2019/12/05 by Reza Soleymanifar, Soleymanifar, Reza, Amber Srivastava +5
Computer Science · Engineering · #Advanced Optical Network Technologies #Cooperative Communication and Network Coding #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Networking and Internet Architecture (cs.NI) #Software-Defined Networks and 5G
paper · pdf · doi:10.48550/arxiv.1912.02915
openalex publication_date 2019/12/05 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
In this work we introduce two novel deterministic annealing based clustering\nalgorithms to address the problem of Edge Controller Placement (ECP) in\nwireless edge networks. These networks lie at the core of the fifth generation\n(5G) wireless systems and beyond. These algorithms, ECP-LL and ECP-LB, address\nthe dominant leader-less and leader-based controller placement topologies and\nhave linear computational complexity in terms of network size, maximum number\nof clusters and dimensionality of data. Each algorithm tries to place\ncontrollers close to edge node clusters and not far away from other controllers\nto maintain a reasonable balance between synchronization and delay costs. While\nthe ECP problem can be conveniently expressed as a multi-objective mixed\ninteger non-linear program (MINLP), our algorithms outperform state of art\nMINLP solver, BARON both in terms of accuracy and speed. Our proposed\nalgorithms have the competitive edge of avoiding poor local minima through a\nShannon entropy term in the clustering objective function. Most ECP algorithms\nare highly susceptible to poor local minima and greatly depend on\ninitialization.\n