Indian Journal of Science and Technology

Abstract

In the recent years Wireless Sensor Networks (WSNs) have gained a significant attention from the research and industrial community.Areal-timeWSNisexpectedtoperformexceptionallywellwithlarge-scale applications across theInternetwork. Most sensor nodes adopt un-rechargeable batteries with limited power supply. The practice of embedded and autonomous interrogation of data collected remotely across WSNs is one among the prime challenges faced in real-time environments. The process of data collection has a major impact in energy consumption across the sensor networks. The work proposes a Cluster-Based Priority Traverse (CBPT) technique where mobile sinks traverse in a dynamic path decided from the prioritized routing table. Also the work proposed emphasizes not only the collection of relevant data, but also a means of conditioning and interpreting these data. Sensed data are categorized into primary data that has a drastic variation from historical data and secondary data that resembles regular sensed data pattern. The WSN is organized as a clustered sensor network. The CBPT prioritizes the clusters in the network based on their Primary Data Count (PDC). The mobile sinks are ensured with high reliable traverse paths that include maximum high priority clusters to collect data in an energy efficient manner. A Minimum Spanning Tree is constructed for each cluster to all other clusters in the network. With repetitive simulations the categorization of data is achieved with high accuracy. Experimentations have been carried out to prove CBPT performs extremely well in terms of message overhead and delay.
Keywords: Cluster-based Priority Traverse (CBPT), Primary Data Count (PDC), Wireless Sensor Networks (WSNs)