Title: Intelligent Networking, AI and Machine Learning for Telecommunications Operators

Progress at last, but challenges remain

Table of Contents:

1  Executive Summary - Key Takeaways and Overview

0.25 page – Beth Cohen  will fill in when the paper is mostly complete.

Since publishing  Intelligent Networking, AI and Machine Learning While Paper in 2021, the telecom industry has seen tremendous growth in both interest and adoption of the Artificial Intelligence and Machine Learning (AI/ML) technologies.  While it is still early days, the industry is now well past the tire kicking, lab testing phases that was the state of the art in 2021.  Intelligent networking is coming into its own as telecoms increasingly use it for operational support, whither that means deploying intelligence into their next generation 5G networks, where it is used for ?? and it continues its spread in operations.

Key Takeaways

• Some intelligent networking has been deployed by Telecoms, but more research and development is needed to establish industry wide best practices.
• Currently the industry is more interested in using artificial intelligence and machine learning to address operational, maintenance, and service assurance issues over network optimization. 
• The Telecom industry needs to develop common AI platforms and intelligent networking frameworks and methodologies to support the delivery of new services quickly and efficiently. 
• A shared understanding of intelligent networking will help to support interoperability.
• Creating commonly understood sources of reliable data is difficult, both within a company across business units and across the Telecom industry
• The Open Source community can play a key role in furthering the development of these frameworks and best practices. Some projects that look promising include, Anuket Thoth, O-RAN, 3GPP SA5 and ITU-T standards organizations, as well as the ONAP, O-RAN and TIP open-source projects.

2  Background Intro/History Beth Cohen 

0.5 page

Telecoms need to be able to incorporate new technologies and next-gen connectivity such as 5G, to customers and end users.  To achieve these ambitious goals, they need to optimize their networks – make them more intelligent if you will.  Some of the tools needed include artificial intelligence (AI), machine learning (ML) and artificial intelligence operations (AIOps). This document will explore what intelligent networking means to telecoms, vendors and customers, and how AI and ML technologies and tools can be used, the cultural shifts the industry needs to make it a success, and what to bear in mind when deploying machine learning across a telecom network.

The LFN (Linux Foundation Networking) End User Advisor Group (EUAG) is publishing this document to identify and highlight the latest thinking and recommendations for building and supporting intelligent networking and the tools needed to achieve it.  It will touch on the state of automation and adoption of intelligent networking tools by telecom operators.  This is a new area for many in the telecom industry, so the focus will be on the requirements, tools and approaches that have been deployed, and some potential futures for intelligent networking and AI/ML tools.  Some of the topics covered will include:

• Fundamental issues and challenges that might be solved by applying Intelligent networking technology
• Establish a definition of intelligent networking
• The promotion of intelligent networking transformations
• Results from a February 2021 survey of telecom operators and vendors in the industry ecosystem about the current state of industry adoption
• How intelligent networking might be incorporated into networks and processes to improve operations and ensure that the solutions work as expected in production networks environments.
• Recommended approaches and the potential for open source projects to contribute to the next generation of intelligent networking tools.

Beth Cohen  can we use drivers for AI side bars using following :

1. Network Planning and Design: Leveraging generative AI for precise placement of small cells, MIMO antennas, beamforming, and efficient backhaul connections to optimize network performance.
2. Predictive Maintenance: Utilizing AI to forecast equipment failures, enhancing maintenance efficiency and network reliability.
3. Automated Closed Loop: Employing AI models trained on operational data to ensure network assurance through automated processes.
4. Network AIOps: Implementing AIOps methodologies to automate and streamline network operations, improving overall efficiency.
5. Self-Organizing Networks (SON): Harnessing AI-based algorithms for autonomous optimization and management of network resources in Self-Organizing Networks.
6. Technical Assistant/Customer Service: Real-time guidance from AI-based tech assistants, enhancing customer service and field operations efficiency.
7. Code Generation for Network Protocols: Enabling co-pilot functionality for generating software implementations of network protocols specifications, facilitating protocol development and deployment.
8. Capacity Forecasting for Access, Edge, and Core Networks: Utilizing AI for accurate load prediction on each RAN site to optimize network capacity and avoid unnecessary upgrades.
9. Traffic Management: Dynamic rerouting of traffic based on AI analysis to efficiently utilize network resources and improve user experience.
10. Shared AI and 5G Infrastructure: Leveraging unused 5G RAN infrastructure resources for training and inference, enhancing AI capabilities and network efficiency.

