Research Of Dynamic Reconfigurable Network Mechanism Under The Condition Of Node Failure

With the development of information technology and the increase of demand for network transmission resources,the traditional method like base station network deployment gradually presents drawbacks in many scenarios.With different demand being needed gradually,reasonable resource allocation has become the main direction of network research.The requirement of network is closely related to users' requirements.Therefore,reconfigurable network has a lot advantages camparing to the traditional network-it can change its network form with the change of service mode.In the application process of reconfigurable network,the network will suffer different degrees of damage due to physical damage such as bad natural conditions,operation accidents,large-scale network intrusion or strong noise interference.As a result of the network follows the multi-hop network mode,the long-distance information transmission will be broken or affected under certain probability physical space,which will seriously interfere with the production,work and communication between nodes,links and networks.In order to optimize the resource allocation of the remaining nodes in the network to meet the needs of the original network,this dissertation takes MWSN(Mobile Wireless Sensor Networks)and MANET(Mobile Ad-hoc Networks)as the research objects,and researchs the optimization algorithm of the behavior of autonomous controllable nodes.In this network state,the path formed between two remote nodes is very sensitive to the impact of unpredictable node movement.At the same time,this network state generally has limited resources.The path used for data transmission must test the link quality and time for data forwarding.In the case of resource constraints,the main contributions of this dissertation are as follows:In view of the relationship between node mobility and topological structure,this dissertation establishes a network model based on graph theory and communication energy consumption.The mobility of nodes is divided into two states: controllable and uncontrollable.Considering both mobile energy consumption and communication energy consumption,this dissertation analyses the network quality and the lifetime of the network.At the same time,this dissertation completes the repair behavior research of dynamic reconfigurable network based on graph theory.The dynamic reconfigurable network node group is modeled from the perspective of communication and graph theory,and the parameters are tested to work in various environments.For network repair and network performance optimization,this dissertation will propose a MMT(Mobile Model Transformation)algorithm,which is based on ant colony algorithm and the algorithm for estimating link quality and link delay of each node.The network repairment and optimization are divided into node location allocation algorithm,link reinforcement algorithm,link disassembly-build algorithm and energy balance algorithm.The algorithm takes link quality and delay as optimization objectives,and optimizes cluster head selection mechanism in energy balance algorithm.While balancing the energy cost and effect of recovery,the algorithm increase the lifetime of dynamic reconfigurable network.In this dissertation,the simulation results will show that the node repairing behavior of the proposed algorithm can get some effect of recovery,and detect the network performance rate of recovery under the specific mobile power of the algorithm in this situation.The simulation results will show that the proposed algorithm can effectively reduce packet loss and delay to improve reliability and increase network lifetime compared with existing protocols.