Large-scale Rechargeable Wireless Sensor Network Mobile Charger’s Path Optimization

With the development of wireless sensor network technology in recent years, various new applications have been widely used in different fields. Because of the high cost of post-maintenance, the traditional wireless sensors which rely on battery power supply becomes more and more difficult to fit the high-speed development of sensor network technology. However, due to the harvesting techniques of ordinary renewable resource will directly influence the stability and security of whole system because of energy-temporal nature, so new method of the energy supplementsno time to delay. The energy transmission fills the gaps in this field and can effectively solve the energy supply problem of wireless sensorsIn recent researches, this technology has been effectively applied to the issue of small scale wireless sensor supply. But it is not suitable in large-scale sensor networks. Mobile chargers carry limited power and they repeatedly return to base station toreplenish energy. Besides, Single mobile charger’s energy cycle is too long to ensure that all sensor nodes work in the charging period. Therefore, So we have to put multi-mobile chargers into consideration.Selecting appropriate charging path can effectively ensure the normal operation of the entire sensor network.Based on reading a large number of existing VRP research results, Main research work and contributions are shown as follows.(1) Constructing the mathematical model of large-scale rechargeable wireless sensor network mobile charger’s path optimization problem.This paper firstly state all segments of mobile charger replenishment problem. Based on integrating Objectives and constraints,and using a loop model, a heterogeneous sensor network mathematical model was constructed.(2) After comparing the existing VRP algorithm, explained heuristic algorithm was necessary and feasible, proposing the greedy idea of the routing algorithm, which could effectively solve our problem.(3)Proposing a engineering assumption based on our problem, and proved the upper bound of performance could achieve under this hypothesis.Meanwhile, it also proved that better than 1.5 times the approximate optimal solution does not exist.(4)Analyzing and comparing the algorithm we proposed through simulation experiment. The analysis results showed that our algorithm had good effect in large scale sensor networks, the algorithm was a good solution to the vehicle routing problem charging for large-scale sensor networks.