Research On 3D Underwater Sensor Networks Coverage Optimal Control Algorithm

In the ocean,there are a large number of scarce resources such as minerals that can be exploited and used by human beings,so how to explore and develop underwater resources has become a major issue that human beings need to solve.With its unique advantages,underwater sensor network has been widely concerned by scholars both at home and abroad.In order to realize the effective monitoring of the target object by the underwater sensor network and the complete acquisition of the objective physical information,it is necessary to ensure that the underwater sensor node can effectively cover the target object.Therefore,in this dissertation,the coverage of underwater sensor networks is the main research content.In order to prolong the network life cycle and save energy efficiently,the coverage deployment,coverage maintenance(high density deployment scenario and dynamic water environment deployment scenario),and interference control of sensor nodes are studied.Theoretical analysis and simulation experiments have verified the effectiveness of the proposed algorithm.The main contributions are as follows:(1)Due to the high dynamic and uncertainty of the monitoring target caused by the mobility of underwater environment during the deployment of underwater sensor nodes,a distributed hybrid fish swarm optimization algorithm based on water flow impact and artificial fish swarm system operation is proposed.By simulating the foraging behavior of the fish,the sensor nodes autonomously tend to monitor the target;by setting the congestion degree,the distribution of the sensor nodes is consistent with the distribution of the monitoring targets.At the same time,by constructing the information pool,all sensor nodes in the connected state can share the perceptual information,thereby enhancing the global search capability of the sensor nodes and reducing the blindness in the process of moving the sensor nodes.Finally,the effectiveness of the proposed algorithm is demonstrated by simulation tests.(2)Considering the problem of underwater sensor network coverage retention in high-density deployment scenarios,a coverage control scheme based on cultural genetic optimization is proposed.Based on the redundancy of deployment nodes,a node sleep-wake scheduling mechanism is established.The scheme mainly includes two parts:(1)node sleep scheduling based on cultural genetic algorithm.Under the premise of ensuring network monitoring performance,only some nodes are scheduled to work and the redundant nodes are in a low-power sleep state,thereby saving energy consumption and prolonging the network lifetime.The goal of node scheduling is to find a minimum set of nodes that can cover the monitoring area,and the cultural genetic algorithm can efficiently solve the problem;(2)wake-up scheme.During network operation,the wake-up scheme wakes up the sleepy node to recover coverage holes caused by dead nodes,thereby maintaining high coverage.This solution can not only reduce network energy consumption but also take into account the monitoring coverage of the network.(3)Based on coverage maintenance of underwater sensor networks in dynamic water environment deployment scenarios,a heuristic sleep scheduling scheme is proposed.This scheme dynamically determines a sufficient number of active nodes in the underwater sensor network at different points to cover the targets to be monitored and allows sensor nodes and autonomous underwater vehicles in the network to dynamically select sleep or work to adapt to the changing environment.A special static scenario of the problem has been shown to be NP-complete.Therefore,the proposed heuristic disjoint coverage set algorithm solve this dynamic problem.Through simulation testing,the algorithm exhibits high performance in extending network life cycle,robustness,and computation time.(4)Considering the problem of interference management in underwater heterogeneous networks,a distributed resource allocation scheme based on cooperative reinforcement learning is proposed.In this scheme,reinforcement learning in intelligent control is introduced into underwater acoustic communication network.Sensor nodes are regarded as agents and sensor networks as multi-agent networks.By dividing the state space and the action space,the income function and search strategy are established.When the channel state information is unknown and the network interference level around the transmitting node is not estimated,the scheme can optimize the resource allocation by continuously interacting with the environment to reduce the network interference level and improve the resource utilization.In addition,the convergence of the proposed algorithm is also proved.Finally,compared with the previous work,the proposed algorithm can increase the system capacity by more than 5 times,and the execution efficiency of the solution can be significantly improved.