Freedom: online activity recognition via dictionary-based sparse representation of RFID sensing data

Abstract

Understanding and recognizing the activities performed by people is a fundamental research topic for a wide range of important applications such as fall detection of elderly people. In this paper, we present the technical details behind Freedom, a low-cost, unobtrusive system that supports independent living of the older people. The Freedom system interprets what a person is doing by leveraging machine learning algorithms and radio-frequency identification (RFID) technology. To deal with noisy, streaming, unstable RFID signals, we particularly develop a dictionary-based approach that can learn dictionaries for activities using an unsupervised sparse coding algorithm. Our approach achieves efficient and robust activity recognition via a more compact representation of the activities. Extensive experiments conducted in a real-life residential environment demonstrate that our proposed system offers a good overall performance (e.g., achieving over 96% accuracy in recognizing 23 activities) and has the potential to be further developed to support the independent living of elderly people.