Searcharxiv⌕ Search

arXiv subjects

Merim Dzaferagic

Publications and source records attributed to Merim Dzaferagic.

At least 19 recordsLinked to original sources
arXiv2026

Energy-Latency Trade-offs in O-RAN with Distributed Baseband Processing and AI Inference

The Open Radio Access Network (O-RAN) architecture introduces flexible functional splits and open interfaces that enable distributed and centralized deployment of baseband processing. While this flexibility offers opportunities for improved resource utilization, it also introduces fundamental trade-offs between energy efficiency and latency. In this paper, we develop a throughput-based end-to-end energy consumption model for O-RAN and extend it by incorporating detailed latency modeling and application-specific Artificial Intelligence/Machine Learning inference costs. The proposed end-to-end modeling framework provides a general representation of processing, transport, and inference-related energy and delay across the access, metro, and long-haul network segments. Building on this general model, we formulate an optimization problem that selects the placement of baseband processing and AI inference tasks across candidate O-RAN configurations to analyze energy-latency tradeoffs under network load, server frequency, and energy-budget constraints. Using representative hardware platforms and realistic traffic assumptions, we evaluate multiple baseband processing placements corresponding to different O-RAN functional configurations. Our results reveal how user quality of service requirements and network load conditions jointly determine the optimal placement of baseband processing and AI inference tasks, highlighting the inherent trade-off between energy efficiency and latency. The analysis provides practical insights for latency-aware and energy-efficient O-RAN deployments supporting emerging AI-driven services.

cs.NI↗
arXiv2026

Is Our Benchmark Enough? An Analysis of Continual Learning for MLLMs

Continual adaptation is essential for multimodal large language models (MLLMs) deployed across evolving domains, but the state-of-the-art MR-LoRA method highly relies on the assumption that a MLLM-based router is necessary to process complex multimodal inputs. This paper revisits this claim on the MLLM-CL benchmark and argues for two claims. \textbf{First}, routing does not require an MLLM: a simple training-free, replay-free ptotypical routing method (\textsc{RePRo}), uses frozen pretrained features and task prototypes to match the MLLM-based router of MR-LoRA at far lower computational cost. \textbf{Second}, shared experts do not improve continual learning for MLLMs, despite their theoretical appeal. We show that these findings arise from two structural limitations of MLLM-CL: (1) its tasks are \textbf{highly separable} in representation space, and (2) its fixed task order makes conclusions \textbf{sensitive to a single curriculum} rather than robust across diverse continual-learning trajectories. As a result, the benchmark primarily rewards learning in isolation rather than genuine continual transfer. This motivates a new design for future benchmarks of continual MLLM learning, with overlapping task manifolds, multiple task orders, fine-grained domain shifts, and evaluation protocols that reward forward transfer as well as retention.

cs.LG↗
arXiv2026

UB-SMoE: Universally Balanced Sparse Mixture-of-Experts for Resource-adaptive Federated Fine-tuning of Foundation Models

Heterogeneous LoRA-rank methods address system heterogeneity in federated fine-tuning of foundation models by assigning client-specific ranks based on computational capabilities. However, these methods achieve only marginal computational savings, as dense feed-forward computations dominate. Sparse Mixture-of-Experts (SMoE) provides a promising alternative through conditional computation, yet we identify that its naive application to heterogeneous federated settings introduces two critical discordances: (i) expert utilization imbalance and (ii) non-differentiability of Top-K routing. Our convergence analysis demonstrates that these discordances lead to degraded convergence, particularly for resource-constrained clients. To address these challenges, we propose Universally Balanced Sparse Mixture-of-Experts (UB-SMoE), which introduces Dynamic Modulated Routing (DMR) to rebalance expert utilization, and Universal Pseudo-Gradient (PG) to reconstruct learning signals for non-activated experts. These mechanisms form a self-reinforcing cycle that maintains expert viability across heterogeneous clients. Experiments on benchmarks show that UB-SMoE achieves up to $45.0\%$ computational reduction on low-resource clients while improving their performance by $8.7 \times$ compared to existing heterogeneous LoRA-rank methods.

