DPD and PA Modeling

AI for digital predistortion and power amplifier modeling

Model a power amplifier (PA) using a neural network. To offset the effects of nonlinearities in a PA, you can use a neural network to apply digital predistortion (DPD) to the input signal.For a description of the workflow, see AI for Digital Predistortion Design (Communications Toolbox).

NN-DPD deployed in a communications system

Topics

AI for Digital Predistortion Design (Communications Toolbox)

Example workflows for training, compressing, and using a deep learning network for digital predistortion design. (Since R2024a)
- STEP 1: Data Preparation for Neural Network Digital Predistortion Design (Communications Toolbox)
- STEP 2: Neural Network for Digital Predistortion Design-Offline Training (Communications Toolbox)
- STEP 3: Neural Network for Digital Predistortion Design - Online Training (Communications Toolbox)
- STEP 4: Structurally Compress Neural Network DPD Using Projection (Communications Toolbox)
- STEP 5: HDL Neural Network Design for Digital Predistorter (Wireless HDL Toolbox)
- STEP 6: Deploy and Verify HDL Neural Network Based DPD on FPGA (SoC Blockset)
Power Amplifier Modeling Using Neural Networks (Communications Toolbox)
Model a power amplifier (PA) using several different neural network (NN) architectures. (Since R2024a)

Featured Examples

New

Compare Residual Recurrent Neural Network Structures for Digital Predistortion Design

Compare residual RNN structures for a digital predistortion (DPD) design.

(Communications Toolbox)

Since R2024b

Data Preparation for Neural Network Digital Predistortion Design

Prepare data for training a neural network digital predistortion (DPD) system. The DPD is used to offset the effects of nonlinearities in a power amplifier (PA).

(Communications Toolbox)

Since R2024a

Neural Network for Digital Predistortion Design-Offline Training

Use a neural network, that is trained offline, to apply digital predistortion to offset the effects of nonlinearities in a power amplifier.

(Communications Toolbox)

Neural Network for Digital Predistortion Design - Online Training

Use a neural network, that is trained online, to apply digital predistortion to offset the effects of nonlinearities in a power amplifier.

(Communications Toolbox)

Structurally Compress Neural Network DPD Using Projection

Structurally compress a neural network DPD to reduce computational complexity and memory requirements using projection and principal component analysis.

(Communications Toolbox)

Since R2024a

Power Amplifier Modeling Using Neural Networks

Model a power amplifier (PA) using several different neural network (NN) architectures.

(Communications Toolbox)

Since R2024a

HDL Neural Network Design for Digital Predistorter

Use neural-network-based (NN-based) digital predistorter (DPD) to compensate effects of nonlinearities in power amplifier.

(Wireless HDL Toolbox)

Deploy and Verify HDL Neural Network Based DPD on FPGA

Deploy neural-network-based digital predistorter (NN DPD) algorithm and verify using OFDM signal.

(SoC Blockset)