A Congestion Aware, Energy Efficient, on Demand Fuzzy Logic Based Clustering Protocol for Multi-hop Wireless Sensor Networks

Node clustering or simply clustering provides an effective way of organizing the network to achieve energy efficiency, efficient distribution of workload, maintain a connected hierarchy and accomplish data aggregation. On the other hand multi-hop communication significantly reduces the energy required for communication and multi-path communication helps in distributing the relay load and it also increases reliability of the network. The proposed methodology in this paper is a combination of these three features. In addition to that, we have used fuzzy logic in cluster head (CH) selection to blend different parameters to select the set of best possible CHs and to handle the uncertainties in parametric quantities. There are five main aspects of the proposed algorithm (a) on demand based clustering (b) fuzzy logic based CH selection (c) dynamic unequal cluster range for CHs’ (d) a cost function based multi-hop relay selections (e) Fibonacci sequence based relay load distribution. The proposed approach creates more balanced cluster by means of fuzzy logic and inequality in cluster range. It saves energy by conducting CH selection on demand basis. Finally, the cost function based multi-hop relay selections along with Fibonacci sequence based relay load distribution create an optimal structure for relaying data.The proposed approach is implemented and compared with the well known approaches like LEACH, ECPF, CHEF and UCR. Simulation results have shown that the proposed approach performs better, in every aspect of network life time and other metrics.