الفهرس Only 14 pages are availabe for public view
 |  | from | 55 |  |  |
| | | from | 55 | | |
Abstract
The importance of wireless sensor networks (WSNs) has been increasing in the last few years. Miniaturization technology caused the emergence of sensor nodes with limited capa¬bilities in computation, energy and memory. WSNs need customized algorithms for routing, topology management and other network algorithms to work with limited resources. One of the main aspects of WSNs is power management. Power management allows sensor nodes to save their power which increases the lifetime of the network.
In this thesis we propose a new power management algorithm for WSNs called Saving Energy by Counting Neighbors (SECN). This algorithm tries to keep coverage and connec¬tivity of the network to a certain degree while saving energy by turning off redundant nodes. Sensor nodes running this algorithm can turn off its hardware while keeping the overall coverage to the required degrees. Redundant nodes are detected using a fully distributed al¬gorithm without global information. Each node needs only local information about its direct neighbors. The information it needs is the number of nodes in its sensing range.
We have proposed two versions of the algorithm. In the first version (SECN-l), all nodes start in working state and redundant nodes turn off. In the second version (SECN-2), all nodes stmt in off state then nodes turn on one by one until deployment area is covered.
We have perfOlmed simulation experiments on ns-2 to compare our algorithm with previ¬ous work. Comparing our algorithm to position aware algorithms we find that our algorithm increases network lifetime by 53%. At the same time coverage ratio decreases by only 3%. We recommend SECN-2 over SECN-l for it scalability and stability.