Antonio De Domenico
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Antonio De Domenico has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Nicola Piovesan
Paris Research Center, Huawei Technologies, Boulogne-Billancourt, France
11 shared publications Harvey Bao
Advanced Wireless Technology Lab, Huawei France, Boulogne-Billancourt, France
4 shared publications- David Lopez-Perez
Universitat Politcnica de Valncia, Valencia, Spain
3 shared publications - Emilio Calvanese Strinati
University Grenoble Alpes and CEA-Leti
3 shared publications - Fadhel Ayed
Paris Research Center, Huawei Technologies, Boulogne-Billancourt, France
3 shared publications - Gourab Ghatak
Department of Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, India
3 shared publications
Latest Publications
- Open Access
- 4 April 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Advancing green mobile networks
Mobile communication is a cornerstone technology in our society, facilitating personal relationships, fostering cooperation, and advancing education. Each generation of this technology has brought new services, from voice calls and messaging to internet access, and now Artificial Intelligence (AI). This transformation has steadily augmented the mobile network resource requirements. Driven by the societal and economical needs to increase the mobile network energy efficiency, reduce its energy consumption and the associated environmental footprint, this comment surveys standardized metrics to assess energy-related network performance and overview data-driven state-of-the-art solutions to accelerate the energy transition of mobile networks. Mobile communication underpins society, evolving from calls to AI and increasing network demands. To cut energy use and environmental impact, this work reviews standardized energy metrics and data-driven solutions for greener mobile networks.
- 4 April 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessAdvancing green mobile networks
Mobile communication is a cornerstone technology in our society, facilitating personal relationships, fostering cooperation, and advancing education. Each generation of this technology has brought new services, from voice calls and messaging to internet access, and now Artificial Intelligence (AI). This transformation has steadily augmented the mobile network resource requirements. Driven by the societal and economical needs to increase the mobile network energy efficiency, reduce its energy consumption and the associated environmental footprint, this comment surveys standardized metrics to assess energy-related network performance and overview data-driven state-of-the-art solutions to accelerate the energy transition of mobile networks. Mobile communication underpins society, evolving from calls to AI and increasing network demands. To cut energy use and environmental impact, this work reviews standardized energy metrics and data-driven solutions for greener mobile networks.
- 30 March 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Technique de gestion de ressources radios pour l'amélioration de l'efficacité énergétique dans les réseaux cellulaires hétérogènes
Les communications sans fil prolifèrent dans presque chaque aspect de la société humaine : puissants ‘smart-phones' et ‘tablets', accès haut débit sans fil, et communications ‘machine-to-machine' ont généré des volumes de trafic de données imprévisibles quelques années en arrière. Dans ce nouveau paradigme, l'industrie des télécommunications se doit de garantir à la fois la durabilité économique des communications sans fil à large bande ainsi que la qualité de son service. En outre, il y a une forte incitation sociale à réduire les émissions de C02 duent aux communications mobiles, qui a augmenté notamment dans la dernière décennie. Dans ce contexte, l'intégration des ‘femtocells' dans les réseaux cellulaires est une solution à faible coût pour offrir une qualité de service élevée et en même temps de décharger le réseau macrocellule. Cependant, le déploiement massif et chaotique des points d'accès femtocell et leurs opérations non coordonnées peuvent conduire à une augmentation de l'interférence co-canal. De plus, un nombre élevé de cellules faiblement chargées augmente la consommation énergétique du réseau. Dans cette thèse, nous avons étudié les effets du déploiement de femtocells sur l'efficacité énergétique du réseau cellulaire. Par ailleurs, nous investiguons sur les mécanismes d'adaptation pour les réseaux des femtocells comme un moyen pour améliorer l'efficacité des communications mobiles. Notre objectif est de répondre dynamiquement à la demande des ressources afin de limiter la consommation d'énergie moyenne et l'interférence co-canal, tout en garantissant la qualité de service. Nous profitons du contexte inhabituel de communication ‘femtocellulaire' pour proposer des mécanismes d'allocation des ressources et des systèmes de gestion de réseau qui coordonne l'activité des points d'accès, la consommation d'énergie et de la couverture. Les résultats des simulations montrent que nos propositions améliorent l'efficacité énergétique et les performances perçues par les utilisateurs du système dans les réseaux ‘femtocellulaires' coopératives et autonomes.
