Research Of Several Key Technologies For Dynamic Sensor Networks

With the development and widely application of wireless sensor network, the technology of MEMS has made it possible of dynamic sensor network with tightly structure and low cost. More and more people get interested in the field of dynamic sensor network. The development of robots makes it become a reality of the practical application of the dynamic sensor network. The tracking of the mobile object requires the whole network follow with it. and keep the object on visible to the network. In dynamic sensor network, the topology also changes with the node for its mobility, and it must assure all nodes in its coverage. A good coverage algorithm could make the mobile node adjust its location only according to the nearer nodes around it so that the whole network has a uniform coverage.(1)According to the advantages of various wireless sensor networks (WSN) topology protocols, exploring a new topology control protocol is a hot spot in WSN research. A cluster-based self-adaptive hybrid topology control protocol-SAHRC for WSN in special application is proposed based on traditional LEACH protocol. It aims at large-scale event-driven network application scenarios, uses self-configuring and trigger mechanism to solve the shortages of LEACH in large-scale networks, such as the lack of self-adaptability, leaving consideration of the node's residual energy, and the difficulty in improving communication efficiency. The performance of the algorithm SAHRC with simulation experiments was evaluated. Compared with existing LEACH algorithm, the algorithm SAHRC provides higher performance in stability and energy-saving.(2)The existing coverage algorithms could be divided into three kinds, the potential field algorithm, the Voronoi diagram and the Delaunay algorithm. Due to these three method's shortcomings, the paper presents a new virtual force coverage algorithm based on boundary contraction algorithm. The effective coverage algorithm has some advantages to MAC protocol, routing protocol, and the energy-saving in some degree. The presented algorithm can really respond to the changes of the network topology in a distributed circumstance. Simulation results show that the proposed algorithm could realize a uniform coverage and have a better performance.(3)Firstly, the paper analyzes the traditional location algorithms and presents an improved Dead Reckoning and Voronoi-Weight based MCL algorithm called DRVW-MCL for dynamic sensor network. This algorithm uses Monte Carlo algorithm as the basic localization algorithm for mobile sensor network, while in prediction phase it imports dead reckoning method to improve the accuracy of the particle's prediction position by reducing the error of the prediction angle, in filtering phase the Voronoi diagram and weight are imported to improve the filtering accuracy by dual filtering of Voronoi diagram and weight. Simulation results show that the proposed algorithm outperforms the MCL algorithm for its localization accuracy is improved to approximately 15%, which is a half of MCL algorithm. Furthermore, DRVW-MCL algorithm has little effect in particle number for the location accuracy will stay at 20% when the particle number is more than 20.(4)Then the paper presents a path dynamic planning algorithm for mobile anchor node in dynamic sensor network. Localization algorithm based on mobile anchor uses a walking anchor to broadcast location information in the network to construct the virtual anchors, which can save positioning cost. But the mobile anchors in the network path determine the network positioning cost and accuracy. In this paper. mobile anchor path planning algorithm is researched in order to minimize the cost of access to adequate location information. Simulation results show that the algorithm for wireless sensor network node can obtain optimal state path and improve positioning accuracy.The research about dynamic sensor networks involve in many fields and several subjects intercrossing such as micro-electronics, robot technology, information technology, computer science and wireless communication and so on except for location, tracking, coverage. There are a great deal of work to be continued on above theories and the application and development of the key technologies.