Wireless Sensor Networks Based On Fuzzy Clustering, Clustering Routing Algorithm

Wireless Sensor Network is a large-scale,limited resources and dynamic topology self-organizing network.The routing algorithm is an important part of the Wireless Sensor Networks, which is responsible for transmiting of data between the source node and the destination node. At present,the research on routing algorithm has become an important issue in Wireless Sensor Networks. This paper first introduces the Wireless Sensor Networks routing algorithms,and analyzes the existing cluster-based routing algorithms advantages and disadvantages in detail. Including two problems: the uneven distribution of cluster-heads and withouting considering the energy of the nodes. Aimed at these problems, the fuzzy clustering approach is applied in Wireless Sensor Networks.Two algorithms are proposed in this paper: Fuzzy C-Means based Clustering Multi-hops Routing Algorithm for Wireless Sensor Networks (FCMC) and Subtractive Clustering based Clustering Routing Algorithm for Wireless Sensor Networks (SCC).In FCMC, at the first round, the cluster-heads are chosen and the nodes are grouped into the fixed clusters by Fuzzy C-Means. In the next round, cluster-heads are chosen by the energy of nodes. This method reduces the cost of clusters and expands the scope of network.In the course of data transmission, single-hop communication mode is used in inner-cluster communication phase, and multi-hop communication mode is used in outer-cluster communication phase, these methods balance the load of nodes and solve the problem of the farest nodes died earlier. In SCC, the Subtractive Clustering method is used to choose the cluster-heads and generate the cluster-heads where have high density; besides, the current non-cluster head attribution mechanism is modified to distribute the expenditure of energy averagely into the whole network.Compared with LEACH (Low-Energy Adaptive Clustering Hierarchy ),LEACH-C(LEACH-Centrialized)algorithms,simulation results show that the FCMC and SCC algorithms can provide the reasonable arrangement of cluster-head, longer lifetime of the first node and even longer lifetime of the network to balance the energy expenditure of all the nodes in the network.