2.1 Autonomous Network Beth Cohen 

2.2 LLM & GenAI Sandeep Panesar 

2.3 LFN & Previous AI WP Beth Cohen 

2.4 LF Data (Thoth)- Sandeep Panesar Beth Cohen 

3  Challenge Statement and Motivation Beth Cohen 

1 page

Focus on Telco pain points only, 

3.1 traditional CSP pain points - Beth Cohen 

• Operational Efficiency –   reduce costs and errors, potentially increase margins
• Network Automation - reduce expenditure

3.2 Emerging CSP Opportunities Sandeep Panesar 

• Converged Infrastructure for the era of AI
• Converged Service for the era of AI
• Business Innovation - New revenue streams (data monetization?)

4.1 network AI explorations so far

1.5 page

• l  Radio Access Network ChangJin Wang 
  • The intelligent evolution of wireless access networks is currently in a phase of rapid evolution and continuous innovation. In June 2022, 3GPP announced the freezing of R17 and described the process diagram of an intelligent RAN in TR37.817, including data collection, model training, model inference, and execution modules, which together form the infrastructure of an intelligent RAN. This promotes the rapid implementation and deployment of 5G RAN intelligence and provides support for intelligent scenarios such as energy saving, load balancing, and mobility optimization.
  • AI and Machine Learning Drive 5G RAN Intelligence

Artificial intelligence and machine learning technologies are playing an increasingly important role in 5G RAN intelligence. The application of these technologies enables the network to learn autonomously, self-optimize, and self-repair, thereby improving network stability, reliability, and performance. For example, by using machine learning algorithms to predict and schedule network traffic, more efficient resource allocation and load balancing can be achieved. By leveraging AI technologies for automatic network fault detection and repair, operation and maintenance costs can be greatly reduced while improving user experience. The intelligence of 5G wireless access networks also provides broad space for various vertical industry applications. For instance, in intelligent manufacturing, 5G can enable real-time communication and data transmission between devices, improving production efficiency and product quality. In smart cities, 5G can provide high-definition video surveillance, intelligent transportation management, and other services to enhance urban governance. Additionally, 5G has played a significant role in remote healthcare, online education, and other fields.

• • Challenges Facing 5G RAN Intelligence Industrialization

However, despite the remarkable progress made in the 5G wireless access network intelligence industry, some challenges and issues remain to be addressed. For example, network security and data privacy protection are pressing issues that require effective measures to be implemented. The energy consumption issue of 5G networks also needs attention, necessitating technological innovations and energy-saving measures to reduce energy consumption. In the future, continuous efforts should be made in technological innovation, market application, and other aspects to promote the sustainable and healthy development of the 5G wireless access network intelligence industry.

• l  Core Network Hui Deng ChangJin Wang 
  • Mobile core network is kind of central  or brain of mobile communication, it is the only and largest part which experienced the transformation from legacy propritary hardware into telco cloud native, almost 100% of today mobile core network has been deployed based on telco cloud architecture by NFV technologies. It mostly consisted of packet core network such as 5GC and UPF which is responsible for packet forwarding, IMS which help operators multimedia communications such as voice, message and video communication of telecomunicaiton operators, and the management of core network including telco cloud infrastructure and 5G network functionalities.  F.or those 3 parts of AI evolvement,

  • Network Intelligence Enables Experience Monetization and Differentiated Operations

    For a long period of time, operators have strived to realize traffic monetization on MBB(Mobile Broadband) networks. However, there are three technical gaps: not assessable user experience, no dynamic optimization, and no-closed-loop operations. To bridge these gaps,  there is strong requirement for Intelligent Personalized Experience solution, aiming to help operators add experience privileges to service packages and better monetize differentiated experiences. In the industry, the user plane on the mobile core network usually processes and forwards one service flow using one vCPU. As heavy-traffic services increase, such as 2K or 4K HD video and live streaming, microbursts and elephant flows frequently occur. It is, therefore, more likely that a vCPU will become overloaded, causing packet loss. To address this issue, Intelligent AI supported 5G core network can deliver ubiquitous 10 Gbps superior experiences.

