5G AI ‐Enabled Automation

Abstract

Abstract Artificial intelligence (AI) has been investigated for more than 60 years. With the development of computing power, innovative algorithms, and gathering of massive data, AI will be rejuvenated in the next decade and become one of the most important enabling technologies for 5G automation. How to use the powerful analysis, judgment, and prediction capabilities provided by the AI algorithms to enable 5G equipment, networks and service systems, and combine them with the planning, construction, maintenance, operation, and optimization of carriers' networks become an important topic in the ICT industry. Based on deep understanding, accumulated experience in telecom field, and long‐term investment in AI strategies, many organizations have been introducing AI technology to telecom networks and contributing to development of standards essential for improving the quality and security of AI products and services, ensuring user safety and creating a fair and open industry ecosystem. 5G networks can be automatically deployed (automatic service deployment and automatic running) with the introduction of AI based on application scenarios. AI algorithms can also be used for self‐healing (automatic fault recovery), self‐optimization (network self‐optimization) and creating autonomous 5G systems (self‐evolving networks), which continuously improve network efficiency and reduce OPEX. This article aims at presenting the most relevant use cases, prominent research outcomes, latest standard developments and related standardization gaps, examples of AI strategy interpretation, AI solutions – including communication intelligent platform, AI model service, and deployment solution – and typical application scenarios. As a part of this framework, this article introduces the most important aspects of AI security, in terms of protecting the integrity and confidentiality of AI models and data, and thus preventing attackers from changing the inference results or stealing data.