Neural-HAR: A Dimension-Gated CNN Accelerator for Real-Time Radar Human Activity Recognition

2025-10-26

AI Summary

To address excessive computational and memory overhead when deploying radar-based human activity recognition (HAR) on resource-constrained edge devices, this paper proposes GateCNN—a lightweight neural network architecture. GateCNN integrates Doppler-time dual-path gated convolutions, residual connections, and a dimension-wise gating mechanism to preserve discriminative capability under ultra-low complexity. Furthermore, it incorporates Doppler vector embedding and co-designs the model with FPGA hardware, enabling efficient acceleration on the Xilinx Zynq-7000 platform without utilizing DSP slices or BRAM. Experimental evaluation on the UoG2020 dataset achieves 86.4% classification accuracy using only 2.7K parameters and 0.28M FLOPs. FPGA implementation yields a measured inference latency of 107.5 μs and dynamic power consumption of 15 mW—significantly outperforming existing CNN- and RNN-based approaches in both efficiency and resource utilization.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments

Abstract

Radar-based human activity recognition (HAR) is attractive for unobtrusive and privacy-preserving monitoring, yet many CNN/RNN solutions remain too heavy for edge deployment, and even lightweight ViT/SSM variants often exceed practical compute and memory budgets. We introduce Neural-HAR, a dimension-gated CNN accelerator tailored for real-time radar HAR on resource-constrained platforms. At its core is GateCNN, a parameter-efficient Doppler-temporal network that (i) embeds Doppler vectors to emphasize frequency evolution over time and (ii) applies dual-path gated convolutions that modulate Doppler-aware content features with temporal gates, complemented by a residual path for stable training. On the University of Glasgow UoG2020 continuous radar dataset, GateCNN attains 86.4% accuracy with only 2.7k parameters and 0.28M FLOPs per inference, comparable to CNN-BiGRU at a fraction of the complexity. Our FPGA prototype on Xilinx Zynq-7000 Z-7007S reaches 107.5 $μ$s latency and 15 mW dynamic power using LUT-based ROM and distributed RAM only (zero DSP/BRAM), demonstrating real-time, energy-efficient edge inference. Code and HLS conversion scripts are available at https://github.com/lab-emi/AIRHAR.

Problem

• Enables real-time radar activity recognition on edge devices
• Reduces computational complexity for resource-constrained platforms
• Achieves energy-efficient inference without DSP or BRAM usage

Innovation

• Dimension-gated CNN accelerator for real-time radar HAR
• GateCNN uses dual-path gated convolutions with residual paths
• FPGA prototype achieves microsecond latency with milliwatt power