Research On Cluster-Based Routing Algorithm For Wireless Sensor Networks

With the development of sensor technology, Micro-Electro-Mechanism System (MEMS), modern networks and wireless communication technology, Wireless Sensor Network (WSN) comes out and develops gradually. After many years' research and development, WSN has been applied in a lot of fields such as environment detecting, medical treatment guarding, traffic management, storage management and military affairs reconnaissance. As research's going deeper, cluster-based routing algorithm in routing layer has become a hot issue lately.In this thesis, we give a detailed introduction to WSN cluster-based routing algorithms proposed by now, and analyze the actuality and shortcomings of existing work. Based on this, we summarize the specialties of WSN cluster-based routing algorithm and with the purpose of saving network's energy and prolonging its lifetime, we propose an Unequal Scaled Cluster-based Routing (USCR) algorithm. USCR uses the method of confining clusters' radius in its clustering process. More clusters are formed in the area nearer to sink node to share the task of delivering routing load. In this way, USCR relieved the area energy unbalancing problem caused by the multi-to-one transmission mode of WSN in earlier cluster-based routing algorithm. At the same time, hops of clusters from the sink node to a cluster are maintained to instruct route selection. USCR selects the neighbor cluster with smaller hop value and whose head has more power as the next cluster and pick the neighbor node with less delivering cost and better energy condition according to the distances between nodes and the nodes' remaining power as the next hop. By the technique of decreasing the number of hops and selecting the proper next hop in a route, USCR algorithm effectively reduces the energy cost of the whole network, balances the loading of nodes and prolongs the lifetime of network.In WSN routing algorithm, the data aggregation process which meets the data-centric requirement of WSN will definitely introduce a lot of delay during data transmission. There is a contradiction between doing data aggregation in routing process and supplying QoS (Quality of Service) guarantee to real-time data. We analyze the feature of USCR algorithm and based on it, we propose a routing algorithm which can both do data aggregation and supply QoS guarantee to real-time data during routing process. In this algorithm, we propose a method to estimate data's maximum transmitting delay with the help of the specialty in cluster structure formed by USCR algorithm. According to the estimated delay value, we choose different routes for data with different service requirements. This algorithm inherits the advantage of energy saving from USCR algorithm, serves real-time data flow and non-real-time data flow at the same time, selects the route that can meet the real-time data's delay requirement and can also do as more aggregation to data as possible.At last, we simulate this two proposed cluster-based routing algorithms by OMNET++ simulator and lucubrate the analysis of the results. It verifies that USCR exceeds other algorithms in saving the network's energy and prolonging the network's lifetime. We can also see from the results that the QoS algorithm we proposed is feasible and can come to an anticipant effect.