Indian Journal of Science and Technology

Abstract

Wireless sensor network consists of densely deployed sensor nodes which have limited resources like energy, node lifetime. Data aggregation is an effective scheme to reduce redundancy of sampled data generated by sensor nodes. Homogeneous sensor networks easily adapt the data aggregation scheme because of easy synchronization of data samples; but heterogeneous sensor network have difficulty to handle data aggregation due to synchronization of different data packets produced by different sensor nodes. In order to perform efficient data aggregation in heterogeneous sensor networks our proposed method Attribute based Spanning Tree (AST) introduced the method of attribute based spanning tree construction over heterogeneous networks. Based on the characteristics of sensor nodes, logical separation of nodes formed then each group constructs Minimum Spanning Tree (MST), aggregation follows this MST to reach sink node. By adapting Kruskal’s algorithm into our proposed method MST is constructed in sensor nodes. Our simulation results shows that AST is more spatially convergent and it uses shortest path cost for aggregation leads to increase node lifetime and saves energy.

Keywords: Data Aggregation, Heterogeneous Sensor Networks, Spanning Tree