Evolutionary learning of link allocation algorithms for 5G heterogeneous wireless communications networks

Wireless communications networks are operating at breaking point during an era of relentless traffic growth. Network operators must utilize scarce and expensive wireless spectrum efficiently in order to satisfy demand. Spectrum on the links between cells and user equipments ('users': smartphones, tablets, etc.) frequently becomes congested. Capacity can be increased by transmitting data packets via multiple links. Packets can be routed through multiple Long Term Evolution (LTE) links in existing fourth generation (4G) networks. In future 5G deployments, users will be equipped to receive packets over LTE, WiFi, and millimetre wave links simultaneously. How can we allocate spectrum on links, so that all customers experience an acceptable quality of service? Building effective schedulers for link allocation requires considerable human expertise. We automate the design process through the novel application of evolutionary algorithms. Evolved schedulers boost downlink rates by over 150% for the worst-performing users, relative to a single-link baseline. The proposed techniques significantly outperform a benchmark algorithm from the literature. The experiments illustrate the promise of evolutionary algorithms as a paradigm for managing 5G software-defined wireless communications networks.