cs.LG↗
arXiv2026

Energy Consumption in Next Generation Radio Access Networks

The radio access network (RAN) accounts for the largest share of energy consumption in mobile networks, making it essential to understand how and where this energy is used, particularly as future networks move toward higher levels of densification. Open radio access networks (O-RAN) have emerged as a promising approach to support this evolution through open interfaces that enable a multivendor environment, support for hierarchical intelligent controls, and simplified, cost-effective radio units that facilitate large-scale deployments. This paper examines the energy consumption in next-generation RAN architectures through transaction-based energy models. The model captures both processing and transmission energy components and evaluates how energy use varies with the placement of baseband processing (BBP) across network nodes and with different levels of network densification. Results indicate that processing energy dominates total consumption and that the location of BBP strongly influences overall energy efficiency. These insights can inform the design of future RAN deployments that balance flexibility, cost, and sustainability.

cs.NI↗
arXiv2026

AI-Native Network Controller: A Modular Framework for Safe Agentic Control of Multi-Domain Network Infrastructure

The convergence of multiple network domains, including radio access, optical transport, and core networks, under unified intelligent control is a fundamental requirement for future 6G systems. This is important because existing network controllers remain largely domain-specific, such as the O-RAN RIC for radio, or they lack native support for AI-driven automation across heterogeneous infrastructure. As a result, safe and coordinated agentic control of multi-domain networks is still an open challenge. In this paper, we present the AI-Native Network Controller (AI-NNC), an open-source and modular framework that enables agentic AI control across diverse network domains. The framework is designed around a protocol-agnostic architecture in which each physical device is integrated through a lightweight Python adapter, while control logic is implemented through domain-specific control applications. Beyond closed-loop control, the framework also supports dataset collection, agentic AI experimentation, and coordinated testbed operation using the same validated control and measurement interfaces. This design enables a safer paradigm for autonomous network management, where AI agents operate through validated applications rather than issuing commands directly to network equipment.

cs.NI↗
arXiv2026

msData: A Millisecond-Resolution Network Dataset for Advancing Time Series Foundation Models

Time series foundation models (TSFMs) require diverse, real-world datasets to adapt across varying domains and temporal frequencies. However, current large-scale datasets predominantly focus on low-frequency time series with sampling intervals, i.e., time resolution, in the range of seconds to years, hindering their ability to capture the nuances of high-frequency time series data. To address this limitation, we introduce a novel dataset, \textbf{msData}, that captures millisecond-resolution wireless and traffic conditions from an operational 5G wireless deployment, expanding the scope of TSFMs to incorporate high-frequency data for pre-training. Further, the dataset introduces a new domain, namely, wireless networks, thus complementing existing more general domains like energy and finance. The dataset also provides use cases for short-term forecasting, with prediction horizons spanning from 1 millisecond (1 step) to 96 milliseconds (96 steps). To demonstrate the utility of the dataset, we benchmark traditional machine learning models, transformer-based deep learning models, and TSFMs on forecasting tasks using representative subsets of the data, including a static mobility pattern within YouTube traffic class and a train mobility pattern within Web Browsing traffic class. Across these data distributions, we demonstrate that most TSFM model configurations perform poorly in both zero-shot and fine-tuned settings. Our work underscores the importance of incorporating high-frequency datasets during pre-training and forecasting to enhance architectures, fine-tuning strategies, generalization, and robustness of TSFMs in real-world applications. Code is available at this repository: https://github.com/khanalsubina/msData. Data is available on: \href{https://huggingface.co/datasets/subinak/Open_RAN_Performance_Measurement_Dataset_with_Traffic_and_Mobility_Labels}{Hugging Face}.

cs.LG↗
arXiv2025

Dynamic Features Adaptation in Networking: Toward Flexible training and Explainable inference

As AI becomes a native component of 6G network control, AI models must adapt to continuously changing conditions, including the introduction of new features and measurements driven by multi-vendor deployments, hardware upgrades, and evolving service requirements. To address this growing need for flexible learning in non-stationary environments, this vision paper highlights Adaptive Random Forests (ARFs) as a reliable solution for dynamic feature adaptation in communication network scenarios. We show that iterative training of ARFs can effectively lead to stable predictions, with accuracy improving over time as more features are added. In addition, we highlight the importance of explainability in AI-driven networks, proposing Drift-Aware Feature Importance (DAFI) as an efficient XAI feature importance (FI) method. DAFI uses a distributional drift detector to signal when to apply computationally intensive FI methods instead of lighter alternatives. Our tests on 3 different datasets indicate that our approach reduces runtime by up to 2 times, while producing more consistent feature importance values. Together, ARFs and DAFI provide a promising framework to build flexible AI methods adapted to 6G network use-cases.

