With the rapid development of embedded computer technology and wireless communication technology, large-scale wireless sensor networks began to spread across the globe. Coverage of wireless sensor networks is directly related to the sensor network perception in the monitoring area. Owing to the forms of arrangement in wireless sensor networks are variety, including aircraft spreading and artificial arrangement, etc., the form of coverage in wireless sensor networks is flexible. Unlike the traditional network coverage, coverage in wireless sensor networks not only focus on how to improve the geographical distribution of wireless sensor nodes to complete the monitoring tasks, but also focus on how to make the distances between nodes no larger than the communication distance for network communication. Moreover, the coverage will also focus on energy and life issues. Good coverage control not only enhances the quality of the perceived information in the area or target monitoring task, but also reduces the energy consumption and prolong the network lifetime.In large-scale wireless sensor networks with a large number of sensor nodes, sensor nodes must cover the targets or certain area in the monitoring area for the perception task. Without the coverage of the target or monitoring area, wireless sensor networks will not be able to complete the goal of perception task and will lose availability. Similarly, if there is no suitable node topology control and duty cycle during the process of monitoring task, the energy of network will soon run out and the survival of the network would be threatened. This dissertation first analyzes the characteristics of wireless sensor network coverage, then the use of evolutionary algorithms to solve several key problems for large-scale wireless sensor network coverage. The main innovations are as follows:(1) In the self-organizing sensor networks, enhancing the target coverage rate with a limited number of sensors and monitoring capabilities in the target coverage area is a key issue in wireless sensor networks target coverage problem. In order to solve the problem, we propose a method based on quantum ant colony evolutionary algorithm for self-organizing wireless sensor networks target coverage, and we build the corresponding system model. The method use the quantum state vectors for ant colony algorithm encoding, and use quantum rotation gate for ant routes dynamically adjusting. By using parallel search with multiple quantum ants, the quantum ant colony evolutionary algorithm could increase the search range and achieve a parallel quantum evolution. In the self-organizing wireless sensor network environment, the proposed method and the target coverage method based on genetic algorithms and simulated annealing are simulated and compared. Simulation results show that with the different number of sensor nodes and perception radius, the target coverage rate of the proposed method are increased around 10 percentage and 20 percentage compared to the genetic and simulated annealing algorithm respectively. The number of targets that successfully detected by the proposed algorithm is 6.90% to 19.09% higher than the method based on genetic algorithm, and 32.67% to 54.27% higher than the method based on simulated annealing. So the proposed method significantly enhanced the monitoring results.(2) Enhancing the network lifetime with the duty cycle method for two-dimensional wireless sensor networks with full coverage of all monitoring targets is a key issue in target coverage problem. However, selecting the order of duty cycle is an NP-hard problem for large-scale wireless sensor networks, and the computational complexity of an exhaustive search for all possible duty cycle schemes is too high. In this paper, a new quantum immune clonal evolutionary algorithm is proposed to solve the duty cycle selection problem with full coverage constraint. In order to improve the coding efficiency, the method maps the issue and its solutions to antigens and antibodies into q-bit forms. By using quantum rotation gate for antibody mutation, the algorithm accelerates the convergence rate. In the wireless sensor network environment, the proposed method is simulated and compared with simulated annealing algorithm based method and genetic algorithm based method. Simulation results show that with the different number of sensor nodes and monitored targets, the proposed method based on quantum immune clonal evolutionary algorithm not only maintains full coverage of all the targets in the monitoring area, but also extend the lifetime of wireless sensor network by 2.85%to 6.62% compared to the method based on genetic algorithm, and extend the lifetime by 3.95% to 8.61% compared to the simulated annealing algorithm based method, which enhances the energy efficiency.(3) With the continuous development of wireless sensor network technology, clustering coverage has become a key issue in the study of wireless sensor networks. In order to reduce communication energy consumption, a fuzzy simulated evolutionary computation clustering method for wireless sensor networks coverage is proposed. In order to achieve a global optimization, the method dynamically selecting the wireless sensor network cluster heads with fuzzy controller and simulated evolutionary computation is proposed. Simulations are conducted by using the proposed method, the clustering methods based on particle swarm optimization and the method based on quantum evolutionary algorithm. Simulation results show that with different number of sensor nodes and the proportion of cluster heads, the energy consumption of the proposed method decreased 2.34% to 36.02% compared to the method based on particle swarm optimization and decreased 18.41% to 61.31% compared to the method based on quantum evolutionary algorithm, which means the proposed method significantly improves the energy efficiency. |