On pareto-koopmans efficiency for performance-driven optimisation in self-organising networks

Abstract

In this paper, a novel Multi-Objective Optimisation (MOO) method has been introduced for Self-Organising Networks (SONs). Meta-heuristic algorithms based on Simulated Annealing (SA) are used to evaluate the Pareto Frontier (PF) of UE throughput vs. fairness index in a simulation of Coverage & Capacity Optimisation (CCO) use-case in SONLTE. We have evaluated the performance optimisation methods through the final optimal set of solutions. The boundaries of the optimal sets are evaluated as PF and compared with the results of the conventional method of Multi-Objective Simulated Annealing (MOSA). We have detected a Pareto improvement for the estimated PF of the proposed method, which outperforms that of MOSA.