Scheduling in Heterogeneous Networks Using Grammar-Based Genetic Programming

Effective scheduling in Heterogeneous Networks is key to realising the benefits from enhanced Inter-Cell Interference Coordination. In this paper we address the problem using Grammar-based Genetic Programming. Our solution executes on a millisecond timescale so it can track with changing network conditions. Furthermore, the system is trained using only those measurement statistics that are attainable in real networks. Finally, the solution generalises well with respect to dynamic traffic and variable cell placement. Superior results are achieved relative to a benchmark scheme from the literature, illustrating an opportunity for the further use of Genetic Programming in software-defined autonomic wireless communications networks.