A Research On Activity Recognition Of Mobile Device Users Based On Wireless Sensor Network

Recent advances in micro-electronic, computing and wireless communication technology have enabled the rapid development of low-power, multifunctional sensor nodes that are small in size and integrated with the functions of information acquisition, data processing and wireless communications and many other features. Currently, wireless sensor networks have already been widely used in the smart space, transportation, military engineering, and many other fields. Smart space is an electronic environment that could feel and respond to the existence of people. It could create the desired environment and functions to improve people's quality of life through providing intelligent services with interconnected systems. In realization of a smart space application, such as smart family and smart office environment, activity recognition plays a very important role.In this paper we present a method for activity recognition of mobile device users based on wireless sensor network. A user's mobile device directly communicates with nodes in WSN and collect user related data for preprocessing and activity recognition.Main work in this paper is as follows:1. Design and implement functions of data collection from wireless sensor network and data reception and preprocessing on mobile sink node.2. Implement non-wearable activity recognition in a smart office scenario. Mobile device receives sensor data and performs activity recognition. The result is reported to server for publication.3. Analyze the application of classifier for wearable activity recognition. Compare the performance of some popular classifiers and analyze the effect of feature fusion in activity recognition with experiment.