  • Service Intelligence Expands the Profitability of Calling Services

    In 2023, New Calling was put into commercial use based on 3GPP specification, it could enhanced intelligence and data channel (3GPP specification)-based interaction capabilities; it is taking user to a multi-modal communication era and helping operators reconstruct their service layout. In addition, 3GPP architecture allow users to control digital avatars through voice during calls, delivering a more personalized calling experience. An enterprise can also customize their own avatar as an enterprise ambassador to promote their branding.

  • O&M Intelligence Achieves High Network Stability and Efficiency

    Empowered by the multi-modal large model, the Digital Assistant & Digital Expert (DAE) based AI technology could reduces O&M workload and improves O&M efficiency. It reshapes cloud-based O&M from "experts+tools" to intelligence-centric "DAE+manual assistance". With DAE, 80% of telecommunication operators trouble tickets can be automatically processed, which is much more efficient than 100% manual processing as it used to be. DAE also enables intent-driven O&M, avoiding manual decision-making. Before, it usually took over five years to cultivate experts in a single domain, however, the multi-modal large model is now able to be trained and updated within just weeks.

• l  Bearer & DC Network - Cisco?
• l  Cross domain network AI platform Lingli Deng //moved to 4.3.1
• l  Telecom Cloud Hui Deng 

4.2 challenges towards fully autonomy Lingli Deng Andrei Agapi 

0.5 page

• High quality structured data Communication networks are different from general human-computer interaction in that a large number of interactions between systems use structured data. However, due to the complexity of network systems and differences in vendor implementation, the degree of standardization of these structured data is very low, causing them to be divided into "information islands" that cannot be uniformly "interpreted". The data in each "island" There is a lack of "standard bridge" to establish correlation between them, and it cannot be used as effective input to incubate modern data-driven AI/ML applications.
• AI trustworthiness Different from traditional AI application scenarios, in order to meet carrier-level reliability requirements, communication network operation management needs to be strict, precise, and prudent. Although operators have introduced various automation methods to the greatest extent in the process of building autonomous networks, legal person organizations composed of human units are still responsible for ensuring the quality of communication services. In other words, AI is still assisting people and cannot replace people. Not only because the loss caused by the AI algorithm itself cannot be defined as the responsible party (the developer or the user of the algorithm), but also because the deep learning model based on the AI/ML algorithm itself is based on the mathematical characteristics of statistics, resulting in uncertainty in its behavior, its credibility is difficult to guarantee.
• Non-economic margin cost According to the analysis and research of the previous white paper [the reference to the previous white paper], there are a large number of potential AI technology application scenarios in communication networks, but to independently build a customized data-driven AI/ML model for each specific scenario, would be uneconomical and hence unsustainable for both the R&D stage and the operation stage. How to build an effective business model between basic computing power providers, general AI/ML capability providers and algorithm application consumers in specific communication network scenarios will be the prerequisite for its effective application in the field of autonomous networks.
• Contextual data sets 
• l  TBA

4.3 Network LLM: the game changer?

Jason Hunt (at least on how foundation models can be applied to network data) Andrei Agapi 

1 page

4.3.1       issues solved/mitigated Lingli Deng Andrei Agapi 

• The advent of transformer models and attention mechanisms [][], and the sudden popularity of ChatGPT [], LLMs, transfer learning and foundation models in the NLP domain have all sparked vivid discussions and efforts to apply generative models in many other domains [].

  Interestingly, all of: word embeddings [], sequence models such as LSTMs [] and GRUs [], attention mechanisms [], transformer models [] and pretrained LLMs [][] have long been around before the launch of the ChatGPT tool in late 2022. Pretrained transformers like BERT[] in particular (especially transformer-encoder models) were very popular and widely used in NLP for tasks like sentiment analysis, text classification [], extractive question answering [] etc, long before ChatGPT made chatbots and decoder-based generative models go viral.

• That said, there has clearly been a spectacular explosion of academic research, commercial activity and ecosystems that have emerged since ChatGPT came out, in the area of both open [][][] and closed source [][][] LLM foundation models, related software, services and training datasets.

