부산대학교 도서관

In this paper, we propose a novel method based on genetic algorithm for constructing the wireless sensor network to extend its functionality and availability. In our proposed method, the structure of the network is dynamically decided and the organization differs after each message transmission round. With the goal of optimizing the lifespan of the entire network, genetic algorithm is employed to search for the most suitable sensor nodes as the cluster heads to relay the messages to base station. Using the chosen cluster heads, sensor clusters are formed that minimize the total inner cluster node-to-cluster head distance. Compared with eight other methods, our experimental results demonstrated that our proposed method greatly extended the network life. The network life improvement rate with respect to the second best cases is in the range of 13% to 43.44%. In each transmission round, the remaining energy of sensor nodes are fairly even with some fluctuations. That is, as a consequence of our proposed method, the variance among remaining energy is quite low, which implies that the sensor nodes shared the burden of relaying messages and, hence, elongated the overall network life.