Research On Energy Hole Avoidance Method Of 3D Underwater Senor Network Based On Position Adjustment

With the continuous advancement of the national marine development strategy,the underwater acoustic sensor network technology has been developed rapidly and has been applied to many fields.At the same time,the technology also faces certain challenges.In the underwater acoustic sensor network system,most of the sensor nodes are equipped with batteries with limited energy,and most of them are placed in deep underwater areas,and the replacement cost is relatively high.Therefore,it is of great significance to improve the energy utilization rate of the nodes.Most of the underwater sensor nodes are randomly distributed in a wide area of water,and the distance between them and the SINK node on the surface is large.Therefore,multi-hop routing is often used to realize the bottom-up transmission of data,which causes the upper nodes to have a large energy load and consumes energy quickly.However,the lower node still has a large amount of energy remaining,so an energy hole problem occurs.The energy hole problem has been extensively studied,but the proposed methods are mostly impractical or unsuitable for underwater 3D network scenarios.Aiming at the randomly deployed three-dimensional underwater sensor network scenario,this paper proposes the use of node depth rotation strategy and node position exchange strategy to avoid the energy hole problem.In order to achieve the above goals,this paper mainly carried out the following research work:(1)In a randomly deployed network,the movement of nodes will bring about topological changes,leading to impaired network connectivity.In order to solve this problem,this paper proposes a grid clustering model,which divides the entire threedimensional underwater area into multiple unit grids.Each unit grid can be used as a network cluster.Through theoretically calculated constrained conditions,the grid clustering model enables the data transmission unit to transform from a single node to a single grid,and from irregular distribution to regular distribution,which solves the problem of routing mixing caused by random distribution of nodes and makes the network structure more clear and easy to manage.(2)In order to avoid the energy hole problem,based on the above model,this paper proposes a deep rotation mechanism based on cluster heads and a mechanism for cluster head exchange based on geographic location.Both mechanisms can achieve position rotation between cluster heads.This makes the load between cluster heads more balanced.Furthermore,in order to ensure the load balance in the cluster,this paper proposes a rotation strategy for the cluster heads in the cluster,which dynamically selects the appropriate node as the cluster head according to the remaining energy.Through the position adjustment mechanism between the cluster heads and the cluster head rotation strategy within the cluster,the load balance of the entire network can be effectively realized.(3)In order to ensure the effective operation of the cluster head rotation mechanism and the cluster head rotation strategy within the cluster,this paper designs related information exchange protocols.The paper design the details of network data packet,then designs the cluster head selection protocol to select the cluster head,and realizes the multi-hop routing of the network through the routing protocol.Because the position rotation between cluster heads and cluster head rotation in the cluster will cause changes in the network topology,this paper designs and implements a network topology reconstruction protocol to ensure the feasibility of the rotation strategy.The paper has carried out extensive performance evaluation of the abovementioned methods.Through experimental comparison,it can be found that when the entire grid clustering model is running,using a round-robin strategy can significantly improve the balance of node load and effectively extend the life of the entire network.