C3HAC: A Controller Placement Approach For SDWSN

Abstract

In Software Defined Wireless Sensor Network (SDWSN), the Controller Placement Problem (CPP) deals with determining the optimal number of controllers and their location in the network. In recent year, many metaheuristic approaches have been proposed for CPP but their solutions are of high time complexity. Therefore, we propose a novel graph-based clustering approach to solve the CPP in SDWSN. Our proposed algorithm Coverage and Controller Capacity-based Hierarchical Agglomerative Clustering (C3HAC) minimize overall network latency. C3HAC aims to determine the optimal number of controllers and divides the network into an equal number of clusters. Moreover, we also formulate multiple objective functions and a fitness function for placement of controllers in each cluster. The simulation results demonstrate that our proposed method outperforms the state-of-art algorithms in the number of controllers, flow setup latency, and network propagation latency.