Research On Multi-factor Dynamical Decision Based Access Selection Algorithm In Heterogeneous Wireless Networks

The coexistence of a variety of wireless access networks is one of the main features of the next-generation wireless communication systems. The networks complement each other, as well as seamlessly integrate into a unified environment for heterogeneous wireless networks, which provide users with the best service experience anywhere. Heterogeneous terminals with multiple wireless interfaces could achieve seamless mobility in the heterogeneous network environment. How to make full use of network resources to get multi-access gain has already been a key issue for heterogeneous network radio resource management. Heterogeneous network access selection algorithm, as one of the sticking technologies of radio resource management, impacts network resource utilization and user perceptions directly, is currently a major research topic.This paper introduces an OFDM based dynamic resource allocation model, which defines a two-dimensional unit constructed by a time slot and a sub-carrier of OFDM systems as the basic resource unit. It is the basic unit of resource allocation. Taking QoS requirements of real time and non-real-time services, link adaptive retransmission mechanism and some other factors into consideration, the resource demand and network payload in packet access are estimated.Aiming at working out the problem that the judging factors are one-sided in heterogeneous wireless networks, multi-objective decision making (MODM) based access selection algorithm is proposed. On the basis of satisfying the multifactor judgment and dynamic applicability, five-grade scale method and entropy of information method are adopted to calculate the subjective and objective weight. Meanwhile, least squares method is used to choose them eclectically. Both the network state and user preferences are taken into consideration in order to improve the accuracy and practicability of the algorithm. Simulation results show that the MODM algorithm promotes the access property and user satisfaction on a small number of access users with the decrease of load balance among networks. When there is a big number of access users, MODM algorithm adjusts the decision indicators dynamically, promotes load balancing, ensures the access of more users on the circumstance of guaranteed network performance and user satisfaction. The algorithm is designed more reasonable and more practical.In order to reduce algorithm complexity, a grey relational analysis (GRA) based access selection algorithm is proposed. The algorithm calculates the correlation coefficient of the optimal solution to do priority queuing for alternative programs and access network after choosing the best network. The performance of GRA algorithm is similar to that of MODM algorithm, among which the access blocking rate and the property of load balance are better than MODM. But user satisfaction is slightly lower than that of MODM algorithm. However, the utilization of discrete sequence analysis reduces calculation complexity. GRA algorithm is designed to reduce system complexity, which ensures the practicality.The research work of this paper could provide references for the design of access selection algorithms for heterogeneous wireless networks.