Main Content

Bluetooth EDR RF-PHY Transmitter Tests for Modulation Accuracy and Carrier Frequency Stability

This example shows you how to perform Bluetooth® enhanced data rate (EDR) radio frequency (RF) physical layer (PHY) transmitter tests specific to modulation accuracy and carrier frequency stability using the Bluetooth® Toolbox. The test measurements compute the initial frequency offset, root mean square (RMS) differential error vector magnitude (DEVM), and peak DEVM values. This example also verifies whether these test measurement values are within the limits specified by the Bluetooth RF-PHY Test Specifications [1].

Objectives of Bluetooth RF-PHY Tests

The Bluetooth RF-PHY Test Specifications [1] defined by the Bluetooth Special Interest Group (SIG) includes RF-PHY tests for the transmitter and receiver. The objectives of these RF-PHY tests are to:

  • Ensure interoperability between all of the Bluetooth devices.

  • Ensure a basic level of system performance for all of the Bluetooth products.

Each test case has a specific test procedure and an expected outcome, that the implementation under test (IUT) must achieve.

RF-PHY Transmitter Tests

The main goal of the transmitter test measurements is to ensure that the transmitter characteristics are within the limits specified by the Bluetooth RF-PHY Test Specifications [1]. This example includes transmitter tests relevant to EDR modulation accuracy and carrier frequency stability. This table shows various RF-PHY transmitter tests that this example performs.

RF-PHY Transmitter Test Procedure

This block diagram summarizes the test procedure for transmitter tests relevant to EDR modulation accuracy and carrier frequency stability of Bluetooth EDR waveforms.

This table shows the packet types and payload lengths required for different PHY modes:

  1. Generate Bluetooth EDR test waveforms by using the bluetoothTestWaveform function.

  2. Add a carrier frequency offset and drift.

  3. Add additive white Gaussian noise (AWGN).

  4. Estimate the initial frequency offset using the basic rate (BR) portion of the waveform.

  5. Compensate the EDR portion with the estimated initial frequency offset.

  6. Perform square root raised cosine filtering using the filter whose coefficients are generated based on the Bluetooth RF-PHY Test Specifications [1].

  7. Divide the EDR portion into blocks of length 50 microseconds each.

  8. For each block, delay the compensated sequence by 1 microsecond and differentiate the delay with actual compensated sequence to get the error sequence.

  9. Compute the RMS DEVM and peak DEVM based on the error sequence and compensated sequence.

  10. Get the test verdict and display the results.

Configure Simulation Parameters

To specify PHY transmission mode, packet type, initial frequency offset, maximum frequency drift, and samples per symbol, set the phyMode, packetType, initialFreqOffset, maxFreqDrift, and sps respectively.

phyMode = 'EDR2M';                    % PHY transmission mode
packetType = "2-DH1";                 % EDR packet type
initFreqOffset = 40000;    % Initial frequency offset in Hz
maxFreqDrift = 0;      % Maximum frequency drift in Hz, must be in the range [-10e3, 10e3]
sps = 8;                                           % Samples per symbol

Generate Test Parameters

Use the preceding configured parameters to generate the test parameters. To get all of the test parameters, use the helperEDRModulationTestConfig.m helper function. To add frequency offset and thermal noise, create and configure comm.PhaseFrequencyOffset and comm.ThermalNoise System objects™, respectively.

[edrTestParams,waveformConfig,filtCoeff] = helperEDRModulationTestConfig(phyMode,packetType,sps);

% Create frequency offset System object
frequencyDelay = comm.PhaseFrequencyOffset('SampleRate',edrTestParams.sampleRate);

% Create thermal noise System object
NF = 12; % Noise figure in dB
thNoise = comm.ThermalNoise('NoiseMethod','Noise figure', ...
                            'SampleRate',edrTestParams.sampleRate, ...
                            'NoiseFigure',NF);

Simulate Transmitter Tests

Using the preceding RF-PHY transmitter test procedure, simulate the transmitter tests.

% Initialize variables 
symDEVM = zeros(1,edrTestParams.requiredBlocks*edrTestParams.blockLength);
[blockRMSDEVM,estimatedBlockFreqDrifts] = deal(zeros(1,edrTestParams.requiredBlocks));
estimatedInitFreqOff = zeros(1,edrTestParams.NumPackets);
blockCount = 0;

% Generate 200 blocks of data as specified in Bluetooth RF-PHY Test Specifications
for packetCount = 1:edrTestParams.NumPackets
    
    % Generate Bluetooth EDR test waveform
    txWaveform = bluetoothTestWaveform(waveformConfig);
    
    % Generate ideal EDR symbols from waveform
    idealTxEDRWaveform = txWaveform(edrTestParams.startIndex*sps+1:end);
    
    % Perform matched filtering
    rxFilt = upfirdn(idealTxEDRWaveform,filtCoeff,1,sps);
    
    % Remove delay and normalize filtered signal
    idealEDRSymbols = rxFilt(edrTestParams.span+1:end,1)/sqrt(sps);
    
