Research On Topology Control Based On Small World For Heterogeneous Wireless Sensor Networks

How to balance the energy of network nodes and reduce the communication delay has been a hot issue in the research of heterogeneous sensor networks. However, the control of network topology is an effective method to solve this problem. The small world network model which is between regular network and random network, has small average shortest path length between any pair of nodes and high values of clustering coefficient, so its application on network topology control can improve the performance of network. And in this dissertation, the small world network is applied to heterogeneous sensor network topology control. The main research contents are as follows:(1) The heterogeneous sensor network infrastructure and the small world theory was introduced briefly. Detailed analysis of the current research status of heterogeneous sensor network topology control as well as the applications of small world theory in wireless sensor network was carried out. Meanwhile, the existing problems and shortcomings was pointed out.(2) In order to solve the problems of the inequality of H-Sensors energy consumption and large communication latency in heterogeneous wireless sensor network, we propose a topology control method called CSWM to construct a heterogeneous wireless sensor network based on the small world theory. Through the area partitioning strategy the network can be divided into multiple clusters, and the cluster head is selected according to the residual energy of the HSensor in the cluster. Then, the shortcut is created among the cluster head by dynamic selfadapting strategy, in order to reduce the average path length from sensor nodes to sink node and the communication latency. The network topology is updated by periodical cluster head election to balance the energy consumption of the H-Sensor. And the results of simulation show that the proposed method can reduce the average path length between the sink node and the sensor nodes by constructing the topology with small world property.(3) To construct the network topology which meets the requirements of the real-time continuous data collection in the heterogeneous sensor, we propose a minimum cost flow based distributed routing strategy(MCFR) and the feed-back control for load balancing. In the network each L-sensor uses the MCFR strategy to find the shortest routes to sink node, which reduces the communication delay and constructs a logical topology with small world characteristics. In order to balance the H-sensors’ load, we propose a feed-back control mechanism for load balancing and design an algorithm called optimal load vector calculation(OLC), management station can calculates the optimal loadvector using OLC algorithm and send it to the corresponding H-Sensor. All the L-Sensors use MCFR routing again to adjust the routes toward the sink, and the logical topology can be updated. Simulation results show that the network topology constructed by the MCFR strategy and the OLC algorithm can reduce the communication delay, balance the network node load and prolong the network lifetime, so it can meet the requirements of real-time continuous data collection.