Workshops Program

Next-generation wireless systems, such as 6G and beyond, are expected to tightly embed artificial intelligence (AI) into their design, creating a new breed of AI-native wireless systems. Remarkably, despite significant academic, industrial, and standardization efforts dedicated to AI-native wireless systems in the past few years, the very definition of such systems remains ambiguous. Moreover, to date, most existing efforts in this space remain as incremental extensions to existing "AI for wireless" paradigms, i.e., using existing tools like autoencoders, diffusion models, or even large-language models to replicate known wireless functions. However, these tools have several limitations including the black-box nature of the AI models, their curve-fitting nature, which can limit their ability to reason and adapt, their reliance on large amounts of training data, their energy inefficiency, and their poor generalizability to new, unseen and out-of-domain/out-of-distribution data points. To overcome these limitations, in this talk, we present a forward-looking vision that addresses these shortcomings by introducing a novel framework for building AI-native wireless networks grounded in the rigorous tools of causal reasoning. The proposed framework will help embed generalizability, explainability, and reasoning into tomorrow's AI-native wireless networks. We start by explaining some of the key concepts of causal reasoning-based AI-native wireless networks, while providing simplistic examples. Then, we illustrate such a design by drawing a connection with the emerging idea of semantic communications. We show how, by embracing causal reasoning with semantic communication, we can usher in a new era of knowledge-driven, reasoning wireless networks that are more sustainable and resilient than today's data-driven, knowledge-agnostic networks. We particularly present our recent key results in this area, with foundations in AI, theory of mind, and game theory, that showcase how the proposed causal reasoning approach for semantic communications can reduce the volume of data circulating in a network while improving reliability, two critical requirements for emerging wireless services, such as connected intelligence and digital twins. We conclude with a discussion on future opportunities in this exciting area.

NA

The dawn of 6G presents an opportunity to redefine our understanding of "Cell-Free" networks. The current focus on merely altering the physical layer barely scratches the surface of its true potential. This talk will introduce our vision of Cell-Free architecture for 6G. We envision a Cell-Free system with a richer protocol stack distribution, going beyond current mainstream perspectives. It's not just an evolution; it's a transformative leap that showcases the depth of what Cell-Free can potentially offer in 6G.

But with bold ideas come significant challenges. As we explore this broader vision of Cell-Free, we encounter complexities of scalability, stringent coordinations demands, and the architecture dynamism, all of which call for innovative solutions. This is where the role of AI can be crucial. It's more than just an enhancement; AI will be an enabler, ensuring our networks are equipped to address these challenge

Title: IoT and Artificial Intelligence Experimental Testbeds: Filling the Gap

Abstract: Internet of Things (IoT) is a widely discussed topic, as it encompasses embedded devices, smart objects, and smart users that are interconnected through a communication infrastructure. It is gaining significant attention and popularity. This leads to the emergence of several paradigms, including smart cities, smart homes, smart transportation systems, Industrial applications, smart grids, and smart logistics, to name just a few. Data processing through IoT smart devices is more significant compared to information processing capacity. Data becomes humongous, even coming from a single source. Besides, unstructured data from production reports or external sources must also be integrated to analyze and optimize the production process. Therefore, when data emanates from all heterogeneous sources, its magnitude makes it harder to process up-to a needed scale. IoT testbeds are sophisticated platforms that enable extensive experimentation, debugging, and deployment of IoT applications on a large scale. Testbeds offer researchers and developers a platform to test and improve new technologies, making it easier to adopt and commercialize them. This tutorial begins with an introduction to IoT, followed by a comprehensive examination of the latest applications of IoT. Next, we will present a comprehensive overview of the requirements and features that IoT testbeds offer. We will explore the different experimental facilities currently available to the research community. In this discussion, we will explore the various capabilities and services offered by each testbed, with a particular emphasis on its AI-related features.

KEYNOTE TITLE: IRS/RIS EMPOWERED WIRELESS NETWORKS: RECENT ADVANCES AND FUTURE TRENDS?

ABC

KEYNOTE TITLE: INTELLIGENT SURFACES FOR WIRELESS COMMUNICATIONS: LIVING AT THE INTERFACE OF ELECTROMAGNETIC AND COMMUNICATION THEORIES

A panel on NTN IoT and Use Cases for 6G and Beyond