Hierarchical clustering techniques for energy-efficient algorithms in WSNs

Abstract

In Wireless Sensor Networks (WSNs) with large number of micro-sensor nodes, successfully receiving all the useful data with unnecessary loss of energy is a major challenge. WSNs are extremely important for many applications in civil and military domains. Despite their use in many successful applications, the sensor nodes are randomly deployed. This creates many operational and design challenges for WSNs that result in reduced network lifetime. This is because of limited energy available in the nodes and increased energy consumption with large sized networks. Therefore, giving attention to large-scale integration and energy consumption helps to improve the energy-efficiency of the sensor node. This plays a fundamental part in increasing the lifetime of WSNs. Despite the development of many WSNs protocols, such as the hierarchal routing protocols to address this issue, the energy-efficiency is still a limiting factor. This thesis, addresses the above-mentioned challenges by proposing a novel set of Energy-Adaptive Clustering Protocols (ECP) for energy-efficiency in WSNs. The proposed routing strategy takes on several approaches to improve energy efficiency in WSNs. This functional set of protocols are integrated with each other. The proposed novel solutions have been implemented using simulation and evaluated for a range of given metrics. The evaluations were based on the most critical metrics in WSNs, such as node density, throughput and network lifetime etc. For validation and evaluation purposes, the proposed techniques have been compared with existing solutions. It is found that our proposed protocols exhibit reduction in energy consumption over well-known protocols such as LEACH, DEEC, EE-LEACH and HT2HL. While the proposed technique network lifetime improved by average of 67 %, 17 %, 21 % and 8 % more than LEACH, DEEC, EE-LEACH and HT2HL protocols respectively. The ECPs were also evaluated to include mobile node detection. It is named Energy-Adaptive Clustering Protocols (d-ECP) for mobile node detection in WSNs. The proposed d-ECP has potential applications in scenarios where RFID type mobile nodes are used. For example wildlife monitoring, condition monitoring, industrial applications and IoT. Keywords: Wireless Sensor Network, Routing Protocols, Network Lifetime, Energy Consumption, Cluster-Head, Radio-frequency identification (RFID).