The Reaserch Of Multi-objectives Optimization For Routing Protocol In Wireless Sensor Networks

Wireless sensor network technology, which attracts attention to internationalresearchers, has broad application background. There exist many indicators andsensitive factors for evaluating the performance of wireless sensor networks. Fortraditional routing protocols, the sensor nodes in energy storage are limited becauseof hardware limitation, thus most researchers aim to optimize the energy consumptionfor the network. But when the energy consumption has been promoted individually,the performance of network may be decreased by the deterioration of the othersfactors. Multi-objective optimization can solve the problem of mutual goalssimultaneously which are constrained each other so that ensure the performance ofwireless sensor networks.In order to improve performance indicators and balance various factors ofwireless sensor networks, this dissertation first investigates the evaluating indicatorsof wireless sensor networks and analyzes the relationship of them. Then we proposepropose the method of optimizing wireless sensor network routings from the view ofmulti-objective problem. Additionally, the model of isolation nodes probability forconnectivity in the cluster-based networks is proposed in this place. This dissertationoptimizes the cluster-based protocol of wireless sensor networks based onNSGAII(Non-dominated Sorting Genetic Algorithm II), SPEAII(Strength ParetoEvolutionary Algorithm II) and BMOPSO(Binary Multi-Objective Particle SwarmOptimization). The simulation results show that the performance of energyconsumption, network latency and network connectivity in routing based onmulti-objective optimization is better than the one in traditional routing protocols.