This session starts with an imaginary (?) scenario where generative AI has taken over the networks. Use cases are generated by fine-tuned models, architectures and code, including test code generated by co-pilots and ChatGPT(-likes); digital twins including scenarios and objects are generated by AI-enabled gaming engines; and MLOps pipelines are triggered autonomously. Experts will present their views on the topic, including important challenges such as data collection and modeling in networks, AI-empowered features in 6G and the current state of MLOps. Discussions will focus on questions such as whether networks ready are for this, and identify some of the major areas which need further study including current standards and gaps.
Questions to answer:
1. Given the recent advances in AI/ML, are networks ready for taking advantage of these?
2. How to bring the best advances to networks so that operators and end-users can benefit? what are the integration strategies?
1. More study is needed in the areas of ........
2. Practical challenges in ......
3. Standards gaps in this area are ....
Speaker abstractUnleash the potential of ML and AI in 6G: Promising methods, applications, and research directions; This talk will explore the potential of machine learning (ML) and artificial intelligence (AI) in the context of 6G radio access networks (RAN). We will discuss promising applications along with corresponding ML and AI methods that can help unlock the full potential of ML and AI in future 6G networks. We will highlight research directions in this exciting area that we are pursuing as part of the 6G Research and Innovation Cluster (6G-RIC). In particular, we will discuss how ML and AI can be used to improve the performance of 6G networks in terms of efficiency, reliability, and security. The talk will also cover some of the key challenges related to the use of ML and AI in 6G networks.Slawomir StanczakProfessorFraunhofer Heinrich Hertz Institute (HHI)
Speaker abstractOpen source opportunities in AI-enabled 6G; While 6G technology is still in its early stages of development, it is clear that open source will continue to play a critical role in its evolution. Open source frameworks and tools have already developed very rapidly in the field of AI and proven to be instrumental in the development of AI-enhanced 5G networks, and it is likely that this trend will continue with 6G. By leveraging open source, developers and organizations can collaborate more effectively, accelerate innovation, and ultimately create more robust and reliable networks. Additionally, open source provides greater transparency and flexibility, enabling organizations to tailor their solutions to their specific needs. As the race to develop 6G heats up, open source will undoubtedly remain a key driver of progress and innovation in this space.Wei MengDirector of standard and open source planningZTE
Speaker abstractThe Role of MLOps in Enabling Successful ML Deployments in the Telecommunications Industry; The telecommunications industry is increasingly relying on machine learning (ML) to improve network performance, reduce costs, and enhance customer experience. However, deploying ML models in this industry comes with its own set of challenges, including the need for low-latency processing, distributed architectures, and the lack of standardization. MLOps is an emerging discipline that applies DevOps practices to ML development and deployment, aiming to address these challenges and enable successful ML deployments. We will discuss the motivations for MLOps, its benefits, and its role in enabling successful ML deployments in the telecommunications industry. We will explore the challenges of ML deployments in telecommunications, the evolution of ML workflows in this industry from the ML Function Orchestrator (MLFO), which was proposed by the ITU focus group on “Machine Learning for Future Networks including 5G (FG-ML5G)”, to pipelining and MLOps, and the benefits of MLOps in terms of improved model quality, faster time-to-market, and increased collaboration between teams. We will also discuss the specific challenges that 5G and 6G networks present for ML deployments in telecommunications, and how MLOps can help overcome them. Finally we will be highlighting the importance of MLOps in enabling successful ML deployments in the telecommunications industry, and the need for telecommunications companies to adopt MLOps practices to stay competitive in the ever-evolving telecommunications landscape.Salih ErgutChief Data Science, R&D, and Strategy OfficerOREDATA
In recent years, we have witnessed several advances in network architectures. Networks are becoming more open and decentralized. On one hand, data is increasingly generated, augmented and processed at the edge of the network. On the other hand, cloud based, scalable compute can enable AI/ML models to infer at high speeds. Sandbox environments would allow for training machine learning models on edge data while keeping it private and secure while controllers and digital twins are spun up on demand. Many algorithms and AI capabilities are now offered as cloud-based services that can be integrated on demand.
