Body Sensor Networks (BSNs) are formed by the equipped or transplanted sensors in the human body, which can sense the physiology and environment parameters. As a novel e-health technology, BSNs promote the deployment of innovative healthcare monitoring applications. In the past few years, most of the related research works have focused on sensor design, signal processing, and communication protocol. This chapter addresses the problem of system design and data fusion technology over a bandwidth and energy constrained body sensor network. Compared with the traditional sensor network, miniaturization and low-power are more important to meet the requirements to facilitate wearing and long-running operation. As there are strong correlations between sensory data collected from different sensors, data fusion is employed to reduce the redundant data and the load in body sensor networks. To accomplish the complex task, more than one kind of node must be equipped or transplanted to monitor multi-targets, which makes the fusion process become sophisticated. In this chapter, a new BSNs system is designed to complete online diagnosis function. Based on the principle of data fusion in BSNs, we measure and investigate its performance in the efficiency of saving energy. Furthermore, the authors discuss the detection and rectification of errors in sensory data. Then a data evaluation method based on Bayesian estimation is proposed. Finally, the authors verify the performance of the designed system and the validity of the proposed data evaluation method. The chapter is concluded by identifying some open research issues on this topic.
Predicting Traffic Dynamics with Driver Response Model for Proactive Variable Speed Limit Control Algorithm
