Wi-Spike: A Low-power WiFi Human Multi-action Recognition Model with Spiking Neural Networks

2026-03-15

AI Summary

This work proposes Wi-Spike, the first low-power framework to integrate spiking neural networks (SNNs) into WiFi-based multi-action recognition, addressing the high energy consumption of existing methods that hinders deployment on resource-constrained edge devices. Wi-Spike employs spiking convolutional layers to extract spatiotemporal features, incorporates a temporal attention mechanism to enhance discriminative representations, and utilizes spiking fully connected layers combined with a voting layer for efficient classification. Experimental results on three benchmark datasets—NTU-Fi-HAR, NTU-Fi-HumanID, and UT-HAR—demonstrate that Wi-Spike achieves 95.83% accuracy in single-action recognition and outperforms state-of-the-art approaches in multi-action scenarios, while reducing energy consumption by at least 50%. This advancement establishes a new benchmark for low-power edge intelligence in wireless sensing.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Learning on the Edge & Model Compression
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

WiFi-based human action recognition (HAR) has gained significant attention due to its non-intrusive and privacy-preserving nature. However, most existing WiFi sensing models predominantly focus on improving recognition accuracy, while issues of power consumption and energy efficiency remain insufficiently discussed. In this work, we present Wi-Spike, a bio-inspired spiking neural network (SNN) framework for efficient and accurate action recognition using WiFi channel state information (CSI) signals. Specifically, leveraging the event-driven and low-power characteristics of SNNs, Wi-Spike introduces spiking convolutional layers for spatio-temporal feature extraction and a novel temporal attention mechanism to enhance discriminative representation. The extracted features are subsequently encoded and classified through spiking fully connected layers and a voting layer. Comprehensive experiments on three benchmark datasets (NTU-Fi-HAR, NTU-Fi-HumanID, and UT-HAR) demonstrate that Wi-Spike achieves competitive accuracy in single-action recognition and superior performance in multi-action recognition tasks. As for energy consumption, Wi-Spike reduces the energy cost by at least half compared with other methods, while still achieving 95.83% recognition accuracy in human activity recognition. More importantly, Wi-Spike establishes a new state-of-the-art in WiFi-based multi-action HAR, offering a promising solution for real-time, energy-efficient edge sensing applications.

Problem

• WiFi-based human action recognition
• energy efficiency
• low-power sensing
• multi-action recognition
• spiking neural networks

Innovation

• Spiking Neural Networks
• WiFi-based HAR
• Low-power Sensing
• Temporal Attention
• Multi-action Recognition