Study on interaction between layered self-organization based control
Hiroshi Yamamoto, Naoki Wakamiya, Masayuki Murata · 2011
Self-organization is considered one of key design principles to establish highly scalable, adaptive, and robust network systems to accommodate dynamic, diverse, and massive nodes and traffic. Although there are many proposals on self-organization based protocols that are useful, effective, and practical, there has never been any in-depth investigation into interaction, interference, and synergetic effects among multiple self-organization based control. In this paper, we show an idea of analysis of mutual interaction among layered self-organization based control. We consider an overlay network that is constructed over an ad-hoc network, both of which adopt adaptive routing protocols based on the attractor selection model, i.e. a mathematical model of adaptive behavior of biological systems. We modified the degree of coupling by changing the way how layered self-organization control shared an objective parameter. Through simulation experiments, we showed that lower layer-aware routing can provide the best performance, while coupling sometimes brings worse results than independent control.