Self-Deploying Mechanism for Sensor Networks

• Main Authors: Chien-Chun Lin, 林建均
• Other Authors: Lih-Chyau Wuu
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/21678217407208005318

碩士 === 國立雲林科技大學 === 資訊工程系碩士班 === 100 === With IEEE 802.15.1/Bluetooth and IEEE 802.15.4/ZigBee development, sensors can be connected as an Ad Hoc wireless network to collect the environmental information (pressure, temperature, carbon dioxide, humidity, sound, etc.), and then transfer the collected information to a specific node, called as the Sink. In general, not all the sensors can transfer the data to the sink directly. Data may be relayed by sensors before it delivers to the sink. It means that sensors near the sink must consume more energy to relay data than those sensors faraway from the sink. It is well known that sensors have limited energy supply. The wireless sensor network may not work properly once some sensors exhaust their energy. It is one of the important issues to prolong the network lifetime by optimizing the energy consumption of sensors. In this paper, we will discuss a self-deploying mechanism for sensor networks to optimize the energy consumption. Our idea is that the area around the Sink should have more sensors being deployed. To do that, the whole sensing area is divided into several hexagon blocks, and the number of sensors needed in each block is computed by our formula. The blocks near the sink will have more sensors than the blocks faraway from the sink. A block state is “Hungry” if it shorts of sensors, and is “Overfull” if it has too much sensors, and is “Neutral” if the number of sensors in the block is equal to the computed one. Sensors in a block will select a sensor to be the leader to communicate with other block leaders. A leader of an “Overfull” block will move sensors to a “Hungry” block. By our scheme, eventually, each block will remain in the “Neutral” state. The simulation results show that our mechanism prolongs the network lifetime. For the best case, our network lifetime is three times of those networks without executing our scheme.