Huawei Unveils Autonomous Network Solution

The drive for autonomous networks stems from the mounting complexity of modern telecommunications. The rollout of 5G, expansion of edge computing, and a surge in IoT devices have created intricate network architectures that are increasingly difficult to manage manually. This complexity can lead to higher maintenance costs, service errors, and slower response times. Huawei's ICNMaster solution aims to tackle these challenges by deeply integrating artificial intelligence into network operations and maintenance. The system leverages a "telecom foundation model," digital twin technology, and a "Toolbank" to automate the entire lifecycle of the core network, from planning and construction to maintenance and optimization. This allows for a shift from reactive problem-solving to proactive prevention. A key feature of the updated ICNMaster is its move towards "single-domain autonomy." This means creating a self-governing system for the core network that can automatically detect issues, optimize performance, and even heal itself with minimal human intervention. For example, intelligent agents can automate tasks like handling customer complaints and resolving network alarms, significantly boosting efficiency. The solution has already seen deployment in various locations with China Mobile. In one application, an alarm handling agent reduced the average time to handle core network alarm tickets from 90 minutes to just 12 minutes. This 87% improvement in efficiency highlights the potential for autonomous networks to free up experienced engineers from routine tasks. The push for network autonomy is an industry-wide trend, with companies like Ericsson and NVIDIA also developing solutions. The ultimate goal, as defined by industry groups like TM Forum, is to achieve different levels of autonomy, with Level 5 representing a fully self-managing and self-adapting network. Huawei's recent solutions are aimed at helping operators transition towards Level 4 autonomy. Underpinning these autonomous systems is the concept of a digital twin, a virtual replica of the network. This allows for online simulation and evaluation to detect risks, such as signaling storms, before they impact the live network. By creating this virtual model, operators can test changes and predict outcomes without affecting service.