Michele Polese

Maxime Elkael, Michele Polese, Yunseong Lee et al.

Ultra Reliable and Low Latency Communications (URLLC) was one of the main motivations behind 5G, with 3GPP advertising 1-10 ms latency targets for applications such as industrial automation, Vehicle-To-Everything (V2X), tactical edge networking, and unmanned-system control. Years on, real 5G Time Division Duplexing (TDD) networks still show median Uplink (UL) round-trip times in the 50-70 ms range, largely because of the Scheduling Request (SR) procedure that a User Equipment (UE) must complete before transmitting UL data. Existing remedies, primarily Configured Grant (CG) scheduling, only eliminate this overhead for strictly periodic traffic and require cross-layer synchronization, which has limited their adoption. We propose AUGUSTE (Anticipatory Uplink Grants for URLLC via Self-Adapting Temporal Estimation), a learning-based Medium Access Control (MAC) scheduling framework that embeds online Machine Learning (ML) models in the UL scheduler to predict packet arrivals and proactively allocate resources before an SR is issued. An adaptive state machine alternates between a learning phase that collects unbiased arrival statistics and a confident phase that exploits the learned predictions to schedule only when traffic is expected. We evaluate AUGUSTE on a real 5G testbed running OpenAirInterface across three URLLC traffic patterns (request-response, ML edge inference, and periodic autonomous reporting), and show that it operates at the best achievable point on the latency-overhead trade-off: it matches always-on scheduling's median Round Trip Time (RTT) (around 10 ms, halving the 20 ms SR-based baseline) at roughly one-tenth its resource cost (7-10 percent overhead).

Leonardo Bonati, Michele Polese, Salvatore D'Oro et al.

Open Radio Access Network (RAN) architectures will enable interoperability, openness and programmable data-driven control in next generation cellular networks. However, developing and testing efficient solutions that generalize across heterogeneous cellular deployments and scales, and that optimize network performance in such diverse environments is a complex task that is still largely unexplored. In this paper we present OpenRAN Gym, a unified, open, and O-RAN-compliant experimental toolbox for data collection, design, prototyping and testing of end-to-end data-driven control solutions for next generation Open RAN systems. OpenRAN Gym extends and combines into a unique solution several software frameworks for data collection of RAN statistics and RAN control, and a lightweight O-RAN near-real-time RAN Intelligent Controller (RIC) tailored to run on experimental wireless platforms. We first provide an overview of the various architectural components of OpenRAN Gym and describe how it is used to collect data and design, train and test artificial intelligence and machine learning O-RAN-compliant applications (xApps) at scale. We then describe in detail how to test the developed xApps on softwarized RANs and provide an example of two xApps developed with OpenRAN Gym that are used to control a network with 7 base stations and 42 users deployed on the Colosseum testbed. Finally, we show how solutions developed with OpenRAN Gym on Colosseum can be exported to real-world, heterogeneous wireless platforms, such as the Arena testbed and the POWDER and COSMOS platforms of the PAWR program. OpenRAN Gym and its software components are open-source and publicly-available to the research community. By guiding the readers through running experiments with OpenRAN Gym, we aim at providing a key reference for researchers and practitioners working on experimental Open RAN systems.

Davide Villa, Mauro Belgiovine, Nicholas Hedberg et al.

The transition of cellular networks to (i) software-based systems on commodity hardware and (ii) platforms for services beyond connectivity introduces critical system-level challenges. As sensing emerges as a key feature toward 6G standardization, supporting Integrated Sensing and Communication (ISAC) with limited bandwidth and piggybacking on communication signals, while maintaining high reliability and performance, remains a fundamental challenge. In this paper, we provide two key contributions. First, we present a programmable, open-source framework for processing PHY/MAC signals through real-time, GPU-accelerated Artificial Intelligence (AI) applications on the edge Radio Access Network (RAN) infrastructure. Building on the Open RAN dApp architecture, the framework interfaces with a GPU-accelerated gNB based on NVIDIA Aerial Testbed (ATB), feeding PHY/MAC data to custom AI logic with a framework overhead of 150 us, multiple inference engines, and support for several AI backends. We evaluate the framework on multiple GPU platforms with and without hardware-level GPU isolation. Second, we demonstrate the framework capabilities through cuSense, an indoor localization dApp that consumes uplink DMRS channel estimates, removes static multipath components, and runs a neural network to infer the position of a moving person. Evaluated on a 3GPP-compliant 5G NR deployment, cuSense achieves a mean localization error of 77 cm, with 75% of predictions falling within 1 meter, without dedicated sensing hardware or modifications to the RAN stack or signals. The framework is released as open source, providing a reference design for future AI-native RANs and ISAC applications.

