Research On Coverage Control Of Mobility-assisted Sensors In Wireless Sensor Networks

The deployment of large groups of autonomous vehicles is becoming possible because of technological advances in networking and in miniaturization of electromechanical systems. These sensor nodes integrate information perception, data processing and wireless communi-cation, and other functions. The sensor nodes in wireless sensor network(WSN) can collabo-ratively work real-time monitoring and gathering all kinds of environmental information, and transfer detailed and valuable data to base station for processing under unsupervised condition.Coverage holes may exist if the number of deployed nodes is not large enough in one random deployment, or some nodes have run out of energy. In such a case we need redeploy the sensors to increase the coverage rate of the network under certain movement strategy. The coverage control is one of the basic problems in WSN. And it considers how to deploy the nodes of the sensor networks in order to maximize the target area covered by the network sensor nodes. There are many applications of coverage control such as tracking unfriendly targets, environment information sampling or monitoring, and area search and rescue. The coverage control of mobility-assisted sensors is also called movement strategy of mobility-assisted sensors. The paper considers the movement strategy of mobility-assisted sensors when the coverage holes exist after initial deployment of the sensor nodes, analyses the movement strategy based on Visual Forces and the movement strategy based on Voronoi Diagram. And we proposed three different movement strategies. the main works are as follows:1. We analyse the Visual Forces algorithm, and we propose a new algorithm. The new algorithm adds an coverage hole guidance force based on two forces. The coverage hole guid-ance force prompts the nodes to move towards the coverage holes. The new algorithm keeps the movement distance of sensors least, and thus it saves the energy consumption. The computer simulation shows that the convergence rate of the new algorithm is the fastest, and the coverage rate is the highest, and the movement distance is the least among the Visual Forces algorithm and Uncovered Region Explore algorithm.2. We analyse the swarm intelligent algorithms including PSO(Particle Swarm Optimiza-tion), GA(Genetic Algorithm), DE(Differential Evolution), which are used to solving optimal problem. In view of the sensor network consisting of heterogeneous sensors which have d-ifferent coverage sensing abilities, we propose a new algorithm——CHDDE(Coverage Hole Directed Differential Evolution). The new algorithm chooses the local coverage rate as the objective function of DE, and introduces the coverage hole guidance force to the evolution pro-cess. The simulation results show that the new algorithm can effectively improve the network coverage rate and reduce the movement distance of the nodes.3. We consider the movement strategy based on Voronoi Diagram. We propose a continu-ously differentiable coverage optimal function, and compute the gradients of the function with the help of Voronoi Diagram. Through the mathematical deduction we prove that positions where the function gradients equal to zero are Voronoi Diagram centroids. In order to reach the uniform distribution of the network, we move the sensors to the centroids. The comput-er simulation shows that the new movement strategy has the very good convergence rate and robustness.