Research On Clustering Routing Protocol Using Fuzzy Logic In Wireless Sensor Network

In the wireless sensor network architecture, the design of routing protocol in network layer for wireless sensor networks has played a decisive role. The main objective of routing protocol design is to reduce energy consumption, balance the network load, prolong lifetime of the entire network. Clustering routing protocol forms cluster by using a distributed and self-organized algorithm, organizes intermediate nodes within a cluster, combines the date from the individual nodes and chooses the right path to send the fusion data to the base station, which could evenly consume the energy consumption of network nodes, effectively extend network lifespan.Firstly, this paper analyzes the structure, features, application fields and challenges of wireless sensor networks. Moreover, through the researches against the clustering routing protocols represented by LEACH and improved LEACH protocol, a detailed contrast and analysis on the performance index have been put up. In consideration of the mechanism of LEACH, we found that the energy consumption of each node is unbalanced due to the random selection of cluster nodes. To solve these disadvantages, a novel clustering routing protocol using fuzzy logic (CRPFL) has been proposed, which includes two phases, the set-up phase and the steady-state phase. During the set-up phase, cluster heads selection based on fuzzy logic are the kernel of CRPFL protocol design. The principle of fuzzy logic approach to cluster head selection algorithm is described as follows: two main factors of cluster selection are remaining energy and the number of neighbor nodes, the remaining energy is divided into four levels according to sensor nodes'energy. Also, the number of neighboring nodes is divided into three levels. Considering the experience and the two factors mentioned above, twelve fuzzy logic rules are defined. These rules are used to gain the probability of each node, of which these nodes with the maximal probability will be selected as clusters. During the steady-state phase, the newly-selected clusters fuse information gathered from nodes within the cluster and then send to the base station. Membership functions and fuzzy logic rules are defined through MATLAB tool box and the optimal selection algorithm is coded with MATLAB. Finally, the experimental simulations on the MATLAB platform have been carried out. The simulation scenario is that 100 nodes are randomly deployed in 100m×100m network area, and the energy was set to 0.25 J, 0.5 J, 1 J, 2 J. Experimental results have showed that CRPFL protocol prolongs the lifetime of the network about 23.1%, 19.1%, 17.7%, 16.0% respectively compared with LEACH protocol. Meanwhile, compared with other clustering routing protocols such as EECH and CEFL, the performances of CRPFL have improved differently in terms of network lifetime. So it is safely concluded that our proposed protocol CRPFL has a relatively better energy-efficient characteristic.