Huawei "Autonomous Driving Network Solution White Paper"

News on May 21, recently, during the global analyst conference HAS 2020, Huawei released the "Autonomous Driving Network Solution White Paper" to the world, systematically explaining the future network architecture, operation and maintenance architecture and its key technologies, through network elements, networks and cloud The collaboration of the three-layer AI capability enables the network to move towards simplicity and ultra-broadband, and the operation and maintenance to move towards human-machine collaboration, providing practical reference for the digital transformation of operators and industry partners.

Huawei autonomous driving network ADN target architecture

Lu Hongju, President of Huawei's Public Development Department, said: "The next ten years will be a golden decade for the booming development of the intelligent era. New technologies represented by 5G, cloud, and AI will give the core momentum of connected intelligence upgrades. Huawei calls on colleagues in the industry to join in Explore practice and build an autonomous network that is automatic, self-healing, and self-optimizing through data and knowledge drive, and seize the new opportunities given by the digital economy. "

The white paper points out that building an autonomous driving network requires two major changes:

First, transform from a "network element-centered" fragmented network construction model to a "service-centered" building block autonomous domain network construction model. Realize single-domain autonomy and real-time closed loop through integrated "management-control-analysis", balancing the cost and speed of intra-domain innovation and inter-domain collaboration;

Second, the industry should work together to define the target architecture and programmable API standards for cross-domain open collaboration, greatly simplifying the complexity of cross-domain business collaboration and assurance, reducing research and operation costs and risks, simplifying integrated agile business, and reducing collaboration costs for the entire industry. .

At the same time, the white paper recommends L4 level (highly autonomous driving network) as the phased goal of future architecture, which should have the following four characteristics: 1. Digitization of network knowledge and expert knowledge, moving from passive manual operation and maintenance to predictive intelligent operation and maintenance ; 2. Minimalist architecture network infrastructure, network elements move towards intelligence; 3. Hierarchical single-domain autonomy and cross-domain collaboration, the network moves towards online real-time closed loop; 4. Unified cloud AI training, knowledge management and operation and maintenance design The platform supports the iterative evolution of telecommunications networks.

The white paper calls on the industry to achieve the ambitious goal of autonomous driving networks, which requires all parties in the industry to reach a consensus and jointly formulate unified standards and hierarchical evaluation systems based on the development generation, research generation, and exploration generation methods to form an efficient and collaborative industrial ecosystem and jointly promote industrial intelligence. Upgrade and healthy sustainability.