\n',device);
    simParameters.Prototype = zeros(0,0,simParameters.DataType,"gpuArray");
    if class(channel) == "nrTDLChannel"
        error("The nrTDLChannel object does not support GPU arrays. Use an nrCDLChannel object instead.")
    end
else
    simParameters.Prototype =  zeros(0,0,simParameters.DataType);
end
```

--------------------------------

### Simulation Results Table Output

Source: https://www.mathworks.com/help/5g/ug/accelerate-link-level-simulations-with-parallel-processing.html

Example output showing the throughput results table.

```MATLAB
    SNR    Simulated bits    Tr Block errors    Number of Tr Blocks    Number of frames    Throughput (%)    Throughput (Mbps)
    ___    ______________    _______________    ___________________    ________________    ______________    _________________

    5.7      1.8749e+06            120                  120                   12                    0                  0      
    5.8      1.8749e+06            108                  120                   12                   10             1.5624      
    5.9      1.8749e+06             67                  120                   12               44.167             6.9006      
      6      1.8749e+06             31                  120                   12               74.167             11.588      
    6.1      1.8749e+06              6                  120                   12                   95             14.843      
    6.2      1.8749e+06              1                  120                   12               99.167             15.494      
```

--------------------------------

### Display Index and Info Output

Source: https://www.mathworks.com/help/5g/ref/nrpdschindices.html

Example output structures for indices and information metadata.

```text
ind = _3900×3 uint32 matrix_

    1    1    1
    2    1    1
    3    1    1
    4    1    1
    5    1    1
    6    1    1
    7    1    1
    8    1    1
    9    1    1
   10    1    1
   11    1    1
   12    1    1
   13    1    1
   14    1    1
   15    1    1
      ⋮
```

```text
info = _struct with fields:_
                G: 15600
               Gd: 3900
        NREPerPRB: 156
    DMRSSymbolSet: 2
    PTRSSymbolSet: [1×0 double]
           PRBSet: [0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24]
```

--------------------------------

### Initialize System-Level Channel Model

Source: https://www.mathworks.com/help/5g/ug/evaluate-3gpp-indoor-reference-scenario.html

Create the channel model object and associate it with gNB sites.

```matlab
channel = h38901Channel(Scenario="InH",OfficeType=officeType);
chcfg.Site = 1:numGNBs;
```

--------------------------------

### Configure SRS Symbols

Source: https://www.mathworks.com/help/5g/ug/nr-sounding-reference-signals.html

Sets the number of SRS symbols and the starting symbol index.

```MATLAB
srs = nrSRSConfig;
srs.NumSRSSymbols = 4;  % Number of SRS symbols in a slot
srs.SymbolStart =8; % Allocate SRS in OFDM symbols 10:13
```

--------------------------------

### Initialize and Connect Channel Model

Source: https://www.mathworks.com/help/5g/ug/nr-interference-modeling-with-toroidal-wrap-around.html

Create the system-level channel model and link it to the network simulator and nodes.

```matlab
% Create the channel model
channel = h38901Channel(Scenario="UMa", UseGPU=gpuFlag);

% Connect the channel model to the simulator
addChannelModel(networkSimulator, @channel.channelFunction);

% Connect the gNB and UE nodes to the channel
connectNodes(channel, networkSimulator, scenario);
```

--------------------------------

### Retrieve Cluster Hostname

Source: https://www.mathworks.com/help/5g/ug/use-matlab-parallel-server-and-experiment-manager-to-optimize-csi-feedback-autoencoder-training.html

Get the job scheduler host name from the cluster object.

```matlab
jobSchedulerHost = c.Host;
```

--------------------------------

### Initialize Site Viewer

Source: https://www.mathworks.com/help/5g/ug/5g-nr-beamforming-with-ray-tracing-channels.html

Opens or clears the site viewer window with the specified OpenStreetMap data for the environment.

```matlab
if exist("viewer","var") && isvalid(viewer) % Viewer handle exists and viewer window is open
    clearMap(viewer);
else
    viewer = siteviewer("Basemap","openstreetmap","Buildings","hongkong.osm");    
end
```
Tokens:
Raw