Andrea Lacava, Michele Polese, Rajarajan Sivaraj et al.

5G and beyond mobile networks will support heterogeneous use cases at an unprecedented scale, thus demanding automated control and optimization of network functionalities customized to the needs of individual users. Such fine-grained control of the Radio Access Network (RAN) is not possible with the current cellular architecture. To fill this gap, the Open RAN paradigm and its specification introduce an open architecture with abstractions that enable closed-loop control and provide data-driven, and intelligent optimization of the RAN at the user level. This is obtained through custom RAN control applications (i.e., xApps) deployed on near-real-time RAN Intelligent Controller (near-RT RIC) at the edge of the network. Despite these premises, as of today the research community lacks a sandbox to build data-driven xApps, and create large-scale datasets for effective AI training. In this paper, we address this by introducing ns-O-RAN, a software framework that integrates a real-world, production-grade near-RT RIC with a 3GPP-based simulated environment on ns-3, enabling the development of xApps and automated large-scale data collection and testing of Deep Reinforcement Learning-driven control policies for the optimization at the user-level. In addition, we propose the first user-specific O-RAN Traffic Steering (TS) intelligent handover framework. It uses Random Ensemble Mixture, combined with a state-of-the-art Convolutional Neural Network architecture, to optimally assign a serving base station to each user in the network. Our TS xApp, trained with more than 40 million data points collected by ns-O-RAN, runs on the near-RT RIC and controls its base stations. We evaluate the performance on a large-scale deployment, showing that the xApp-based handover improves throughput and spectral efficiency by an average of 50% over traditional handover heuristics, with less mobility overhead.

Leonardo Bonati, Michele Polese, Salvatore D'Oro et al.

Softwarization, programmable network control and the use of all-encompassing controllers acting at different timescales are heralded as the key drivers for the evolution to next-generation cellular networks. These technologies have fostered newly designed intelligent data-driven solutions for managing large sets of diverse cellular functionalities, basically impossible to implement in traditionally closed cellular architectures. Despite the evident interest of industry on Artificial Intelligence (AI) and Machine Learning (ML) solutions for closed-loop control of the Radio Access Network (RAN), and several research works in the field, their design is far from mainstream, and it is still a sophisticated and often overlooked operation. In this paper, we discuss how to design AI/ML solutions for the intelligent closed-loop control of the Open RAN, providing guidelines and insights based on exemplary solutions with high-performance record. We then show how to embed these solutions into xApps instantiated on the O-RAN near-real-time RAN Intelligent Controller (RIC) through OpenRAN Gym, the first publicly available toolbox for data-driven O-RAN experimentation at scale. We showcase a use case of an xApp developed with OpenRAN Gym and tested on a cellular network with 7 base stations and 42 users deployed on the Colosseum wireless network emulator. Our demonstration shows the high degree of flexibility of the OpenRAN Gym-based xApp development environment, which is independent of deployment scenarios and traffic demand.

Reshma Prasad, Maxime Elkael, Gabriele Gemmi et al.

As networks evolve towards 6G, Mobile Network Operators (MNOs) must accommodate diverse requirements and at the same time manage rising energy consumption. Integrated Access and Backhaul (IAB) networks facilitate dense cellular deployments with reduced infrastructure complexity. However, the multi-hop wireless backhauling in IAB networks necessitates proper routing and resource allocation decisions to meet the performance requirements. At the same time, cell densification makes energy optimization crucial. This paper addresses the joint optimization of routing and resource allocation in IAB networks through two distinct objectives: energy minimization and throughput maximization. We develop a novel capacity model that links power levels to achievable data rates. We propose two practical large-scale approaches to solve the optimization problems and leverage the closed-loop control framework introduced by the Open Radio Access Network (O-RAN) architecture to integrate the solutions. The approaches are evaluated on diverse scenarios built upon open data of two months of traffic collected by network operators in the city of Milan, Italy. Results show that the proposed approaches effectively reduces number of activated nodes to save energy and achieves approximately 100 Mbps of minimum data rate per User Equipment (UE) during peak hours of the day using spectrum within the Frequency Range (FR) 3, or upper midband. The results validate the practical applicability of our framework for next-generation IAB network deployment and optimization.

