Towards Scalability of Dense Sensor Networks: A Software-Defined Networking Approach

Abstract

This work proposes an efficient controller placement algorithm in SDN to address scalability challenges in dense sensor networks. An important question must be answered: given network topology, how many controllers are needed, and where should they be placed to satisfy network-specific requirements? This requires the decision-maker to define a set of objectives that must be considered during placement. In this work, the Min-Max weight determination optimization was implemented to address the trade-offs involved in a two-objective optimization problem where a cost function of multiple constraints is minimized over a single variable. Our proposed approach guaranteed QoS (in terms of latency and controller load) in dense sensor network scenarios compared to state-of-The-Art methods.