EfficientFi: Towards Large-Scale Lightweight WiFi Sensing via CSI Compression

Jianfei Yang; Xinyan Chen; Han Zou; Dazhuo Wang; Qianwen Xu; Lihua Xie

Source-linked AI summary

EfficientFi: Towards Large-Scale Lightweight WiFi Sensing via CSI Compression

Jianfei Yang, Xinyan Chen, Han Zou, Dazhuo Wang, Qianwen Xu, Lihua Xie

arXiv:2204.04138v2Apr 8, 2022cs.NIcs.AIcs.HC

TL;DR

Large-scale WiFi sensing is difficult because edge devices have limited computation while high-dimensional CSI creates substantial transmission demands. EfficientFi uses edge-cloud quantized representation learning to compress CSI, restore it, and perform recognition jointly. Experiments across activity recognition and person identification show substantial communication reduction while preserving recognition performance.

Problem
Existing WiFi sensing systems mainly target local or single-user settings, while edge computation limits and high-dimensional CSI transmission impede large-scale deployment.
Method
EfficientFi jointly learns a quantized compressed CSI representation, cloud-side restoration, and recognition across edge and cloud components.
Results
EfficientFi reduces communication burdens by over 1,700 times without degrading recognition performance across WiFi activity-recognition and human-identification experiments.
Takeaways & Limitations
The framework supports large-scale WiFi sensing by preserving classification performance after feature compression and transmission.
Takeaways & Limitations
The supported scope is large-scale WiFi sensing based on CSI, evaluated through human activity recognition and person identification applications.

Abstract

from arXiv · show

WiFi technology has been applied to various places due to the increasing requirement of high-speed Internet access. Recently, besides network services, WiFi sensing is appealing in smart homes since it is device-free, cost-effective and privacy-preserving. Though numerous WiFi sensing methods have been developed, most of them only consider single smart home scenario. Without the connection of powerful cloud server and massive users, large-scale WiFi sensing is still difficult. In this paper, we firstly analyze and summarize these obstacles, and propose an efficient large-scale WiFi sensing framework, namely EfficientFi. The EfficientFi works with edge computing at WiFi APs and cloud computing at center servers. It consists of a novel deep neural network that can compress fine-grained WiFi Channel State Information (CSI) at edge, restore CSI at cloud, and perform sensing tasks simultaneously. A quantized auto-encoder and a joint classifier are designed to achieve these goals in an end-to-end fashion. To the best of our knowledge, the EfficientFi is the first IoT-cloud-enabled WiFi sensing framework that significantly reduces communication overhead while realizing sensing tasks accurately. We utilized human activity recognition and identification via WiFi sensing as two case studies, and conduct extensive experiments to evaluate the EfficientFi. The results show that it compresses CSI data from 1.368Mb/s to 0.768Kb/s with extremely low error of data reconstruction and achieves over 98% accuracy for human activity recognition.

I. INTRODUCTION

Large-scale WiFi sensing is constrained by edge computation and CSI transmission demands, while cloud processing also requires reconstructible data. EfficientFi addresses these constraints by jointly compressing, restoring, and recognizing CSI across edge and cloud components.

Motivation: CSI sensing streams have high dimensionality and sampling rates, creating communication burdens that can hinder basic WiFi Internet access.The paper identifies compression as necessary for transmitting large-scale sensing data from users to cloud servers.
Motivation: Cloud-based sensing additionally requires both accurate recognition and CSI reconstruction for IoT data logging and incremental model learning.These requirements extend beyond simply transmitting compressed CSI for recognition.
EfficientFi Framework: EfficientFi provides CSI compression, CSI restoration, and CSI-based recognition to bridge single-user sensing and large-scale WiFi sensing.The framework assigns feature extraction and compression to WiFi routers, while cloud servers perform recognition and reconstruction.
EfficientFi Framework: EfficientFi uses a quantized low-dimensional CSI representation and joint learning to make compressed features discriminative for recognition.Its multi-task, end-to-end design jointly supports compression, restoration, and recognition.
Evaluation: Over 1,700× communication-burden reduction was achieved without degrading recognition performance in WiFi activity-recognition and human-identification experiments.The experiments evaluate the framework in real-world WiFi sensing settings.

