Abstract: Due to the limited resources of the network transmission of large sized data can be a bottleneck in sensor network. The objective of sensor network differs from computer network in a sense that in sensor network the goal is to get an estimate of the environment and not to receive all packets. As such the network provides the ability for In-Network data processing. The data delivery is done by selecting appropriate cluster heads. This kind of region diversity may be utilized to achieve higher energy efficiency. We divide the network into four regions. Node with highest energy in each region is selected as cluster head. Remaining nodes forwards their data to cluster head which aggregates the data and forwards only one packet instead of all the packets received from other nodes. Therefore the overall energy consumption is reduced in such a network. In our proposed method we recalculate cluster head after every round of transmission and bypass nodes with lower energy to preserve more energy. We compare the performance with non-energy centric pure region based cluster head and prove that the proposed technique fares better in terms of low energy consumption, fewer node deaths, more packet delivery ratio and improved packets par node. Here by compressing redundant data we ensure better degree of compression without compromising on mean square error and in the process extending the lifetime of the network. The design is simulated using NS2 network simulator and results are analyzed using trace analysis feature. A result shows that proposed system performs better in terms of all the performance metrics in comparison to conventional uncompressed sensing.
Keywords: Wireless Sensor Network, Clustering, Energy model, Energy conserving, Normalized Routing Load, Throughput.
Title: Cluster Based Energy Conserving Wireless Sensor Network with In-Network Data Aggregation
Author: Priya Kinikar, Bharati S. Pochal
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
