Scientists at Incheon National University (INU) in South Korea have developed an AIoT framework, to enhance smart home security using wifi-based human activity recognition.

By leveraging deep learning and multimodal frequency fusion, the system overcomes environmental interference for improved accuracy. This technology can improve home automation, elderly care and remote health monitoring.

The combination of AI and IoT (AIoT) is becoming popular because of its widespread applications. In a study, researchers from INU have presented an AIoT framework called MSF-Net for accurately recognising human activities using wifi signals. The framework uses a novel approach that combines different signal processing techniques and a deep learning architecture to overcome problems from environmental interference and achieve high recognition accuracy.

In contrast to typical IoT setups, wherein devices collect and transfer data for processing at some other location, AIoT devices acquire data locally and in real time, enabling them to make smart decisions. This technology has found applications in intelligent manufacturing, smart home security and healthcare monitoring.

In smart home AIoT technology, accurate human activity recognition helps smart devices identify various tasks, such as cooking and exercising. Based on this information, the system can tweak lighting or switch music automatically, thus improving the user experience while ensuring energy efficiency. In this context, wifi-based motion recognition is quite promising; wifi devices are ubiquitous, ensure privacy, and tend to be cost-effective.

Recently, in a research article, a team of researchers, led by Gwanggil Jeon from INU’s College of Information Technology, has come up with an AIoT framework called multiple spectrogram fusion network (MSF-Net) for wifi-based human activity recognition. Their findings were published in f the IEEE Internet of Things Journal (ieeexplore.ieee.org/document/10530186).

“As a typical AIoT application, wifi-based human activity recognition is becoming increasingly popular in smart homes,” said Jeon. “However, wifi-based recognition often has unstable performance due to environmental interference. Our goal was to overcome this problem.”

In this view, the researchers developed the MSF-Net deep learning framework, which achieves coarse as well as 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, finding it achieved Cohen’s Kappa scores of 91.82%, 69.76%, 85.91% and 75.66% on SignFi, Widar 3.0, UT-HAR and NTU-HAR datasets, respectively. These values highlight the performance of MSF-Net compared with the latest techniques for wifi data-based coarse and fine activity recognition.

“The multimodal frequency fusion technique has significantly improved accuracy and efficiency compared with existing technologies, increasing the possibility of practical applications,” said 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.”

Overall, activity recognition using wifi, the convergence technology of IoT and AI proposed in this work is expected to improve people’s lives through everyday convenience and safety.

INU (www.inu.ac.kr) was founded in 1979 and given university status in 1988. One of the largest universities in South Korea, it houses nearly 14,000 students and 500 faculty members. In 2010, INU merged with Incheon City College to expand capacity and open more curricula.