Michele Polese, Rajeev Gangula, Tommaso Melodia

The convergence of communication, sensing, and Artificial Intelligence (AI) in the Radio Access Network (RAN) offers compelling economic advantages through shared spectrum and infrastructure. How can inference and sensing be integrated in the RAN infrastructure at a system level? Current abstractions in O-RAN and 3GPP lack the interfaces and capabilities to support (i) a dynamic life cycle for inference and Integrated Sensing and Communication (ISAC) algorithms, whose requirements and sensing targets may change over time and across sites; (ii) pipelines for AI-driven ISAC, which need complex data flows, training, and testing; (iii) dynamic device and stack configuration to balance trade-offs between connectivity, sensing, and inference services. This paper analyzes the role of a programmable, software-driven, open RAN in enabling the intelligent edge for 5G and 6G systems. We identify real-time user-plane data exposure, open interfaces for plug-and-play inference and ISAC models, closed-loop control, and AI pipelines as elements that evolutions of the O-RAN architecture can uniquely provide. Specifically, we describe how dApps - a real-time, user-plane extension of O-RAN - and a hierarchy of controllers enable real-time AI inference and ISAC. Experimental results on an open-source RAN testbed demonstrate the value of exposing I/Q samples and real-time RAN telemetry to dApps for sensing applications.

Neagin Neasamoni Santhi, Davide Villa, Michele Polese et al.

Artificial Intelligence (AI) has become a powerful tool for model-free Radio Access Network (RAN) signal processing and optimization. However, designing a single model that generalizes across all radio environments is challenging. Specialized AI models outperform conventional algorithms only under specific conditions, while their higher compute and energy cost makes unconditional execution impractical at the base station. This creates a need for real-time expert switching: dynamically activating the most appropriate AI or conventional expert based on current network conditions. To address this, we propose ARCHES (Adaptive Real-time CUDA Hot-swapping of Experts in the RAN Stack), a framework hosting multiple AI-based and conventional signal processing experts within a GPU-accelerated PHY pipeline, dynamically selecting the most appropriate expert at slot-boundary granularity without dropping or corrupting in-flight data. ARCHES includes a lightweight CUDA switch kernel for zero-gap output selection, a dApp-based control plane that collects cross-layer telemetry and drives the switching policy, and a reusable process for policy design based on controlled perturbation, monotonicity filtering, and hierarchical clustering. We validate ARCHES on UL channel estimation, switching between an AI-based and a Minimum Mean Square Error (MMSE) estimator under changing propagation and interference conditions. Implemented on the X5G platform with NVIDIA Aerial and OpenAirInterface (OAI), ARCHES achieves median UL PHY throughput gains of 5.32% and 7.23% under good and poor conditions, with a control-loop latency of ~140 us and sub-microsecond decision inference. Under good conditions, defaulting to MMSE saves 15.8 W of GPU power (9.6%) and 17 percentage points of GPU utilization versus unconditional AI execution, validating the performance-per-watt tradeoff that motivates adaptive expert selection.

Gabriele Gemmi, Michele Polese, Tommaso Melodia

The large-scale deployment of 5G networks has not delivered the expected return on investment for mobile network operators, raising concerns about the economic viability of future 6G rollouts. At the same time, surging demand for Artificial Intelligence (AI) inference and training workloads is straining global compute capacity. AI-RAN architectures, in which Radio Access Network (RAN) platforms accelerated on Graphics Processing Unit (GPU) share idle capacity with AI workloads during off-peak periods, offer a potential path to improved capital efficiency. However, the economic case for such systems remains unsubstantiated. In this paper, we present a techno-economic analysis of AI-RAN deployments by combining publicly available benchmarks of 5G Layer-1 processing on heterogeneous platforms -- from x86 servers with accelerators for channel coding to modern GPUs -- with realistic traffic models and AI service demand profiles for Large Language Model (LLM) inference. We construct a joint cost and revenue model that quantifies the surplus compute capacity available in GPU-based RAN deployments and evaluates the returns from leasing it to AI tenants. Our results show that, across a range of scenarios encompassing token depreciation, varying demand dynamics, and diverse GPU serving densities, the additional capital and operational expenditures of GPU-heavy deployments are offset by AI-on-RAN revenue, yielding a return on investment of up to 8x. These findings strengthen the long-term economic case for accelerator-based RAN architectures and future 6G deployments.