cs.LG↗
arXiv2025

Modular and Integrated AI Control Framework across Fiber and Wireless Networks for 6G

The rapid evolution of communication networks towards 6G increasingly incorporates advanced AI-driven controls across various network segments to achieve intelligent, zero-touch operation. This paper proposes a comprehensive and modular framework for AI controllers, designed to be highly flexible and adaptable for use across both fiber optical and radio networks. Building on the principles established by the O-RAN Alliance for near-Real-Time RAN Intelligent Controllers (near-RT RICs), our framework extends this AI-driven control into the optical domain. Our approach addresses the critical need for a unified AI control framework across diverse network transport technologies and domains, enabling the development of intelligent, automated, and scalable 6G networks.

cs.NI↗
arXiv2024

Decentralized Multi-Party Multi-Network AI for Global Deployment of 6G Wireless Systems

Multiple visions of 6G networks elicit Artificial Intelligence (AI) as a central, native element. When 6G systems are deployed at a large scale, end-to-end AI-based solutions will necessarily have to encompass both the radio and the fiber-optical domain. This paper introduces the Decentralized Multi-Party, Multi-Network AI (DMMAI) framework for integrating AI into 6G networks deployed at scale. DMMAI harmonizes AI-driven controls across diverse network platforms and thus facilitates networks that autonomously configure, monitor, and repair themselves. This is particularly crucial at the network edge, where advanced applications meet heightened functionality and security demands. The radio/optical integration is vital due to the current compartmentalization of AI research within these domains, which lacks a comprehensive understanding of their interaction. Our approach explores multi-network orchestration and AI control integration, filling a critical gap in standardized frameworks for AI-driven coordination in 6G networks. The DMMAI framework is a step towards a global standard for AI in 6G, aiming to establish reference use cases, data and model management methods, and benchmarking platforms for future AI/ML solutions.

cs.NI↗
arXiv2024

ML-based handover prediction over a real O-RAN deployment using RAN Intelligent controller

O-RAN introduces intelligent and flexible network control in all parts of the network. The use of controllers with open interfaces allow us to gather real time network measurements and make intelligent/informed decision. The work in this paper focuses on developing a use-case for open and reconfigurable networks to investigate the possibility to predict handover events and understand the value of such predictions for all stakeholders that rely on the communication network to conduct their business. We propose a Long-Short Term Memory Machine Learning approach that takes standard Radio Access Network measurements to predict handover events. The models were trained on real network data collected from a commercial O-RAN setup deployed in our OpenIreland testbed. Our results show that the proposed approach can be optimized for either recall or precision, depending on the defined application level objective. We also link the performance of the Machine Learning (ML) algorithm to the network operation cost. Our results show that ML-based matching between the required and available resources can reduce operational cost by more than 80%, compared to long term resource purchases.

cs.NI↗
arXiv2024

Cross-Domain AI for Early Attack Detection and Defense Against Malicious Flows in O-RAN

Only the chairs can edit In the fight against cyber attacks, Network Softwarization (NS) is a flexible and adaptable shield, using advanced software to spot malicious activity in regular network traffic. However, the availability of comprehensive datasets for mobile networks, which are fundamental for the development of Machine Learning (ML) solutions for attack detection near their source, is still limited. Cross-Domain Artificial Intelligence (AI) can be the key to address this, although its application in Open Radio Access Network (O-RAN) is still at its infancy. To address these challenges, we deployed an end-to-end O-RAN network, that was used to collect data from the RAN and the transport network. These datasets allow us to combine the knowledge from an in-network ML traffic classifier for attack detection to bolster the training of an ML-based traffic classifier specifically tailored for the RAN. Our results demonstrate the potential of the proposed approach, achieving an accuracy rate of 93%. This approach not only bridges critical gaps in mobile network security but also showcases the potential of cross-domain AI in enhancing the efficacy of network security measures.

cs.CR↗
arXiv2023

Machine Learning-based Early Attack Detection Using Open RAN Intelligent Controller

