Shahab Hamidi-Rad | AI/ML for Wireless Communications

Principal Engineer · AI/ML Researcher · Wireless Communications

I am a Principal Engineer at InterDigital, where I work on AI/ML solutions for next-generation wireless communication systems, including 3GPP and 6G-related research. My interests span physical-layer simulation, AI-native air interfaces, neural network compression, channel estimation, signal intelligence, and quantum approaches to communication systems.

Featured current project

Research-grade wireless simulation in Python.

NeoRadium is a Python library for 3GPP 5G NR physical-layer simulation. It simplifies the simulation of physical-layer communication pipelines based on the 3GPP 5G NR standard, with an object-oriented architecture that abstracts the complexity of different stages of the pipeline and enables rapid end-to-end simulations on standard computers.

Developed from scratch as an open-source research tool, NeoRadium provides a practical foundation for generating data, prototyping algorithms, and testing receiver-side AI/ML ideas in a realistic wireless communication context.

GitHub Documentation Playground

Animated NeoRadium beam sweeping trajectory example for a ray-tracing wireless simulation

Go to this example

3GPP 5G NR Physical layer End-to-end simulation AI/ML wireless research Jupyter workflows

Research areas

AI/ML for wireless systems

Machine learning methods and experiments for communication pipelines, receiver design, and next-generation 3GPP systems including 6G.

Signal intelligence

Modulation classification, channel estimation, CSI feedback, beamforming, and wireless signal understanding.

Efficient AI

Neural network compression and deployment-aware model design for practical communication systems.

Quantum communications research

Exploring quantum-assisted approaches to channel decoding and communication problems.

Publications

2026

Physics-Based Channel Transformation for Wireless Configurations

Physics-informed transformation of wireless channel data to reduce the cost of adapting ML models across wireless configurations.

2026

A Self-Refining Multi-Layer Receiver Pipeline

A receiver pipeline that refines channel estimation and equalization to improve throughput under practical 5G NR resource tradeoffs.

2025

A Dataset Similarity Evaluation Framework for Wireless Communications and Sensing

A model-agnostic framework for evaluating dataset similarity and realism in wireless communications and sensing scenarios.

2023

A Deep Learning Method for Joint Compression and Unsupervised Denoising of CSI Feedback

Deep learning for massive-MIMO CSI feedback that jointly compresses and denoises channel information using noisy observations.

2022

Quantum Channel Decoding

Explores quantum computing approaches to channel decoding for communication systems.

2021

MCformer: A Transformer Based Deep Neural Network for Automatic Modulation Classification

A transformer-based neural architecture for automatic modulation classification of complex-valued wireless signals.

2021

Low Rank Based End-to-End Deep Neural Network Compression

A low-rank framework for compressing deep neural networks for computationally constrained deployment.

View all papers on Google Scholar

Patents

Wireless systems

Methods for improving throughput in wireless systems

Methods for improving wireless throughput by adapting receiver-side processing and resource usage in communication systems.

Channel state information

Method and system for determining residual channel state information based on estimated and predicted values of channel state information

Determines residual CSI from estimated and predicted CSI values to support more efficient wireless channel feedback and reconstruction.

Neural network compression

Contact

I welcome thoughtful messages related to AI/ML for wireless systems, NeoRadium, research collaboration, publications, patents, and professional opportunities.

Send a Message Resume PDF LinkedIn Google Scholar