Study Of Clustering Optimization Algorithm In Wireless Sensor Networks

Wireless Sensor Networks(WSNs) have gained worldwide attention in recent years. Particularly, with the development of Communication technology, Chip technology, Internet of Things technology and Computer technology, WSNs are widely used in various industries of society, which provides great convenience for people’s life.However, due to the energy of sensor node is limited, and it also be deployed where the environment is bad, once the node’s storage energy is depleted, the subsequent energy complement is almost impossible to achieve, this is one of the bottlenecks restricting of the development of WSNs. So, how to efficiently use the limited energy of the node as far as possible to extend the life cycle of WSNs has become the research focus in domestic and foreign researchers. This paper studied the clustering algorithm of WSNs, in view of the existing WSNs protocols energy efficiency is not high, and easy to produce the problem such as energy hole, we put forward the following three kinds of improved algorithms:(1) Based on the idea of energy efficient, this paper designed a kind of energy efficient clustering routing optimization algorithm. During the period of cluster head election, this algorithm selects the high residual energy node as the cluster head from the global perspective, resulting in the equilibrium of energy consumption of each node. In the routing phase, this algorithm constructs a efficient and stable multi-hop route with the help of linear equation, and it also considers the residual energy and forwarding cost of the cluster head when choosing the next-hop relay node, therefore the elected next-hop route is more suitable for undertaking the task of data forwarding. The simulation results show that the proposed algorithm can effectively improve the network energy efficiency.(2) For the energy hole is easily emerged in the equal clustering network, a kind of Energy awareness Unequal Clustering Routing optimization algorithm(EUCR) is put forward in this paper. This algorithm selects the node with larger residual energy as the cluster head as much as possible through the local energy cluster head election strategy. At the same time, EUCR algorithm creates a neighbor nodes set by searching for neighbor nodes, once the cluster head in the network is determined, each cluster head can directly determine its cluster members according to itself neighbor nodes set, it doesn’t need to send broadcast messages to search for cluster members, which greatly reduces the energy costs in clustering stage, and improves the network energy efficiency. In data transmission phase, EUCR algorithm defines the ratio of residual energy and forwarding cost of the node as the basis of searching for routing, making the selected cluster head is more suitable for data forwarding. Through the simulation, the proposed algorithm can effectively balance the energy consumption among nodes and prolong the life cycle of the network.(3) According to the existing shortcomings of using genetic algorithm to optimize the clustering WSNs protocols, a algorithm that using jumping genes to improve this type protocols is put forward in this paper. The simulations indicate that the proposed method in this paper can effectively improve the convergence rate of genetic algorithm.