Huawei ADN Lab is a joint innovation practice center and communication platform built based on openness and win-win cooperation by Huawei and its partners from the vendor, academic, research, and customer sectors to advance the ADN industry.
The ADN Lab aims to incubate key technologies, facilitate standards collaboration, and promote industry cooperation. The lab focuses on technological researches into network automation, intelligence, cloudification, man-machine interaction, trustworthiness, and other fields; it offers open software and hardware resources for joint verification of key service scenarios with carriers, enterprises, academic institutions, and developers to address crucial ADN challenges.
Here, industry partners can collaboratively explore future networks and share the benefits of the ICT digital economy.
Joint Innovation with Customers
Joint Innovation in Standards
Solution Experience
Open Data Platform
Model Training Platform
Training and Education Platform
Technological Information Sharing
Annual ADN Forum
Technology Workshops in Research Centers
Management and control units have become the main center of ADNs. Attackers can find the weak points of a management and control unit from multiple paths and continuously penetrate into the unit to control a large number of network devices without intruding them individually. Attacks such as 0-day vulnerabilities and APTs are rapidly increasing as cyber attack techniques evolve. How can we eliminate these 'undefined threats' which cannot be perceived and identified by rule-based detection?
O&M management objects and data on 5G networks are increasing at an exponential rate. Carriers require large networks to be centrally managed by province-specific management and control units or a nation-wide management and control unit. They also expect network management experience to meet the 2-5-8 principle. How can we ensure efficient data governance and deliver optimal user experience on ultra-large networks?
The accuracy and quality of telecom network AI models deteriorate with time or site conditions. How can we normalize different data change features to relatively consistent mathematical models and develop basic, common detection algorithms? How can we use these algorithms in a detection mechanism to identify what additional feature data is required?
The evolution towards autonomous driving telecom networks is a process that involves reducing manual operations and increasing machine operations. A suitable man-machine trust model can build better trust between humans and machines. How can we evaluate the man-machine interaction level of an ADN and develop suitable man-machine trust models to ensure explainable and predictable machine behavior? This is key to ensuring the stable running of an AI system.
Industry communication requires a transformation from best effort to E2E deterministic SLA for network services. Cloudifying network communication devices requires deterministic SLAs to be broken down layer by layer and be continuously guaranteed. How can we break down E2E deterministic high-level SLAs layer by layer for a complex network to ensure these SLAs are effective?
On an ultra-large complex network (one that features a large number of devices, high bandwidth, and complex service requirements), how can we develop high-precision functionality and SLA simulation capabilities to verify configuration changes online in real time?
Telecom network faults often occur in different domains (such as wireless and transmission domains) and have mutual impacts, causing a long MTTR. How can we use prior knowledge to automatically locate root causes of cross-layer faults, and infer and visualize the fault propagation chain for a complex network to achieve proactive fault prediction, prevention, automatic locating, and rectification?
Wireless networks often need to be optimized based on the physical environment or user experience. These optimization activities are challenging due to a large number of targets, objectives, and parameters, especially on 5G networks. How can we develop multi-target, multi-object, and multi-parameter fast optimization capabilities for ultra-large networks?
The power consumption of global cloud computing DCs continues to increase and accounts for a large proportion of cloud computing O&M costs. These DCs are facing challenges such as power supply and environmental protection. How can we maximize computing power per watt by scheduling services and optimizing cooling control?
Huawei proposes an ADN strategy to deploy automatic, self-healing, and self-optimized autonomous networks. This strategy enables agile innovation and automatic O&M, delivers optimal experience for telecom services, and improves resource and energy efficiency. The evolution towards ADNs requires long-term efforts from all parties. The Huawei ADN Forum is held at Songshan Lake Campus, Xi Liu Bei Po Cun in Q4 every year to focus on key ICT ADN O&M challenges, offering an opportunity for professors and industry experts to share innovative ideas. All parties from the vendor, academic, and research sectors are welcome to join the forum and accelerate the evolution towards ADNs.
Huawei ADN Lab is dedicated to collaborating with academic and research institutions in various modes, including but not limited to delivery cooperation, sponsored research, consultancy, gift funding, academic conference sponsorship, and joint labs.
If you are interested in partnering with ADN Lab, feel free to email vivienne.guan@huawei.com for collaboration solutions and modes.