- 30 March 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Technique de gestion de ressources radios pour l'amélioration de l'efficacité énergétique dans les réseaux cellulaires hétérogènes
Les communications sans fil prolifèrent dans presque chaque aspect de la société humaine : puissants ‘smart-phones' et ‘tablets', accès haut débit sans fil, et communications ‘machine-to-machine' ont généré des volumes de trafic de données imprévisibles quelques années en arrière. Dans ce nouveau paradigme, l'industrie des télécommunications se doit de garantir à la fois la durabilité économique des communications sans fil à large bande ainsi que la qualité de son service. En outre, il y a une forte incitation sociale à réduire les émissions de C02 duent aux communications mobiles, qui a augmenté notamment dans la dernière décennie. Dans ce contexte, l'intégration des ‘femtocells' dans les réseaux cellulaires est une solution à faible coût pour offrir une qualité de service élevée et en même temps de décharger le réseau macrocellule. Cependant, le déploiement massif et chaotique des points d'accès femtocell et leurs opérations non coordonnées peuvent conduire à une augmentation de l'interférence co-canal. De plus, un nombre élevé de cellules faiblement chargées augmente la consommation énergétique du réseau. Dans cette thèse, nous avons étudié les effets du déploiement de femtocells sur l'efficacité énergétique du réseau cellulaire. Par ailleurs, nous investiguons sur les mécanismes d'adaptation pour les réseaux des femtocells comme un moyen pour améliorer l'efficacité des communications mobiles. Notre objectif est de répondre dynamiquement à la demande des ressources afin de limiter la consommation d'énergie moyenne et l'interférence co-canal, tout en garantissant la qualité de service. Nous profitons du contexte inhabituel de communication ‘femtocellulaire' pour proposer des mécanismes d'allocation des ressources et des systèmes de gestion de réseau qui coordonne l'activité des points d'accès, la consommation d'énergie et de la couverture. Les résultats des simulations montrent que nos propositions améliorent l'efficacité énergétique et les performances perçues par les utilisateurs du système dans les réseaux ‘femtocellulaires' coopératives et autonomes.
- Top Cited Papers
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-7
TeleMath: A Benchmark for Large Language Models in Telecom Mathematical Problem Solving
Large Language Models (LLMs) achieve near-human performance on general-purpose benchmarks such as MATH and GSM8K, yet their ability to solve domain-specific numerical problems remains underexplored. In telecommunications, many core tasks—from signal processing to network optimization—require precise quantitative reasoning, where errors directly impact engineering practice. Existing benchmarks evaluate LLMs on general mathematics or telecom-related language tasks, but none assess llM capability to solve numerical mathematical problems in the telecom domain. To address this gap, we introduce TeleMath, the first benchmark dataset specifically designed to evaluate LLM performance in solving mathematical problems with numerical solutions in the telecommunications domain. TeleMath is built through a novel data generation methodology that starts from a compact seed of expert-crafted problems and expands it via problem decomposition, code- and symbolic math–driven blueprint generation, parameterized synthesis, and semantic validation. The resulting benchmark comprises 500 expert-verified question– answer pairs spanning diverse topics in the telecom domain with different levels of difficulty, each requiring a numerical solution. We then conduct a systematic benchmarking study of leading open-source LLMs, comparing reasoning-oriented models against general-purpose ones. On TeleMath, the best reasoning model, Qwen3-32B, achieves 69.51% pass@1 accuracy, substantially outperforming the best non-reasoning baseline, Qwen2.5-Math-72B-Instruct, which achieves 39.99% average accuracy with 2× more parameters, underscoring the unique challenges of telecom mathematics. We release TeleMath and its evaluation framework publicly to foster progress in domain-specific reasoning for telecommunications and beyond.
