A new wireless sensor networks deployment strategy using hybrid particle swarm optimization

Sensor Deployment (SN) is one of the major challenges in wireless sensor network architectural. One of the most fundamental issues in wireless sensor deployment is to find a trade-off between two conflicting network objectives; Packet Delivery Ratio (PDR) and lifetime, under certain coverage and connectivity constraint. Although the approach of meta-heuristic searching optimization has been commonly applied, it has failed in addressing several issues related to multiple objectives and intricate optimization surface. However, the multi-objective nature of this problem and the complicated optimization surface requires developing customizable multi-objective meta-heuristic searching optimization. This thesis proposes a Lagged Multi-Objective Jumping Particle Swarm Optimization (LMOJPSO) approach that aims to find the Pareto front that maximizes the packet delivery ratio and minimizes the sensor energy consumption for prolonging network lifetime. The proposed LMOJPSO framework for improving the performance of the meta-heuristic search optimization process, is done by combining two different searching techniques. The first optimization technique carries out its searches with the help of Extreme Learning Machine (ELM), whereas the second search optimization uses a wireless sensor network simulator. In this thesis, the proposed method is examined in a given wireless sensor network test instances and the evaluation of its performance is carried out by using a wireless sensor networks performance metric. The results indicated that the proposed model is superior to the Non-dominated Sorting Genetic Algorithm (NSGA-II).