Enhancing Fourier-based Doppler Resolution with Diffusion Models

TL;DR

This paper introduces a novel AI-based method using diffusion models to enhance Doppler resolution in radar systems, surpassing traditional FFT limitations and improving target separation in range-Doppler maps.

Contribution

The paper presents a new approach combining zero-padded FFT with diffusion neural networks to significantly improve Doppler resolution in radar data.

Findings

Enhanced separation of closely spaced targets

Overcomes traditional FFT resolution limits

Effective in cluttered radar environments

Abstract

In radar systems, high resolution in the Doppler dimension is important for detecting slow-moving targets as it allows for more distinct separation between these targets and clutter, or stationary objects. However, achieving sufficient resolution is constrained by hardware capabilities and physical factors, leading to the development of processing techniques to enhance the resolution after acquisition. In this work, we leverage artificial intelligence to increase the Doppler resolution in range-Doppler maps. Based on a zero-padded FFT, a refinement via the generative neural networks of diffusion models is achieved. We demonstrate that our method overcomes the limitations of traditional FFT, generating data where closely spaced targets are effectively separated.