PROJECT TITLE :

Compressive Representation for Device-Free Activity Recognition with Passive RFID Signal Strength - 2018

ABSTRACT:

Understanding and recognizing human activities could be a basic research topic for a big selection of vital applications such as fall detection and Remote Health Monitoring and intervention. Despite active analysis in Human Activity Recognition over the past years, existing approaches based mostly on pc vision or wearable sensor technologies present many vital issues such as privacy (e.g., using video camera to monitor the elderly at home) and practicality (e.g., not potential for an older person with dementia to recollect wearing devices). In this Project, we present a coffee-value, unobtrusive, and sturdy system that supports freelance living of older folks. The system interprets what someone is doing by deciphering signal fluctuations using radio-frequency identification (RFID) technology and Machine Learning algorithms. To deal with noisy, streaming, and unstable RFID signals, we tend to develop a compressive sensing, dictionary-based approach that may learn a group of compact and informative dictionaries of activities using an unsupervised subspace decomposition. In specific, we tend to devise a range of approaches to explore the properties of sparse coefficients of the learned dictionaries for totally utilizing the embodied discriminative info on the activity recognition task. Our approach achieves economical and strong activity recognition via a a lot of compact and robust representation of activities. In depth experiments conducted in a real-life residential surroundings demonstrate that our proposed system offers a good overall performance and shows the promising practical potential to underpin the applications for the freelance living of the elderly.