Research On Resource Management Of Telecom Wlan For Service QoS Guarantee

As the mobile Internet is more and more accepted and relied on, users of the thirdgeneration wireless cellular network have sharply increased. As a result, it witness an rapidlyexplosive growth on data flow in part of WIFI regions. Therefore, the three major domestictelecom operators are starting a large-scale construction of WLAN network, preparing to takethe advantage of low cost and high bandwidth of WLAN to relieve the pressure on mobilenetwork caused by packet data service. At the same time, as the globalization of domestictelecommunication market speed up, the competition of telecommunication industry graduallyshifts from a simple price competition to improving service quality and customer satisfaction,so each major operator has separately put forward a service level agreement (SLA, ServiceLevel Agreement). Moreover, the QoS guarantee is an important factor to realize SLA.Therefore, how to realize QoS guarantee in WLAN network has become an urgent problem tosolve.However, the hardware constraints of bottom devices and protocol parameters of staticconfiguration method from the traditional has led the capability of network end-to-end QoSguarantee to a bottleneck. At present, with the rapid development of microelectronicstechnology, the bottom hardware constraints of equipment is gradually weakening, there is antendency that equipments are willing to present intelligent, the adjustment parameters ofintelligent network, which consists of intelligent devices nodes is more broad, communicationis more flexible, as a more powerful basic platform, it helps to realize QoS guarantee. However,new situation comes, the relationship between the adjusted system parameters and the systemperformance became more complex. It is an urgent problem to be resolved that how people canselect to adjust the target parameters from several tunable parameters and weigh the interactionamong parameters depending on different input and constraints helps the adjusted networkperformance to meet the QoS requirements of different businesses.This article focuses on the QoS-guarantee reconfigurable network resource managementfor this purpose. The research begins with four aspects, that is the model description ofreconfigurable network resource management, the way to present an overall evaluation onservice quality, how to manage the reconfigurable network resource with end to endQos-guarantee, the construction of simulation platform and the algorithm validation. Taking themodel description of reconfigurable network resource management as an example, this article elaborate on the topic of characterization, physical description and working mechanism whichhas provided a model support for the resource management based on the basic theory. Withregard to the overall evaluation methods to service quality, this article has gave a detaileddescription on Qos requirements of various businesses based on reconfigurable networkresource management model which combines the attribution classification of network carriagebiz, it also has established the model of overall evaluation on service quality and provided thedetailed reference model in evaluation for Qos-guarantee network resource managementmethod. Based on the two mentioned parts of work above, the article conduct a research on thereconfigurable network resource management methods with end to end Qos-guarantee, to beginwith a formalized mathematical description on resource management model, then set up therelationship between tunable parameters and perception parameters by the study of deduction.The final decision is also made by the study of deduction based on the combined intelligentheuristic algorithm between non-dominated sorting genetic algorithms and fuzzy logic methodto present day environment and business historical data.In addition, this article designed and established MATLAB and OPNET joint simulationplatform, and gives a detailed process-related variables to parameters and procedure design, itelaborated on the establishment process of simulation platform in the aspects of applicationmodule design, MATLAB Interface module, ip module, design and definition ofwireless_lan_mac module.Finally the established network simulation platform is utilized toconduct a simulation verification to proposed end to end Qos-guarantee reconfigurable networkresource management.