Guest Editorial: AI-Powered Telco Network Automation: 5G Evolution and 6G

Abstract

The fifth generation (5G) of cellular networks is significantly more complex than its predecessors due to several factors, such as increased cell density, differentiated service requirements, and coexistence with legacy networks. As a result, traditional operation and management (O&M) solutions, which heavily rely on human intervention, are no longer feasible to support such complex networks at reasonable operating expense (OPEX). Over the past few years, the telecommunication industry has come to the realization that leveraging artificial intelligence (AI) technology to enable a fully automated network O&M is a must to lowering OPEX and enhancing network key performance indicators (KPIs) for 5G, Beyond 5G (B5G), and the sixth generation (6G) of cellular networks. There have been numerous research efforts from both industry and academia to develop AI-powered network automation solutions. Many telecommunication operators and vendors have already adopted AI technology to automate some repetitive operational tasks and reduce reliance on personnel experience, such as cell planning, network deployment simplification, fault detection, and KPI optimization. While there has been notable progress for certain network O&M applications, the development of network automation solutions still faces several unique technical challenges that arise from telecommunication fields, including overwhelming network complexity, massive and diverse proprietary data, lack of industry-wide standards for radio access network (RAN) interfaces, and scarcity of labeled datasets.