Frequency Matching in Spiking Neural Networks for mmWave Sensing

2026-05-11

AI Summary

This work addresses the challenges of deploying efficient perception models on edge devices for millimeter-wave (mmWave) sensing, where signals exhibit sparsity, irregular temporal patterns, and high-frequency noise. Conventional artificial neural networks (ANNs) often require complex preprocessing or deep architectures, hindering their practicality in resource-constrained settings. To overcome this, the authors propose an efficient sensing approach based on spiking neural networks (SNNs), leveraging the low-pass filtering property of leaky integrate-and-fire (LIF) neurons and a spectral matching mechanism. By tuning the membrane potential decay factor, the effective bandwidth of LIF neurons aligns with the discriminative spectral components of mmWave signals, revealing the mechanistic conditions under which SNNs outperform ANNs. Evaluated on four mainstream mmWave datasets, the method achieves an average accuracy improvement of 6.22% and a 3.64× reduction in theoretical energy consumption compared to ANN baselines.

Abstract

Millimeter-wave (mmWave) sensing enables privacy-preserving, always-on edge perception, but its measurements are often sparse, temporally irregular, and corrupted by high-frequency noise. Existing mmWave pipelines predominantly rely on artificial neural networks (ANNs), which achieve robustness through extensive preprocessing or deep architectures, thereby limiting their efficiency on edge devices. In this work, we study spiking neural networks (SNNs) for mmWave sensing from a mechanism-data alignment perspective. By leveraging the low-pass filtering behavior of leaky integrate-and-fire (LIF) dynamics, we analyze how their implicit temporal filtering interacts with the frequency structure of mmWave signals. Our analysis shows that when discriminative information resides in low-to-mid frequencies, LIF dynamics can inherently suppress high-frequency noise, clarifying when and why SNNs outperform ANNs. Based on this insight, we derive a principled criterion for configuring the membrane decay factor by matching the effective bandwidth of LIF dynamics to the data's discriminative spectral content. Experimental results across four widely used mmWave datasets validate the proposed frequency-matching hypothesis, yielding an average test-accuracy improvement of 6.22% and a 3.64$\times$ reduction in theoretical energy consumption relative to ANN baselines, under a unified evaluation protocol.

Problem

• mmWave sensing
• spiking neural networks
• high-frequency noise
• edge perception
• temporal irregularity

Innovation

• Spiking Neural Networks
• Frequency Matching
• mmWave Sensing
• Leaky Integrate-and-Fire
• Energy Efficiency