Edge & 5G
Engage with our experts on Edge & 5G topics
66 Discussions

Intel AI in the RAN

Cristina_Rodriguez
1 0 1,160

AI in the RAN: A Game Changer for Optimizing and Monetizing Networks with New Services

With the advent of 4G, the mobile industry developed an ambitious vision for self-organizing networks that could configure, manage, and heal themselves automatically and dynamically based on changing conditions. We imagined a world where networks could balance traffic workloads across multiple cells and correct coverage and capacity problems on the fly—all without human intervention.

Sadly, all the envisioned benefits of self-organizing networks weren’t fully attainable with 4G. Processing power couldn’t meet the challenge, and prevailing artificial intelligence and machine learning (AI/ML) solutions couldn’t make timely decisions or handle the explosion of data. The challenge today is compounded by the more complex network architecture of 5G. However, the vision for an intelligent, automated RAN is rapidly approaching reality thanks to considerable hardware and software advances in AI/ML technologies. The standardization of intelligent RAN by industry bodies like 3GPP, the O-RAN Alliance, and the Telecom Infra Project (TIP) provides the foundation for this vision.

By virtualizing their 5G radio networks, communications service providers (CoSPs) can significantly expand their range of network resource control using methods such as dynamic network re-configuration, traffic steering, and optimizing spectrum allocation. Ultimately, this will lead to automated, dynamic network optimization that will reduce total cost of ownership—and enable opportunities to monetize the network with new revenue-generating services.

 

AI optimizes networks to generate total cost of ownership savings

By employing AI in the RAN, service providers can manage their network operations and resources more granularly and in real-time. This can deliver very real operational and capital cost savings and maximize their return on investment. For example, a closed-loop system backed by reinforcement learning can automatically create the best settings or configuration to meet a given workload requirement with minimum resources. This results in more productivity from infrastructure like radios and base stations, and eliminates the time and effort required to perform resource planning manually.

AI/ML can also provide real-time response by dynamically adjusting to changing network demands for right-fit resource allocation. One AI-powered solution, Capgemini’s Project Marconi deploys ML to improve spectral efficiency with real-time predictive analytics for faster insight and action. In testing, the AI/ML solution, optimized with AI software and 3rd Gen Intel® Xeon® Scalable processors, achieved 15 percent better spectral efficiency by increasing the amount of traffic each cell can handle.

Of course, AI/ML isn’t only about reacting to changing network conditions—it also predicts them. Predictive analytics can model usage data and trends over time to preemptively optimize network resources and minimize idle or underutilized resources. Imagine a situation where user demand consistently skyrockets in a specific location on a particular day, such as at a professional football stadium on a Sunday. AI algorithms would recognize a pattern of need and proactively adjust network resources for increased bandwidth in the stadium during the game. When the crowds have left, network resources are reduced in that area or reallocated to another high-demand area.

There’s an environmental benefit to this as well. Dynamic network resource management, enabled by predictive and inferencing algorithms, improves a network’s energy efficiency and power management. Network sustainability is an industry-wide priority, and AI/ML will play a considerable role in fulfilling operators’ goals.  

 When it comes to enhancing network performance reliability, AI/ML offers big advantages. It can significantly improve network operations by enabling problem detection and automated intervention at scale. AI/ML-enabled systems will learn and continuously model normal behavior at every dimension, then respond to detected anomalies in areas like energy usage, quality assurance, and load balancing. In addition, AI and autonomous security management can harden the security of mobile networks by countering potential threats and live attacks.

 

AI enables unprecedented value-added services

With AI in the RAN, CoSPs can create new revenue streams by offering value-added services to enterprise and high-end consumers in ways they couldn’t with 4G. Organizations such as media providers, factories, retailers, healthcare systems, and smart cities can benefit from guaranteed throughput, latency, reliability, and increased quality of experience.

AI-powered virtualized RAN makes it possible to deliver on service level agreements (SLAs) and to facilitate services at scale. For example, an automated factory assembling car parts requires ultra-reliable, low-latency communications (URLLC) to ensure that it meets production targets without costly downtime. With AI-powered networks, CoSPs can bolster delivery of performance-related SLAs and sell that as a service to the manufacturer.