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-7
Top Cited PapersTeleMath: A Benchmark for Large Language Models in Telecom Mathematical Problem Solving
Large Language Models (LLMs) achieve near-human performance on general-purpose benchmarks such as MATH and GSM8K, yet their ability to solve domain-specific numerical problems remains underexplored. In telecommunications, many core tasks—from signal processing to network optimization—require precise quantitative reasoning, where errors directly impact engineering practice. Existing benchmarks evaluate LLMs on general mathematics or telecom-related language tasks, but none assess llM capability to solve numerical mathematical problems in the telecom domain. To address this gap, we introduce TeleMath, the first benchmark dataset specifically designed to evaluate LLM performance in solving mathematical problems with numerical solutions in the telecommunications domain. TeleMath is built through a novel data generation methodology that starts from a compact seed of expert-crafted problems and expands it via problem decomposition, code- and symbolic math–driven blueprint generation, parameterized synthesis, and semantic validation. The resulting benchmark comprises 500 expert-verified question– answer pairs spanning diverse topics in the telecom domain with different levels of difficulty, each requiring a numerical solution. We then conduct a systematic benchmarking study of leading open-source LLMs, comparing reasoning-oriented models against general-purpose ones. On TeleMath, the best reasoning model, Qwen3-32B, achieves 69.51% pass@1 accuracy, substantially outperforming the best non-reasoning baseline, Qwen2.5-Math-72B-Instruct, which achieves 39.99% average accuracy with 2× more parameters, underscoring the unique challenges of telecom mathematics. We release TeleMath and its evaluation framework publicly to foster progress in domain-specific reasoning for telecommunications and beyond.
- 8 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 2334-2347
Telco-oRAG: Optimizing Retrieval-Augmented Generation for Telecom Queries via Hybrid Retrieval and Neural Routing
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.
- 8 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 2334-2347
Telco-oRAG: Optimizing Retrieval-Augmented Generation for Telecom Queries via Hybrid Retrieval and Neural Routing
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.
- Top Cited Papers
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 40 (2) , 253-260
TeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledge
We introduce TeleQnA, the first benchmark dataset designed to evaluate the knowledge of Large Language Models (LLMs) in telecommunications. Comprising 10,000 questions and answers, this dataset draws from diverse sources, including standards and research manuscripts. This paper outlines the automated question generation framework responsible for creating this dataset and provides details on how we integrated human input at various stages to ensure the quality of the questions. Afterwards, using the provided dataset, an evaluation is conducted to assess the capabilities of several LLMs, including GPT-3.5, GPT-4, and Mixtral. The results highlight that these models struggle with complex standards-related questions but exhibit proficiency in addressing general telecom-related inquiries. Additionally, our results showcase how incorporating telecom knowledge context significantly enhances their performance, thus shedding light on the need for a specialized telecom foundation model. Finally, the dataset is shared with active telecom professionals, whose performance is subsequently benchmarked against that of the LLMs. The findings illustrate that LLMs can rival the performance of active professionals in telecom knowledge, thanks to their capacity to process vast amounts of information, underscoring the potential of LLMs within this domain. The dataset has been made publicly accessible on GitHub1.
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 40 (2) , 253-260
Top Cited PapersTeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledge
We introduce TeleQnA, the first benchmark dataset designed to evaluate the knowledge of Large Language Models (LLMs) in telecommunications. Comprising 10,000 questions and answers, this dataset draws from diverse sources, including standards and research manuscripts. This paper outlines the automated question generation framework responsible for creating this dataset and provides details on how we integrated human input at various stages to ensure the quality of the questions. Afterwards, using the provided dataset, an evaluation is conducted to assess the capabilities of several LLMs, including GPT-3.5, GPT-4, and Mixtral. The results highlight that these models struggle with complex standards-related questions but exhibit proficiency in addressing general telecom-related inquiries. Additionally, our results showcase how incorporating telecom knowledge context significantly enhances their performance, thus shedding light on the need for a specialized telecom foundation model. Finally, the dataset is shared with active telecom professionals, whose performance is subsequently benchmarked against that of the LLMs. The findings illustrate that LLMs can rival the performance of active professionals in telecom knowledge, thanks to their capacity to process vast amounts of information, underscoring the potential of LLMs within this domain. The dataset has been made publicly accessible on GitHub1.
