Skip to content

Raj Bhattacharjea Twitter Profile LinkedIn Profile ResearchGate Profile Google Scholar Profile StackExchange Profile

Picture of Raj Bhattacharjea I am an Electrical Engineer working as the Director of Machine Learning for DeepSig Inc. I'm also an Adjunct Principal Research Engineer with the School of Electrical and Computer Engineering at Georgia Tech. My goals include conducting, publishing, and productizing research in machine learning for communication and sensing systems. You'll find a bit about me here, and once in a blue moon I'll post other content about topics I'm interested in.


Research Interests

The common threads that tie together my work are:

  • Electromagnetics / Computational EM
  • Signal processing
  • Machine learning

At various times I've been interested in applying these techniques to the following domains:

  • Commercial wireless communications
  • Free space optical communications
  • Fiber optic communications
  • Photonic signal processing
  • Near-field inductive communications
  • Backscatter communications

Courses Taught (Georgia Tech)

ECE3025 Electromagnetics, Spring 2019

ECE2026 Intro to Digital Signal Processing, Summer 2018 and Fall 2018

ECE4012 Senior Capstone Design, Spring 2016


Pre-Prints, Papers, and Presentations

Last updated February 2025.

Pre-Print: 3D Geometry Extraction for Sionna RT and Sensitivity Analysis

OpenGERT: Open Source Automated Geometry Extraction with Geometric and Electromagnetic Sensitivity Analyses for Ray-Tracing Propagation Models

Site-Specific Propagation Measurements Improve Learned 5G Processing Algorithms

How Critical is Site-Specific RAN Optimization? 5G Open-RAN Uplink Air Interface Performance Test and Optimization from Macro-Cell CIR Data

An arXiv preprint is available for those without IEEE Xplore access

Pre-Print: Encoder-Decoder Networks for Signal Bandwidth Estimation on Arrays

This one was accepted for publication at IEEE Array 2022 and my co-author, Nathan West, presented results of this work at the conference. However, we missed the deadline for submitting the final camera-ready files for archival on IEEE Xplore (oops). So, here's the arXiv version. Encoder-Decoder Networks for Self-Supervised Pretraining and Downstream Signal Bandwidth Regression on Digital Antenna Arrays

Paper: Feedback Magnitude Pruning

Neural network compression with feedback magnitude pruning for automatic modulation classification

Paper: Modulation Pattern Detection Using Complex Convolutions

Modulation Pattern Detection Using Complex Convolutions in Deep Learning

Pre-Print: High-Capacity Complex Convolutional Neural Networks Modulation Classification

Pre-Print: High-Capacity Complex Convolutional Neural Networks For I/Q Modulation Classification

Paper: Complex-Valued Convolutions for Modulation Recognition

Complex-Valued Convolutions for Modulation Recognition using Deep Learning

Presentation: Photonic ICs for Channelization and Downconversion

Presentation: Photonic ICs for Channelization and Downconversion

Paper: Photonic ICs for Channelization and Downconversion

Paper: Photonic ICs for Channelization and Downconversion

Presentation: Army Signal Classification Challenge

Presentation: Army Signal Classification Challenge

Paper: ReMoRa

Paper: ReMoRa

Paper: Perfect Pulses

Paper: Perfect Pulses

Presentation: Accelerated Signal Processing on Embedded Platforms

Presentation: Accelerated Signal Processing on Embedded Platforms

Presentation: Open Source SDR on Embedded Platforms

Presentation: Open Source SDR on Embedded Platforms

Dissertation: Refractive Near-Earth UHF Channels

Dissertation: Refractive Near-Earth UHF Channels

Paper: Near-Ground Vertical Dipoles in Atmospheric Refractivity

Paper: Near-Ground Vertical Dipoles in Atmospheric Refractivity

Presentation: Near-Ground Vertical Dipoles in Atmospheric Refractivity

Presentation: Near-Ground Vertical Dipoles in Atmospheric Refractivity

Paper: Rician Fading Estimation

Paper: Rician Fading Estimation