  Beyond typical chatbot-style applications, LLMs have been extended to generate code [][], solve Math problems (stated either formally or informally) [], pass science exams [][], or act as incipient "AGI"-style agents for different tasks, including advising on investment strategies, or setting up a small business [][]. Recent advancements to the basic LLM text generation model include instruction finetuning [], retrieval augmented generation using external vector stores [][], using external tools such as web search [], external knowledge databases or other APIs for grounding models [], code interpreters, calculators and formal reasoning tools [][]. Beyond LLMs and NLP, transformers have also been used to handle non-textual data, such as images [], sound [] and arbitrary sequence data [].

• A natural question arises on how the the power of LLMs can be harnessed for problems and applications related to Intelligent Networking, network automation and for operating and optimizing telecommunication networks in general, at any level of the network stack.

  Datasets encountered in telco-related applications have a few particularities. For one, data one might encounter ranges from fully structured (e.g. code, scripts, configuration, or time series KPIs), to semi-structured (syslogs, design templates etc), to unstructured data (design documents and specifications, Wikis, Github issues, emails, chatbot conversations).

• Another issue is domain adaptation. Language encountered in telco datasets can be very domain specific (including CLI commands and CLI output, formatted text, network slang and abbreviations, syslogs, RFC language, network device specifications etc). Off-the-shelf performance of LLM models strongly depends on whether those LLMs have actually seen that particular type of data during training (this is true for both generative LLMs and embedding models). There exist several approaches to achieve domain adaptation and downstream task adaptation of LLM models. In general these either rely on 1) In-context-learning, prompting and retrieval augmentation techniques; 2) Finetuning the models; or 3) Hybrid approaches. For finetuning LLMs, unlike for regular neural network models, several specialized techniques exist in the general area of PEFT (Parameter Efficient Fine Tuning), allowing one to only finetune a very small percentage of the many billions of parameters of a typical LLM. In general, the best techniques to achive domain adaptation for an LLM will heavily depend on: 1) the kind of data we have and how much domain data we have available, 2) the downstream task, and 3) the foundation model we start from. In addition to general domain adaptation, many telcos will have the issue of multilingual datasets, where a mix of languages (typically English + something else) will exist in the data (syslogs, wikis, tickets, chat conversations etc). While many options exist for both generative LLMs [] and text embedding models [], not many foundation models have seen enough non-English data in training, thus options in foundation model choice are definitely restricted for operators working on non-English data.

• In conclusion, while foundation models and transfer learning have been shown to work very well on general human language when pretraining is done on large corpuses of human text (such as Wikipedia, or the Pile[]), it remains an open question to be answered whether domain adaptation and downstream task adaptation work equally well on the kinds of domain-specific, semi-structured, mixed modality datasets we can find in telco networks. To enable this, telcos should very likely focus on standardization and data governance efforts, such as standardized and unified data collection policies and high quality structured data, as discussed earlier in this whitepaper.

• Deploying large models such as LLMs in production, especially at scale, also raises several other issues in terms of: 1) Performance, scalability and cost of inference, especially when using large context windows (most transformers scale poorly with context size); 2) Deployment of models in the cloud, on premise, multi-cloud, or hybrid; 3) Issues pertaining to privacy and security of the data for each particular application; 4) Issues common to many other ML/AI applications, such as ML-Ops, continuous model validation and continuous re-training.

• high quality structured data Large language models can be used to understand large amounts of unstructured operation and maintenance data (for example, system logs, operation and maintenance work orders, operation guides, company documents, etc., which are traditionally used in human-computer interaction or human-to-human collaboration scenarios), from which effective knowledge is extracted to provide guidance for further automatic/intelligent operation and maintenance, thereby effectively expanding the scope of the application of autonomous mechanism.
• AI trustworthiness moved to 4.3.2
• non-economic margin cost Although equipment manufacturers can provide many domain AI solutions for professional networks/single-point equipment, these solutions are limited in "field of view" and cannot solve problems that require a "global view" such as end-to-end service quality assurance and rapid response to faults. . Operators can aggregate management and maintenance data in various network domains by building a unified data sharing platform, and based on this, further provide a unified computing resource pool, basic AI algorithms and inference platform (i.e. cross-domain AI platform) for various scenario-specific AI for end-to-end scenarios and intra-domain scenarios. Applied reasoning platform.
• TBA

