What Do Level 4 Autonomous Driving Networks Mean for Operators?

Huawei’s Autonomous Driving Network solution has fully achieved L3 and is accelerating evolution to L4. With its promising progress in the field of network intelligence, the ADN solution garnered much attention at MWC 2023 last month.

L4 highly autonomous networks build on L3 capabilities to accommodate more complex cross-domain environments. They achieve predictive or active closed-loop management of service and customer experience-driven networks. Operators can then resolve network faults prior to customer complaints, reduce service outages, and ultimately, improve customer satisfaction.

ADN leverages key technologies such as digital twins and intent APIs, and exposes network capabilities to build a solid Network as a Solution (NaaS) foundation for the digital transformation of Communication Service Providers (CSPs).

CSPs embarking on the digital transformation journey used MWC as an opportunity to gain  better understanding of ADN principles, commercial values, and development trends.

Here’s why the ADN solution is in the spotlight:

• According to a TM Forum survey, more than 90% of CSPs have added or plan to incorporate network automation into their corporate strategy.
• The ADN solution has fully achieved L3 and will accelerate CSPs’ evolution toward L4, embracing the 5.5G era.
• Huawei collaborates with tier-1 CSPs in China and many CSPs around the world, such as MTN, to contribute to successful practices in intelligent network automation.

Related: MTN Group Boosts User Experience with PremiumWi-Fi

So, why does Huawei offer ADN?

Since Huawei proposed evolution to the 5.5G era in 2022, L4 ADN and many other key technologies, including 5.5G, F5.5G, and NeT5.5G, have become key capabilities for the success of CSPs in the 5.5G era.

From the perspective of network O&M, ADN eliminates the need for O&M engineers to handle a significant number of network alarms and perform repetitive work. It allows them to focus on innovation, significantly improving the innovation success rate.

From the perspective of customer experience, ADN features revolutionary improvements from service provisioning to service usage, and enables CSPs to improve customer loyalty, brand image, and ARPU, thus facilitating innovation in the 5.5G era.

Three reasons why CSPs can benefit from ADN

1. Wide Applicability

At 2019 PT Expo China, Lu Hongju, President of Huawei General Development Department, released the future-oriented ADN architecture, which offered AI at the network element (NE), network, and service layers.

Years of development have yielded a systematic ADN target architecture that covers many domains, including network, computing, and storage. Aligned with Huawei’s commitment to Intelligence for ICT, we systematically apply AI technologies to ICT infrastructure to overcome key challenges and enable CSPs to deploy next-generation telecom networks.

ADN is applicable to nine domains:

• Optical access networks for homes
• Wireless and cloud core networks for terminals
• IP and optical transmission networks for WAN
• Private lines for governments and enterprises, campus networks, and data center networks
• Intelligent O&M services for a business close loop

2. Strong reputation

In 2021, three tier-1 CSPs in China announced the goal of reaching L4 Autonomous Networks, as defined by TM Forum, by 2025. Vodafone, MTN, Zain, and AIS also proposed the same goal, showing that ADN is the target network that enables the digital transformation of CSPs and that L4 ADN is a key feature in the 5.5G era.

3. Many use cases

As the parties that first proposed the concept of ADN and launched innovative practices, China Mobile and Huawei have reached a consensus on the vision, objective, industry standards, target architecture, and implementation path of ADN. Huawei helped China Mobile improve its Autonomous Network (AN) level from L2.1 in 2021 to L2.7 in 2022. China Mobile and Huawei have jointly deployed three benchmark projects, seven demonstration sites, and numerous pilot programs across 21 provinces in China.

Based on the systematic and strategic collaboration with China Mobile, Huawei has summarized four steps to facilitate continuous ADN evolution:

1. Formulate a top-level design, including the vision, strategy, and target architecture of ADN.

2. Define ADN levels, including level standards and the evaluation method.

3. Select effectiveness indicators that cover user experience, network O&M, and other domains.

4. Launch pilot programs and practices, including use cases and projects for continuous improvement.

What value does ADN deliver?

The scenario-specific value of ADN lies in advancing network automation capabilities to improve quality and efficiency, reduce costs, and increase revenue.

Value 1

Automatic service provisioning is implemented to achieve “zero-wait”.

Scenario: SLA visualization and automatic provisioning of private line services

Network-level interfaces and standard RESTful APIs that transmit inventory, performance, alarms, and configuration data in the northbound direction are used to interconnect with OSSs and BSSs, implementing service provisioning and experience assurance for customers in e-commerce mode based on NaaS. The time to market (TTM) required for private line service provisioning is shortened from weeks to days. SLA visualization shortens the time required for resource check and planning from hours to minutes.

Total labor costs are cut by 85%.

Value 2

Self-healing is implemented to achieve “0-1-3-5”.

Scenario: Alarm compression for wireless intelligent fault management

Potential risks can be predicted. Faults can be detected within 1 minute, located within 3 minutes, and rectified within 5 minutes, achieving multi-objective collaborative network optimization for optimal performance and energy efficiency. Assuming 20,000 base stations, the number of daily alarms is reduced by 90% from 14,000 to 1,400, and the number of customer trouble tickets is reduced by 23%.

Proactive fault prevention is achieved based on the analysis of frequent fault points.

Value 3

Self-optimizing experience management improves satisfaction by 50%.

Scenario: Premium broadband experience improvement and precision marketing

• Intelligent algorithms are used to fully identify weak coverage and interference issues of unmanaged Wi-Fi routers, providing a clear picture of home Wi-Fi experience. This helps CSPs attract potential high-value customers and increase the sales success rate among potential customers from 5% to 15%.
• Second-level correlation analysis is performed on 110 types of application KPIs and 300+ types of network KPIs to quickly identify 30+ types of root causes of poor home broadband experience, including Wi-Fi network congestion, 100BASE-T cables, and weak optical signals. The accuracy of identifying poor-QoE users is increased to 95%.
• The trouble ticket system implements efficient collaboration and configuration optimization to quickly eliminate bandwidth bottlenecks, ensure bandwidth experience, and perform closed-loop optimization for poor-QoE problems, and continuously improves user experience. User satisfaction rate is improved by 101%.

Value 4

Energy saving is implemented to achieve “zero-bit, zero-watt”.

Scenario: Energy-saving for wireless base stations

Models are developed for base station energy consumption and frequency bands, coverage, and performance based on numerous practices. These models dynamically generate hour-level, site-specific power saving policies to achieve precise energy saving, thereby ensuring optimal network experience and energy efficiency. The average power consumption of a single site is reduced by 25%.

As Autonomous Networks become a clear industry direction, 91% of CSPs around the world have added network automation to their strategy, and tier-1 CSPs have unanimously announced the goal of reaching L4 by 2025. Huawei is looking forward to collaborating with industry partners to evolve toward L4 over the next two years. The evolution from L3 to L4 is an important quantitative to qualitative change for ADN. AI machine learning will enable this evolution by 2025, facilitating the transformation from machines assisting humans to humans assisting machines.

Read more about Huawei Autonomous Driving Networks.

Disclaimer: Any views and/or opinions expressed in this post by individual authors or contributors are their personal views and/or opinions and do not necessarily reflect the views and/or opinions of Huawei Technologies.