Juxtaposing AI-enabled feature advances in future networks such as beam selection, semantic awareness, RIS-enabled radio optimizations, with the architecture evolution, the standards organizations face the difficult task of choosing the right network architectures, placement of functionalities and level of exposure of data while providing application development capabilities to boost opportunities for innovation.
These considerations present new challenges for future networks such as 6G. To benefit from open networks, edge data, and on-demand models, standards organizations need to carefully consider what are the collaborative working models, especially for proof of concepts which can answer some of these questions. Discussions in this session would focus on some of the important considerations in AI-native design and network architectures which can host those designs.
Questions to answer:
1. what are the evolutions in network architectures needed for enabling the integration of AI/ML?
2. What are the techniques to enable generation, collection, utilization of data in networks beyond 5G?
1. Collaborative analysis is needed in the areas of ........with ......
2. PoC studies can mitigate risks in ....
3. The relevant open source studies are .....
4. use cases such as ... has data from ... part of the network, which may be extracted via ... interface, models may be trained in ... and deployed in ...
Speaker abstractHow to build a Network Digital Twin? Simulation vs Emulation vs ML - Network Digital Twins (NDT) are a key technology for future telecommunication networks. As exemplifying features, NDT are expected to estimate future traffic load and automatically optimize the network to use minimal resources while fulfilling stringent SLAs. NDT should also be able to predict failures before happening, and take the appropriate actions. Overall, NDT offers unprecedented performance with ultra-efficient use of the hardware resources, resulting in very low CAPEX and OPEX. However, there is a certain lack of specificity in the literature that describes how to build it. In this talk we will describe how we can build a NDT and we will compare different technologies: simulation, emulation and ML techniques as well as the advantages and disadvantages that they provide.Albert CabellosAssociate ProfessorUniversitat Politècnica de Catalunya
Speaker abstractTowards "Native Network Intelligence" (5G-A baseline and 6G perspective); The application of AI/ML methods is considered for all parts of the 3GPP 5G-Advanced network (air interface, disaggregated RAN, system architecture, security and management) across a wide range of use cases. For 6G it is anticipated that AI/ML will be an integral part of the system, raising new requirements towards AI/ML-driven network function design and distributed implementation ("native AI"), Management of AI/ML and AI/ML governance ("Trustworthy AI"). The talk presents an outline of the evolution from 5G-A to 6G focusing on the RAN and its management, arguing that the 6G network can become "natively intelligent" by increasing the level of cognition of network functions through AI/ML.Henning SanneckChief Architect AI/MLNokia Standards
Speaker abstractEfficient and Trustworthy AI With Applications to 5G Networks; In this talk, I will discuss efficient and trustworthy AI techniques for optimizing the performance of 5G wireless networks. In particular, to address the efficiency and robustness of the solution generated from data driven models, I will discuss recent progress on the convergence analysis of adaptive gradient algorithms, distributed optimization algorithms, and robust adversarial training of neural networks. Application results of these techniques to optimize real-world 5G networks will be reported.Tom LuoProfessorThe Chinese University of Hong Kong, Shenzhen
Speaker abstractDeep Learning in the Physical Layer: Transforming 5G and 6G Performance with Data and AI; Wireless systems, access schemes, and band allocations are becoming increasingly complex and heterogeneous. Deep learning-based solutions within the physical layer allow for a rapid and accurate state of the art approach for both spectrum sensing as well as channel access and physical layer design. In this talk, we’ll provide an overview of the background and enablers for this trend of ML in the physical layer of communications and highlight how we are building software solutions based on these approaches and techniques at DeepSig. OpenRAN based wireless solutions are also becoming increasingly prevalent within 5G RAN deployments and provide a key opportunity for rapid deployment of these technologies and to improve their performance in real-world deployments. We’ll highlight how we are using both DL driven PHY techniques and ORAN software to bring these capabilities into ORAN deployments in order to improve energy efficiency and to improve spectral efficiency, capacity, and user experience. Finally, we’ll discuss where we believe these technologies are leading future RAN technologies and standards, and provide highlights from our own trials and partnerships, highlighting the maturity of the technology and its likelihood for adoption in next generation wireless deployments.Tim O’SheaCo-Founder & CTODeepSig Inc
The 6G vision is shaping up with emerging use cases from metaverse and other user-centric apps. Standards Developing Organizations are marching on with requirements and feature analysis for future networks. This spurs the need for network controllers and agents to manage bandwidth and latency demands from billions of devices. Network platforms spanning software, hardware, cloud and edge with/without GPUs, TPUs, FPGAs and sensors are emerging. Network operators now face a dilemma: what are the right platforms for 6G? Where does enterprise networks come into the picture? Do the major value-adding algorithms, including the network service optimization, become AI-ed and autonomous and move to cloud? The key questions to ponder over are: What platform architectures do we need for this network evolution? What role does open source play in these architectures?
