| Summary: | 碩士 === 逢甲大學 === 資訊工程所 === 94 === Wireless sensor networks consist of a large number of low-power, small sensors with limited computation capabilities which are oriented towards applications. Most of these applications are to gather the sensing data for observations, analysis, and other usages, such as civilian, military, health, and natural ecology. Therefore, data gathering is one of the most important operations in wireless sensor networks, and convergecast is just the communication operation which collects sensing data from all or many nodes to sinks. It is the opposite operation of broadcast or multicast.
In wireless sensor networks, sensing devices are usually powered by batteries. Therefore, one of the most important issues in wireless sensor networks is to increase the network efficiency under power limitation. Many researches employ multi-hop transmission to reduce energy consumption which leads to long transmission delay. Besides, most of them are centralized and require the whole network information. It is not practical in real world. Moreover, the clustering technique is used to provide an efficient architecture to manage and maintain the network. However, some existing schemes have to consume extra energy for reclustering each round. Hence, in this thesis, we propose various efficient distributed convergecast according to different application environments and the position of the sink. We use a simple clustering technique to make a hierarchical network, and reduce energy consumption through multi-hop transmission. Our schemes achieve better balance between energy consumption and delay under different environments, and prolong network lifetime to improve network efficiency.
|
|---|
