A Dynamic Load Balancing Scheme for IEEE 802.11s Mesh Networks

碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 95 === IEEE 802.11 wireless Local Area Networks (LANs) have been widely deployed for a broad range of users in public venues such as campuses, Internet Cafes, or enterprises.The increasing demand for wireless networks has spurred the need for new innovative technolog...

Full description

Bibliographic Details
Main Authors: Hung-I Hsieh, 謝鴻益
Other Authors: Kuang-Hui Chi
Format: Others
Language:zh-TW
Published: 2007
Online Access:http://ndltd.ncl.edu.tw/handle/14710181044745795285
Description
Summary:碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 95 === IEEE 802.11 wireless Local Area Networks (LANs) have been widely deployed for a broad range of users in public venues such as campuses, Internet Cafes, or enterprises.The increasing demand for wireless networks has spurred the need for new innovative technologies. IEEE 802.11s specifying Wireless Mesh Networks (WMNs) is an emerging supplement to the IEEE 802.11 standard that meets the demand. WMNs cover broader regions, decrease the deployment cost, and provide multi-channels and higher bandwidth to users. When users communicate with nodes across the Internet, traffic is generally forwarded through access points (APs) that are one-hop away from some gateway. These APs are likely to form bottlenecks in the system. Given an IEEE 802.11s WMN, this thesis proposes an approach to balancing traffic load of bottleneck APs. We are concerned with a multi-channel environment where each AP is fit a single network interface using the Destination-Sequenced Distance-Vector Routing protocol. The proposed approach operates by designating the gateway to periodically collect load information from bottleneck APs. With such information in place, we determine to adjust routing paths for downstream APs such that load differences seen by bottleneck APs are reduced. Simulation results show that our proposed approach contributes to an improvement of 1.3% to 9.9% in terms of load balance index. Further, our approach results in an increase of throughput of 1.3% to 8.3%. Evaluation results indicate that our proposal facilitates operations of the whole network system.