In a new study, researchers from Incheon National University are presenting a new Artificial Intelligence of Things (AIoT) framework known as MSF-Net for recognising human activities using Wi-Fi signals. The framework utilises a novel approach that combines different signal processing techniques and a deep learning architecture to address challenges such as environmental interference, achieving high recognition accuracy.
AIoT, which combines the advantages of AI and IoT technologies, has grown in popularity in recent years. In contrast to IoT set-ups where devices collect and transfer data for later processing, AIoT devices acquire data locally and in real-time. This technology has found applications in intelligent manufacuring, smart home security, and healthcare monitoring.
In smart homes, for example, accurate human activity recognition is important. It supports smart devices with identify tasks such as cooking and exercising. Based on this information, the AIoT system can tweak lighting or switch music automatically. Wi-Fi based motion recognition is promising in this scenario, as Wi-Fi devices are ubiquitous and usually cost-effective.
Researchers, led by by Professor Gwanggil Jeon from the College of Information Technology at Incheon National University, South Korea, have come up with a new AIoT framework called multiple spectrogram fusion network (MSF-Net) for WiFi-based human activity recognition.
“As a typical AIoT application, Wi-Fi-based human activity recognition is becoming increasingly popular in smart homes. However, Wi-Fi-based recognition often has unstable performance due to environmental interference. Our goal was to overcome this problem,” said Professor Jeon.
Researchers developed the framework, which achieves coarse and fine activity recognition via channel state information (CSI). MSF-Net has three main components: a dual-stream structure comprising short-time Fourier transform along with discrete wavelet transform, a transformer, and an attention-based fusion branch. While the dual-stream structure pinpoints abnormal information in CSI, the transformer extracts high-level features from the data efficiently. Lastly, the fusion branch boosts cross-model fusion.
The researchers performed experiments to validate the performance of their framework, with findings showing it achieves Cohen’s Kappa scores of 91.82%, 69.76%, 85.91%, and 75.66% on SignFi, Widar3.0, UT-HAR, and NTU-HAR datasets, respectively. These values highlight the superior performance of MSF-Net compared to state-of-the-art techniques for Wi-Fi data-based coarse and fine activity recognition.
“The multimodal frequency fusion technique has significantly improved accuracy and efficiency compared to existing technologies, increasing the possibility of practical applications,” said Professor Jeon. “This research can be used in various fields such as smart homes, rehabilitation medicine, and care for the elderly. For instance, it can prevent falls by analysing the user’s movements and contribute to improving the quality of life by establishing a non-face-to-face health monitoring system.”
Activity recognition using Wi-Fi, the convergence technologies of AI and IoT proposed in this work is expected to significantly improve people’s lives.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by commenting below or visiting our LinkedIn page.