II. RELATED WORK

Related WiFi sensing work spans applications, model-based and learning-based recognition, and deep learning architectures. Deep models learn more complex CSI patterns and improve sensing granularity, but require substantial data and computation, motivating EfficientFi’s compression approach.

WiFi Sensing Systems and Applications: WiFi sensing systems use CSI for applications including occupancy detection, crowd counting, activity recognition, respiration monitoring, person identification, and gesture recognition.CSI can be extracted from WiFi hardware at increasing subcarrier resolutions and directly at access points.
Recognition Models: Model-based methods use propagation and signal-processing theories, whereas learning-based methods learn statistical or deep-neural-network patterns from data.The paper focuses more directly on learning-based recognition models.
Deep Learning Models for WiFi Sensing: Deep learning models capture more complicated WiFi sensing patterns and improve recognition granularity, but require massive data collection and greater computation overhead.These requirements commonly motivate use of cloud servers for WiFi sensing.
Deep Learning Models for WiFi Sensing: Prior deep-learning approaches apply temporal, spatial, recurrent, convolutional, and Siamese architectures to activity recognition, gesture recognition, human identification, and localization.EfficientFi combines deep CSI compression, quantization, and recognition within a large-scale sensing framework.

C. CSI Compression in Communication

CSI compression for communication reduces feedback overhead by encoding channel information into a lower-dimensional representation while preserving signal characteristics needed for recovery. Learned autoencoder methods improve reconstruction beyond handcrafted priors, but the communication setting differs from EfficientFi’s sensing scenario.

Traditional CSI feedback methods use compressive sensing or LASSO, while learned autoencoders improve recovery quality from training samples.
CSI records amplitude attenuation and phase shifts of multipath components for each antenna pair and OFDM subcarrier.
WiFi sensing can compress only task-relevant CSI components, such as subcarrier amplitudes for human activity recognition, rather than the whole matrix.
EfficientFi extends low-dimensional CSI compression by requiring the compressed representation to be discrete for communication and cloud restoration.

B. WiFi Sensing and Limitations

WiFi sensing can support large-scale applications because CSI captures activity-related propagation changes, but its deployment must bridge edge constraints and cloud processing. EfficientFi addresses this gap with a discriminative compressed feature space and distributed processing.

WiFi Sensing: Human activities alter wireless-signal propagation, producing CSI patterns that learning or analytic methods can use for recognition.
WiFi Sensing: EfficientFi bridges edge and cloud computing by learning a discriminative compressed CSI feature space for large-scale WiFi sensing.
System Overview: At the WiFi AP, a lightweight CNN extracts features and quantizes them by nearest-vector lookup in a CSI codebook.
System Overview: EfficientFi defines a feature extractor, decoder, and task classifier, jointly optimizing extraction and classification while reconstructing the original CSI sample.

B. Discrete Representation Learning for Efficient Communication

EfficientFi uses discrete feature learning to make CSI compact, reconstructable, and useful for recognition. Its end-to-end objectives address codebook learning, non-differentiable quantization, reconstruction, and task discrimination.

Quantization: Each continuous encoder feature is assigned to its nearest CSI codebook vector, producing a one-hot quantized representation.
Restoration: At the cloud server, the quantized representation is reversed into a reconstructed feature and decoded into CSI for database storage.
Recognition: The classifier uses reconstructed CSI features to predict task labels such as human activity categories.
Optimization Challenges: Training must learn a comprehensive codebook and overcome the non-differentiability of quantized feature generation.
Optimization: VQ-VAE addresses these difficulties by learning codebook parameters with back-propagation and using straight-through gradient estimation for end-to-end training.
Learning Objectives: The reconstruction objective trains encoder and decoder parameters through mean squared error, with a straight-through estimator linking continuous and discrete features.
Learning Objectives: The codebook objective minimizes the distance between continuous encoder outputs and quantized features using vector quantization.
Learning Objectives: Joint feature-extractor and classifier training makes the compressed feature space discriminative for WiFi sensing tasks.

E. Algorithm Summary

EfficientFi trains its extractor, decoder, classifier, and codebook offline, then deploys the extractor at WiFi APs and the decoder and classifier at the cloud. Its design combines codebook learning and multi-task feature learning for large-scale sensing.

