Study Of Wireless Sensor Network Clustering Routing Algorithm

With the development of “smart city” and “Internet+”, researchers enhance the requirements of perception ability, computing ability and communication capability to process information, at the same time, strengthen the application of wireless sensor network. The research of routing algorithm is the core technology of wireless sensor network. As the traditional wireless sensor network routing algorithms make the “transfer-data” as a center point, and do not consider the energy consumption of battery-powered sensor nodes, they cannot meet the demands in the processing information area. Therefore, the study of wireless sensor network routing algorithm based on energy balance is becoming the research mainstream.The classical clustering routing algorithms, Low Energy Adaptive Clustering Hierarchy and Power-Efficient Gathering in Sensor Information Systems are analyzed. At the same time, simulates and compares the two algorithms by using software NS3. By combining the character of chain and clustering, some improvements on the way of forming clustering and choice of cluster-heads are improved. A clustering routing algorithm based on divided-regions is proposed. In the proposed routing algorithm, the whole communication area is divided into several regions. In each region, all nodes are arranged to form a net-cluster by using the greedy algorithm. Even though some nodes die, the consumption of energy of the whole network restructuring can be reduced. A cluster-head is selected, and receives other node's data in each region. At last, the cluster-head communicate with BS directly in each region. The First Order Radio Model is used to consider the consumption of energy, which can reduce the consumption of energy.At last, the proposed algorithm is simulated in MATLAB, and the PEGASIS, original improved and the CDEB algorithm are compared in energy consumption and life cycle. The experiment results show that the CDEB algorithm balances the energy consumption effectively extends the life cycle of the whole network.