Research On Network Selection And Resource Allocation In Heterogeneous Wireless Networks

Higher transmission speed, larger system capacity, wider coverage of telecommunication service and more rational resource allocation are the future directions of current mobile communication system. Effective cooperation of the heterogeneous networks is the inexorable trend of network technology. In heterogeneous wireless networks, radio resource management has become one of the most important problems. Researches on network selection and resource allocation scheme have attracted lots of attention from the academia community.This paper focuses on the study of network selection and resource allocation algorithms in heterogeneous wireless networks. They are resource management architecture, vertical handover algorithms, decision tree based network selection and non-cooperative game theory based approach for bandwidth allocation in heterogeneous wireless networks.With respect to the radio resource management architecture in heterogeneous wireless networks, this paper mainly analyzes the goals that future heterogeneous wireless networks aimed and corresponding requirements. In addition, the paper analyzes the main radio access technologies standards and 3 GPP and IEEE protocols, finding that the current radio access technologies standards all support the handover between different radio access technologies, which provide a good foundation for the future heterogeneous wireless networks convergence.Considering the study of network selection algorithm study in heterogeneous wireless networks, this paper discusses the traditional statistic network selection algorithms and dynamic network selection algorithms in heterogeneous wireless networks. Generally speaking, statistic network selection algorithms generally include rule based network selection algorithm, fuzzy logic based network selection algorithm and vertical multi-attribute based network selection algorithm. At the same time, dynamic network selection algorithms usually includes Markov decision process based network selection algorithm for heterogeneous wireless networks and Q-learning based radio resource management. Besides, advantages and disadvantages of these network selection algorithms, with respect to the suitable conditions and applicable network environment, are also analyzed.In respect of decision tree based network selection in heterogeneous wireless networks, this paper proposes a decision tree based network selection algorithm in heterogeneous wireless networks, which utilizes the multi-attribute of network and traffic as the attribute nodes of decision tree. Differed from traditional multi-attribute decision algorithms, the proposed algorithm only requires part of the key attributes, which can bring significant improvement of decision efficiency.Then, this paper proposes a distributed bandwidth allocation algorithm based on non-cooperative game for the problem of bandwidth allocation in heterogeneous wireless networks. Proof of the existence of Nash Equilibrium solution is given. Each type of network can reach its stable state using the proposed algorithm by maximizing its utility function. The simulation results show that the proposed algorithm is effective.Finally come the conclusions of the paper. Future directions are also discussed in the final part of the paper.