A new approach for distance vector-Hop localization algorithm improvement in wireless sensor networks

Abstrak

This article shows a new range-free localization technique based on a metaheuristic algorithm (MA) dedicated to wireless sensor network (WSN), named sequential online-grey wolf optimization-distance vector-Hop (SOGWO-DVHOP). Indeed, we use the improved GWO based on selective opposite learning to improve GWO in order to enhance the traditional DVHOP localization algorithm. In reality, we choose GWO due to its better outcomes compared to other meta-heuristics, which leads us to improve this algorithm further. In the literature, the improvement works of GWO try to reconstruct the hierarchy of GWO or improve specifically the role of omega individuals. In our contribution, we opt for opposition-based learning (OBL) to ameliorate GWO, aiming to further enhance the quality of localization made by DVHOP. On the other hand, we make an empirical comparison of DVHOP and its improved versions in terms of accuracy. The results of the simulation demonstrate that SO-GWO-DVHOP gives the best performance when we vary the anchor ratio and the density of nodes.