Node Similarity Abstraction for Wireless Sensor Networks

Abstract

Wireless Sensor Networks are being used for applications of many areas in today's world. Energy of sensor nodes is a critical factor that determines the lifetime and usefulness of sensor networks specially that are large-scale networks. Mostly the large- scale sensor networks tend to have similarly behaving nodes and many approaches are introduced recently to identify such similar node sets, but they are of the major limitation of being in exible to changes after deployment. In this research, we explore the use of combination of SQL and SQL-like acquisitional query language in database abstractions for WSNs as the language for the user to de ne and maintain similar node sets and gather data from relevant representatives by being in the client-side of the network and thus provide the node similarity abstraction by combining an RDMS and a database abstraction for WSNs which is also capable of reducing energy con- sumption of nodes by reducing data transmissions between nodes. We also provide a theoretical analysis of number of messages being passed in the network with respect to our proposed solution and thus we provide the design for a node similarity ab- straction for WSNs by combining a client-side database and a database abstraction for WSNs. We also provide an implementation of the proposed solution and we also provide an evaluation for the proposed solution using a simulation platform where we have explored the ability in our approach for providing user a node similarity abstraction for wireless sensor networks while reducing energy consumption in nodes and network congestion.