Tamerlan Aghayev, Maxime Elkael, Michele Polese et al.

Cellular research and development (R&D) is throttled by six structural processes that each consume months of manual engineering work per iteration: (i) synthesizing new features from standards or research papers into production code; (ii) conformance and interoperability testing; (iii) hardening against field anomalies and diverse deployment environments; (iv) data-driven optimization of network functionalities; (v) discovering and prototyping novel waveforms, functionalities, and capabilities for future standards; and (vi) securing the stack against vulnerabilities. Although Large Language Models (LLMs) have compressed comparable R&D work in general software engineering from days to minutes, their known pitfalls worsen on Radio Access Network (RAN) use cases: they hallucinate Application Programming Interfaces (APIs) and mis-read specifications, which kills interoperability of RAN components at the first mistake, and they heavily rely on simulations for designing algorithms, which is notorious for breaking when transferred to real hardware. To address these challenges, we present GENESIS, an agentic Artificial Intelligence (AI) framework that converts intents (e.g., a specification clause, a telemetry anomaly, or a research hypothesis) into solutions validated with over-the-air experiments, fed back into a persistent knowledge base. GENESIS is built on three composable primitives (agents, skills, hooks) and a knowledge layer (SYNAPSE) that doubles as the source of ground truth and the recipient of every artifact the framework produces, making capabilities compound across runs.

Maria Tsampazi, Paolo Testolina, Michele Polese et al.

Frequency Range 3 (FR3), encompassing frequencies between 7.125 and 24.25 GHz, is an emerging frequency band for 6th generation (6G) applications. The upper mid-band, as it is frequently referred to, represents the sweet spot between coverage and capacity, providing better range than mmWaves and higher bandwidth than the sub-6 GHz band. Despite these advantages, the spectrum is already occupied by incumbent systems such as satellites (e.g., Starlink), and sharing it with terrestrial cellular applications results in spectrum conflicts, only exacerbating the existing spectrum scarcity. This article investigates the impact of two state-of-the-art methods, namely Quality of Service (QoS)-Aware Power Control and Interference Nulling, as well as their joint application, on interference mitigation toward non-terrestrial links while maintaining acceptable QoS on terrestrial networks. Our simulation results demonstrate the advantages and disadvantages of each method, pinpointing how interference nulling can maintain high average performance and how power control is more appropriate for risk-averse scenarios to enhance fairness in terrestrial QoS. Finally, we showcase how the two can complement each other to enhance fairness in terrestrial QoS and increase the Next Generation Node Base (gNB)'s energy efficiency, while suppressing interference toward incumbents.

MohammadErfan Jabbari, Abhishek Duttagupta, Claudio Fiandrino et al.

Deep reinforcement learning (DRL) promises adaptive control for future mobile networks but conventional agents remain reactive: they act on past and current measurements and cannot leverage short-term forecasts of exogenous KPIs such as bandwidth. Augmenting agents with predictions can overcome this temporal myopia, yet uptake in networking is scarce because forecast-aware agents act as closed-boxes; operators cannot tell whether predictions guide decisions or justify the added complexity. We propose SIA, the first interpreter that exposes in real time how forecast-augmented DRL agents operate. SIA fuses Symbolic AI abstractions with per-KPI Knowledge Graphs to produce explanations, and includes a new Influence Score metric. SIA achieves sub-millisecond speed, over 200x faster than existing XAI methods. We evaluate SIA on three diverse networking use cases, uncovering hidden issues, including temporal misalignment in forecast integration and reward-design biases that trigger counter-productive policies. These insights enable targeted fixes: a redesigned agent achieves a 9% higher average bitrate in video streaming, and SIA's online Action-Refinement module improves RAN-slicing reward by 25% without retraining. By making anticipatory DRL transparent and tunable, SIA lowers the barrier to proactive control in next-generation mobile networks.

