Fault-Tolerant Mechanism on ZigBee Wireless Sensor Networks

碩士 === 逢甲大學 === 資訊工程學系 === 104 === ZigBee is an emerging open standard protocol for wireless communication networks. In that protocol, it uses IEEE 802.15.4 as its physical and media access control layer standards. Based on the two layers, the network and application layers are further developed. Zi...

Full description

Bibliographic Details
Main Author: 林恩頡
Other Authors: 黃志銘
Format: Others
Language:zh-TW
Published: 2016
Online Access:http://ndltd.ncl.edu.tw/handle/7djwxx
Description
Summary:碩士 === 逢甲大學 === 資訊工程學系 === 104 === ZigBee is an emerging open standard protocol for wireless communication networks. In that protocol, it uses IEEE 802.15.4 as its physical and media access control layer standards. Based on the two layers, the network and application layers are further developed. ZigBee is with some salient characteristics, for instance, low cost, low complexity, and low power consumption. It also supports many different network topologies, and used in a variety of applications, such as smart home, healthcare, and wireless sensor networks. However, in wireless sensor networks, since all nodes usually operate with battery power, when a node uses up its energy, the network might malfunction. Until now, there are only a few research articles discussed the fault tolerance topics concerning with the ZigBee wireless sensor networks. In this thesis, we therefore propose a fault-tolerant mechanism for ZigBee wireless sensor networks to promote the network reliability and lifetime. When a node dies, we first use the spare nodes to replace the faulty node rapidly, and thus recover the network. While, if there exist no redundant nodes, we do neighboring node migration to mitigate the number of orphan nodes, such that the network is always in action. In addition to the proposed recovery scheme, we also conduct several extensive simulations with JAVA programming. The simulation results demonstrate that our proposed scheme is superior to the native ZigBee standard and its counterparts, in terms of data deliver rate, making about 8% ~ 40% improvements.