    % Add frequency offset    
    driftRate = maxFreqDrift/length(txWaveform); % Drift rate
    freqDrift = driftRate*(0:1:(length(txWaveform)-1))'; % Frequency drift for the packet
    frequencyDelay.FrequencyOffset = freqDrift + initFreqOffset; % Frequency offset, includes initial frequency offset and drift
    transWaveformCFO = frequencyDelay(txWaveform(1:length(txWaveform)));
    
    % Add thermal noise
    noisyWaveform = thNoise(transWaveformCFO);
            
    % Compute initial frequency offset specified in Bluetooth RF-PHY Test Specifications
    estimatedInitFreqOff(packetCount) = helperEstimateInitialFreqOffset(noisyWaveform,sps);
    
    % Compensate initial frequency offset in the received waveform
    pfOffset = comm.PhaseFrequencyOffset('SampleRate',edrTestParams.sampleRate,'FrequencyOffset',-estimatedInitFreqOff(packetCount));
    freqTimeSyncRcv = pfOffset(noisyWaveform);

    % Remove access code, packet header, and guard time from packet
    rxEDRWaveform = freqTimeSyncRcv((edrTestParams.startIndex)*sps+1:end);
    
    % Perform matched filtering
    rxFilt = upfirdn(rxEDRWaveform,filtCoeff,1,sps);
    receivedEDRSymbols = rxFilt(edrTestParams.span+1:end,1)/sqrt(sps);
    
    % Compute DEVM values
    [rmsDEVM,rmsDEVMSymbol,samplingFreq] = ...
        helperEDRModulationTestMeasurements(receivedEDRSymbols,idealEDRSymbols,edrTestParams);
    
    % Accumulate measured values for 200 blocks as specified in Bluetooth RF-PHY Test Specifications 
    blockCount = blockCount + edrTestParams.numDEVMBlocks;
    symDEVM(((packetCount-1)*edrTestParams.numDEVMBlocks*edrTestParams.blockLength)+1:(packetCount)*edrTestParams.numDEVMBlocks ...
        *edrTestParams.blockLength) = rmsDEVMSymbol(1:edrTestParams.numDEVMBlocks*edrTestParams.blockLength);
    blockRMSDEVM(((packetCount-1)*edrTestParams.numDEVMBlocks)+1:((packetCount)*edrTestParams.numDEVMBlocks)) = ...
        rmsDEVM(1:edrTestParams.numDEVMBlocks);  
    estimatedBlockFreqDrifts(((packetCount-1)*edrTestParams.numDEVMBlocks)+1:((packetCount)*edrTestParams.numDEVMBlocks)) = ...
        samplingFreq(1:edrTestParams.numDEVMBlocks); 
end

Use the helperEDRModulationTestVerdict.m helper function to verify whether the measurements are within the specified limits and display the verdict.

helperEDRModulationTestVerdict(phyMode, ...
    edrTestParams,estimatedInitFreqOff,symDEVM,blockRMSDEVM,estimatedBlockFreqDrifts)
Modulation Accuracy Test Results: 

       Expected peak DEVM for all pi/4-DQPSK symbols is less than or equal to 0.35
       Result: Pass
       Percentage of pi/4-DQPSK symbols with DEVM less than or equal to 0.3 is 100
       Expected percentage of pi/4-DQPSK symbols with DEVM less than or equal to 0.3 is 99 % 
       Result: Pass

       Expected RMS DEVM for all pi/4-DQPSK blocks is less than or equal to 0.2
       Result: Pass
Carrier Frequency Stability Test Results: 
       Expected initial frequency offset range: [-75 kHz, 75 kHz]
       Do estimated initial frequency offsets for all the packets fall under expected values?
       Result: Yes
       Expected sampling frequencies range: [-10 kHz, 10 kHz]
       Do estimated sampling frequencies for all the blocks fall under expected values?
       Result: Yes
% Plot the constellation diagram
if strcmp(phyMode,'EDR2M')
    refSymbols = dpskmod(0:edrTestParams.M-1,edrTestParams.M,pi/4,'gray'); % Perform pi/4-DQPSK modulation
else
    refSymbols = dpskmod(0:edrTestParams.M-1,edrTestParams.M,0,'gray'); % Perform 8-DPSK modulation
end
constDiag = comm.ConstellationDiagram('ReferenceConstellation',refSymbols, ...
    'Title','Received EDR Constellation');
constDiag(receivedEDRSymbols);
release(constDiag);

This example demonstrates the Bluetooth EDR transmitter test measurements specific to modulation accuracy and carrier frequency stability. The simulation results verify that these computed test measurement values are within the limits specified by the Bluetooth RF-PHY Test Specifications [1].

Appendix

The example uses these helpers:

Selected Bibliography

1 - Bluetooth Special Interest Group (SIG). “Bluetooth RF-PHY Test Specification”, v1.2/2.0/2.0, EDR/2.1/2.1, EDR/3.0/3.0, HS (), RF.TS/3.0.H.1, Section 4.5. 2009. https://www.bluetooth.com

2 - Bluetooth Special Interest Group (SIG). "Core System Package [BR/EDR Controller Volume]". Bluetooth Core Specification. Version 5.3, Volume 2. https://www.bluetooth.com

See Also

Functions

Objects

Related Topics