AI in Network Seminar Powered by Beta Labs Keynote Xiongyan - PowerPoint PPT Presentation

智慧网 络论坛 AI in Network Seminar – Powered by Beta Labs

Keynote 主 题演讲 Xiongyan Tang 唐雄燕 Chief Scientist, Research Institute 网 络技术研究院首席科学家 China Unicom 中国 联 通

AI-enabled Network Transformation and Service Innovations Dr. Xiongyan Tang CTO ， Intelligent Network ， China Unicom 2019-06-27

AI-enabled intelligent network is the future network IP Network 1 2 TDM Network 4 3G/4G, DSL/FTTx AI-enabled Intelligent 2G/PSTN Network 3 SDN/NFV-based 5G/6G, FTTH 4G/5G, FTTH Cloud Network 1 2 3 4 Voice Traditional Intelligent Cloud Services Communications Internet Services 55

The core value of AI for telecom networks ： Intelligent network operation  Along with the expansion of the network and the increased number of connections, as well as the virtualization of the network, the network operation faces great challenges. It becomes essential to realize the intelligent network with AI and big data.  SDN/NFV laid the foundation for applying AI in telecom networks. Intelligent Network Traditional Network Intelligent operation & maintenance + Manual operation & maintenance+ Autonomous evolution System assistance Real-time Reasoning and Control AI model library & off-line training Network Operation and Artificial Auxiliary tools maintenance personnel AI model Data platform Intelligence strategy • • Network self-operation Artificial maintenance • • Automatic inspection, self-evolution Post-optimization • • Closed-loop for planning, construction and maintenance Open-loop of planning, construction and maintenance 56

Network transformation and 5G development need AI  Increasing network traffic and connections make the size of network is greatly expanded. Network management and OPEX face challenges. Challenges  SDN/NFV/Cloud greatly increases the complexity and difficulty of network AI is a must for management and operation. 5G and network  The flexibility and complexity of 5G networks bring challenges to network transformation planning, maintenance and optimization.  Service innovations put forward higher requirements for automation and intelligence of network operation.  Data : A large amount of data generated by network, terminals Data and services lay the foundation for AI mining and analyzing. Opportunities  Computing Capability : Operators have abundant computing resources for AI applications including DCs, edge computing, Computing Algorithm and network connectivity. AI Factors  Algorithm : Matured ML and DL algorithms provide convenient tools for network AI applications.  Model Scenario : AI can be not only applied in network operational Scenario scenarios, but also applied to service innovations. 57

China Unicom’s new generation intelligent network: CUBE -Net 2.0+ CUBE-Net 2.0+ ： AI-powered Intelligent Network based on SDN/NFV/Cloud 应用层 应用组件 应用组件 应用组件 北向开放 API API AI 模型注入 NaaS 网络即服务 ④云化网络服务 大数据分析 AI 训练平台 AI 推理分析平台 业务协同和编排 … 虚拟网 资源 网络 云化 … 东西向集成 络功能 管理 控制 业务 东西向集成 面向信息交换 南向控制 面向用户中心的网络（ UoN ） 面向数据中心的网络（ DoN ） 的网络（ IoN ） 固网 泛 区域 区域 +AI 云数据中心 +AI 云数据中心 云数据中心 云数据中心 +AI 城域汇聚 宽带 IP+ 光网承载平面 在 无线 云互联 接 边缘 +AI 边缘 宽带 基础 入 网元 移动 云数据中心 +AI 云数据中心 +AI 云数据中心 移动回传 云数据中心 区域 +AI 区域 宽带 ①超宽带弹性管道 ③云接入与互联服务 ②泛在宽带接入 Introducing AI+SDN/NFV/Cloud to build the next generation intelligent, agile, intensive and open network 58

Today: AI Introduced into current network operation Applying AI to network planning, construction, operation and optimization to accelerate network transformation and improve network intelligence. Network Planning Network Maintenance • • Performance analysis • Intelligent diagnostics Automatic inspection • • • Control & scheduling Automatic troubleshooting MR grid assessment of Work Order (WO) • Root-tracing of network • Development prediction • Automatic verification alarm • of WO Deployment suggestion Network Optimization Network Construction • • Network adjustment and optimization Big data analysis • • RF self-optimization and automatic correction Intelligent design 59

Use Case 1: IPRAN intelligent alarm recognition  This system realizes the roots-tracing of network alarm and has been deployed in some provincial branches of China Unicom.  It has a high accuracy of root-cause location, and plays a supporting and assistant role in the maintenance and planning of IPRAN network. IPRAN intelligent alarm recognition Rule analyzing and Mining Statistical analysis of data Alarm data • Association Rule Mining Average compression rate Algorithms Recognition of offline alarm: ~90% knowledge base Association Rule Templates result Topological Association Rules Knowledge Base data • Average compression rate Blacklist rule knowledge On-line alarm filtering recognition feedback information Root-derived Association of online alarm: ~80% Service data User-side port matching Frequent alarm 60

Use Case 2: Intelligent data analysis for customer care  This system integrates user’s authentication and consumption information, terminal capability and service behaviors with network data to do collaborative intelligent analysis.  It realizes the intelligent diagnosis of customer complaints and assists the maintenance engineer to solve the problem quickly. Customers with complaints Telephone operator WeChat Subscription Mobile network first call resolution Network One key Fault- Capability Fault locating NPS Forecast Removing Packaging Intelligent Diagnosis Fixed network first call resolution Supervision&Scheduling instruction resources MR/TRACE HSS info …… BSS/CBSS XDR OSS equipment OSS service BSS data data data 61

Tomorrow: AI-enabled network re-architecture AI+SDN/NFV SDN （ Software-Defined Networking ） SDN （ Self-Driving Network ） On different layers of the network (infrastructure and measurement, network control and real-time analysis, overall intelligence and orchestration), AI capability could be introduced step by step and embedded into the network system. 62

5G+AI ： AI for 5G intelligence and 5G for AI applications  AI promotes the intelligence of 5G network and improve the overall network performance, as well as maintenance and operation efficiency. AI-based wireless network optimization AI-powered beam management Optimize parameter adjustment; AI-powered beam management Improve the utilization rate of wireless enables quick adaptation to resources and network capacity; Predict environment change, enhancing user trajectory/service requirements; access experience Optimize content cache and enhance QoE of users. AI to manage network slicing AI and 5G MEC • Automatic network slice configuration Edge computing provides key • Automatic network slice fault recovery capabilities for AI applications, and edge • Optimizing network slice performance AI provides support for the edge computing applications. Intelligent network slicing Edge computing and AI 63

China Unicom is developing the Network AI platform CUBE-AI Technical service platform CubeAI based on the design concept 01 to meet the needs of of the Linux AI Foundation (FLAI) network AI applications and project Acumos business innovation CubeAI platform integrates AI Industry cooperation platform 02 model development, model sharing to create an multi-win AI and capability opening ecosystem Technology sharing platform for open source contributions, technology exchange, and application demo 64

Summary and Prospect: AI-enabled intelligent operators  AI-enabled intelligent network is a new trend of telecom network Build Achieve development, which has far-reaching Intelligent Intelligent impact and great potential. Network Operation  The application of AI in telecom network is still in the early stage and needs continuous attention and exploration.  China Unicom will actively conduct Provide Create network AI applications and apply AI Intelligent Intelligent to improve network operation Services Ecosystem efficiency and service intelligence.  China Unicom is looking forward to working with industry partners to promote the development of network Moving towards AI and co-create the new era of the Age of Intelligence network intelligence. 65

Thanks!