Unique identity and localization based replica node detection in hierarchical wireless sensor networks

Abstract

Clustering in Wireless sensor networks (WSN) is a prevalent Hierarchical network management technique. Though disjoint clusters are generally preferred, overlapping clusters find its prominence in some applications of inter-cluster routing, time-synchronization and node localization. Replica node detection is a major challenge in overlapping clusters. This paper aims at replica node detection in overlapping clusters based on two methods, Replica detection based on RFID (RDBRFID) and Replica detection based on Localization techniques (RDBLT). The first method uses RFID for unique node identification and the second method detects replica by identifying its locality based on received signal strength (RSSI) and Triangulation method. These methods are implemented and their performance is compared with Multicast and non clustered methods: Randomized multicast (RM), Line selected multicast (LSM), Fault tolerant virtual back bone tree (FTVBT) and K-coverage WSN. It is observed that RDBRFID exhibits better detection rate and lesser communication overhead due to its deterministic approach.