Relay Node Placement (RNP) is a critical design problem in Wireless Sensor Networks (WSNs). Network performance and energy efficiency can be highly affected by the placement strategy of RNs. It is an NP-hard optimization problem which can be effectively addressed with multi-objective formulation and solved using metaheuristics. The main aim of this research work is to efficiently optimize the unconstrained deployment of energy-harvesting RNs in a pre-established stationary WSN. The focus is on optimizing different conflicting objectives, namely network coverage, energy consumption, and deployment cost. This was approached in this paper using a novel integrated approach combining the Genetic Algorithm (GA) with a greedy technique. The greedy algorithm plays a vital role in the optimization process as it effectively guides GA during convergence. The outcome of the greedy algorithm is used to limit the search space complexity of GA and realize the critical places at which RNs might be placed. The greedy-integrated GA approach was evaluated extensively with different WSN scenarios, in comparison to the classic GA when used independently. The experimental results
Paper
Full text
A Genetic Algorithm Integrated Approach for Efficient Relay Nodes Placement
Semantic Scholar · Engineering · 2020
Abstract
Relay Node Placement (RNP) is a critical design problem in Wireless Sensor Networks (WSNs). Network performance and energy efficiency can be highly affected by the placement strategy of RNs. It is an NP-hard optimization problem which can be effectively addressed with multi-objective formulation and solved using metaheuristics. The main aim of this research work is to efficiently optimize the unconstrained deployment of energy-harvesting RNs in a pre-established stationary WSN. The focus is on optimizing different conflicting objectives, namely network coverage, energy consumption, and deployment cost. This was approached in this paper using a novel integrated approach combining the Genetic Algorithm (GA) with a greedy technique. The greedy algorithm plays a vital role in the optimization process as it effectively guides GA during convergence. The outcome of the greedy algorithm is used to limit the search space complexity of GA and realize the critical places at which RNs might be placed. The greedy-integrated GA approach was evaluated extensively with different WSN scenarios, in comparison to the classic GA when used independently. The experimental results