Emerging practical deployments for the Internet of Things (IoT) trigger a need to integrate and inter-operate a variety of heterogenous networks to realize real business values. Several applications require the integration of wireless sensor networks (WSN), WiFi, and Radio Frequency Identifiers (RFID) into one single network to fulfil business requirements. As most of such deployments are characterized as being large-scale and heterogeneous, special algorithms and techniques are needed in order to deal with data collection, processing, and transmission in such networks. Results reported in the literature confirm that clustering techniques can be very efficient in dealing with routing in large-scale networks. However; due to the heterogeneity of IoT networks, the use of conventional clustering techniques may not result in an efficient clustering. Accordingly, in this paper, we attempt to address this problem by studying the use of evolutionary clustering algorithms in integrated WSN-RFID networks. In particular, the performance of two evolutionary algorithms; namely the Genetic Algorithms (GA) and the Harmony Search (HS), is analyzed and compared. It is shown that, the GA outperforms the HS significantly in the cluster formation process for integrated WSN-RFID networks.
CCS Concepts•Networks → Network design and planning algorithms;