Research On Heterogeneous Network Access Selection Algorithm Based On Multiple Attribute Decision Making

The integration of heterogeneous wireless networks is an inevitable trend in the future development of mobile communication network.The target of terminal access selection in the heterogeneous wireless network environment is as follows.The most suitable and appropriate wireless network is choosing to be accessed according to the performance status of the current network,the characteristics of the terminal itself,the type of mobile application,and the user's preference information,etc.,so that users can get the most satisfactory service,and make the radio resources of whole network to be effectively used.Access network selection is not only an important part of switching management,but also is important to the realization of maintain the best connection to the users continuously.This paper would have this topic as the main research content.Base on the summary of the existing network access selection algorithms,in view of the shortcomings of the effectiveness and robustness of the network selection algorithm,this paper proposed and implemented two kinds of heterogeneous wireless network access selection algorithms.The weight calculation method of these two algorithms is considering both subjective weight and objective weight.The subjective weight of each attribute in the four types of services is determined by AHP.The entropy weight method and the CRITIC algorithm are used to calculate the objective weight of each attribute,and the final weight is obtained by the comprehensive weighting method based on the minimization of the difference.This kind of comprehensive weighting method makes the decision result satisfy both the user's preference and the objective state of the network.The first algorithm is a network selection algorithm based on chi-square distance.In this algorithm,the distance between the candidate network and the ideal network is the scoring standard,and the candidate network of smallest chi-square distance is the best access network.The second algorithm is a network selection algorithm based on relative entropy,is also an improvement of TOPSIS method.In this algorithm,the relative entropy is used to calculate the distance between candidate network and the positive and negative ideal network,and the relative closeness degree is used to evaluate candidate networks.The candidate network of the biggest closeness degree is the best access network.The simulation results show that these two algorithms can select the network which is suitable for the current service requirements under different service types.Furthermore,both of these two algorithms can avoid getting into the partial optimal situation when the candidate network appears extreme data,and choose the correct optimal access network.