Energy-Efficient Wireless Sensor Network Clustering Routing Algorithm

Internet of Things is the third wave of the world information industry following the computer and the Internet. It is a fully interconnected network which realizes the connection from human to thing and thing to thing through sensing devices according to the protocols. Wireless sensor network (WSN) is the perception layer for the Internet of Things. WSN completes the real-time sensing, monitoring and gathering tasks of a variety of environments or objects through the deployment and collaboration of a large number of micro sensor nodes. Then WSN could transmit the information to the sink node and eventually to user terminals by multi-hop self-organization. Therefore, WSN achieves the integration of information and real world and is changing the way people interact with nature.However, WSN still confronts many problems. For example, limited energy has been the bottleneck for WSN. While node hardware is limited in the reduction of energy consumption, software-based routing algorithm more adapts to the energy-saving development. With the continuous expansion of node number and network size, conventional planar network is no longer suitable. So WSN needs clustering structure to extend the network life. This paper focuses on the WSN clustering routing algorithm.In this paper, it proposes three clustering algorithms based on two network models and makes analysis.(1) PSO Clustering Routing Algorithm based on Non-Linear Decrease Weight with Dynamical Adaptation (PSO-NLDIW)The study constructs an isomorphic network model and proposes PSO Clustering Routing Algorithm based on Non-Linear Decrease Weight with Dynamical Adaptation for the homogeneous network. To realize energy efficiency and balance, this paper considers three factors to optimize the cluster head, average residual energy of cluster heads, energy distance of cluster nodes, and the maximum Euclidean distance from cluster heads to the sink node. Then the paper introduces a nonlinear self-adaptive decreasing weight to balance the local and global search capabilities. Result shows that this method effectively balances network energy consumption and increases the lifetime of network.(2) Uneven Clustering Routing Algorithm based on Weight for heterogeneous WSN (UCRAW)This paper constructs heterogeneous network model and proposes Uneven Clustering Routing Algorithm based on Weight for heterogeneous WSN. In selecting cluster head, there are five factors:node residual energy, elected numbers of cluster head, distance from sink node, nodes density, and node energy consumption. Analytic Hierarchy Process is used to compute coefficients for avoiding artificially inconsequence. Finally, the least weight node would be selected as a cluster head. The idea of uneven clustering and cluster head rotation with energy threshold protects the cluster head near the sink node, which helps to reduce unnecessary energy consumption by frequent election of clusters. The simulation results show that the introduction of energy-heterogeneous nodes and optimization of clusters have improved network transmission performance and network lifetime significantly.(3) Double Cluster-Heads and Secure Clustering Algorithm based on Relay Cluster Transmission for heterogeneous WSN (DCSRT)In order to adapt to attack happened in large-scale network, this paper proposes Double Cluster-heads and Secure Clustering Algorithm based on Relay Cluster Transmission for heterogeneous WSN. Main cluster head collects data in the clusters, member nodes join clusters by clustering cost formula; assistant cluster heads transmit data in the form of relay cluster. In the meanwhile, the study increases security measures for encrypt data transmission with cluster keys and keys between clusters. Results show that the scenario adapts to the large-scale networks with densely node distribution and strengthens the protocol security to a certain extent.