WiAR : Wi-Fi-based human activity recognition using time-frequency analysis and lightweight deep learning for smart environments

Key Points

• Human activity recognition was effectively achieved using a lightweight deep learning model, enhancing real-time response.
• The model's accuracy reached up to 92% through time-frequency analysis of Wi-Fi signal variations.
• Analysis focused on diverse activity categories, making it suitable for varied smart home applications.
• Highlighting the potential for integration in smart environments, current work lays groundwork for future advancements.