WiFi Channel State Information (CSI)-based human activity recognition (HAR)
enables contactless, long-range sensing in spatially constrained environments
while preserving visual privacy. However, despite the presence of numerous
WiFi-enabled devices around us, few expose CSI to users, resulting in a lack of
sensing hardware options. Variants of the Espressif ESP32 have emerged as
potential low-cost and easy-to-deploy solutions for WiFi CSI-based HAR. In this
work, four ESP32-S3-based 2.4GHz directional antenna systems are evaluated for
their ability to facilitate long-range through-wall HAR. Two promising systems
are proposed, one of which combines the ESP32-S3 with a directional biquad
antenna. This combination represents, to the best of our knowledge, the first
demonstration of such a system in WiFi-based HAR. The second system relies on
the built-in printed inverted-F antenna (PIFA) of the ESP32-S3 and achieves
directionality through a plane reflector. In a comprehensive evaluation of
line-of-sight (LOS) and non-line-of-sight (NLOS) HAR performance, both systems
are deployed in an office environment spanning a distance of 18 meters across
five rooms. In this experimental setup, the Wallhack1.8k dataset, comprising
1806 CSI amplitude spectrograms of human activities, is collected and made
publicly available. Based on Wallhack1.8k, we train activity recognition models
using the EfficientNetV2 architecture to assess system performance in LOS and
NLOS scenarios. For the core NLOS activity recognition problem, the biquad
antenna and PIFA-based systems achieve accuracies of 92.0$pm$3.5 and
86.8$pm$4.7, respectively, demonstrating the feasibility of long-range
through-wall HAR with the proposed systems.
The article discusses the use of WiFi Channel State Information (CSI)-based human activity recognition (HAR) for contactless sensing in spatially constrained environments. This method allows for long-range sensing while preserving visual privacy. However, the lack of available sensing hardware options has been a challenge. Fortunately, the Espressif ESP32, a low-cost and easy-to-deploy solution, has shown promise for WiFi CSI-based HAR.
This work evaluates four ESP32-S3-based 2.4GHz directional antenna systems for their effectiveness in facilitating long-range through-wall HAR. Two systems, in particular, show promise. The first combines the ESP32-S3 with a directional biquad antenna, which marks the first demonstration of such a system in WiFi-based HAR. The second system utilizes the built-in printed inverted-F antenna (PIFA) of the ESP32-S3 and achieves directionality through a plane reflector.
To evaluate the performance of these systems, a comprehensive study is conducted in an office environment spanning 18 meters across five rooms. The researchers collect the Wallhack1.8k dataset, consisting of 1806 CSI amplitude spectrograms of human activities, which is made publicly available. Using this dataset, they train activity recognition models using the EfficientNetV2 architecture.
The results of the evaluation show that both the biquad antenna and the PIFA-based systems achieve high accuracies for non-line-of-sight (NLOS) activity recognition, with the biquad antenna system achieving higher accuracy (92.0±3.5) compared to the PIFA-based system (86.8±4.7). This demonstrates the feasibility of long-range through-wall HAR with these proposed systems.
Expert Analysis
This research delves into multi-disciplinary concepts, combining expertise in WiFi technology, antenna design, and machine learning for HAR. The use of WiFi CSI as a sensing mechanism is innovative, as it leverages the existing infrastructure of WiFi-enabled devices for contactless activity recognition. This opens up possibilities for applications in various domains, including smart homes, healthcare monitoring, and security systems.
The evaluation of multiple ESP32-S3-based antenna systems highlights the importance of hardware selection in achieving accurate and reliable HAR. The use of directional antennas, such as the biquad antenna and the PIFA-based system with a plane reflector, enables long-range sensing through walls. This is particularly important in environments where visual privacy is a concern or where physical access is restricted.
The collection and public availability of the Wallhack1.8k dataset is a valuable contribution to the research community. This dataset serves as a benchmark for future HAR studies, allowing researchers to compare their models and algorithms against a standardized dataset. It promotes transparency and reproducibility, essential in advancing the field.
The high accuracies achieved in NLOS activity recognition with the proposed systems indicate their effectiveness in practical scenarios. However, further research is necessary to explore the robustness and generalizability of these systems in different environments and real-world conditions. Additionally, the incorporation of more sophisticated machine learning techniques, such as deep learning or reinforcement learning, may further improve the accuracy and versatility of WiFi CSI-based HAR.
Conclusion
The article showcases the potential of WiFi Channel State Information (CSI)-based human activity recognition (HAR) using ESP32-S3-based antenna systems. The evaluation and experimentation demonstrate the feasibility of long-range through-wall HAR in both line-of-sight and non-line-of-sight scenarios. The research contributes to the understanding of multi-disciplinary concepts in WiFi technology, antenna design, and machine learning for HAR applications. The availability of the Wallhack1.8k dataset further promotes transparency and reproducibility in the field. With further advancements and research, WiFi CSI-based HAR has the potential to revolutionize contactless sensing in various domains, enhancing privacy, efficiency, and security.