Summary
International Symposium on Nonlinear Theory and Its Applications
2016
Session Number:A4L-E
Session:
Number:A4L-E-3
A Hierarchical Routing Algorithm for MANET Based on Multi-Agent Learning
Yuki Hoshino, Hidehiro Nakano, Arata Miyauchi,
pp.-
Publication Date:2016/11/27
Online ISSN:2188-5079
Summary:
Mobile Ad-hoc Networks (MANETs) can construct impromptu networks by wireless mobile nodes without fixed infrastructure. A routing algorithm for MANETs based on multi-agent cost learning has been proposed. In this algorithm, multiple agents learn the costs of the communication paths by calculating weighted values, and select the effective paths based on the costs. This algorithm is robust for the movements of wireless mobile nodes. In order to develop this algorithm into larger-scale environments, this paper proposes a hierarchical routing algorithm for MANETs. The proposed algorithm constructs hierarchical networks based on address information, and does not require cluster-head nodes. In the simulation experiments, the effectiveness of the proposed algorithm can be confirmed by evaluating the delivery rate and control packet counts.