A modified Manhattan distance with application for localization algorithms in ad-hoc WSNs

Abstract : Building an efficient node localization system in wireless sensor networks is facing several challenges. For example, calculating the square root consumes computational resources and utilizing flooding techniques to broadcast nodes location wastes bandwidth and energy. Reducing computational complexity and communication overhead is essential in order to reduce power consumption, extend the life time of the battery operated nodes, and improve the performance of the limited computational resources of these sensor nodes. To that end, we propose a novel modified Manhattan distance norm and employ it in a previous localization system so-called TALS (Trigonometric Ad-hoc Localization System), such that we optimize TALS. Furthermore, an analysis and an extensive simulation for the optimized TALS (OTALS) is presented showing its cost, accuracy, and efficiency, thus deducing the impact of its parameters on performance. Our novel similarity measure formula can be used in many other domains such as clustering and classification. However, we present its efficiency only for a particular problem in this work. Thus, the major contribution of this work can be summarized as follows: (1) Proposing and employing a novel modified Manhattan distance norm in the TALS localization process. (2) Analyzing and simulating of OTALS showing its computational cost and accuracy and comparing them with other related work. (3) Studying the impacts of different parameters like anchor density, node density, noisy measurements, transmission range, and non-convex network areas. (4) Extending our previous joint work, TALS, to consider base anchors to be located in positions other than the origin and analyzing this work to illustrate the possibility of selecting a wrong quadrant at the first iteration and how this problem is overcome. Through mathematical analysis and intensive simulation, OTALS proved to be iterative, distributed, and computationally simple. It presented superior performance compared to other localization techniques.
Type de document :
Article dans une revue
Ad Hoc Networks, Elsevier, 2015, 33, pp.168-189. 〈10.1016/j.adhoc.2015.05.003〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01260942
Contributeur : Sara Alouf <>
Soumis le : vendredi 22 janvier 2016 - 17:30:42
Dernière modification le : samedi 27 janvier 2018 - 01:31:43

Identifiants

Collections

Citation

Mohamed Shaheen Elgamel, Abdulhalim Dandoush. A modified Manhattan distance with application for localization algorithms in ad-hoc WSNs. Ad Hoc Networks, Elsevier, 2015, 33, pp.168-189. 〈10.1016/j.adhoc.2015.05.003〉. 〈hal-01260942〉

Partager

Métriques

Consultations de la notice

96