Research On Energy Balance Problem In Wireless Sensor Network

Wireless sensor network (WSN) is a self-organized system composed of a largenumber of tiny sensors, which has the characteristic such as large-scaled deployment,low energy, dynamics, self-organization and so on. In recent years there are a lot ofapplications in different kinds of areas.Energy balance is an important issue in WSN because the sensors are so tiny thatthey have limited energy. How to take advantage of sensors’ limited energy andprolong network’s life time is the goal persued by researchers.According to current researching situation and existing solutions, in this paper,firstly we propose an unequal cluster algorithm based hybrid rotation manner inwireless sensor network, combining the time-driven rotation manner andenergy-driven rotation manner. This algorithm uses energy-driven rotation manner inlocal so that it can avoid of a large number of wasted controlling messages, whileusing time-driven rotation manner in total network, namely, reconstructing thenetwork periodically to keep the energy balanced. On the meanwhile, unequalclustering concept is introduced which has an advantage of balancing energy. In thesimulation we compare the algorithm proposed with Leach and EDUC, and validatethe effectiveness of our algorithm.Secondly, due to the shortage that when the number of mobile sensors and eventsis large, the algorithm is not effective in previous literature, in this paper, we proposean improved energy balanced dispatch algorithm of mobile sensors in hybrid wirelesssensor network. In this algorithm, dispatch problem of mobile sensors is switched intothe maximum matched problem, while a mechanism of local energy balancedprediction is introduced, improving the effectiveness of the competing between twomobile sensors in collision. In addition, we improve the clustering category in whichwe define the concept of local balance so that the clusters generated are more balanced when the number of mobile sensors is less than that of events. In thesimulation, we compare our algorithm with the algorithm in literature, and validatethe effectiveness of our algorithm.