Wireless sensor networks (WSNs) have recently been extensively investigated due to their numerous applications in processes that have to be spread over a large area. One of the important issues of WSN is node localization; node localization capability is highly desirable for the performance evaluation in monitoring applications. Localization is defined as estimating the locations of sensors with initially unknown location information as most sensors do not know their locations due to the cost and size of sensors. The main objective of localization is to find the nodes' positions in a short time with a low energy cost; for this reason, recent approaches relying on swarm intelligence techniques are utilized, and node localization is considered an optimization problem in a multidimensional space. Recently, the meta-heuristic Bat algorithm was proposed as a solution for the node localization problem. This paper proposes an effective Bat algorithm for the node localization problem, the effectiveness of which is based on the adaptation of velocity of the Bats by hybridization, with Doppler effect for improving the performance, aptly termed Dopeffbat. Hence, Dopeffbat computes (through evolution) the nodes' positions iteratively through the Euclidian distance as fitness. Deploying this algorithm on a large WSN with hundreds of sensors demonstrates decent performance in terms of node localization. Moreover, the Dopeffbat parameters are simulated and interpreted in different scenarios of simulation; in addition, a comparative study was performed to further demonstrate the performance of the proposed algorithm, and the simulation results prove that Dopeffbat has a high convergence rate and greater precision compared with the original Bat algorithm and particle swarm optimization (PSO) algorithms.

No abstract

Recently developments in wireless sensor networks (WSNs) have raised numerous challenges, node localization is one of these issues. The main goal in of node localization is to find accurate position of sensors with low cost. Moreover, very few works in the literature addressed this issue. Recent approaches for localization issues rely on swarm intelligence techniques for optimization in a multi-dimensional space. In this article, we propose an algorithm for node localization, namely Moth Flame Optimization Algorithm (MFOA). Nodes are located using Euclidean distance, thus set as a fitness function in the optimization algorithm. Deploying this algorithm on a large WSN with hundreds of sensors shows pretty good performance in terms of node localization. Computer simulations show that MFOA converge rapidly to an optimal node position. Moreover, compared to other swarm intelligence techniques such as Bat algorithm (BAT), particle swarm optimization (PSO), Differential Evolution (DE) and Flower Pollination Algorithm (FPA), MFOA is shown to perform much better in node localization task.

Recently, Node localization has played a vital role in wireless sensor networks (WSN), where the node's coordinates have mostly required before sensing operation. Sensor devices are randomly deployed in the network for monitoring task. The crucial objective of localization is to determine the position of the sensor nodes based on a few landmark nodes. Node localization problem is formulated as an optimization problem and metaheuristic techniques are called to deal with the issue in multidimensional space, in this paper, the Whale Optimization Algorithm (WOA) is proposed to deal with the challenge, furthermore, the capability of auto localization is highly requested. WOA computes iteratively the node's positions based on a multistage approach to browse the whole region of the network. Deploying this algorithm on a large WSN with hundreds of sensors shows good performance in terms of node localization. the simulation result proved the good performance of WOA in term of localization error and computing time . CCS CONCEPTSCCS → Networks → Network types → Wireless access networks.