Noemi Giustini, Andrea Lacava, Leonardo Bonati et al.

5G networks provide low-latency, high throughput, and massive connectivity, yet the control plane remains exposed to several security threats. Among the most common and impactful threats are Denial-of-Service (DoS) attacks, with Radio Resource Control (RRC) signaling storms being particularly effective and difficult to mitigate. In this attack, a malicious User Equipment (UE) aims to exhaust Next Generation Node Base (gNB) resources, preventing legitimate UEs from establishing a connection. Existing defenses are typically limited to detection, only evaluated through numerical simulations, and cannot discern between high-load network conditions and attacks. Most of them also assume static setups and do not take mobility into account. In this paper, we first evaluate the feasibility of the signaling storm attack by using the OpenAirInterface(OAI) 5G protocol stack. Then, we propose StormShield, a signaling storm attack detection and mitigation technique implemented as an xApp on an O-RAN Near-Real-Time (near-RT) RAN Intelligent Controller (RIC). It fingerprints and blocks Malicious UEs (MUEs) before gNB resources are exhausted. We prototyped our solution on an Over-The-Air (OTA) testbed with OAI, NVIDIA Aerial, and two different gNB setups. The first one leverages an USRP X410 Software-defined Radio (SDR) with 8.1 functional split; the second a commercial Foxconn Radio Unit (RU) with 7.2 functional split. Our experimental evaluation demonstrates that StormShield effectively prevents gNB resource exhaustion, identifying and blocking MUEs with an average detection accuracy of 97.6% within 106.5 ms from the beginning of the attack.

Ravis Shirkhani, Reshma Prasad, Leonardo Bonati et al.

Software-driven O-RAN architectures enable rapid innovation through frequent, independent updates to virtualized components. However, attributing performance variations to specific software changes is challenging due to the stochastic nature of wireless systems, where channel conditions, interference, and hardware variability confound analysis. Traditional threshold-based monitoring and manual troubleshooting do not scale with modern software evolution. This paper presents RANalyzer, an automated test analysis framework that quantifies the performance impact of software updates beyond what can be explained by wireless channel conditions. RANalyzer combines LLM-assisted semantic extraction with residuals analysis. The first categorizes code changes by affected protocol layers and functional components, while the second provides insights on the effect of load, channel, or code changes on the test performance. We contribute an extensive dataset collected over more than two years of continuous over-the-air testing on an experimental O-RAN testbed, comprising over 8,600 automated tests across 69 releases of the OAI stack. By modeling expected performance and interpreting deviations as software-induced effects, we identify degraded instances attributable to code changes and correlate them with specific change categories. The framework can be integrated into CI/CD/CT pipelines for automated, continuous evaluation of software updates at scale.

Leonardo Bonati, Salvatore D'Oro, Michele Polese et al.

Next Generation (NextG) cellular networks will be natively cloud-based and built upon programmable, virtualized, and disaggregated architectures. The separation of control functions from the hardware fabric and the introduction of standardized control interfaces will enable the definition of custom closed-control loops, which will ultimately enable embedded intelligence and real-time analytics, thus effectively realizing the vision of autonomous and self-optimizing networks. This article explores the disaggregated network architecture proposed by the O-RAN Alliance as a key enabler of NextG networks. Within this architectural context, we discuss the potential, the challenges, and the limitations of data-driven optimization approaches to network control over different timescales. We also present the first large-scale integration of O-RAN-compliant software components with an open-source full-stack softwarized cellular network. Experiments conducted on Colosseum, the world's largest wireless network emulator, demonstrate closed-loop integration of real-time analytics and control through deep reinforcement learning agents. We also show the feasibility of Radio Access Network (RAN) control through xApps running on the near real-time RAN Intelligent Controller, to optimize the scheduling policies of co-existing network slices, leveraging the O-RAN open interfaces to collect data at the edge of the network.

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.

