Markov Approximation Based Access Selection And Resource Allocation Algorithm In Heterogeneous Networks

With the rapid development of communication technology,the heterogeneous network based on the coexistence of multiple wireless network and services has attracted huge interest from industry and academia.The access technologies,resource management mechanisms and quality assurance of service of different wireless networks are different,so multiple wireless networks must work together to meet the needs of users.In heterogeneous networks,how to select network to access and allocate resources to make full use of network resources is a heated research.Most of these access selection and resource allocation problems are combinatorial problems.For such Non-deterministic Polynomial(NP)complete problems,a Markov approximation method is introduced.According to the constructed Markov chain,we could design a distributed algorithm.As the Markov chain converges,the system will probabilistically jump and achieve the approximate optimal system performance when reaching the stationary distribution.The main research contents of this paper are as follows:To meet users' requirements of quality of service(Qo S)and achieve load balancing between base stations(BSs),we study a joint access selection,resource allocation and interference mitigation problem in heterogeneous cellular networks.In order to solve this complex joint optimization problem,we design a distributed algorithm,which uses Markov approximation method.During the implementation process of the algorithm,dynamic resource allocation is carried out according to the change of users' access selection and Qo S requirements,which is based on the spectrum conflict and spectrum reuse between BSs.In addition,we set the power charge in the optimization target to achieve load balancing.Simulation results show that,the generated distributed algorithm can not only coordinate the three sub-problems well to ensure users' Qo S requirements,but also effectively achieve load balancing between BSs.To meet the requirement of user experience,we propose an access selection algorithm based on quality of experience in heterogeneous wireless networks.When making network access selections,users need to consider the influence of various attributes.In addition,users in different scenarios have different preferences for various attributes.By using the analytic hierarchy process(AHP),we could determine the attribute weights under different service demands and obtain the utility value of accessing to different networks.Users make network access selections to maximize the system utility,which is a combinatorial optimizationproblem.The Markov approximation method can effectively obtain the approximate optimal selection by generating a distributed algorithm.Simulation results show that the combination of AHP and Markov approximation can effectively solve the problem of network access selection with considering users' experience.To meet the requirements of users for different services in heterogeneous wireless networks,we propose a multi-service resource allocation algorithm.Based on practical application requirements,two typical services of constant bit rate(CBR)and dynamic bit rate(DBR)are considered.Since the constraints of bandwidth resources and power resources,it is very important to select network access and allocate resources to guarantee users' multi-service requirements and achieve optimal network performance.We transfer this complex optimization problem into network topology updating problem,which is caused by the change of network access selection,and resource allocation problem under fixed access selection.These two problems are solved by Markov approximation method and Lagrange primal dual method respectively.Simulation results show that the combination of the two methods can not only ensure users' multi-service requirements but also effectively achieve the optimal network performance.