- Open Access
- 10 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (2) , 2989-3005
Cell-Free Processing for 5G Massive MIMO Deployments: Sleep Modes and User Stream Management
Cell-free massive MIMO (CF-M-MIMO) is heralded as a technological pillar for 6G mobile networks, outperforming the classical cellular-based M-MIMO architecture of 5G by bringing infrastructure closer to end users and leveraging interference cancellation through network-wide joint processing. However, the major drawback is the significant expense required to deploy and maintain numerous infrastructure sites and connect them to a central processing unit (CPU) for baseband processing. This paper proposes utilizing CF-M-MIMO processing atop existing micro/macrocellular deployments in current 5G networks. It suggests connecting all base stations (BSs) to a CPU via fronthaul links to enable joint processing of user equipment (UE) without the costly deployment of many scattered access points (APs). By jointly managing all transmit/receive points (TRPs) in the network, a novel switch on/off strategy is derived to maximize energy efficiency by exploiting the sleep mode capabilities of current baseband/RF hardware, adapting to changes in UE behavior and operator requirements. Additionally, the paper considers multiantenna user terminals, common in current 5G setups, and proposes a technique to manage the number of streams directed to each UE, balancing multiplexing gains and increased pilot contamination.
- 10 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (2) , 2989-3005
Open AccessCell-Free Processing for 5G Massive MIMO Deployments: Sleep Modes and User Stream Management
Cell-free massive MIMO (CF-M-MIMO) is heralded as a technological pillar for 6G mobile networks, outperforming the classical cellular-based M-MIMO architecture of 5G by bringing infrastructure closer to end users and leveraging interference cancellation through network-wide joint processing. However, the major drawback is the significant expense required to deploy and maintain numerous infrastructure sites and connect them to a central processing unit (CPU) for baseband processing. This paper proposes utilizing CF-M-MIMO processing atop existing micro/macrocellular deployments in current 5G networks. It suggests connecting all base stations (BSs) to a CPU via fronthaul links to enable joint processing of user equipment (UE) without the costly deployment of many scattered access points (APs). By jointly managing all transmit/receive points (TRPs) in the network, a novel switch on/off strategy is derived to maximize energy efficiency by exploiting the sleep mode capabilities of current baseband/RF hardware, adapting to changes in UE behavior and operator requirements. Additionally, the paper considers multiantenna user terminals, common in current 5G setups, and proposes a technique to manage the number of streams directed to each UE, balancing multiplexing gains and increased pilot contamination.
- Open Access
- 18 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 5, 4839-4867
Waste Factor and Waste Figure: A Unified Theory for Modeling and Analyzing Wasted Power in Radio Access Networks for Improved Sustainability
This paper introduces Waste Factor (W), also denoted as Waste Figure (WF) in dB, a promising new metric for quantifying energy efficiency in a wide range of circuits and systems applications, including data centers and Radio Access Networks (RANs). Creating and managing 5G and future 6G networks that are energy-efficient is of paramount importance as the wireless industry evolves to become a major consumer of energy. Also, the networks used to connect data centers and artificial intelligence (AI) computing engines with users for machine learning (ML) applications must become more power efficient. This paper illustrates the limitations of existing energy efficiency metrics that inadequately capture the intricate energy dynamics of RAN components. We show here that W provides a generalized analysis of power utilization and energy waste at both the component and system levels for any source-to-sink communication system. We delineate the methodology for applying W across various network configurations, including multiple-input single-output (MISO), single-input and multiple-output (SIMO), and multiple-input multiple-output (MIMO) systems, and demonstrate the effectiveness of W in identifying energy optimization opportunities. Our findings reveal that W not only offers nuanced insights into the energy consumption of RANs but also facilitates informed decision-making for network design and operational efficiency. Furthermore, we show how W can be integrated with other key performance indicators (KPIs) and key value indicators (KVIs) to guide the development of optimal strategies for enhancing network energy efficiency under different operational conditions. Additionally, we present simulation results for a distributed multi-user MIMO (MU-MIMO) system at 3.5, 17, and 28 GHz, demonstrating overall network power efficiency on a per square kilometer basis, and show how overall W decreases (e.g., energy efficiency increases over the entire network) with an increasing number of base stations and increasing carrier frequency. This paper shows that adopting W as a figure of merit (FoM) can enable the design of more sustainable next-generation wireless communication networks, paving the way for greener and more sustainable, energy-efficient 5G and 6G technologies.
