TinySense: Effective CSI Compression for Scalable and Accurate Wi-Fi Sensing

TL;DR

TinySense tackles the high data-rate challenge of Wi-Fi sensing by introducing a VQGAN-based CSI compression framework that partitions processing between edge devices and the cloud. It employs a dynamically resizeable codebook via K-means and a second-stage Transformer to recover lost VQ indices, achieving ultra-low bitrate transmission while preserving HPE accuracy. The framework is trained with a composite objective that combines reconstruction, adversarial, and keypoint losses, enabling robust sensing performance under network constraints. Experimental validation on MM-Fi and Wi-Pose, plus a real-world Raspberry Pi/Jetson Nano testbed, shows up to 1.5x gains in PCK20 at the same bitrate and up to 5x latency and 2.5x bandwidth reductions, highlighting TinySense’s potential for scalable, privacy-preserving Wi-Fi sensing in edge environments.

Abstract

With the growing demand for device-free and privacy-preserving sensing solutions, Wi-Fi sensing has emerged as a promising approach for human pose estimation (HPE). However, existing methods often process vast amounts of channel state information (CSI) data directly, ultimately straining networking resources. This paper introduces TinySense, an efficient compression framework that enhances the scalability of Wi-Fi-based human sensing. Our approach is based on a new vector quantization-based generative adversarial network (VQGAN). Specifically, by leveraging a VQGAN-learned codebook, TinySense significantly reduces CSI data while maintaining the accuracy required for reliable HPE. To optimize compression, we employ the K-means algorithm to dynamically adjust compression bitrates to cluster a large-scale pre-trained codebook into smaller subsets. Furthermore, a Transformer model is incorporated to mitigate bitrate loss, enhancing robustness in unreliable networking conditions. We prototype TinySense on an experimental testbed using Jetson Nano and Raspberry Pi to measure latency and network resource use. Extensive results demonstrate that TinySense significantly outperforms state-of-the-art compression schemes, achieving up to 1.5x higher HPE accuracy score (PCK20) under the same compression rate. It also reduces latency and networking overhead, respectively, by up to 5x and 2.5x. The code repository is available online at here.

Figures (8)

• Figure 1: Overview of Cloud-based Wi-Fi Sensing.
• Figure 2: Illustration of the proposed $\mathtt{TinySense}$ framework: (a) the overall compression pipeline, (b) the K-means clustering algorithm to adjust the codebook size for variable bitrates, and (c) a transformer approach for predicting lost VQ indices.
• Figure 3: The generative transformer process deployed in the second stage.
• Figure 4: Experimental setup for performance evaluation to evaluate the resource consumption.
• Figure 5: T-SNE visualizations of CSI representations: (a) raw CSI with action labels, (b) quantized features with action labels, (c) raw CSI with subject labels, and (d) quantized features with subject labels.