Roberto Morabito

Sasu Tarkoma, Roberto Morabito, Jaakko Sauvola

The evolution towards 6G architecture promises a transformative shift in communication networks, with artificial intelligence (AI) playing a pivotal role. This paper delves deep into the seamless integration of Large Language Models (LLMs) and Generalized Pretrained Transformers (GPT) within 6G systems. Their ability to grasp intent, strategize, and execute intricate commands will be pivotal in redefining network functionalities and interactions. Central to this is the AI Interconnect framework, intricately woven to facilitate AI-centric operations within the network. Building on the continuously evolving current state-of-the-art, we present a new architectural perspective for the upcoming generation of mobile networks. Here, LLMs and GPTs will collaboratively take center stage alongside traditional pre-generative AI and machine learning (ML) algorithms. This union promises a novel confluence of the old and new, melding tried-and-tested methods with transformative AI technologies. Along with providing a conceptual overview of this evolution, we delve into the nuances of practical applications arising from such an integration. Through this paper, we envisage a symbiotic integration where AI becomes the cornerstone of the next-generation communication paradigm, offering insights into the structural and functional facets of an AI-native 6G network.

Alaa Saleh, Praveen Kumar Donta, Roberto Morabito et al.

Human biological systems sustain life through extraordinary resilience, continually detecting damage, orchestrating targeted responses, and restoring function through self-healing. Inspired by these capabilities, this paper introduces ReCiSt, a bio-inspired agentic self-healing framework designed to achieve resilience in Distributed Computing Continuum Systems (DCCS). Modern DCCS integrate heterogeneous computing resources, ranging from resource-constrained IoT devices to high-performance cloud infrastructures, and their inherent complexity, mobility, and dynamic operating conditions expose them to frequent faults that disrupt service continuity. These challenges underscore the need for scalable, adaptive, and self-regulated resilience strategies. ReCiSt reconstructs the biological phases of Hemostasis, Inflammation, Proliferation, and Remodeling into the computational layers Containment, Diagnosis, Meta-Cognitive, and Knowledge for DCCS. These four layers perform autonomous fault isolation, causal diagnosis, adaptive recovery, and long-term knowledge consolidation through Language Model (LM)-powered agents. These agents interpret heterogeneous logs, infer root causes, refine reasoning pathways, and reconfigure resources with minimal human intervention. The proposed ReCiSt framework is evaluated on public fault datasets using multiple LMs, and no baseline comparison is included due to the scarcity of similar approaches. Nevertheless, our results, evaluated under different LMs, confirm ReCiSt's self-healing capabilities within tens of seconds with minimum of 10% of agent CPU usage. Our results also demonstrated depth of analysis to over come uncertainties and amount of micro-agents invoked to achieve resilience.

Roberto Morabito, Mallik Tatipamula, Sasu Tarkoma et al.

The network edge's role in Artificial Intelligence (AI) inference processing is rapidly expanding, driven by a plethora of applications seeking computational advantages. These applications strive for data-driven efficiency, leveraging robust AI capabilities and prioritizing real-time responsiveness. However, as demand grows, so does system complexity. The proliferation of AI inference accelerators showcases innovation but also underscores challenges, particularly the varied software and hardware configurations of these devices. This diversity, while advantageous for certain tasks, introduces hurdles in device integration and coordination. In this paper, our objectives are three-fold. Firstly, we outline the requirements and components of a framework that accommodates hardware diversity. Next, we assess the impact of device heterogeneity on AI inference performance, identifying strategies to optimize outcomes without compromising service quality. Lastly, we shed light on the prevailing challenges and opportunities in this domain, offering insights for both the research community and industry stakeholders.

Su Wang, Roberto Morabito, Seyyedali Hosseinalipour et al.

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, and (ii) there may be significant overlaps in devices' local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization methodology aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy while minimizing data processing and D2D communication resource consumption subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using these results, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and D2D data offloading to maximize FedL accuracy. Through evaluation on popular datasets and real-world network measurements from our edge testbed, we find that our methodology outperforms popular device sampling methodologies from literature in terms of ML model performance, data processing overhead, and energy consumption.

Lauri Lovén, Abhishek Kumar, Alexander Engelhardt et al.

Large language models (LLMs) can serve as the semantic-matching engine of a content-based publish/subscribe broker for agentic AI across the edge-cloud computing continuum, bridging the vocabulary and modality gaps that defeat keyword and embedding filters. Framed as offline multi-label retrieval over three public datasets spanning social-media, legal, and smart-home sensor domains (six LLMs, seven baselines), our central contribution is a two-crossover cost-accuracy characterisation: an analytical context-window crossover below which a CoverAndMerge compression pipeline reduces LLM invocations, and an empirical discrimination-capacity crossover above which matching accuracy collapses independently of context budget, by a model-dependent factor of parameter count and training generation. Two findings carry practical weight: above the discrimination crossover, compression cannot recover accuracy and only frontier-scale models clear large subscription sets; and there backend choice dominates configuration choice, so model selection, not pipeline tuning, is the primary operator lever. We accompany this with three composable algorithms and a per-cluster Quality-of-Experience framework for autonomic LLM-tier selection.

