Telcos Must Take Lead to Develop Autonomous Driving Mobile Networks with AI: Huawei

In the age of Artificial Intelligence, the telecom industry must be prepared to take the lead to drive development of autonomous driving mobile networks, according to David Wang, Executive Director of the Board at Huawei in a keynote at the Ninth Global Mobile Broadband Forum (MBBF) in London.

“Over the past three centuries, three industrial revolutions have taken place, moving humanity from the mechanical and electrical eras to the information age,” Wang noted. “Each revolution has ushered in a new age and today rapid development of AI is triggering a Fourth Industrial Revolution ushering humanity into an era of intelligence.”

Wang began his speech by reviewing Huawei’s AI strategy and full-stack, all-scenario AI portfolio, which the company released in October. "Full stack" refers to the functionality of Huawei technology. Huawei's full-stack portfolio includes chips, chip enablement, a training and inference framework, and application enablement. By "all scenarios", Huawei means different deployment scenarios for AI, including public clouds, private clouds, edge computing in all forms, industrial IoT devices, and consumer devices.

5G is now. Growing network complexity presents three major challenges for mobile network deployment. In an era in which 2G, 3G, 4G, and 5G co-exist, the complexities of network management cannot be addressed manually. New technologies and architecture such as Massive MIMO and heterogeneous networking make it more difficult to maximize network potential. As 5G enables more diverse mobile network services, the management of user experience is increasingly complicated. To address these challenges, Wang noted, we should deeply integrate AI with mobile networks to deliver a superior user experience, increase operating efficiency, and ultimately make autonomous driving mobile networks a reality.

Autonomous driving networks go far beyond the innovation of a single product, and are more about innovating system architecture and business models. Huawei calls for all industry players to work together to clearly define standards and guide technological innovation and rollout. With the aim of improving service experience and operating efficiency, Huawei has proposed to build autonomous driving mobile networks step-by-step, from network planning and deployment, to operation and maintenance (O&M) optimization and service provisioning. Huawei has defined five levels of autonomous driving mobile networks and set 12 goals for the five levels, including assisted scenario-specific planning, automatic equipment configuration, automatic alarm monitoring, and coverage-based service provisioning.

Autonomous driving mobile networks are part of Huawei's overall architecture for autonomous driving telecom networks – SoftCOM AI. According to Wang, intelligent mobile network capabilities must be built at the site, network, and cloud levels, due to the unique characteristics of mobile networks.

At the site level, Huawei focuses on building capabilities such as scenario mapping, data collection and extraction, and low latency intelligent algorithms. The result is real-time data analysis and low latency intelligent inference.

At the network level, Huawei shifts its focus from network element-centric O&M to scenario-centric O&M and from network management to the convergence of management and control. This is to enable automated management and control of mobile networks by building risk prediction and intelligent identification capabilities into networks.

At the cloud level, intelligent models and training capabilities are built to deliver new-generation intelligent services. In the future, Huawei will utilize its global experience in automated network management to provide AI model development and training services for operators and continue to develop new AI services and update AI models while delivering AI as a Service (AaaS).

Wang highlighted that Huawei serves over 400 global operators and has nearly 30 years of experience in the O&M of over 150 telecom networks. It also boasts an expert team specializes in telecom networks, which is one of the key strengths for the company to make autonomous driving mobile networks a reality.

Autonomous driving mobile networks are not built overnight. The value of automation can only be unleashed step by step, scenario by scenario. Huawei released a white paper on autonomous driving mobile networks during the event that goes into the details of how to automate mobile networks in seven scenarios, including site deployment and energy savings. As research advances, Huawei will update the use cases and publicize relevant research results.

"The age of AI is here. Huawei looks forward to working with all industry players to apply AI in mobile networks and develop innovative methods for network management," said Wang. "With the aim of higher efficiency, better performance, and more agile services, Huawei is committed to taking complexity itself while creating simplicity for our customers. Together, we will make autonomous driving mobile networks a reality in the 5G era."