Practical Wi-Fi-based Motion Recognition Under Variable Traffic Patterns

2026-05-08

AI Summary

This study addresses the performance degradation of Wi-Fi sensing under variable network traffic, where fluctuating sampling rates and intervals hinder recognition accuracy and limit the generalizability of existing methods. To tackle this challenge, the work introduces, for the first time, sampling rate versatility into Wi-Fi sensing and proposes a Transformer-based Sampling Rate Versatile Neural Network (SRV-NN). This architecture, combined with a dynamic sampling rate data augmentation strategy, enables robust activity and gesture recognition from Channel State Information (CSI) inputs with varying lengths and sampling rates. Extensive experiments on a newly curated SRV dataset and two public benchmarks demonstrate that the proposed method significantly outperforms current baselines, achieving higher average accuracy and substantially reduced performance variance across diverse sampling rates.

Abstract

Wi-Fi sensing detects human motions and activities by analysing the channel state information (CSI) derived from Wi-Fi transmissions. However, the impact of variable transmission traffic, which dictates the effective sampling rate and interval, is often overlooked. Existing Wi-Fi sensing systems are trained with fixed input size and sampling rate, which suffer from poor sampling rate generalisation. This paper proposes a novel Wi-Fi sensing approach for motion recognition applications, e.g., gesture and activity recognition, under variable traffic patterns. A sampling rate versatile neural network (SRV-NN) based on the transformer is proposed to efficiently handle variable input-sized sensing signals. A dynamic sampling rate augmentation is employed for variable sampling rates and intervals. To validate our approach, we have carried out extensive experimental evaluation, using two self-collected datasets, namely SRV activity and SRV gesture, as well as two publicly available datasets. Our method demonstrated exceptional performance and stability under variable sampling rates, with substantial improvements in average accuracy compared to baseline models without augmentation. The proposed approach significantly enhances stability by greatly reducing accuracy variance across different sampling rates.

Problem

• Wi-Fi sensing
• motion recognition
• variable traffic patterns
• sampling rate generalisation
• channel state information

Innovation

• Wi-Fi sensing
• variable sampling rate
• transformer-based neural network
• motion recognition
• data augmentation