Questions to answer:
1. what are the platform architectures needed for this?
2. whats the role of open source in such architectures?
Speaker abstractFrom Multimodal Sensing to Digital Twin-Assisted Communications; Large-scale MIMO is a key enabler for 5G, 6G, and beyond. Scaling up MIMO systems, however, is subject to critical challenges, such as the large channel acquisition/beam training overhead and the sensitivity to blockages. These challenges make it difficult for MIMO systems to support applications with high mobility and strict reliability/latency constraints. In this talk, I will first motivate the use of multi-modal sensing data to address some of these challenges. Then, I will present a vision where multi-modal sensing, real-time ray-tracing, and machine learning can be integrated to construct real-time digital twins of the communication environments and comprehensively assist all the layers of the communication systems. I will discuss some of the open questions to realize this vision, present a research platform for investigating the digital twin problems, and highlight some initial results.Ahmed AlkhateebAssistant ProfessorArizona State University
Speaker abstractArchitectures and Platforms Enabling AI/ML in 6G; This presentation explores necessary technology transformations to address challenges and opportunities in operators’ evolving roles and value propositions. It focuses on the critical role of AI/ML in shaping the future of telecommunications, specifically within the context of 6G. While marching ahead with the ongoing 5G-Adv evolution towards a potential 6G revolution, the vision and courage to embrace advanced architectures and platforms enabled by the deep ICDT convergence has become increasingly essential. The journey from today’s limited AI/ML capabilities as an over the top add-on to the network to the next generation network with native AI embedded into its very fabric in 2030 is exciting yet challenging. We’ll examine the significant progress in HW/SW decoupling, network data collection, edge computing, decentralized processing, and distributed AI in the past 10 years. Then, growing recognition of O-RAN in creating an open, flexible, efficient, and scalable network infrastructure that seamlessly integrates AI/ML will be highlighted. Various practical and valuable use cases utilizing such architecture will be shared to illustrate how O-RAN is considered as a potential starting point of 6G network. Finally, the emergence of communication-computation integrated networks (CCIN), a paradigm shift that empowers the radio access networks with distributed native computing capabilities, will be elaborated. CCIN native computing platform, together with O-RAN native AI architecture, will further enhancing the adaptability, efficiency, and security of 6G system to drive a more connected, intelligent, and sustainable digital future globally.Chih-Lin IChina Mobile Chief Scientist, Wireless TechnologiesChina Mobile Research Institute
Speaker abstractForging the Networks of Tomorrow: Leveraging Deep Learning for Mobile Network planning and operation - The networking community is actively engaged in the search for the key technologies that will drive the success of 6G networks. In this exciting landscape, Deep Learning can be a game-changer in propelling such a revolution, especially for processing the vast amounts of data collected in networks, uncovering intricate patterns in that data, and making complex decisions in real time. This talk will present some opportunities and ongoing efforts to apply Deep Learning for planning and operation in mobile networks. We will introduce some use cases where we leverage Deep Learning for achieving unprecedented levels of connectivity and user experience. Next, we will discuss some key open challenges to achieve mature Deep Learning solutions for networks, with a particular focus on energy efficiency. Finally, we will outline some future research directions that may help materialize Deep Learning-based solutions for the Mobile Networks of Tomorrow.José Suárez-VarelaResearcherTelefonica Research
Speaker abstractWhy Telco Must Stop Solving Problems to Survive; Whether from external demands (more users and non-human behaviours) or internal pressures (new technologies, network failure, buisness priorities), telecommunication networks are the most complex that they have ever been. In response to this, network operators have been embracing the role of ML in network operation and management to address copmlexity with automation. Unfortunately, depsite the significant advantages that they bring, human-centric design (and subsequent maintenence) of these ML apporaches applied per usecase per context across the network simply does not scale. In this talk i will discuss a new mindset to the design of software for networks that addresses the scalabliity of software design and maintence for ML-powered systems and beyond. I will also discuss how this work aligns with the efforts of the ITU-T Focus Group on Autonomous Networks, and give a breif overview of connected activies.Paul HarveyLecturerUniversity of Glasgow