Roberto Morabito, Bivek Pandey, Paulius Daubaris et al.

Digital Twins (DTs) are set to become a key enabling technology in future wireless networks, with their use in network management increasing significantly. We developed a DT framework that leverages the heterogeneity of network access technologies as a resource for enhanced network performance and management, enabling smart data handling in the physical network. Tested in a Campus Area Network environment, our framework integrates diverse data sources to provide real-time, holistic insights into network performance and environmental sensing. We also envision that traditional analytics will evolve to rely on emerging AI models, such as Generative AI (GenAI), while leveraging current analytics capabilities. This capacity can simplify analytics processes through advanced ML models, enabling descriptive, diagnostic, predictive, and prescriptive analytics in a unified fashion. Finally, we present specific research opportunities concerning interoperability aspects and envision aligning advancements in DT technology with evolved AI integration.

Adarsh Prasad Behera, Paulius Daubaris, Iñaki Bravo et al.

On-device inference holds great potential for increased energy efficiency, responsiveness, and privacy in edge ML systems. However, due to less capable ML models that can be embedded in resource-limited devices, use cases are limited to simple inference tasks such as visual keyword spotting, gesture recognition, and predictive analytics. In this context, the Hierarchical Inference (HI) system has emerged as a promising solution that augments the capabilities of the local ML by offloading selected samples to an edge server or cloud for remote ML inference. Existing works demonstrate through simulation that HI improves accuracy. However, they do not account for the latency and energy consumption on the device, nor do they consider three key heterogeneous dimensions that characterize ML systems: hardware, network connectivity, and models. In contrast, this paper systematically compares the performance of HI with on-device inference based on measurements of accuracy, latency, and energy for running embedded ML models on five devices with different capabilities and three image classification datasets. For a given accuracy requirement, the HI systems we designed achieved up to 73% lower latency and up to 77% lower device energy consumption than an on-device inference system. The key to building an efficient HI system is the availability of small-size, reasonably accurate on-device models whose outputs can be effectively differentiated for samples that require remote inference. Despite the performance gains, HI requires on-device inference for all samples, which adds a fixed overhead to its latency and energy consumption. Therefore, we design a hybrid system, Early Exit with HI (EE-HI), and demonstrate that compared to HI, EE-HI reduces the latency by up to 59.7% and lowers the device's energy consumption by up to 60.4%.

Adnan Shahid, Adrian Kliks, Ahmed Al-Tahmeesschi et al.

This white paper discusses the role of large-scale AI in the telecommunications industry, with a specific focus on the potential of generative AI to revolutionize network functions and user experiences, especially in the context of 6G systems. It highlights the development and deployment of Large Telecom Models (LTMs), which are tailored AI models designed to address the complex challenges faced by modern telecom networks. The paper covers a wide range of topics, from the architecture and deployment strategies of LTMs to their applications in network management, resource allocation, and optimization. It also explores the regulatory, ethical, and standardization considerations for LTMs, offering insights into their future integration into telecom infrastructure. The goal is to provide a comprehensive roadmap for the adoption of LTMs to enhance scalability, performance, and user-centric innovation in telecom networks.

Alaa Saleh, Praveen Kumar Donta, Roberto Morabito et al.

This article introduces Follow-Me AI, a concept designed to enhance user interactions with smart environments, optimize energy use, and provide better control over data captured by these environments. Through AI agents that accompany users, Follow-Me AI negotiates data management based on user consent, aligns environmental controls as well as user communication and computes resources available in the environment with user preferences, and predicts user behavior to proactively adjust the smart environment. The manuscript illustrates this concept with a detailed example of Follow-Me AI in a smart campus setting, detailing the interactions with the building's management system for optimal comfort and efficiency. Finally, this article looks into the challenges and opportunities related to Follow-Me AI.

SiYoung Jang, Roberto Morabito

The widespread adoption of Language Models (LMs) across industries is driving interest in deploying these services across the computing continuum, from the cloud to the network edge. This shift aims to reduce costs, lower latency, and improve reliability and privacy. Small Language Models (SLMs), enabled by advances in model compression, are central to this shift, offering a path to on-device inference on resource-constrained edge platforms. This work examines the interplay between edge and cloud deployments, starting from detailed benchmarking of SLM capabilities on single edge devices, and extending to distributed edge clusters. We identify scenarios where edge inference offers comparable performance with lower costs, and others where cloud fallback becomes essential due to limits in scalability or model capacity. Rather than proposing a one-size-fits-all solution, we present platform-level comparisons and design insights for building efficient, adaptive LM inference systems across heterogeneous environments.

