Conventional time-series forecasting methods typically aim to minimize overall prediction error, without accounting for the varying importance of different forecast ranges in downstream applications. We propose a training methodology that enables forecasting models to adapt their focus to application-specific regions of interest at inference time, without retraining. The approach partitions the prediction space into fine-grained segments during training, which are dynamically reweighted and aggregated to emphasize the target range specified by the application. Unlike prior methods that predefine these ranges, our framework supports flexible, on-demand adjustments. Experiments on standard benchmarks and a newly collected wireless communication dataset demonstrate that our method not only improves forecast accuracy within regions of interest but also yields measurable gains in downstream task performance. These results highlight the potential for closer integration between predictive modeling and decision-making in real-world systems.

Artificial intelligence will be one of the key pillars of the next generation of mobile networks (6G), as it is expected to provide novel added-value services and improve network performance. In this context, large language models have the potential to revolutionize the telecom landscape through intent comprehension, intelligent knowledge retrieval, coding proficiency, and cross-domain orchestration capabilities. This paper presents Telco-oRAG, an open-source Retrieval-Augmented Generation (RAG) framework optimized for answering technical questions in the telecommunications domain, with a particular focus on 3GPP standards. Telco-oRAG introduces a hybrid retrieval strategy that combines 3GPP domain-specific retrieval with web search, supported by glossary-enhanced query refinement and a neural router for memory-efficient retrieval. Our results show that Telco-oRAG improves the accuracy in answering 3GPP-related questions by up to 17.6% and achieves a 10.6% improvement in lexicon queries compared to baselines. Furthermore, Telco-oRAG reduces memory usage by 45% through targeted retrieval of relevant 3GPP series compared to baseline RAG, and enables open-source LLMs to reach GPT-4-level accuracy on telecom benchmarks.

This paper presents a new system model to evaluate the capacity and power consumption of multi-layer 6 G networks utilising the upper mid-band (FR3). The model captures heterogeneous$(4 \mathrm{G}, 5 \mathrm{G})$, and 6 G deployments, analyzing their performance under different deployment strategies. Our results show that strategic 6 G deployments, non-co-located with existing 5G sites, significantly enhance throughput, with median and peak user rates of 300 Mbps and exceeding 1 Gbps, respectively. We also emphasize the importance of priority-based cell reselection and beam configuration to fully leverage 6 G capabilities. While 6 G implementation increases power consumption by 33%, non-colocated deployments strike a balance between performance and power consumption.

In this article, we introduce a method to optimize 5G massive multiple-input multiple-output (mMIMO) connectivity for unmanned aerial vehicles (UAVs) on aerial highways through strategic cell association. UAVs operating in 3D space encounter distinct channel conditions compared to traditional ground user equipment (gUE); under the typical line of sight (LoS) condition, UAVs perceive strong reference signal received power (RSRP) from multiple cells within the network, resulting in a large set of suitable serving cell candidates and in low signal-to-interference-plus-noise ratio (SINR) due to high interference levels. Additionally, a downside of aerial highways is to pack possibly many UAVs along a small portion of space which, when taking into account typical LoS propagation conditions, results in high channel correlation and severely limits spatial multiplexing capabilities. In this paper, we propose a solution to both problems based on the suitable selection of serving cells based on a new metric which differs from the classical terrestrial approaches based on maximum RSRP. We then introduce an algorithm for optimal planning of synchronization signal block (SSB) beams for this set of cells, ensuring maximum coverage and effective management of UAVs cell associations. Simulation results demonstrate that our approach significantly improves the rates of UAVs on aerial highways, up to four times in achievable data rates, without impacting ground user performance.

The application of Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) systems in the telecommunication domain presents unique challenges, primarily due to the complex nature of telecom standard documents and the rapid evolution of the field. The paper introduces Telco-RAG,1an open-source RAG framework designed to handle the specific needs of telecommunications standards, particularly 3rd Generation Partnership Project (3GPP) documents. Telco-RAG addresses the critical challenges of implementing a RAG pipeline on highly technical content, paving the way for applying LLMs in telecommunications and offering guidelines for RAG implementation in other technical domains.

