Font Size: a A A

Research On Resources Allocation Mechanism Based On Dynamic Policy For Mobile IP Handover

Posted on:2008-06-23Degree:MasterType:Thesis
Country:ChinaCandidate:Z WenFull Text:PDF
GTID:2178360212497018Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
With the development of wireless communication networks, mobile terminal and multimedia communicating technology, people can visit networks using wireless methods any time, any place even under fast moving mode. It is necessary to provide mobility in the next generation networks .The development of mobile IP technology makes more and more WLAN begin to connect to Internet, and more and more multimedia services like interactive voice and video frequency applications are needed under mobile state. The mobile IP networks will gradually exceed telecommunication network and other networks, and will also become the key part of the information networks compared with others. However, due to the limited bandwidth of wireless channel in the mobile networks, the whole bandwidth resources seem rather precious, at the same time, the mobility of terminals makes the dynamic requirement change greatly, therefore,it is more complex to ensure mobile users'QoS in wireless state than in wire networks.In order to ensure mobile users'QoS, how to increase the rate of bandwidth usage inevitably becomes the primary task in dire need of settlement. Among all the management schemes, Policy-Based Network Management is an effective one. The Policy-Based Network Management separates the control of management behavior from the concrete execution policy, thus enhances the abstract level of management tasks, and also hides the specialization of network objects to be managed. And at the same time it can relieve the burden of network manager. Here comes the definition of policy from IETF, policy is a set of management rules which are composed of conditions and actions. The satisfaction of some conditions may cause the execution of a group of actions. However, traditional policies are created by network administrator through policy editor, so it is static and could not be changed during its life time, while mobile users'movement are stochastic and uncertain, which causes that users'requires are changed.This paper provided a model to adapt to satisfy dynamic requirements in mobile state. In the model it needs to ensure mobile users'QoS in wireless connection networks, especially fast connection and providing satisfying multimedia QoS. Because users could not put up with call dropping compared with call block. At present, for ensuring handover QoS, most researches are based on two projects: protect channels and queue handover requirements. Protecting channels need to remain some parts of channel for handover .Hence ensure that handover's priority is higher than new call, but it lacks adaptability to load changes. For queue method, blocked handover calls are put in queue, and they will not be served until resources are available. Obviously, this will increase delay of multimedia service. In this paper, policy based management is combined with QoS management, so guarantee is given to the process of IP handover, also this mechanism provides a reasonable usage of the bandwidth resources, what is more, the efficiency of dynamic resource allocation is improved.Traditional methods for ensuring the application QoS are usually allocating resource in advance, and in this process, the resource allocated to certain application can not be used by other applications, thus applications will not affect each other. But one problem comes up, it will waste resource, and reduce efficiency of bandwidth usage if beforehand distributed bandwidth is not used completely. For this reason, we put forward the idea of dynamic policy.Firstly, concept of dynamic policy is presented and applied for the first time. Secondly, the model of a dynamic policy scheme of resource management for mobile IP handover is designed based on IETF framework of policy management.In the model, as to those disciplinary mobile users in local areas, their next actions can be predicted on basis of prediction information of mobile node, and then corresponding policies are provided.This paper is mainly about works on generation mechanism of dynamic policy algorithm of dynamic policy generation is designed from the view of users'QoS. It can automatically produce dynamic policies by prediction information and network states. Corresponding parameters of dynamic policies could be automatically updated by the changes of network states. Three priorities are taken into account: handover priority, user priority and application priority. It is proved that the algorithm possesses the advantages of easy to implement and strong adaptability .It can distribute the whole bandwidth reasonably, and consider more users'application, as well as enhance the whole network bandwidth using efficiency. Thus the quality of on line service is ensured, and also the algorithm's efficiency is considered. In this process some bandwidth is remained for use of emergency. Network bandwidth state and data provided, an alternation interface experiment for this arithmetic can show that high priority users and high priority applications will obtain higher satisfy resource when handover mobile nodes gradually increase.Resource allocation on basis of dynamic policy is different from traditional ways of allocating fixed bandwidth. Using forecast information to produce dynamic policies automatically for admission control and preparing resources for handover before it happens. When the resources of access domain can not afford mobile nodes'requisition, it applies"step aside"arithmetic to adjust agilely resource possession according to the priority with assuring the business credibility. It is shown to ensure the QoS of users and services with higher priority and improved rate of handover, meanwhile, the management mode is more flexible than other methods. Also it can be used for reference for other network management.
Keywords/Search Tags:Allocation
PDF Full Text Request
Related items