All-Optical ADN: All-Optical Connectivity with Deterministic Experience
[Shenzhen, China, May 7, 2021] At the Intelligent Optical Networks Webinar on AI and optical networks, Chu Tao, president of Huawei NCE Transport and Access domains, shared his ideas about how to enable all-optical connectivity with deterministic experience, how to activate gigabit optical networks, and how to achieve business agility and efficient O&M using the all-optical autonomous driving network (ADN) solution.
president of Huawei NCE Transport and Access domains Tao Chu
The following are some of the key points from his speech:
The telecommunications industry has developed rapidly since the 1980s, evolving from generation to generation and enabling connectivity across a vast number of households and industries. The Fifth-Generation Fixed Network (F5G) technology, which runs parallel to the wireless 5th Generation (5G) technology, has also been widely applied throughout the telecommunications industry, functioning as root technology for unlocking the all-optical network era. Behind this inter-generational evolution, three different types of driving forces exist. First of all, before 3G and F3G were established, overall user penetration rates were consistently increasing. This led to continued revenue growth within the industry, with carriers as the main beneficiaries of demographic dividends. In the 4G and F4G eras, services such as HD video, online gaming, and Internet social networking developed rapidly as network traffic steadily increased. In the traffic dividend era, a large number of Internet enterprises, such as Tencent, Baidu, Facebook, and Google, generated massive business profits. Today, we embrace a new, fully connected era (fiber to everything) of F5G. Optical connections extend from enterprises and homes to rooms and smart devices, and have penetrated into the production flow of numerous industries. According to the F5G definition provided by the European Telecommunications Standards Institute (ETSI), the core of F5G is to ensure the high-quality connections required for digital home and informatization innovation of various industries through deterministic experience. This represents a new era of experience dividends, and the start of a new phase for carrier profit models and opportunities.
According to PwC's Global Consumer Insights Survey, customers are willing to pay a premium of over 16% for a better experience. In 2019, Thailand's 3BB released the "Smart Mesh" smart home services. With an extra monthly payment of about US$1.7, customers could enjoy full-home Wi-Fi coverage and low-latency (within 100 ms) gaming experience. This enticing offer attracted more than 10,000 subscribers within the first month of launch. This willingness to pay a higher price for premium services is especially true in government and enterprise scenarios. In China, government and enterprise customers of China Telecom, Shanghai paid eight times the standard private line rental fee to obtain low-latency (0.6 ms) private lines, while in the U.S., the stock exchange paid 10 times the standard rent to reduce latency by 1 ms. As such, we can see that deterministic experience comes at a premium in the telecommunications industry. Huawei's all-optical ADN solution makes it possible to monetize deterministic experience. All-optical ADN is divided into three layers, with the basic layer being NE intelligence with edge intelligence. Huawei all-optical network devices provide second-level data analysis with more than 1000 built-in optical parameters to detect the status changes of devices, networks, and services in real time. Meanwhile, the intelligent AI training architecture on the cloud provides Artificial Intelligence for IT Operations (AIOps) capabilities. (Intelligent O&M: applies AI to the O&M field and uses machine learning to further solve the problems automatic O&M cannot based on existing O&M data.) Through data services with ultra-high computing power and AI training with continuous iteration and update, automation for carrier services is achieved and continuously optimized. iMaster NCE, serving as a connecting system, is the core component of the all-optical ADN architecture. Based on the Manager, Controller, and Analyzer engines, iMaster NCE provides real-time full-lifecycle closed-loop management capabilities for all-optical networks. This all-optical ADN solution with cloud-local synergy enables carriers to achieve agile business for services such as home broadband as well as government and enterprise private lines, as they transition from selling bandwidth to selling committed SLAs.
As of March 2021, over 700 sets of all-optical ADN solutions with iMaster NCE as the core were commercially deployed worldwide, taking the lead within the industry. Let's take a look at two typical cases, starting with China Telecom in Hebei. Through Premium Broadband 1.0, represented by committed bandwidth, China Telecom, Hebei acquired 150,000 subscribers in just four months by providing value-added services with full-home Wi-Fi coverage based on comprehensive home broadband bottleneck analysis capabilities. Through massive data training, up to 95% of third-party Wi-Fi router models can be evaluated, also taking the lead within the industry. The second case involves China Mobile, Beijing. For Premium Private Line 1.0, which is aimed at the government and enterprise industry and represented by latency monetization, Huawei helped China Mobile, Beijing release the "Smart Link" premium optical private line service, which features a unique latency measurement solution offering 0.1 ms precision and intuitive visualization capabilities. In just 18 months, more than 7000 services were provisioned, achieving great business success for the company. To date, China Mobile, China Telecom, China Unicom, and their branches have deployed more than 70 optical private line projects, with each choosing Huawei as their exclusive primary equipment partner.
