Open AccessA Distributed Target Localization Algorithm for Mobile Adaptive Networks
Author(s) -
Amin Lotfzad Pak,
Azam Khalili,
Md. Kafiul Islam,
Amir Rastegarnia
Publication year - 2016
Publication title -
ecti transactions on electrical eng. / electronics and communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.148
H-Index - 7
ISSN - 1685-9545
DOI - 10.37936/ecti-eec.2016142.171139
Subject(s) - computer science , node (physics) , adaptation (eye) , algorithm , distributed computing , computer network , engineering , physics , structural engineering , optics
Adaptive networks with mobile nodes possess the distributed adaptation abilities in addition to the collective patterns of motion. Thus, mobile adaptive networks have been used in new applications such as modeling the biological networks and source localization. The original mobile adaptive networks needs full cooperation among the neighbor nodes meaning that each node gathers the information from all of its neighbour nodes. This strategy requires large amount of communications per iteration per node. To address this problem, in this paper we propose a mobile diffusion adaptive network with selective cooperation. In the proposed algorithm each node selects a subset of its neighbors so that its steady-state performance be as close as possible to the traditional mobile diffusion network. Since the selective cooperation reduces the learning rate we also use affine projection algorithm (APA) as the learning rules at the nodes. Our simulation results reveal that the proposed algorithm is able to achieve the same mean-square deviation (MSD) as the original mobile adaptive network but with a lower communication per iteration per node.