The increasing interest in Large Language Models (LLMs) within the telecommunications sector underscores their potential to revolutionize operational efficiency. However, the deployment of these sophisticated models is often hampered by their substantial size and computational demands, raising concerns about their viability in resource-constrained environments. Addressing this challenge, recent advancements have seen the emergence of small language models that surprisingly exhibit performance comparable to their larger counterparts in many tasks, such as coding and common-sense reasoning. Phi-2, a compact yet powerful model, exemplifies this new wave of efficient small language models. This paper conducts a comprehensive evaluation of Phi-2’s intrinsic understanding of the telecommunications domain. Recognizing the scale-related limitations, we enhance Phi-2’s capabilities through a Retrieval-Augmented Generation approach, meticulously integrating an extensive knowledge base specifically curated with telecom standard specifications. The enhanced Phi-2 model demonstrates a profound improvement in accuracy, answering questions about telecom standards with a precision that closely rivals the more resource-intensive GPT-3.5. The paper further explores the refined capabilities of Phi-2 in addressing problem-solving scenarios within the telecom sector, highlighting its potentials and limitations.

The rapidly evolving communication landscape, with the advent of 6G technology, brings new challenges to the design and operation of wireless networks. One of the key concerns is the energy efficiency of the Radio Access Network (RAN), as the exponential growth in wireless traffic demands increasingly higher energy consumption. In this paper, we assess the potential of integrating a High Altitude Platform Station (HAPS) to improve the energy efficiency of a RAN, and quantify the potential energy conservation through meticulously designed simulations. We propose a quantitative framework based on real traffic patterns to estimate the energy consumption of the HAPS-integrated RAN and compare it with the conventional terrestrial RAN. Our simulation results elucidate that HAPS can significantly reduce energy consumption by up to almost 30% by exploiting the unique advantages of HAPS, such as its self-sustainability, high altitude, and wide coverage. We further analyze the impact of different system parameters on performance, and provide insights for the design and optimization of future 6G networks. Our work sheds light on the potential of HAPS-integrated RAN to mitigate the energy challenges in the 6G era, and contributes to the sustainable development of wireless communications.

The fifth generation of the Radio Access Network (RAN) has brought new services, technologies, and paradigms with the corresponding societal benefits. However, the energy consumption of 5G networks is today a concern. In recent years, the design of new methods for decreasing the RAN power consumption has attracted interest from both the research community and standardization bodies, and many energy savings solutions have been proposed. However, there is still a need to understand the power consumption behavior of state-of-the-art base station architectures, such as multi-carrier active antenna units (AAUs), as well as the impact of different network parameters. In this paper, we present a power consumption model for 5G AAUs based on artificial neural networks. We demonstrate that this model achieves good estimation performance, and it is able to capture the benefits of energy saving when dealing with the complexity of multi-carrier base stations architectures. Importantly, multiple experiments are carried out to show the advantage of designing a general model able to capture the power consumption behaviors of different types of AAUs. Finally, we provide an analysis of the model scalability and the training data requirements.

The energy consumption of the fifth generation (5G) of cellular technology is concerning for the mobile industry and the entire society. To minimize the environmental footprint and economic costs of 5G, it is necessary to adapt the transmission capabilities of networks to end-users’ quality of service requirements. In this paper, we focus on the carrier shutdown approach that enables a base station (BS) to autonomously switch off during low traffic periods, by transferring its load to neighbouring active BSs. More specifically, we propose a data-driven framework, constructed through real network measurements, which statistically characterizes the user equipment (UE) transfer across neighbouring BSs, when carrier shutdown operates. The implementation of this framework allows the 5G system to determine a poor load distribution due to energy saving mechanisms, prevent drastic reductions in UE performance, and ultimately estimate energy savings when activating carrier shutdown.

This work considers a multi-cell, multi-carrier massive MIMO network with carrier aggregation, and tackles the rate versus power consumption trade-off, by jointly optimizing the number of employed component carriers, active antennas, base station density, and transmit power. A provably convergent algorithm is developed together with closed-form results for the individual optimization of the considered resources. Numerical results show how carrier aggregation can effectively reduce the power consumption without sacrificing the rate performance.