- 18 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 5, 4839-4867
Open AccessWaste Factor and Waste Figure: A Unified Theory for Modeling and Analyzing Wasted Power in Radio Access Networks for Improved Sustainability
This paper introduces Waste Factor (W), also denoted as Waste Figure (WF) in dB, a promising new metric for quantifying energy efficiency in a wide range of circuits and systems applications, including data centers and Radio Access Networks (RANs). Creating and managing 5G and future 6G networks that are energy-efficient is of paramount importance as the wireless industry evolves to become a major consumer of energy. Also, the networks used to connect data centers and artificial intelligence (AI) computing engines with users for machine learning (ML) applications must become more power efficient. This paper illustrates the limitations of existing energy efficiency metrics that inadequately capture the intricate energy dynamics of RAN components. We show here that W provides a generalized analysis of power utilization and energy waste at both the component and system levels for any source-to-sink communication system. We delineate the methodology for applying W across various network configurations, including multiple-input single-output (MISO), single-input and multiple-output (SIMO), and multiple-input multiple-output (MIMO) systems, and demonstrate the effectiveness of W in identifying energy optimization opportunities. Our findings reveal that W not only offers nuanced insights into the energy consumption of RANs but also facilitates informed decision-making for network design and operational efficiency. Furthermore, we show how W can be integrated with other key performance indicators (KPIs) and key value indicators (KVIs) to guide the development of optimal strategies for enhancing network energy efficiency under different operational conditions. Additionally, we present simulation results for a distributed multi-user MIMO (MU-MIMO) system at 3.5, 17, and 28 GHz, demonstrating overall network power efficiency on a per square kilometer basis, and show how overall W decreases (e.g., energy efficiency increases over the entire network) with an increasing number of base stations and increasing carrier frequency. This paper shows that adopting W as a figure of merit (FoM) can enable the design of more sustainable next-generation wireless communication networks, paving the way for greener and more sustainable, energy-efficient 5G and 6G technologies.
- Open Access
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Machine Learning in Communications and Networking
- Vol. 2, 780-804
Data-Driven Energy Efficiency Modeling in Large-Scale Networks: An Expert Knowledge and ML-Based Approach
The energy consumption of mobile networks poses a critical challenge. Mitigating this concern necessitates the deployment and optimization of network energy-saving solutions, such as carrier shutdown, to dynamically manage network resources. Traditional optimization approaches encounter complexity due to factors like the large number of cells, stochastic traffic, channel variations, and intricate trade-offs. This paper introduces the simulated reality of communication networks (SRCON) framework, a novel, data-driven modeling paradigm that harnesses live network data and employs a blend of machine learning (ML)- and expert-based models. These mix of models accurately characterizes the functioning of network components, and predicts network energy efficiency and user equipment (UE) quality of service for any energy carrier shutdown configuration in a specific network. Distinguishing itself from existing methods, SRCON eliminates the reliance on expensive expert knowledge, drive testing, or incomplete maps for predicting network performance. This paper details the pipeline employed by SRCON to decompose the large network energy efficiency modeling problem into ML- and expert-based submodels. It demonstrates how, by embracing stochasticity, and carefully crafting the relationship between such submodels, the overall computational complexity can be reduced and prediction accuracy enhanced. Results derived from real network data underscore the paradigm shift introduced by SRCON, showcasing significant gains over a state-of-the-art method used by a operator for network energy efficiency modeling. The reliability of this local, data-driven modeling of the network proves to be a key asset for network energy-saving optimization.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Machine Learning in Communications and Networking
- Vol. 2, 780-804
Open AccessData-Driven Energy Efficiency Modeling in Large-Scale Networks: An Expert Knowledge and ML-Based Approach
The energy consumption of mobile networks poses a critical challenge. Mitigating this concern necessitates the deployment and optimization of network energy-saving solutions, such as carrier shutdown, to dynamically manage network resources. Traditional optimization approaches encounter complexity due to factors like the large number of cells, stochastic traffic, channel variations, and intricate trade-offs. This paper introduces the simulated reality of communication networks (SRCON) framework, a novel, data-driven modeling paradigm that harnesses live network data and employs a blend of machine learning (ML)- and expert-based models. These mix of models accurately characterizes the functioning of network components, and predicts network energy efficiency and user equipment (UE) quality of service for any energy carrier shutdown configuration in a specific network. Distinguishing itself from existing methods, SRCON eliminates the reliance on expensive expert knowledge, drive testing, or incomplete maps for predicting network performance. This paper details the pipeline employed by SRCON to decompose the large network energy efficiency modeling problem into ML- and expert-based submodels. It demonstrates how, by embracing stochasticity, and carefully crafting the relationship between such submodels, the overall computational complexity can be reduced and prediction accuracy enhanced. Results derived from real network data underscore the paradigm shift introduced by SRCON, showcasing significant gains over a state-of-the-art method used by a operator for network energy efficiency modeling. The reliability of this local, data-driven modeling of the network proves to be a key asset for network energy-saving optimization.
