Research On Intelligent Optimization Technology Of MANET Routing Strategy

As a wireless transmission,self-organizing and infrastructure-independent distributed multihop network,the mobile ad hoc network has the advantages of low-cost,fast and flexible deployment,and is widely used in military,rescue and outdoor fields.However,due to the limitations of its dynamic topology,limited resources and time-varying links,the routing protocol of the wireless communication network controlled by the traditional center is no longer fully applicable to the decentralized mobile ad hoc network,which makes the research of mobile ad hoc network routing protocols has received extensive attention.Currently,many routing protocols have been proposed for application scenarios of mobile ad hoc networks.However,these routing protocols do not have the ability to dynamically adapt in large-scale networks in complex scenarios,and it is difficult to meet the quality of service(QoS)requirements of wireless communication networks such as low latency,high throughput and load-balancing.The focus of this dissertation is to design the intelligent optimization routing strategy of mobile ad hoc network to achieve this requirement.On the one hand,in response to the low latency and high throughput requirements of largescale complex dynamic networks,this paper combines deep reinforcement learning technology to design a prioritized replay double-deep Q network(PRD-DQN)based intelligent routing scheme.This scheme explores and learns the network status by designing routing packets,and designs a routing reward function according to the congestion level of the node and the link quality,so that the node can adaptively obtain the best routing strategy in the dynamically changing network.Through numerical simulation and performance analysis,this scheme has obtained better QoS guarantee under different network scales,which provides an effective solution for the existing intelligent routing strategy cannot be effectively applied to the decentralized mobile ad hoc network.On the other hand,in response to the requirements of low latency and load balancing in largescale high-speed transmission networks,this paper proposes an optimized link state routing(OLSR)based main and standby optimized link state routing(MS-OLSR)optimized routing scheme.The scheme first designs an adaptive packet sending strategy that is aware of mobile node location and topology changes to reduce the cost of route establishment.Next,the program designed a main-standby routing strategy calculation algorithm based on the main-standby multipoint relay(MPR)node,which aims to ensure the load balance of the network by avoiding the MPR node in OLSR from becoming a hot node in the network.Finally,based on the analysis of the results of software simulation and hardware testing,this solution effectively reduces routing overhead while having lower delay and fairer traffic distribution compared to other routing algorithms.