Mobile Edge Computing Clustering Algorithms for Pedestrian Mobility Scenarios

The purpose of this study is to provide a general framework for the latest trends of mobile network architectures. Edge computing brings the computing capability of the cloud to the edge of the network for minimizing the latency. In the D2D mode, Fog Nodes interact with each other. With the help of clustering, Fog nodes are categorized into two types: Fog Cluster Head (FCH) and Fog Cluster Member (FCM). In each cluster, FCMs offload the task towards their respective FCHs for computation. The characterization of the performance of system model taking into account the average energy consumption, average task delay, fairness, and packet loss. We provide results based on the numerical simulation performed in Matlab in order to show the difference in the performance of the network using different policies and clustering and cluster update frequencies. In this thesis, a theoretic framework is presented that aims to characterize the performance of Fog network with pedestrian mobility without priority approach and also pedestrian mobility with priority approach using clustering approach and compare the results. The simulation results show how the priority approach has the profound impact on the energy consumption, task delay, and packet loss and solve the problem of coverage constraint.