Reshma Prasad, Michele Polese, Tommaso Melodia

Radio Access Network (RAN) configuration has traditionally required significant manual effort due to indirect causal dependencies between observable Key Performance Indicators (KPIs), and context-dependent characteristics, where the optimal configurations vary with network conditions. Although recent data-driven approaches improve parameter tuning, they remain limited in distinguishing causal direction from statistical correlation and in generalizing across diverse operating contexts. To address these challenges, we propose BLINC (Bayesian Large Language Model (LLM)-Driven Intelligent Network Configuration), an LLM-assisted Bayesian Network framework that integrates telecommunications domain knowledge into causal structure learning. Trained and validated on a private 5G deployment, our method achieves throughput improvement of 63.5% with 19.7% reduction on block error rate over data-only baselines through joint optimization of power control and link adaptation parameters. The framework provides interpretable causal structure, while also quantifying prediction uncertainty. We also demonstrate the ability of the Bayesian Network framework to adapt to different deployment scenarios and propose an incremental Conditional Probability Distribution (CPD) update mechanism with learning rate for continuous model adaptation as network conditions evolve.

Eduardo Baena, Paolo Testolina, Michele Polese et al.

Satellite networks are rapidly evolving, yet most \glspl{ntn} remain isolated from terrestrial orchestration frameworks. Their control architectures are typically monolithic and static, limiting their adaptability to dynamic traffic, topology changes, and mission requirements. These constraints lead to inefficient spectrum use and underutilized network capacity. Although \gls{ai} promises automation, its deployment in orbit is limited by computing, energy, and connectivity limitations. This paper introduces Space-O-RAN, a distributed control architecture that extends Open RAN principles into satellite constellations through hierarchical, closed-loop control. Lightweight \glspl{dapp} operate onboard satellites, enabling real-time functions like scheduling and beam steering without relying on persistent ground access. Cluster-level coordination is managed via \glspl{spaceric}, which leverage low-latency \glspl{isl} for autonomous decisions in orbit. Strategic tasks, including AI training and policy updates, are transferred to terrestrial platforms \glspl{smo} using digital twins and feeder links. A key enabler is the dynamic mapping of the O-RAN interfaces to satellite links, supporting adaptive signaling under varying conditions. Simulations using the Starlink topology validate the latency bounds that inform this architectural split, demonstrating both feasibility and scalability for autonomous satellite RAN operations.

Joshua Groen, Simone Di Valerio, Imtiaz Karim et al.

5G and beyond cellular systems embrace the disaggregation of Radio Access Network (RAN) components, exemplified by the evolution of the fronthaul (FH) connection between cellular baseband and radio unit equipment. Crucially, synchronization over the FH is pivotal for reliable 5G services. In recent years, there has been a push to move these links to an Ethernet-based packet network topology, leveraging existing standards and ongoing research for Time-Sensitive Networking (TSN). However, TSN standards, such as Precision Time Protocol (PTP), focus on performance with little to no concern for security. This increases the exposure of the open FH to security risks. Attacks targeting synchronization mechanisms pose significant threats, potentially disrupting 5G networks and impairing connectivity. In this paper, we demonstrate the impact of successful spoofing and replay attacks against PTP synchronization. We show how a spoofing attack is able to cause a production-ready O-RAN and 5G-compliant private cellular base station to catastrophically fail within 2 seconds of the attack, necessitating manual intervention to restore full network operations. To counter this, we design a Machine Learning (ML)-based monitoring solution capable of detecting various malicious attacks with over 97.5% accuracy.

Rahul Soundrarajan, Claudio Fiandrino, Michele Polese et al.

Open RAN introduces a flexible, cloud-based architecture for the Radio Access Network (RAN), enabling Artificial Intelligence (AI)/Machine Learning (ML)-driven automation across heterogeneous, multi-vendor deployments. While EXplainable Artificial Intelligence (XAI) helps mitigate the opacity of AI models, explainability alone does not guarantee reliable network operations. In this article, we propose a lightweight verification approach based on interpretable models to validate the behavior of Deep Reinforcement Learning (DRL) agents for RAN slicing and scheduling in Open RAN. Specifically, we use Decision Tree (DT)-based verifiers to perform near-real-time consistency checks at runtime, which would be otherwise unfeasible with computationally expensive state-of-the-art verifiers. We analyze the landscape of XAI and AI verification, propose a scalable architectural integration, and demonstrate feasibility with a DT-based slice-verifier. We also outline future challenges to ensure trustworthy AI adoption in Open RAN.