Guanghan Wu, Sasu Tarkoma, Roberto Morabito

The evolving requirements of Internet of Things (IoT) applications are driving an increasing shift toward bringing intelligence to the edge, enabling real-time insights and decision-making within resource-constrained environments. Tiny Machine Learning (TinyML) has emerged as a key enabler of this evolution, facilitating the deployment of ML models on devices such as microcontrollers and embedded systems. However, the complexity of managing the TinyML lifecycle, including stages such as data processing, model optimization and conversion, and device deployment, presents significant challenges and often requires substantial human intervention. Motivated by these challenges, we began exploring whether Large Language Models (LLMs) could help automate and streamline the TinyML lifecycle. We developed a framework that leverages the natural language processing (NLP) and code generation capabilities of LLMs to reduce development time and lower the barriers to entry for TinyML deployment. Through a case study involving a computer vision classification model, we demonstrate the framework's ability to automate key stages of the TinyML lifecycle. Our findings suggest that LLM-powered automation holds potential for improving the lifecycle development process and adapting to diverse requirements. However, while this approach shows promise, there remain obstacles and limitations, particularly in achieving fully automated solutions. This paper sheds light on both the challenges and opportunities of integrating LLMs into TinyML workflows, providing insights into the path forward for efficient, AI-assisted embedded system development.

Roberto Morabito, SiYoung Jang

The rapid adoption of generative AI (GenAI), particularly Large Language Models (LLMs), has exposed critical limitations of cloud-centric deployments, including latency, cost, and privacy concerns. Meanwhile, Small Language Models (SLMs) are emerging as viable alternatives for resource-constrained edge environments, though they often lack the capabilities of their larger counterparts. This article explores the potential of collaborative inference systems that leverage both edge and cloud resources to address these challenges. By presenting distinct cooperation strategies alongside practical design principles and experimental insights, we offer actionable guidance for deploying GenAI across the computing continuum.

Maximilian Abstreiter, Sasu Tarkoma, Roberto Morabito

The rapid rise of Language Models (LMs) has expanded the capabilities of natural language processing, powering applications from text generation to complex decision-making. While state-of-the-art LMs often boast hundreds of billions of parameters and are primarily deployed in data centers, recent trends show a growing focus on compact models-typically under 10 billion parameters-enabled by techniques such as quantization and other model compression techniques. This shift paves the way for LMs on edge devices, offering potential benefits such as enhanced privacy, reduced latency, and improved data sovereignty. However, the inherent complexity of even these smaller models, combined with the limited computing resources of edge hardware, raises critical questions about the practical trade-offs in executing LM inference outside the cloud. To address these challenges, we present a comprehensive evaluation of generative LM inference on representative CPU-based and GPU-accelerated edge devices. Our study measures key performance indicators-including memory usage, inference speed, and energy consumption-across various device configurations. Additionally, we examine throughput-energy trade-offs, cost considerations, and usability, alongside an assessment of qualitative model performance. While quantization helps mitigate memory overhead, it does not fully eliminate resource bottlenecks, especially for larger models. Our findings quantify the memory and energy constraints that must be considered for practical real-world deployments, offering concrete insights into the trade-offs between model size, inference performance, and efficiency. The exploration of LMs at the edge is still in its early stages. We hope this study provides a foundation for future research, guiding the refinement of models, the enhancement of inference efficiency, and the advancement of edge-centric AI systems.

Alaa Saleh, Roberto Morabito, Sasu Tarkoma et al.

As 6G networks evolve into increasingly AI-driven, user-centric ecosystems, traditional reactive handover mechanisms demonstrate limitations, especially in mobile edge computing and autonomous agent-based service scenarios. This manuscript introduces WAAN, a cross-layer framework that enables intent-aware and proactive handovers by embedding lightweight TinyML agents as autonomous, negotiation-capable entities across heterogeneous edge nodes that contribute to intent propagation and network adaptation. To ensure continuity across mobility-induced disruptions, WAAN incorporates semi-stable rendezvous points that serve as coordination anchors for context transfer and state preservation. The framework's operational capabilities are demonstrated through a multimodal environmental control case study, highlighting its effectiveness in maintaining user experience under mobility. Finally, the article discusses key challenges and future opportunities associated with the deployment and evolution of WAAN.

Hung T. Nguyen, Roberto Morabito, Kwang Taik Kim et al.

Edge computing has revolutionized the world of mobile and wireless networks world thanks to its flexible, secure, and performing characteristics. Lately, we have witnessed the increasing use of it to make more performing the deployment of machine learning (ML) techniques such as federated learning (FL). FL was debuted to improve communication efficiency compared to conventional distributed machine learning (ML). The original FL assumes a central aggregation server to aggregate locally optimized parameters and might bring reliability and latency issues. In this paper, we conduct an in-depth study of strategies to replace this central server by a flying master that is dynamically selected based on the current participants and/or available resources at every FL round of optimization. Specifically, we compare different metrics to select this flying master and assess consensus algorithms to perform the selection. Our results demonstrate a significant reduction of runtime using our flying master FL framework compared to the original FL from measurements results conducted in our EdgeAI testbed and over real 5G networks using an operational edge testbed.

Su Wang, Mengyuan Lee, Seyyedali Hosseinalipour et al.

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, while (ii) there may be significant overlaps in devices' local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using this result, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and resulting offloading that maximizes FedL accuracy. Through evaluation on real-world datasets and network measurements from our IoT testbed, we find that our methodology while sampling less than 5% of all devices outperforms conventional FedL substantially both in terms of trained model accuracy and required resource utilization.