Research On Key Technologies Of Energy-efficient Routing In Wireless Sensor Network

Wireless sensor network is a new multi-disciplinary research field from the end of 1990's. It's one of hot points in IT field at present and has an extensive application prospect. Efficiently using of nodes'limited energy and extending network lifetime as long as possible is the basic content of application research on wireless sensor network. On the basic research content, several key technologies about energy-efficient routing are studied.A distributed neural network routing algorithm is proposed. The algorithm is an energy-efficient routing algorithm based on geographical information. It finds optimum path according to the global information of wireless sensor network and distributed represents the finding process on each node using neural network algorithm. The algorithm solves the problems that existing routing algorithms may produce and trap in routing void easily, for they find optimum path according to local information. Simulation results show the algorithm can accomplish routing process correctly and energy-efficiently and avoid the early production of routing void to a great extend. As a result, the influence of routing void on routing process is reduced, and network lifetime is extended.A Positioning algorithm based on particle dynamics is proposed. The algorithm is a rage-free nodes positioning algorithm with high precision. The algorithm achieves node position by comprehensively considering all the global information of wireless sensor network based on principles of particle dynamics in Physics. It solves the low precision problem of existing rage-free positioning algorithms in which positioning process are completely based on beacon nodes. The algorithm can also accomplish node positioning process in three dimensions conditions with high precision which existing range-free node positioning algorithms can not. Simulation results show the algorithm's positioning error is less than 16% if the node's communication distance in two dimensions and a bit more than 20% in three dimensions when there is proper node density. Besides, the positioning results of the algorithm can 100% satisfy the neighbor relationship between nodes in most conditions.A hmax-ηnode deployment model applied to wireless sensor network is proposed. The model changes the even distribution of nodes commonly adopted by researchers at present, and reduces the serious influence on network lifetime caused by uneven nodes energy consumption, as a result extending network lifetime evidently. Simulation results show, on the condition that there are same node number in networks, when using the model, network lifetime can be increased 5-7 times.A forest fire monitoring and controlling system based on wireless sensor network is designed which integrates fireproof monitoring and fire monitoring and controlling. In the system, network nodes are deployed by use of hmax-ηnode distribution model and positioned by use of positioning algorithm based on particle dynamics and optimum paths are chosen by use of distributed neural network routing algorithm. Various network functions are accomplished by connecting distributed neural network routing algorithm with other relevant technologies. Simulation results show the system can keep monitoring forest fire on a large scale for a long time.