The energy consumption of cellular networks, despite the high energy efficiency of the fifth generation (5G) of mobile technology, is still a challenge. The fundamental problem arises due to the complexity of optimising the operation of the available rich set of energy efficiency features in large-scale deployments. To assist such optimisation, a large body of research -with the resulting understanding and algorithms-exists, particularly on the energy efficiency of single-cell massive multiple-input multiple-output systems. However, other funda-mental cellular features, such as those relating to multi-carrier systems, remain largely unexplored. In this paper, we show how multi-carrier features, such as carrier aggregation, can play a significant role in energy savings, and question the need for hundreds of antennas and transceiver chains at the base stations as an urgent solution to increase the energy efficiency of next generation networks.

The energy consumption and carbon footprint of the fifth-generation (5G) of mobile technology is a current concern to mobile network operators (MNOs). These are currently attempting to lower both their carbon emissions and electricity bills by investigating new schemes that allow adapting the network transmission capabilities to the end-users' quality of service (QoS) requirements. Many of such schemes rely on accurate traffic forecasting, and as a consequence, there is a large effort on investigating novel machine learning (ML) algorithms, which fed by network measurement data and empowered by the computing capabilities of dedicated hardware, can help modelling and predicting users' behaviours. Most of the works in the literature, however, focus on predicting the traffic when energy saving features, e.g. carrier shutdown, are not implemented or activated. However, the prediction task becomes much more challenging when energy saving features are adopted due to their impact to the actual measured traffic. In this paper, we consider a scenario in which part of the base stations implement energy saving schemes, which allow them to dynamically switch off part of their hardware to reduce their power consumption. Then, we present a ML framework based on graph convolutional networks (GCNs) for traffic forecasting in such dynamic scenarios, and compare its performance with other statistical and ML prediction algorithms. The proposed GCN framework provides significant accuracy gains. Moreover, we provide an analysis of the impact of spatial correlation-captured by the GCN model-on the achieved performance.

In this paper, we focus on the design of energy self-sustainable mobile networks by enabling intelligent energy management that allows the base stations to mostly operate off-grid by using renewable energy. We propose a centralized control algorithm based on Deep Reinforcement Learning. The single agent is able to learn how to efficiently balance the energy inflow and spending among base stations observing the environment and interacting with it. In particular, we provide a study on the performance achieved by this approach when considering different representations of the environment. Numerical results demonstrate that using a good level of abstraction in the choice of the representation variables may enable a proper mapping of the environment into actions to take, so as to maximize the numerical reward.

In this paper, we propose an unsupervised method to learn hidden features of the solar energy generation from a PV system that may give a more accurate characterization of the process. In a first step, solar radiation data is converted into instantaneous solar power through a detailed source model. Then, two different approaches, namely PCA and autoencoder, are used to extract meaningful features from the traces of the solar energy generation. We interpret the latent variables characterizing the solar energy generation process by analyzing the similarities of 67 cities in Europe, North-Africa and Middle-East through an agglomerative hierarchical clustering algorithm. This analysis provides also a comparison between the feature extraction capabilities of the PCA and the autoencoder.

Flexible functional split in Cloud Radio Access Network (CRAN) greatly overcomes fronthaul capacity and latency challenges. In such architecture, part of the baseband processing is done locally and the remaining is done remotely in the central cloud. On the other hand, Energy Harvesting (EH) technologies are increasingly adopted due to sustainability and economic advantages. Power consumption due to baseband processing has a huge share in the total power consumption breakdown of smaller base stations. Given that such base stations are powered by EH, in addition to QoS constraints, energy availability also conditions the decision on where to place each baseband function in the system. This work focuses on determining the performance bounds of an optimal placement of baseband functional split option in virtualized small cells that are solely powered by EH. The work applies Dynamic Programming (DP), in particular, Shortest Path search is used to determine the optimal functional split option considering traffic requirements and available energy budget.

In this paper, we propose an optimal direct load control of renewable powered small base stations based on Dynamic Programming. The optimization is represented using Graph Theory and the problem is stated as a Shortest Path problem. The proposed optimal algorithm is able to adapt to the varying conditions of renewable energy sources and traffic demands. We analyze the optimal ON/OFF policies considering different energy and traffic scenarios. Then, we evaluate network performance in terms of system drop rate and grid energy consumption. The obtained results are compared with a greedy approach. This study allows to elaborate on the behavior and performance bounds of the system and gives a guidance for approximated policy search methods.