Wireless Sensor Networks For Pervasive Medical Research

Following by the wireless communications technology, computer technology and microsystems technology development, Wireless sensor networks (WSNs) have aroused attention widely. WSNs are consisted by sensing, data processing and short-distance wireless communication function of the sensors, which has wide application propects, such as military, environment monitoring, biomedical and commercial etc al. now, WSNs is an extremely important innovation issues in the network.In this study, the drawbacks of WSNs are focused. Basedon the superior features of immune algorithm and evolutionary algorithm, and considering the balance mechanisms in process of computation. The target coverage, route problems are study by building models, and applying in the target coverage, route of complex computation applications in WSNs. Simultaneously, the energy consumption, target tracking and navigation problems of WSNs are analyze by KD trees and P2P mechanism, through a combination of pervasive health-care, we proposed the automatic ECG detection and workflow optimization program based on pervasive health-care.First, a general study of WSNs, pervasive healthcare and service mechanism are given to explain the value and significance. For WSNs, the target coverage and route optimization are emphasized. Energy consumption of WSNs is based on KD-tree algorithm, and target tracking and navigation problems are studied on P2P framework respectively.Then, in the five categories, some problems are abstracted from WSNs, whose models are built for inspiring algorithms and system framework. And they are applied in the fields of target coverage, route optimization, collbration energy optimization and object tarck and navigation. 1) The artificial immune algorithm is applied to research target coverage problem of WSNs, through antibodies initialization, antibody proliferation, high-frequency variation, immune memory and immune elimination research the target coverage of WSNs. 2) The energy consumption of wireless sensor network's routing problem is studied, the traditional method of the route is the single-objective optimization for WSNs, the survival period and speed of reconstruction are the main objectives of WSNs, the multiple-objective energy routing approach based on multi-objective evolutionary algorithm is proposed, we use Pareto-solutions, through the establishment of non-dominated set to maintain multi-objective's balance. 3) For signal collection protocol, which are insufficient reliability of WSNs, real-time and performance are required, a Kernel-based estimates and KD-tree coding method of data transmission is proposed, the method use kernel estimation and data encoding collect data, the coordination of communication between different nodes in order to achieve lower energy consumption and network bandwidth by hybird method. 4) In the aspect of target tracking and navigation, the action mode of WSNs is data-centric applications. In order to ensure that the target tracking and navigation of the service quality, the Voronoi convex polygons, Delaunay triangles and the P2P distributed architecture is propoesed to solve navigation and target tracking problem. 5) In pervasive healthcare applications oriented WSNs, Automatic ECG analysis model is proposed based on rough sets and hidden Markov, the model could identify different heart disease through different heart signal, for pervasive healthcare at the workflow optimization, the time automata workflow model is proposed.Finally, a summary of the thesis is made, and the deficiency of the study and the further development are narrated respectively.