Research On Routing Optimization Method In Intelligent Multimode Network Communication System

With the rapid development of information technology today,Mobile Ad hoc Network has been widely used in civil environment and military fields.At the same time,artificial intelligence has gradually become a new driving force for the information technology industry.The intelligent multimode network communication system is based on the network communication system widely used in the market,applies software radio technology,and introduces machine learning algorithms.As an adaptive mobile ad hoc network communication system,it is used in emergency communication scenarios.However,in various emergency communication scenarios,the energy of intelligent multimode network nodes is limited,the network topology changes greatly,and the communication delay requirements are relatively high.Based on the OLSR(Optimized Link State Routing)protocol with low latency,this paper studies the optimization of routing protocols for intelligent multimode network communication systems.The main research contents of this dissertation are summarized as follows:1)The MPR(Multi-Point Relay)set selection optimization algorithm of the OLSR protocol.In order to solve the problem of time delay and control overhead performance degradation in the intelligent multi-mode network application in the selection algorithm of OLSR protocol MPR set,this paper proposes an OLSR protocol OLSR-M which is an optimization algorithm for MPR set selection based on fuzzy logic comprehensive evaluation of link quality,the optimization algorithm that considers link stability,node load size and remaining energy comprehensively to improve the stability of the local network topology of the MPR set,balance node load and energy consumption,and overcome the difficulties of intelligent multimode network topology changes and energy constraints,to improve the overall performance of the network.The simulation results show that,in terms of delay,packet delivery rate,throughput and energy consumption,the OLSR-M protocol is superior to the OLSR protocol,but in low dynamic scenarios,its control overhead is obviously insufficient.2)TC(Topology Control)message sending interval optimization algorithm of OLSR protocol.In order to reduce the control overhead of the OLSR-M protocol in low dynamic scenarios and adapt to the characteristics of intelligent multimode network topology changes,this paper proposes an OLSR-M-T protocol based on the OLSR-M protocol using the TC message adaptive transmission optimization algorithm.When the topology change is small,the optimization algorithm increases the sending interval of TC messages and reduces the routing control overhead;when the topology changes greatly,it reduces the sending interval of TC messages to adapt to the rapid change of topology.The simulation results show that the performance of the OLSR-M-T protocol is comparable to the OLSR-M protocol in terms of delay,packet delivery rate,throughput and energy consumption,but the control overhead of the OLSR-M-T protocol is better than that of the OLSR-M protocol in low dynamic scenarios protocol.3)Research on OLSR path optimization based on Q-learning.Intelligent multimode network communication systems are caused by unstable or constantly changing OLSR routing paths due to differences in network topology changes,unbalanced load,and limited energy.In response to this problem,this paper introduces a reinforcement learning mechanism,which takes link stability,load size and remaining energy as the basis for path selection,and proposes a routing strategy based on Q-Learning to calculate routing paths with higher global stability and reliability in intelligent multimode networks.The simulation results show that the delay,packet delivery rate,throughput,energy consumption and control overhead performance of the OLSR protocol based on Q-Learning(OLSR-Q protocol)and the OLSR-M-T protocol based on Q-Learning(OLSR-M-T-Q protocol)are better than the OLSR protocol.