How does AI make this possible? CoSPs can employ real-time information on RAN resource utilization—powered by AI-based algorithms used for traffic routing, location prediction, channel quality prediction, and user selection—to create and maintain end-to-end network slices relative to customer SLAs. Network slicing can provide a cost-effective means of supporting multiple services that require high bandwidth, low latency, and mass connectivity on shared infrastructures. In addition, AI-powered traffic steering can automatically route traffic to meet SLAs based on how the customer prioritizes the type of traffic, such as media, voice, or text.

 

Intel delivers the flexibility you need to bring AI to the RAN

There’s a reason why nearly every vRAN deployment worldwide runs on Intel. Over the last 10 years, we’ve been an industry leader in virtualizing the network core, bringing that expertise and experience to the RAN. And our advancements in AI/ML are driving the industry toward realizing the vision of a fully automated and intelligent RAN.

Our 3rd Gen Intel® Xeon® Scalable processors are the only server CPU with built-in AI acceleration. Intel’s next-gen Intel® Xeon® Scalable processors—previously codenamed Sapphire Rapids—will build on these capabilities and set new standards in processing power. We also offer a broad array of AI/ML tools, software development kits, networking applications software solutions, and ecosystem partnerships to enable CoSPs to realize their business goals.

The opportunities to improve radio networks with AI also extend into other innovations that will make radio networks even more intelligent.

In my next blog, I’ll highlight the mobile industry’s emerging RAN Intelligent Controller (RIC) technology and the Open RAN ecosystem software developers who are spearheading new AI innovations. I’ll also reveal more about how Intel and its partner ecosystem enable new use cases and service opportunities in the distributed unit (DU) based on RIC technology.

For more information about AI innovations based on 3rd Gen Intel® Xeon® Scalable platform, read our guide, “Accelerate Your AI Today.”

Notices and Disclaimers:

Performance varies by use, configuration, and other factors. Learn more at www.Intel.com/PerformanceIndex

​No product or component can be absolutely secure.  ​

​Your costs and results may vary.​​​

© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others​.​

About the Author
Vice President Network & Edge Group General Manager Wireless Access Network Division. Cristina Rodriguez is vice president in the Network & Edge Group, general manager of the group's Wireless Access Network Division (WAND), and general manager of the Austin Design Center at Intel Corporation. She leads Intel’s efforts to provide innovative wireless access solutions in both traditional and cloud native networks, enabling the RAN of the future by bringing together 5G, the build out of the Edge and Artificial Intelligence. Rodriguez has full PnL ownership, responsible for strategy and business plans, technical product marketing, roadmaps, product definition, forecasting, customer relationships, investment decisions and GTM. Rodriguez joined Intel in 2014 with the acquisition of LSI Corp.'s Axxia Networking Business from Avago Technologies Ltd., where she led and carried out software strategy. In her first Intel role, Rodriguez managed the division's worldwide software engineering and product development efforts. She also led the successful integration of the Axxia team into Intel and aligned the team's focus with Intel's networking business. Her team has been recognized multiple times with Intel’s highest honor, the Intel Achievement Award, for their contribution to the company’s 5G leadership efforts. Before her tenure at LSI and Avago, Rodriguez spent seven years holding leadership roles in software development at Agere Systems Inc. and its parent company, Lucent Technologies Inc. Earlier in her career, she worked at Zilog Inc. and at two startup ventures. Rodriguez earned her bachelor's degree in electrical engineering from Instituto Superior Politécnico José Antonio Echeverría in Havana, Cuba; her master's degree in computer science from Universidad Politécnica de Madrid in Madrid, Spain; and completed the Stanford Executive Program at Stanford University. Rodriguez is also a member of the Intel Latinx Leadership Council and was recognized as one of HiTEC’s “100 Most Influential Hispanic Leaders in Technology” in 2021 and 2022.