Conflict-Aware Distributed Coordination Framework for Self-Organizing Networks

초록

Self-organizing network (SON) has been introduced as a collection of functions to automatically manage and optimize heterogeneous cellular networks. For network-wide deployment of SON, coordination among SON functions is indispensable as their operational objectives differ. However, since naive coordination among SON functions may lead to high complexity and policy conflicts, a carefully designed coordination framework is required. In this paper, we propose a novel conflict-aware distributed coordination (CADC) framework, employing deep reinforcement learning (DRL)-based coordinators for base station (BS) clusters to manage SON functions at the local level. To minimize inter-cluster dependency, CADC employs a graph-based clustering algorithm that models potential conflict relationships among BSs and groups BSs accordingly, thereby enabling global coordination through localized coordination within clusters. For performance evaluation, CADC is implemented on an ns-3 network simulator integrated with ns3-gym, and extensive simulation results demonstrate that CADC effectively resolves policy conflicts between SON functions compared to baseline schemes while achieving faster convergence.

키워드