Adaptive-Frequency Resonate-and-Fire Neurons for Spectral Estimation of Streaming Radar Signals

2026-06-11

AI Summary

This work addresses the high memory consumption and latency of conventional FMCW radar systems that rely on FFT-based spectral estimation, which hinders their deployment in resource-constrained environments. The authors propose a neuromorphic-inspired Adaptive Resonance Firing (ARF) neuron approach formulated as a discrete-time dynamical system, enabling streaming sample processing to dynamically tune internal frequencies and directly estimate target range and velocity. By incorporating a multi-neuron feedback mechanism, the method allows parallel locking onto distinct frequency components with memory usage scaling only with the number of targets. Experimental results on both simulated and real-world data demonstrate effective multi-target tracking with substantially reduced memory overhead, thereby transcending traditional spectral computation paradigms and offering a viable solution for edge and low-power radar applications.

Abstract

Frequency Modulated Continuous Wave (FMCW) radar systems traditionally rely on Fourier-based methods, such as the Fast Fourier Transform (FFT), to estimate target range and velocity. While computationally efficient, these approaches require storing and processing large blocks of data, which can become a bottleneck in memory-constrained or low-latency applications. In this work, we propose a neuromorphic-inspired signal processing method based on adaptive resonate-and-fire (ARF) neurons formulated as a discrete-time dynamical system. Each neuron dynamically adjusts its internal frequency to match dominant frequency components of the input radar signal, enabling direct estimation of target ranges and velocities without computing the full frequency spectrum. The proposed model operates in a sample-by-sample fashion, resulting in memory requirements that scale with the number of tracked targets rather than the signal length. A feedback mechanism is also introduced to enable multiple neurons to lock on distinct frequency components in multi-target cases. Results on simulated and experimental data demonstrate that the method can successfully track multiple targets. Compared to conventional FFT-based approaches, the proposed method offers reduced memory usage proportional only to the number of tracked targets, making it suitable for resource-constrained and edge-based radar applications.

Problem

• FMCW radar
• spectral estimation
• memory-constrained
• low-latency
• streaming signals

Innovation

• neuromorphic computing
• adaptive resonate-and-fire neuron
• FMCW radar
• spectral estimation
• low-latency signal processing