Maxime Elkael, Salvatore D'Oro, Leonardo Bonati et al.

The Open RAN movement has catalyzed a transformation toward programmable, interoperable cellular infrastructures. Yet, today's deployments still rely heavily on static control and manual operations. To move beyond this limitation, we introduce AgenRAN, an AI-native, Open RAN-aligned agentic framework that generates and orchestrates a fabric of distributed AI agents based on Natural Language (NL) intents. Unlike traditional approaches that require explicit programming, AgentRAN's LLM-powered agents interpret natural language intents, negotiate strategies through structured conversations, and orchestrate control loops across the network. AgentRAN instantiates a self-organizing hierarchy of agents that decompose complex intents across time scales (from sub-millisecond to minutes), spatial domains (cell to network-wide), and protocol layers (PHY/MAC to RRC). A central innovation is the AI-RAN Factory, an automated synthesis pipeline that observes agent interactions and continuously generates new agents embedding improved control algorithms, effectively transforming the network from a static collection of functions into an adaptive system capable of evolving its own intelligence. We demonstrate AgentRAN through live experiments on 5G testbeds where competing user demands are dynamically balanced through cascading intents. By replacing rigid APIs with NL coordination, AgentRAN fundamentally redefines how future 6G networks autonomously interpret, adapt, and optimize their behavior to meet operator goals.

Michele Polese, Niloofar Mohamadi, Salvatore D'Oro et al.

The proliferation of data-intensive Artificial Intelligence (AI) applications at the network edge demands a fundamental shift in RAN design, from merely consuming AI for network optimization, to actively enabling distributed AI workloads. This paradigm shift presents a significant opportunity for network operators to monetize AI at the edge while leveraging existing infrastructure investments. To realize this vision, this article presents a novel converged O-RAN and AI-RAN architecture that unifies orchestration and management of both telecommunications and AI workloads on shared infrastructure. The proposed architecture extends the Open RAN principles of modularity, disaggregation, and cloud-nativeness to support heterogeneous AI deployments. We introduce two key architectural innovations: (i) the AI-RAN Orchestrator, which extends the O-RAN Service Management and Orchestration (SMO) to enable integrated resource and allocation across RAN and AI workloads; and (ii) AI-RAN sites that provide distributed edge AI platforms with real-time processing capabilities. The proposed system supports flexible deployment options, allowing AI workloads to be orchestrated with specific timing requirements (real-time or batch processing) and geographic targeting. The proposed architecture addresses the orchestration requirements for managing heterogeneous workloads at different time scales while maintaining open, standardized interfaces and multi-vendor interoperability.

Eduardo Baena, Paolo Testolina, Michele Polese et al.

Lunar surface operations impose stringent requirements on wireless communication systems, including autonomy, robustness to disruption, and the ability to adapt to environmental and mission-driven context. While Space-O-RAN provides a distributed orchestration model aligned with 3GPP standards, its decision logic is limited to static policies and lacks semantic integration. We propose a novel extension incorporating a semantic agentic layer enabled by the Model Context Protocol (MCP) and Agent-to-Agent (A2A) communication protocols, allowing context-aware decision making across real-time, near-real-time, and non-real-time control layers. Distributed cognitive agents deployed in rovers, landers, and lunar base stations implement wireless-aware coordination strategies, including delay-adaptive reasoning and bandwidth-aware semantic compression, while interacting with multiple MCP servers to reason over telemetry, locomotion planning, and mission constraints.

Salvatore D'Oro, Leonardo Bonati, Michele Polese et al.