4.3.2       issues remained/added/aggravated Lingli Deng 

• AI trustworthiness
• non-economic margin cost Compared with traditional data-driven dedicated AI/ML models in specific scenarios, the R&D and operation of large language models has higher requirements for pre-training data scale, training computing power cluster scale, fine-tuning engineering and other resource requirements, energy consumption, management and maintenance, etc. It is bound to become a playground for the few. How to open up the R&D, application, operation and maintenance chain of large models for the communications network industry so that it can serve the general public may become a stepping stone for operators to realize high-end autonomous networks.
• TBA

5  How could Open Source Help? Ranny Haiby 

2-3 pages

When considering the role of open source software in addressing the challenges of Network AI it is important to understand the current landscape of projects and initiatives and how they came into existence. Several such initiatives have already laid down the ground work for building Network AI solutions, or are actively working on creating them. Building on these foundations, it is possible to envision what role open source software will play in unleashing the power of AI for the future generations of networks. Some of the required technologies required of Network AI are unique to the Networking industry, and will have to be addressed by the existing OSS projects on the landscape, or by creation of additional ones. Some of the other pieces of technology are more generic, and will come from the broad landscape of AI OSS. Here is a rough outline of the different layers of Networking AI and the source of the required technology:

5.1 Related Open Source Landscape

1-2 pages

• Network communities

Open Source Software (OSS) communities have been successfully building projects that provide the building blocks of networks for over decade now. OSS projects provide the underlying technology for all layers of the network, including the data/forwarding plane, control plane, management and orchestration. A vibrant ecosystem of contributing companies exists around these projects, consisting of organizations that realized the value in the principles OSS for networking:

• • Shared effort to develop the foundation layers of the technology, freeing up more resources to develop the value-add layers
  • A neutral platform for innovation where individuals and organizations can exchange ideas and come up with best of breed technology solutions
  • An opportunity to demonstrate thought leadership and domain expertise 
  • A collaboration space where producers and consumers of technology can freely interact and create business opportunities

Many of the networking OSS projects are hosted by the Linux Foundation. Most of them are part of LF Networking, LF Connectivity and LF Broadband.

• AI communities

The same principles are now being applied to the shared development of Network AI technologies, where the open source community fosters innovation and stimulates business growth. AI innovation has been strongly propelled by OSS projects that were initiated following the same principles mentioned above. It is hard to imagine doing any modern AI development without heavily relying on OSS. OSS AI and ML projects range from anything between the framework for developing , Libraries and programming tools. Data scientists who develop domain specific models do not have to start from scratch, instead they can leverage OSS projects to jump start their work and focus on creation of innovation. It would be almost impossible to mention all the relevant OSS AI projects here as there are already so many of them, and the list keeps growing quickly. The Linux Foundation AI&Data provides a useful dynamic landscape here.

Successful development of AI models for use in Networking relies on the availability of data that could be shared under a common license. The Linux Foundation created the Community Data License Agreement (CDLA) for this purpose. Using this license, end users can share data and make it available for researches, who in turn develop the necessary models and applications that benefit the end users.

When it comes to the popular subject of LLMs, there is a lot of debating going on currently about what an "Open AI model" really means. While it is out of the scope of this paper to try and settle any of those debates, it is obvious that there is a clear need to create the definition of "open LLM". The sooner such definition is created and blessed by the industry, the faster innovation can happen.

• Integration blueprints

In recent years The Linux Foundation Networking launched a set of Open Source Networking "Super Blueprints" that outline architectures for common networking use cases and are built using OSS technology. Those blueprints consists of collections of OSS projects and commercial products, integrated together by the open source community and documented for free use by any interested party. Several of these blueprints started incorporating AI technologies and we expect that trend to accelerate, as existing blueprints inspire additional AI-driven solutions. One promising area is intent-based network automation. There is current work on blueprints that use NLP and LLM to translate user intent, expressed in natural spoken language to network requirements, and generate full network configuration based on those requirements. Such approach can significantly simplify existing network operation processes, and enable new services that can be directly consumed by the end users in an automated manner (known as Network-as-a-Service - NaaS)

• Verification programs

Experience with OSS in other domains shows that whenever there is an OSS technology that powers commercial products or offerings, there is a need to validate the products to make sure they are properly using the OSS technology and are ready to serve the end users in a predictable manner. Such validation/verification programs have existed for a while as part of OSS ecosystems. They are often created and maintained by the same OSS community that develops the OSS projects themselves. The Cloud Native Computing Foundation (CNCF) has a successful "Certified Kubernetes" program that helps vendors and end users ensure that Kubernetes distributions provide all the necessary APIs and functionality. A similar approach should be applied to any OSS Networking AI projects. End users should have a certain level of confidence, knowing that the OSS based AI Networking solution they use will behave as expected. 