We design and demonstrate a method for early detection of Denial-of-Service attacks. The proposed approach takes advantage of the OpenRAN framework to collect measurements from the air interface (for attack detection) and to dynamically control the operation of the Radio Access Network (RAN). For that purpose, we developed our near-Real Time (RT) RAN Intelligent Controller (RIC) interface. We apply and analyze a wide range of Machine Learning algorithms to data traffic analysis that satisfy the accuracy and latency requirements set by the near-RT RIC. Our results show that the proposed framework is able to correctly classify genuine vs. malicious traffic with high accuracy (i.e., 95%) in a realistic testbed environment, allowing us to detect attacks already at the Distributed Unit (DU), before malicious traffic even enters the Centralized Unit (CU).

cs.NI↗
arXiv2023

ML Approach for Power Consumption Prediction in Virtualized Base Stations

The flexibility introduced with the Open Radio Access Network (O-RAN) architecture allows us to think beyond static configurations in all parts of the network. This paper addresses the issue related to predicting the power consumption of different radio schedulers, and the potential offered by O-RAN to collect data, train models, and deploy policies to control the power consumption. We propose a black-box (Neural Network) model to learn the power consumption function. We compare our approach with a known hand-crafted solution based on domain knowledge. Our solution reaches similar performance without any previous knowledge of the application and provides more flexibility in scenarios where the system behavior is not well understood or the domain knowledge is not available.

cs.LG↗
arXiv2022

Analysis of Temporal Robustness in Massive Machine Type Communications

The evolution of fifth generation (5G) networks needs to support the latest use cases, which demand robust network connectivity for the collaborative performance of the network agents, like multi-robot systems and vehicle to anything (V2X) communication. Unfortunately, the user device's limited communication range and battery constraint confirm the unfitness of known robustness metrics suggested for fixed networks, when applied to time-switching communication graphs. Furthermore, the calculation of most of the existing robustness metrics involves non-deterministic polynomial-time complexity, and hence are best-fitted only for small networks. Despite a large volume of works, the complete analysis of a $\textit{low-complexity}$ temporal robustness metric for a communication network is absent in the literature, and the present work aims to fill this gap. More in detail, our work provides a stochastic analysis of network robustness for a massive machine type communication (mMTC) network. The numerical investigation corroborates the exactness of the proposed analytical framework for temporal robustness metric. Along with studying the impact on network robustness of various system parameters, such as cluster head (CH) probability, power threshold value, network size, and node failure probability, we justify the observed trend of numerical results probabilistically.

cs.NI↗
arXiv2022

Near Real-Time Distributed State Estimation via AI/ML-Empowered 5G Networks

Fifth-Generation (5G) networks have a potential to accelerate power system transition to a flexible, softwarized, data-driven, and intelligent grid. With their evolving support for Machine Learning (ML)/Artificial Intelligence (AI) functions, 5G networks are expected to enable novel data-centric Smart Grid (SG) services. In this paper, we explore how data-driven SG services could be integrated with ML/AI-enabled 5G networks in a symbiotic relationship. We focus on the State Estimation (SE) function as a key element of the energy management system and focus on two main questions. Firstly, in a tutorial fashion, we present an overview on how distributed SE can be integrated with the elements of the 5G core network and radio access network architecture. Secondly, we present and compare two powerful distributed SE methods based on: i) graphical models and belief propagation, and ii) graph neural networks. We discuss their performance and capability to support a near real-time distributed SE via 5G network, taking into account communication delays.

cs.LG↗
arXiv2022

Minimizing the Signaling Overhead and Latency based on Users' Mobility Patterns

We demonstrate a distributed and a centralized 4G/5G compliant approach to minimize signaling and latency related to user mobility in cellular networks. This is crucial due to the densification of networks and the additional signaling introduced by the new 5G service based architecture. By exploiting standardized protocols, our solutions dynamically reorganize the association between nodes in Radio Access Network (RAN) and the core. We validated the proposed approaches using real user mobility datasets. Our results show that both our distributed and centralized solutions significantly reduce the signaling between core and RAN compared to the traditional approach based on geographical proximity. As a result, both approaches significantly reduce the average handover procedure processing time. Moreover, by relying on locally available information, the distributed approach can quickly adapt to changes in the user movement patterns as they happen.

cs.NI↗