Autonomous Driving Network (ADN)-Huawei

Autonomous Driving Network (ADN)

Simplified Network with Intelligent O&M

Watch Video White Paper ADN Lab

Network Challenges
  • 01

    Comprehensive business

    Enter multi-play network business by evolving from B2C to B2X covering myriad of vertical markets like drones, IoV, and AR/VR and more.

  • 02

    Complex O&M

    Coexistence of 2G, 3G, 4G and 5G mobile networks has created multifold of complexity in all aspects of network O&M.

  • 03

    Extensive connections

    Intelligent connectivity has enabled extension to connection density by 100 times, connection volume by 1000 times and mobility speed of up to 150 km/h.

What Is ADN

"Take Complexity, Create Simplicity" is the supreme principle of autonomous driving network (ADN), which is practiced throughout product planning, design, and development at Huawei. Collaborative evolution of network and O&M intelligence is the core of ADN, and in this regard Huawei has produced a three-layered open architecture delivering intelligence for networks and platforms for O&M, enabling telecom operators to accelerate their digital transformation.

The three layers are cloud intelligence, network intelligence, and NE intelligence, which combine to achieve effective ADN.

Cloud intelligence: Telecom knowledge assets are aggregated in the cloud, generating an intelligent platform for data training as well as model generation and optimization. The results are then synchronized to the network and NE layers, ensuring the optimal utilization of up-to-date models.

Network intelligence: At the network management and control layer, big data analysis, intelligent algorithms, and service APIs are adopted to achieve service intent automation, network O&M intelligentization, and network servitization.

NE intelligence: A lightweight intelligent inference framework is embedded at the device layer to provide NE-level short-period awareness analysis and inference capabilities, with inference completed in microseconds.

  • 21

    Enhance O&M efficiency

    Scenario-specific and open programmable APIs improve O&M integration efficiency, reduce job nodes that need manual intervention, shorten service time, and minimize errors caused by manual operations.

  • 22

    Optimize resource utilization

    A network traffic forecast model is generated through AI training to implement real-time network resource scheduling and topology management based on the network traffic trend, optimizing network resource utilization.

  • 23

    Increase energy efficiency

    AI training is utilized to forecast network load and generate energy consumption models for base stations and data centers, enabling dynamic energy provisioning based on network loads.

  • 24

    Improve user experience

    The cloud and network collaborate under the user-unaware configuration to implement real-time service provisioning, on-demand selection, SLA assurance, and zero interruption, achieving optimal user experience with the right service at the right location.

Products and Solutions

  • Network management and control unit (FBB) iMaster NCE

    Focuses on fixed network scenarios such as Mobile Backhaul , premium broadband, premium private line, DCN, and enterprise campus; integrates manager and controller to translate business intents into network operations, and provides open APIs to achieve quick IT integration, making networks simple, smart, open, and secure.

  • Network intelligent unit iMaster NAIE

    Integrates intelligent knowledge at the network layer, provides multiple cloud services such as data lake, model development and training, as well as model inference, and empowers ADN with simplified network intelligent application development.

  • Intelligent cross-domain O&M unit iMaster AUTIN

    Runs on the digital O&M platform OWS, and utilizes AI, big data, and cloud technologies to help carriers implement all-online, automated, and intelligent O&M, as well as enhance O&M personnel skills.

Use Cases