AI/ML is becoming more and more pervasive in 5G and future networks but the realization options for AI/ML integrated use cases appear fragmented from a method, platforms and design view point. This is partly due to heterogeneity in use case requirements, datasets and models. In addition, we face challenges in bringing innovations from lab to networks due to different perspectives between innovators vs. incumbents, open vs. black boxes, models vs. people (developers, service providers, maintainers, operational engineers), architecture evolution vs. leap-frogging across generations, and cloud vs. edge. This brings to focus the unenviable position of Standards Developing Organizations such as ITU which is required to come up with roadmaps, work items for integration of AI/ML in future networks including 6G.
Questions to answer:
1. what should ITU and other SDOs do?
2. Are there pre-standard, activities to look at?
3. Is the Challenge activities serving the audiences well? what can be done better there?
1. In collaboration with other SDOs, ITU should study ....
2. Challenge should focus on ...type of use cases, expand to ....type of data and models.
3. Papers and code should be published ....more/less... frequently
Speaker abstractThe future of standards for AI/ML in future networks including 6G - Derived from the analysis and experience in ITU FG ML5G and ITU FG AN, this talk focuses on the importance and key learnings from such initiatives, especially to the MENA region. Various use cases, relevant to the AI/ML in 5G are explained and their directions to future networks are indicated. In addition to use case analysis, the importance of practical studies, especially in the form of Proof of concept studies, and our learning from the ITU FG AN Build-a-thon are given. As next steps, based on the conclusions of the ITU CxO meeting in Dubai, Dec 2022, a call for collaboration for analysing data, use cases, and the importance of ML Sandboxes as the meeting-points for collaboration are highlighted. We conclude with our wishlist, where a collaborative study of use cases for AI/ML in future networks including 6G, with the help of regional, interoperable ML Sandboxes such as MENA Sandbox which can help create trusted, validated models is called for.Fathi AbdeldayemLead Technology Standardization and Evolving Technologiesdu UAE
Speaker abstractAI/ML integration in Wi-Fi: an overview; The unlicensed spectrum is (and will continue to be) an essential part of wireless communications, as it provides ease of access and great flexibility to fulfill a plethora of use cases. Wi-Fi is one of the flagship technologies operating in the unlicensed bands and it is now facing fundamental changes toward unprecedented requirements such as ultra-high reliability. This talk will provide an overview of the current evolution path of Wi-Fi towards a deeper integration of AI/ML functionalities, highlighting activities discussing unified architecture, common standardized interfaces and strategical functionalities.Francesc WilhelmiPostdoctoral Researcher, Mobile Networks DepartmentNokia Bell Labs
Speaker abstractDriving Telco Innovation & Business Growth with Cloud-Native & AI; Telcos are undergoing a full transformation concerning their business model & services portfolio. Cloud is perceived as the primary catalyst for digital transformation, enabling new technologies such as Open Ran & 5G and unlocking new monetization opportunities. The role of AI and the contribution towards the Cloud Native platform are embraced by most of the Telcos. Join us for an exciting workshop exploring the technological evolution of cloud computing in the Telco domain and how it converges with other technologies to accelerate the telcos' digital transformation. We will examine the role of Artificial Intelligence in the Telco Cloud, the potential use cases, and the intersection with concepts such as Edge Computing & Cloud-Native.Karim RabieICT Chief ArchitectRed Hat
- James AgajoAssociate Professor and Head of Department of Computer EngineeringFederal University of Technology Minna
- Fidelis Ikechukwu OnahSenior MemberIEEE