Main Content

Generalize Memory Polynomial Power Amplifier Model Fitting

This example shows you how to:

  • Identify the power amplifier (PA) model coefficients matrix using the characterization waveform with the largest dynamic range and then verify the quality of the fitted model using the other waveforms.

  • Experiment by changing the harmonic order and memory depth of the model. Use grid-search algorithms for optimal model identification.

Determine PA Model Coefficient Matrix From Measured Input/Output Signals

Identify a generalized memory polynomial model, using a memory length of five and a degree of nonlinearity of three.

memLen = 5;
degLen = 3;
modType = 'memPoly';

Load the in, out, fs, ovs, g0, and Tstep variables from the Preprocess Measured Data for PA Model Identification example.


To compute the model coefficient matrix, use a predistorted waveform as it has a larger dynamic range. You later use the coefficient matrix to verify the quality of the fitted model.

fitCoefMat = helperPACharMemPolyModel('coefficientFinder',              ...

Alternatively, you can use a subset of the data range to fit the model

fitCoefMat = helperPACharMemPolyModel('coefficientFinder',            ...
in(14e4:15e4,2), out(14e4:15e4,2),memLen,degLen,modType);

Verify the quality of the PA model by computing the RMS error of the fitting, as well as by plotting the fitted and measured waveforms in the time domain, the power transfer function, the spectrum, and the ACLR spectral measurement.

First, verify the quality of the fitted model by comparing the measurement and the prediction using the original data.

Signal standard deviation = 7.8385%
ACPR data = -36.9763 -37.6772
ACPR fit  = -37.4842 -38.3254

Then, verify the quality of the model using the measured data with the iterative direct digital predistortion (DPD) approach. This data has also been used to identify the PA model.

Signal standard deviation = 3.3434%
ACPR data = -38.0929 -38.5962
ACPR fit  = -38.5555 -38.8375

Finally, save the PA model and relevant parameters for later use.

close all;
save('PA_model', 'Tstep','degLen','memLen','modType','fitCoefMat', 'g0');

Use Grid Search Function to Assess Best Order and Memory Depth

Verify the order and memory depth of the PA model that minimize the time domain RMS error. Use a grid search algorithm that you can parallelize on multicore machines.

rmsErrorFreq = helperPACharGridSearchTime(in(:,2),out(:,2),modType,7);
Starting parallel pool (parpool) using the 'Processes' profile ...
Connected to parallel pool with 4 workers.