| Summary: | 碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 96 === In recent years, the advance of technology and rapid development of Micro-Electro-Mechanical Systems (MEMS) have facilitated the rapid growth of Wireless Sensor Networks (WSNs). Therefore, variety of applications in WSNs are paid attention and developed. The sensor node is fairly limited by energy resource, memory, the computation of processor, the communication capability and so on. The limitations of hardware that make the control of energy consumption to become the primary considered factor in various kinds of relevant research of WSNs. Hierarchical clustering is a key technology for reducing energy consumption. In addition, the different phases of network to elect cluster manager and to clustering in WSN are considered. It includes the process of initial phase of network, the process of nodes join or leave a group, and the process of Cluster Manager (CM) fail.
In addition, the stability is very important especially in WSN, so the evacuated mechanism of CM’s workload is proposed to assign the part of works to the Backup Manager (BM) to execute when the CM is overload, and retrieve work authority of BM when the CM returns to normal performance for improving the stability and performance of WSN. The proposed mechanism is called the Load-Balance Clustering Mechanism (LBCM in short).
To summarize the main points that have been made in this thesis, the proposed mechanism that the BM assists to process the workload of the CM by the concept of “slow-start” can balance the overload of a CM. Due to the proposed mechanism that can balance the CM load, keeps a stable network topology and prolongs the lifetime of WSN.
In brief, this study includes: (1) design a stable clustering and election mechanism to form a hierarchal wireless sensor network topology; (2) balance the workload of CM and reduce the energy consumption of CM; (3) enhance the stability and performance of the whole network; (4) prolong the lifetime of the whole network.
|
|---|