Training: EfficientFi’s overall loss comprises separate optimization processes for the model parameters and codebook.
Deployment: Deployment places the feature extractor at the edge, the decoder and classifier at the cloud, and stores the codebook at both sides.
Design Rationale: EfficientFi’s learned codebook captures CSI patterns for reconstruction, while multi-task learning integrates recognition and reconstruction objectives to learn semantic features.
Algorithm: The training algorithm extracts and quantizes features, reconstructs CSI, predicts sensing results, and updates network parameters and the codebook iteratively.
Architecture: The experimental network architecture includes convolution, transposed-convolution, max-pooling, and max-unpooling operations.

A. Experimental Setup

The evaluation uses real-world WiFi AP experiments to measure compression, restoration, and recognition across human-sensing tasks.

System Design: Experiments use two TP-Link N750 APs with three antenna pairs, 40 MHz bandwidth, and 114 CSI subcarriers per timestamp.A local server simulates cloud computing.
Data Collection: 500 Hz sampling produces 3 × 114 × 500 CSI samples per second for six activities performed by 20 subjects.Each activity category contains 400 samples.
Network Implementation: The network uses four convolutional or transposed-convolutional layers for feature extraction and decoding, plus a two-layer softmax classifier.The codebook uses D = 256, while K varies from 64 to 1024 and λ = 0.5.
Criterion: Evaluation compares communication cost, NMSE reconstruction error, and recognition accuracy.The original communication cost is 1.368Mb/s, and NMSE is reported in decibels.
Baselines: Compression baselines include LASSO, BM3D-AMP, CSINet, and a vanilla auto-encoder, while HAR and gait identification use task-specific comparison methods.E-eyes and CARM support HAR comparison; WiWho and AutoID support gait recognition.

B. Evaluation on HAR and Human-ID

EfficientFi preserves recognition performance while compressing CSI for human activity recognition and human identification, outperforming the cited compression and sensing baselines.

Overall comparison: 334 compression rate is achieved with K = 256 while maintaining satisfactory reconstruction error for both HAR and human identification.EfficientFi outperforms LASSO, BM3D-AMP, and CSINet in the reported comparisons.
HAR: 20% around improvement over E-eyes and CARM is reported for human activity recognition.The uncompressed encoder-only accuracy is 98.3%, with accuracy declining as compression increases.
Human-ID: 83% around accuracy is retained for person identification after compression, compared with 67.3% for WiWho and 77.6% for AutoID.The best reported compressed result occurs at compression rate 66.8, while higher compression reduces accuracy.
Summary: EfficientFi achieves high-fidelity reconstruction and high recognition accuracy for both wireless activity recognition and person identification.The conclusion summarizes preservation of classification performance after feature compression and transmission.

C. Incremental Learning via Reconstructed Data

Reconstructed CSI can support incremental classifier training, improving human-identification accuracy when high-confidence samples are used for fine-tuning.

Incremental learning: The reconstructed data is used to fine-tune the classifier with high-confidence samples for improved recognition performance.The result is reported for the Human-ID evaluation in Table III.
Incremental learning: 89.5% accuracy is obtained after fine-tuning with reconstructed CSI and a 90% confidence threshold.Half of the original testing data is used for fine-tuning and half for evaluation.

D. Hyperparameter Sensitivity

The study examines hyperparameter sensitivity and compression efficiency, while visualizing how quantized and reconstructed features preserve discriminative structure.

Hyperparameter sensitivity: With K = 64, increasing embedding dimension D reduces accuracy variance across settings, with D = 64 producing 95%–98% accuracy variation.The analysis also varies the weight λ and focuses on recognition accuracy because NMSE is acceptable across settings.
Inference Time: 2.1ms is required to compress one second of CSI at 500Hz, versus 251ms for LASSO and 747ms for BM3D-AMP.CSINet requires 5.1ms per CSI timestamp, so its cost scales with sampling rate.
Feature visualization: T-SNE shows raw CSI activities can overlap, whereas quantized features form separated discriminative clusters.The reconstructed feature preserves discriminability before the classifier further aligns the latent space.
Conclusion: EfficientFi combines quantized representation learning with joint recognition learning in an end-to-end offline framework for online edge-cloud sensing.Experiments report state-of-the-art performance against classic compressive sensing and deep compression methods.