Research On Routing Technology Of Wireless Monitoring Network In Urban Integrated Pipe Gallery

The urban integrated pipe gallery is a typical long and narrow semi-closed scene.Its wireless monitoring network communication faces challenges such as reliability and real-time.It reduces the reliability of point-to-point communication such as multipath fading,and reduces the reliability and real-time of end-to-end communication such as multi-hop transmission.This paper combines the characteristics and needs of wireless monitoring network of urban integrated pipe gallery,researches routing technology,and improves QoS indicators such as system reliability and real-time performance through optimized routing.China’s urban integrated pipe gallery is still in its infancy.The location and number of wireless communication nodes deployed in the corridor are limited,so it is suitable to adopt a simple and effective line-star network.In order to improve the reliability of routing,this paper designs a dual-line network with redundant routing based on linear network,proposes a multi-hop link optimization retransmission time slot method,and uses genetic algorithm to solve slot allocation scheme to improve end-to-end routing reliability.Aiming at the randomness of genetic algorithm mutation operator,a genetic mutation operator and a chromosome mutation operator based on adaptive selection strategy are proposed to accelerate the convergence speed of the algorithm.Furthermore,considering the heterogeneity of communication network and the difference of communication cycle,multi-cycle scheduling is adopted in star network,combined with the difference of communication transaction scheduling period and the priority transmission of communication transaction,a multi-cycle scheduling model is established.We propose a new genetic algorithm,RLGA by combining reinforcement learning,and a multi-period scheduling strategy based on RLGA is designed to allocate time slots effectively and improves the real-time communication.With the large-scale development of the urban integrated pipe gallery,there will be more and more internal equipment,the topology of the corridor network should be more universal,and the QoS constraints are more diverse.Therefore,this paper further discusses the multipath routing of multi-QoS constraints in the random network topology.According to the internal resources and QoS constraints of the integrated pipeline monitoring network,the communication transaction is modeled.We propose a new genetic algorithm,GA-VNS by combining variable neighborhood search,which generates a set of optimal solutions as a multipath routing scheme.Furthermore,A SAW-AHP algorithm based on QoS equilibrium model is designed to evaluate the quality of redundant routes,routing algorithms and select the optimal routing scheme.In view of the poor performance of individual indicators of the optimal solution,the idea of feedback control is introduced,and the Q-learning method in reinforcement learning is used to optimize the weight of indicators.