The next generation of cellular networks will be characterized by softwarized, open, and disaggregated architectures exposing analytics and control knobs to enable network intelligence. How to realize this vision, however, is largely an open problem. In this paper, we take a decisive step forward by presenting and prototyping OrchestRAN, a novel orchestration framework that embraces and builds upon the Open RAN paradigm to provide a practical solution to these challenges. OrchestRAN has been designed to execute in the non-real-time RAN Intelligent Controller (RIC) and allows Network Operators (NOs) to specify high-level control/inference objectives (i.e., adapt scheduling, and forecast capacity in near-real-time for a set of base stations in Downtown New York). OrchestRAN automatically computes the optimal set of data-driven algorithms and their execution location to achieve intents specified by the NOs while meeting the desired timing requirements. We show that the problem of orchestrating intelligence in Open RAN is NP-hard, and design low-complexity solutions to support real-world applications. We prototype OrchestRAN and test it at scale on Colosseum. Our experimental results on a network with 7 base stations and 42 users demonstrate that OrchestRAN is able to instantiate data-driven services on demand with minimal control overhead and latency.

Michele Polese, Leonardo Bonati, Salvatore D'Oro et al.

In spite of the new opportunities brought about by the Open RAN, advances in ML-based network automation have been slow, mainly because of the unavailability of large-scale datasets and experimental testing infrastructure. This slows down the development and widespread adoption of Deep Reinforcement Learning (DRL) agents on real networks, delaying progress in intelligent and autonomous RAN control. In this paper, we address these challenges by proposing practical solutions and software pipelines for the design, training, testing, and experimental evaluation of DRL-based closed-loop control in the Open RAN. We introduce ColO-RAN, the first publicly-available large-scale O-RAN testing framework with software-defined radios-in-the-loop. Building on the scale and computational capabilities of the Colosseum wireless network emulator, ColO-RAN enables ML research at scale using O-RAN components, programmable base stations, and a "wireless data factory". Specifically, we design and develop three exemplary xApps for DRL-based control of RAN slicing, scheduling and online model training, and evaluate their performance on a cellular network with 7 softwarized base stations and 42 users. Finally, we showcase the portability of ColO-RAN to different platforms by deploying it on Arena, an indoor programmable testbed. Extensive results from our first-of-its-kind large-scale evaluation highlight the benefits and challenges of DRL-based adaptive control. They also provide insights on the development of wireless DRL pipelines, from data analysis to the design of DRL agents, and on the tradeoffs associated to training on a live RAN. ColO-RAN and the collected large-scale dataset will be made publicly available to the research community.

Leonardo Bonati, Pedram Johari, Michele Polese et al.

Colosseum is an open-access and publicly-available large-scale wireless testbed for experimental research via virtualized and softwarized waveforms and protocol stacks on a fully programmable, "white-box" platform. Through 256 state-of-the-art software-defined radios and a massive channel emulator core, Colosseum can model virtually any scenario, enabling the design, development and testing of solutions at scale in a variety of deployments and channel conditions. These Colosseum radio-frequency scenarios are reproduced through high-fidelity FPGA-based emulation with finite-impulse response filters. Filters model the taps of desired wireless channels and apply them to the signals generated by the radio nodes, faithfully mimicking the conditions of real-world wireless environments. In this paper, we introduce Colosseum as a testbed that is for the first time open to the research community. We describe the architecture of Colosseum and its experimentation and emulation capabilities. We then demonstrate the effectiveness of Colosseum for experimental research at scale through exemplary use cases including prevailing wireless technologies (e.g., cellular and Wi-Fi) in spectrum sharing and unmanned aerial vehicle scenarios. A roadmap for Colosseum future updates concludes the paper.

Michele Polese, Rittwik Jana, Velin Kounev et al.

The fifth generation of cellular networks (5G) will rely on edge cloud deployments to satisfy the ultra-low latency demand of future applications. In this paper, we argue that such deployments can also be used to enable advanced data-driven and Machine Learning (ML) applications in mobile networks. We propose an edge-controller-based architecture for cellular networks and evaluate its performance with real data from hundreds of base stations of a major U.S. operator. In this regard, we will provide insights on how to dynamically cluster and associate base stations and controllers, according to the global mobility patterns of the users. Then, we will describe how the controllers can be used to run ML algorithms to predict the number of users in each base station, and a use case in which these predictions are exploited by a higher-layer application to route vehicular traffic according to network Key Performance Indicators (KPIs). We show that the prediction accuracy improves when based on machine learning algorithms that rely on the controllers' view and, consequently, on the spatial correlation introduced by the user mobility, with respect to when the prediction is based only on the local data of each single base station.