Intelligent Localization Method for Hollow Nodes in Mobile Communication Networks Based on Deep Reinforcement Learning
When locating hollow nodes in mobile communication networks,it is difficult to achieve an ideal location effect due to the local optimal problem in the path selection stage.Therefore,an intelligent location method of hollow nodes in mobile communication networks based on deep reinforcement learning is proposed.In the stage of determining the location information between different nodes in the mobile communication network,the distance information of each hop is quantized by weighted processing,and then the positioning step of the hollow node in the mobile communication network is determined by combining the weighted average hop distance and the distance between the node and the anchor node.In the stage of location of hollow nodes,the grey wolf algorithm in deep reinforcement learning is introduced,and the original grey wolf algorithm is optimized by simulated annealing and chaotic mapping from the point of view of rapid convergence,and the final grey wolf running around is taken as the location result of hollow nodes in mobile communication network.The test results show that the design of location method can reduce the path overhead in the location stage of hollow nodes,and has a good performance in the location accuracy,and the specific location efficiency is also significantly improved.
deep reinforcement learninghollow nodes in mobile communication networksintelligent positioningsimulated annealingchaotic mapping
万方数据
维普
