CBLM: Cluster-Based Location Management for 5G Small Cell Network Under Stochastic Environment

Abstract

Internet of Things (IoT) applications are connecting several Mobile Stations (MSs) to form the Mobile Communication Network (MCN). MSs are moving through macrocell and picocell recurrently due to the discontinuous deployment of the picocell over the macrocell. Each time they cross the cell boundary, they need to register with the Mobility Management Entity (MME). It incurs a high registration cost and elevates the signaling overhead ratio. To reduce the signaling overhead ratio, earlier techniques were attempted either by forming closed subscriber groups or by delaying the registration process, but that may create inconsistency in user location information. With this motivation, in this paper, MSs with similar mobility pattern which are moving from the picocell to macrocell, are grouped in clusters. A representative MS, called Cluster Representative (CR), performs the location management task on behalf of the whole cluster. The stochastic process has been employed to analyze the model. This study proposes Cluster-Based Location Management (CBLM) scheme and compares it with two state-of-the-art algorithms, named Group Mobility Management (GMM) and Closed Subscriber Group (CSG). On average, the CBLM scheme exhibits approximately 70% better performance than the CSG scheme whereas, 30% better than the GMM algorithm to mitigate the signaling overhead.