Home // EMERGING 2013, The Fifth International Conference on Emerging Network Intelligence // View article

Proposal and Evaluation of a Predictive Mechanism for Ant-based Routing

Authors:
Naomi Kuze
Naoki Wakamiya
Daichi Kominami
Masayuki Murata

Keywords: self-organization; prediction; routing; Ant Colony Optimization (ACO)

Abstract:
To tackle problems emerging with rapid growth of information networks in scale and complexity, bio-inspired self-organization is considered one of promising design principles of a new generation network, which is scalable, robust, adaptive, and sustainable. However, self-organizing systems would fall into a local optimum or converge slowly under some environmental conditions. Therefore, it may take a long time for self-organizing systems to adapt to environmental changes. In order to adapt to dynamically changing conditions of information networks, each component needs to predict the future state of its neighbors from their past behaviors and to adapt its movement to conform to the predicted states. There are several investigations into self-organization with prediction in the field of biology, but its application to information network systems and technologies needs more discussion. In this paper, we take AntNet, an ant-based routing protocol, as an example and consider a mechanism to accelerate path convergence with prediction. The proposed mechanism is compared with AntNet from viewpoints of the recovery time, path length, and control overhead. Simulation results show that our predictive mechanism can accelerate path convergence after environmental changes.

Pages: 7 to 12

Copyright: Copyright (c) IARIA, 2013

Publication date: September 29, 2013

Published in: conference

ISSN: 2326-9383

ISBN: 978-1-61208-292-9

Location: Porto, Portugal

Dates: from September 29, 2013 to October 3, 2013