A self-learning heterogeneous wireless network access selection method for differentiating services

Abstract

The invention discloses a self-learning heterogeneous wireless network access selection method for distinguishing services, and belongs to the technical field of wireless network access selection. When the mobile terminal service arrives, the present invention accesses a suitable wireless network for the terminal service through the access selection manager, which can not only meet various service requirements of the terminal, but also comprehensively consider network service quality and user experience, and target different types of The business adopts different access selection criteria, and through the self-learning mechanism of the slot machine model, it can reflect the dynamic changes of the network status information without interrupting the communication, and adjust the access selection in time.

Description

technical field [0001] The invention belongs to the technical field of wireless network access selection, and in particular relates to a self-learning heterogeneous wireless network access selection method for distinguishing services. Background technique [0002] With the trend of diversification of wireless network technology becoming more and more obvious, the heterogeneous wireless network composed of various wireless network cooperation and complementarity, especially the heterogeneous wireless network composed of 3G / 4G cellular mobile communication network, IEEE802 series wireless local area network and wireless metropolitan area network, etc. Wireless network has become an inevitable trend of wireless network development and integration. Appropriate access selection method is the key to realizing heterogeneous wireless network resource management, improving network capacity and user experience. [0003] The existing heterogeneous wireless network access selection met...

AI Technical Summary

Problems solved by technology

[0003] The existing heterogeneous wireless network access selection methods have the following problems: the development of network services tends to be diversified, the individual needs of users are becoming more and more prominent, and the resulting different types of service requirements are also increasingly obvious; the existing access selection methods Most of them focus on using QoS as the network selection standard, emphasizing the overall network performance, and ignoring the specific needs of different users. Although the overall network performance is generally optimal, it cannot guarantee the experience of specific users; the availability and dynamic changes of network status information are not guaranteed. It has been fully considered that the traditional method is to access the optimal network once. Once the access is maintained for the entire service duration, it cannot actually guarantee that the entire service duration is accessed to the optimal network.

Hierarchy-based analysis proposed in the literature (Q.Song, A.Jamalipour.Network selection in an integrated wireless LAN and UMTS environment using mathematical modeling and computing techniques[J].IEEE Wireless Communications,2005,12(3):42-48.) In the gray-scale associated access selection algorithm of the method, the typical multi-attribute decision-making method is used to determine the weight of the service attribute, and the optimal access network is selected by fitting the correlation between the service weight and the network attribute. At the same time, it also sets targeted selection criteria for different types of services, but it has the defect that it only makes one decision based on the current network status, and cannot ensure that services are always connected to the optimal network.

Literature (C.Sun, E.Stevens-Navarro, V.Shah-Mansouri, et al.A constrained MDP-based vertical handoff decision algorithm for 4Gheterogeneous wireless networks[J].Wireless Networks,2011,57(4):1243- 1254.) Use other access selection targets as the constraints of the current decision-making target to achieve continuous decision-making for all current targets, and use the value iterative algorithm to achieve the network that maximizes the total return, that is, the selection of the optimal network. This method achieves the optimal The continuous decision-making of the optimal network, dynamic perception of network status changes without interrupting communication, and access feedback through total rewards, but its disadvantage is that the definition of the reward function only focuses on the Qos attribute parameters of the network side, which cannot reflect the quality of user experience

Images