Research On Intelligent Routing Model And Algorithm In Cognitive Networks

Cognitive networks(CN)refer to the networks with learning and reasoning capabilities.It is of great value to promote cognitive networks to actual application in diverse communication scenarios.However,constrained by factors such as harsh geographic environment,strong electromagnetic interference,and limited node energy,cognitive networks in complicated environments are prone to be destroyed due to excessive energy consumption of individual nodes.Hence,how to effectively increase the remaining energy of nodes and extend the lifetime of the networks has become an essential issue in the design of cognitive networks in complicated environments.At the same time,the existing cognitive routing technologies are mainly dedicated to optimizing Qo S,and there is still room for improvement in terms of energy consumption.Therefore,the purpose of this thesis is to optimize the energy consumption of cognitive networks by designing routing algorithm.In response to the above problems,this thesis focuses on the following three aspects.First,a cognitive network model oriented to complicated environments is proposed.In view of the problems and characteristics of cognitive networks in complicated environments,the communication devices distributed in fixed places in the network are defined as Primary Nodes(PN),and mobile communication devices are defined as Secondary Nodes(SN),that is,cognitive nodes,explaining the role and working mechanism of nodes in the network.The spatial distribution of cognitive nodes is also demonstrated.Secondly,this thesis analyzes consumption of cognitive nodes in specific network scenarios.The Gamma distribution model is introduced into the energy consumption statistics to analyze the energy consumption of the two-hop and multi-hop cognitive network respectively,and derive the energy consumption statistical model.In the experiment,the network energy consumption is analyzed based on the proposed energy model.The experimental results show that the node communication distance is the main factor affecting the energy consumption of the network.This model provides a basis for future research on optimizing performance of cognitive networks.Furthermore,this thesis proposes an energy optimization routing algorithm for cognitive networks based on Q-learning.Considering the energy balance and Qo S requirements of cognitive routing in complicated environments,the routing problem is modeled as a partially observable Markov decision process in continuous time,and the reward function is designed based on the remaining energy of the nodes and the communication rate to make the network routing decision process achieve energy balance between nodes and maximize remaining energy.In the simulation experiment,the influence of discount factor and weighting factor on the remaining energy of the network is evaluated.Besides,performance of several Q-routing algorithms is compared in the experiment.It is shown in the results that in the application scenario of this thesis,the proposed algorithm can achieve better performance in terms of end-to-end delay and network lifetime,which is of practical significance.