5.2 Common Vision: intelligence plane for XG Ranny Haiby Muddasar Ahmed 

1 page

In the dynamic realm of communication technologies, the fusion of artificial intelligence (AI) with networks promises to redefine connectivity, ushering in an era of unprecedented intelligence, efficiency, and adaptability. As we embark on the journey towards 6G and embrace the vision outlined by the International Telecommunication Union (ITU) for IMT-2030, it becomes clear that AI will play a pivotal role in reshaping network operations.

At the heart of this transformation lies the concept of the intelligence plane, where AI-driven systems leverage natural intent interaction to seamlessly bridge the gap between users and networks. Cognitive smart orchestration ensures optimal resource allocation and dynamic adaptation to evolving demands, thereby enhancing network performance and user experiences. Real-time meta network verification, powered by AI algorithms, ensures continuous monitoring and validation of network behavior, preempting potential issues and maintaining operational integrity. Additionally, built-in knowledge open-loops enable networks to autonomously learn and adapt, fostering resilience and responsiveness.

Moreover, the concept of ubiquitous intelligence envisaged by IMT-2030 underscores the pervasive presence of AI across the communication system. From autonomous network management to context-aware devices, AI will imbue every aspect of network operations with intelligence, efficiency, and adaptability. AI-enabled air interfaces, coupled with distributed computing, will pave the way for end-to-end AI applicability, fostering convergence between communication and computing domains.

Drawing from the IMT-2030 vision, the integration of AI into mobile networks unlocks transformative capabilities. Advanced human-machine interfaces, such as extended reality (XR) displays and haptic sensors, offer immersive experiences, blurring the boundaries between physical and virtual realms. Intelligent machines, equipped with machine perception and autonomous capabilities, facilitate seamless interactions and drive digital economic growth and societal changes.

Furthermore, the Linux Foundation's role in joining the AI revolution underscores the importance of open-source collaboration in advancing network capabilities. By harnessing open-source technologies, network operators can leverage the collective expertise of the community to accelerate innovation and adoption of AI-driven solutions. This collaborative approach not only democratizes access to advanced AI capabilities but also fosters interoperability and scalability across diverse network environments.

In conclusion, the future of networks in the era of 6G and beyond hinges on the transformative power of AI, fueled by open-source collaboration. By embracing AI-driven intelligence, networks can enhance situational awareness, performance, and capacity management, while enabling quick reactions to undesired states. As we navigate this AI-powered future, the convergence of technological innovation and open collaboration holds the key to unlocking boundless opportunities for progress and prosperity in the telecommunications landscape.

6  Call for Action Beth Cohen 

0.25 page

The future of OSS Networking AI is in the hands of the individuals and organizations who are already contributing to projects and initiatives, and those who will join them. If you are involved in building and operating networks, developing network technology or consuming network services, you should probably consider getting involved. Engaging with OSS communities is a way to shape the future of Networking. Your contribution could be small or large, and does not necessarily involve writing code. Some of the ways Ito contribute include:

• Attending Project meetings
• Attending Developer events
• Joining approved Projects
• Proposing a Project
• Writing documentation
• Contributing use cases
• Analyzing requirements
• Defining tests / processes
• Reviewing and submitting code patches
• Building upstream relationships
• Contributing upstream code
• Start or join a User Group
• Hosting and staffing a community lab

• Answering questions
• Giving a talk / training
• Creating a demo
• Evangelizing the  projects

Ways to get involved in OSS Networking AI:

• Collaborate on Network Super Blueprints or develop new ones:

https://wiki.lfnetworking.org/x/ArAZB

• Join LFN AI mailing list:

https://lists.lfnetworking.org/g/lfn-ai-taskforce