Research On Energy Efficiency And Security Optimization Of Heterogeneous Wireless Networks

With the rapid development of wireless communication,the network scale is getting larger and larger.Especially in heterogeneous networks,as the number of users increases,the number of base stations increases significantly,and the wireless network environment becomes more and more complicated.Energy efficiency optimization and resource allocation issues put forward to an urgent problem.Deep reinforcement learning algorithm can effectively allocate resource dynamically and reduce computational complexity.At the same time,with the development of wireless access technologies in heterogeneous networks and communication security attention,users need reasonable performance and security requirements.Therefore,this thesis studies energy efficiency optimization and joint optimization of user data rate and communication security level based on the heterogeneous wireless network scenario.The main work of the thesis is as follows:Firstly,the thesis studies the power allocation and energy efficiency optimization problem in the scenario where consists of macro base station and multiple cellular base stations.Such the problem is an NP-hard problem.Unlike traditional optimization decomposition algorithms,this thesis uses deep reinforcement learning algorithm to solve this problem.Our goal is to maximize dynamically the system energy efficiency of the entire network when macro base station and multiple cellular base stations are randomly distributed and interference between each other.Firstly,we build a system energy efficiency optimization model based on the scenario where macro base station coexists with multiple cellular base stations,and then discretize the size of the transmit power and restrain macro base station throughput to reduce the excessive state space and deep neural network dimensions in deep reinforcement learning,and improve the deep learning module in Nature DQN.Finally,the simulation results show that Nature DQN not only improves the energy efficiency of the system,but also increases the convergence speed.Secondly,in order to study user wireless access selection and communication security level optimization in heterogeneous wireless network scenario,we consider heterogeneous wireless network where consists of multiple wireless access technologies and transmission modes.We consider the user number limit and user service quality requirement to construct a joint optimization model for user data rate and base station communication security level that exist in multiple transmission modes.Then,in the multiple objective optimization function,we use weight parameter to represent need preference of users.Finally,we re-linearize the mathematical model and useCPLEX commercial optimization software to solve this optimization problem.The simulation results show that the heterogeneous wireless network with various wireless transmission modes can meet the need of users,and the network throughput and outage probability are significantly improved.The weight parameter can represent the need preference of users.