A Data Driven Coordination between Load Balancing and Interference Cancellation

Having multiple optimization functions in a mobile network brings demanding challenges in terms of coordination of potentially conflicting objectives. Typically each function aims at optimizing specific utilities modifying parameters that are coupled to other functions, which jeopardizes the stability of the system, specially if the policies followed by each function are dissonant. There are two commonly accepted approaches for Self-organized networks Function (SF) coordination: on the one hand side, there are heading (tailing) orchestration techniques in which a priori (posteriori) individual parameter conciliation takes place. On the other hand, it is possible address the conflict among SFs using global algorithms and optimize the network performance as a whole in a centralized manner. In this study, we aim at a solution of the second kind, interleaving the dynamic of two well-known SFs, namely Mobility Load Balancing (MLB) and Inter-Cell Interference Cancellation (ICIC), into one global technique and optimizing the global network performance using a method that combines fixed point algorithms and machine learning.