Let's focus now on the upgrade to Premium Broadband 2.0, which transitions from committed bandwidth to committed scenario-based service experience. Following the widespread popularization of home broadband, users are posing increasingly high requirements for scenario-based services. In Thailand, up to 20% users complain about poor online gaming experiences. Meanwhile, according to statistics from China Mobile, more than 19% of online courses experience frame freezing. Given such figures, how can we help carriers ensure user experience of scenario-based home broadband services? This is the original intention of this upgrade, which requires three core capabilities. First of all, based on the In-situ Flow Information Telemetry (IFIT) data model analysis, iMaster NCE identifies various scenario-based applications (such as OTT, games, and online courses) through data training of millions of user data flow features, without impacting user privacy. On this basis, second-level service KQI data analysis can be used to identify poor service quality risks in seconds. In addition, based on time-slice-based service KQIs and spatio-temporal association analysis of network KPIs, iMaster NCE enables carriers to identify network bottlenecks that cause poor service experience in real time, while also achieving accurate poor-QoE demarcation and location. Finally, network optimization is implemented through policy-based optimization methods. Alternatively, mobile apps allow operations personnel to instruct users on how to resolve network problems impacting service experience. Thanks to these three capabilities, iMaster NCE enables carriers to build scenario-based broadband services with deterministic experience and create digital homes that integrate entertainment and production. In 2020, upgrade to Premium Broadband 2.0 was carried out by the Huzhou branch of China Mobile, Zhejiang, resulting in poor-QoE identification accuracy exceeding 95%. Through tracking and closed-loop management, the poor-QoE rate was reduced by 80%, while the marketing success rate was improved by 30%.
Next, let's take a look at the upgrade to Premium Private Line 2.0, which focuses on government and enterprise connections. Many government and enterprise customers have specific requirements on the availability of private lines. In China, the price of a private line with 99.99% availability is 1.5 times that of a standard private line. The factors affecting the availability of private lines are optical fibers, devices, and service protection. iMaster NCE measures the service availability rate based on the calculation of these factors. With the measurement data and the service-oriented capability of iMaster NCE, carriers can evaluate the availability before sales, calculate the route availability during sales, and monitor and optimize the availability after sales in real time, achieving full-lifecycle monetization. The core of the implementation is the availability measurement mechanism. In this regard, Huawei has developed a complete set of fiber length calculation solutions to avoid uncontrollable errors in manual fiber length evaluation. iMaster NCE can work with Huawei OTN devices to intelligently and accurately identify co-cable fiber issues, avoiding various errors caused by time-consuming and inefficient manual checks.
Network infrastructure is also a basis for constructing services with deterministic experience, as networks typically involve numerous problems and high processing costs. With this in mind, how can we reduce unnecessary door-to-door service costs? The answer lies with iMaster NCE, which provides second-level data collection and analysis capabilities and implements three-layer fault diagnosis and layer-by-layer filtering. First, segment-based speed tests are used to quickly demarcate problems and focus on the network location where the problems occur. In cases involving high-cost home network issues, iMaster NCE also provides a one-click detection function which allows the call center to comprehensively analyze seven types of typical home network faults within 30 seconds, enabling it to quickly and accurately locate faults. Finally, through the utilization of policy-based optimization methods, iMaster NCE automatically rectifies faults or displays messages to remind users of manual intervention, achieving self-service recovery. In this way, the number of door-to-door services can be reduced by more than 30%. According to statistics from China Mobile, more than 68% of WDM network faults are related to optical fibers, and more than half of these are incipient faults. iMaster NCE interworks with the built-in optical sensors of devices (such as oDSP, OTDR, and OFDR) to implement second-level sampling and analysis. Based on training and comparison involving a large number of fiber quality deterioration data models, carriers can accurately identify subhealthy fibers and predict the possible time, cause, and location of fiber faults. As a result, carriers can identify fiber risks in advance and perform pre-check and pre-repair, reducing the fiber fault rate by more than 30%. This solution has already been fully utilized by China Unicom in Jiangsu, where it only accurately predicts optical network faults in advance, but also greatly improves the timeliness of fault rectification. The duration for locating network optical fiber faults is reduced from 90 minutes to less than 15, and the fault rectification time is shortened by more than 30%.
These home and enterprise connection scenarios showcase the value delivered by deterministic experience to new home networks, thanks to the integration of entertainment and production and the renewed drive towards digital upgrade. In addition, zero-touch O&M capabilities are provided to help carriers comprehensively improve the quality of all-optical networks. This full-stack intelligent all-optical ADN solution empowers carriers to build all-optical connectivity with deterministic experience and explore a new industry space worth US$60 billion in the F5G era. Truly, deterministic experience is premium.
