Research On Energy Optimization Routing Algorithm Of WSNs Based On Mobile Sink

Wireless Sensor Networks(WSNs)are communication network,which are composed of a large number of low-cost micro-sensors through self-organization and cooperative interconnection.They have the characteristics of high reliability and low deployment cost.The introduction of mobile sink alleviates transmission hotspot caused by static sink node,and broadens WSNs application scenarios,making WSNs more extensively used in military,medical,civil and other related fields.However,the limited energy of nodes,frequent location updates of mobile sink,data transmission congestion and other factors will lead to the degradation of network monitoring performance and network lifespan.Therefore,it is of great significance to design efficient routing algorithms to optimize node energy,balance network load and prolong network life cycle.Based on the research background of WSNs energy optimization,the adaptive cluster head selection strategy is proposed respectively in combination with the changes of energy consumption of nodes and the parameter changes of the moving state of the mobile sink,so as to make the selection of cluster head more dynamic and energy-saving.Meanwhile,a dynamic hierarchical transmission multi-hop path optimization strategy is designed to optimize the data transmission path.Then,based on the idea of Qlearning,the hierarchical transmission multi-hop path optimization strategy is constructed.The major contributions of this paper are as follows:Aiming at high network energy consumption and data delay induced by mobile sink in WSNs,this thesis proposes a cluster-based energy optimization algorithm called CEOMS(Cluster-based Energy Optimization with Mobile Sink).Firstly,according to the moving characteristics of the moving sink node,the motion performance function is constructed to increase the probability that the remote sink node becomes the cluster head.Then,combining the average residual energy rate and node distribution density of sensor nodes within the neighborhood radius,the energy density function is designed,so that the nodes with high residual energy are preferred as cluster heads.Secondly,the initial cluster head selection threshold including energy density function and motion performance function is constructed based on traditional clustering routing algorithm.Finally,the adaptive cluster head selection threshold is adjusted adaptively by considering the density of surviving nodes and calculating the mortality of nodes.The simulation results show that the proposed algorithm increases the adaptability of cluster head selection,alleviates data delay,increases network life,and reduces network energy consumption.Aiming at frequent position updates of mobile sink results in excessive energy consumption of sensor nodes and discontinuity of data transmission in WSNs,this thesis proposes a dynamic spanning tree-based routing algorithm called DSTMS(Dynamic Spanning Tree with Mobile Sink).According to the motion parameters of mobile sink,DSTMS algorithm designs the adaptive position updating strategy to select the sensor node in the local position updating region as the rendezvous point to receive the position information of the mobile sink firstly.Then,based on the LEACH structure,the maximum path energy efficient function is introduced to continuously add the unoptimized cluster tree to the optimized spanning tree,so as to construct the multi-hop transmission path tree.Finally,a dynamic spanning tree of mobile sink nodes under the moving state is constructed to reduce the energy consumption of long-distance transmission network.The simulation results show that the proposed DSTMS algorithm can efficiently prolong the network life,improve the stability of data transmission,save network resources and balance network workload.Aiming at high energy consumption caused by segmental optimization in the construction process of dynamic multi-hop transmission path tree in wireless sensor networks,this thesis proposes an improved dynamic multi-hop routing algorithm based on Q-learning dynamic spanning tree-QLMS(Q-Learning with Mobile Sink)by combining the algorithm framework of Q-learning and comprehensively considering the mobile performance parameters of mobile sink,the current remaining energy of the network node and the energy consumption of data transmission path.Firstly,the algorithm considers the data transmission relationship among the mobile sink,rendezvous point and cluster head,and constructs the adjacency matrix to limit the hierarchical transmission.Secondly,by considering the energy consumption and residual energy of data transmission between nodes,the reward function is constructed to calculate the reward value corresponding to the appropriate next hop node for each node.Then,the Q-learning algorithm framework is adopted to obtain the next hop node with the largest response value,so as to construct the dynamic multi-hop transmission path tree between cluster head,rendezvous point and mobile sink,which can optimize the data transmission path,effectively reduce the energy consumption of the entire network transmission,and increase the network life cycle.