Autonomous Networks is not a new concept. This term was first introduced in 2017, to describe the future when artificial intelligence (AI) capabilities are integrated  at the network element (NE), network, and cloud layers. The main benefits of the autonomous networks are the hyper-automated digital services and operations across industries.

With the ongoing development and adoption of next-generation technologies such as the Internet of Things (IoT) and 5G, we are quickly approaching an intelligent world where all things are connected, all things are sensing, and all things are intelligent. To support this new world, network operations and maintenance (O&M) will undergo a qualitative change.

According to Huawei’s Global Industry Vision (GIV) 2025, 97% of large enterprises will be using AI by 2025. By integrating AI capabilities, the vision of self-organizing, self-healing, and self-optimizing autonomous networks can become a reality. So what is the status of this integration?

AI in Network Elements (NEs)

 

AI is introduced to NEs in order to create digital intelligent NEs.

Tier-1s are leading this “race”, but small innovative vendors is also launching gradually NEs that are using AI in specific use cases. In this way, each NE and the entire network are endowed with more precise insights, processing, and inference execution.

AI in Networks

AI is used to reinvent the network management system (NMS).

There is limited progress in this area. Tier-1s are working actively in this area, but as their NMS systems are “large beasts”, the progress is slow. The target is to build a digital twin of a network through an intelligent management and control system, the network can be controlled dynamically and managed in a closed-loop manner. Key enablers include automatic deployment, pre-event simulation, post-event verification, prevention and prediction, and proactive optimization.

AI in the Cloud

Network AI training and model services are provided on the cloud.

There is good progress in this area, as Cloud ia an enabler of AI. The software systems and AI models on devices and networks (once progress is made in this layer) will be continuously upgraded to boost the autonomous driving capabilities. As a result, the more we use the network, the more intelligent it will become.

 

The next decade will be the golden age for intelligence. Telecom operators are expected to transition to the autonomous networks era, bringing intelligent connectivity within reach.