Wireless Communications

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).

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

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

Train an autoencoder as an end-to-end simulation system.

Generate GPU specific C++ code for a pretrained TensorFlow™ channel state feedback autoencoder.

Model an end-to-end orthogonal frequency division modulation (OFDM) communications system with an autoencoder to reliably transmit information bits over a wireless channel.

Compress downlink channel state information (CSI) for 5G systems by using an autoencoder neural network.

Train a deep Q-network (DQN) reinforcement learning agent for beam selection in a 5G new radio communications system.

Use a neural network to reduce overhead in the beam selection task by using the location of the receiver rather than knowledge of the communication channels.

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

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

Train a semantic segmentation network using deep learning for spectrum monitoring. One of the uses of spectrum monitoring is to characterize spectrum occupancy. The neural network in this example is trained to identify 5G NR and LTE signals in a wideband spectrogram.

Model an end-to-end communications system with an autoencoder to reliably transmit information bits over a wireless channel.

Use a convolutional neural network (CNN) for modulation classification. You generate synthetic, channel-impaired waveforms. Using the generated waveforms as training data, you train a CNN for modulation classification. You then test the CNN with software-defined radio (SDR) hardware and over-the-air signals.

Deploy a pretrained convolutional neural network (CNN) for modulation classification to the Xilinx^® Zynq^® UltraScale+™ MPSoC ZCU102 Evaluation Kit.

Generate signals and channel impairments to train a neural network, called LLRNet, to estimate exact log likelihood ratios (LLR).

Design a radio frequency (RF) fingerprinting convolutional neural network (CNN) with simulated data. You train the CNN with simulated wireless local area network (WLAN) beacon frames from known and unknown routers for RF fingerprinting. You then compare the media access control (MAC) address of received signals and the RF fingerprint detected by the CNN to detect WLAN router impersonators.

Train a radio frequency (RF) fingerprinting convolutional neural network (CNN) with captured data. You capture wireless local area network (WLAN) beacon frames from real routers using a software defined radio (SDR). You program a second SDR to transmit unknown beacon frames and capture them. You train the CNN using these captured signals. You then program a software-defined radio (SDR) as a router impersonator that transmits beacon signals with the media access control (MAC) address of one of the known routers and use the CNN to identify it as an impersonator.