- Open Access
Linguistic Intelligence in Large Language Models for Telecommunications
Large Language Models (LLMs) have emerged as a significant advancement in the field of Natural Language Processing (NLP), demonstrating remarkable capabilities in language generation and other language-centric tasks. Despite their evaluation across a multitude of analytical and reasoning tasks in various scientific domains, a comprehensive exploration of their knowledge and understanding within the realm of natural language tasks in the telecommunications domain is still needed. This study, therefore, seeks to evaluate the knowledge and understanding capabilities of LLMs within this domain. To achieve this, we conduct an exhaustive zero-shot evaluation of four prominent LLMs-Llama-2, Falcon, Mistral, and Zephyr. These models require fewer resources than ChatGPT, making them suitable for resource-constrained environments. Their performance is compared with state-of-the-art, fine-tuned models. To the best of our knowledge, this is the first work to extensively evaluate and compare the understanding of LLMs across multiple language-centric tasks in this domain. Our evaluation reveals that zero-shot LLMs can achieve performance levels comparable to the current state-of-the-art fine-tuned models. This indicates that pretraining on extensive text corpora equips LLMs with a degree of specialization, even within the telecommunications domain. We also observe that no single LLM consistently outperforms others, and the performance of different LLMs can fluctuate. Although their performance lags behind fine-tuned models, our findings underscore the potential of LLMs as a valuable resource for understanding various aspects of this field that lack large annotated data.
Open AccessLinguistic Intelligence in Large Language Models for Telecommunications
Large Language Models (LLMs) have emerged as a significant advancement in the field of Natural Language Processing (NLP), demonstrating remarkable capabilities in language generation and other language-centric tasks. Despite their evaluation across a multitude of analytical and reasoning tasks in various scientific domains, a comprehensive exploration of their knowledge and understanding within the realm of natural language tasks in the telecommunications domain is still needed. This study, therefore, seeks to evaluate the knowledge and understanding capabilities of LLMs within this domain. To achieve this, we conduct an exhaustive zero-shot evaluation of four prominent LLMs-Llama-2, Falcon, Mistral, and Zephyr. These models require fewer resources than ChatGPT, making them suitable for resource-constrained environments. Their performance is compared with state-of-the-art, fine-tuned models. To the best of our knowledge, this is the first work to extensively evaluate and compare the understanding of LLMs across multiple language-centric tasks in this domain. Our evaluation reveals that zero-shot LLMs can achieve performance levels comparable to the current state-of-the-art fine-tuned models. This indicates that pretraining on extensive text corpora equips LLMs with a degree of specialization, even within the telecommunications domain. We also observe that no single LLM consistently outperforms others, and the performance of different LLMs can fluctuate. Although their performance lags behind fine-tuned models, our findings underscore the potential of LLMs as a valuable resource for understanding various aspects of this field that lack large annotated data.
- Open Access
FlexTrain: A Dynamic Training Framework for Heterogeneous Devices Environments
As deep learning models become increasingly large, they pose significant challenges in heterogeneous devices environments. The size of deep learning models makes it difficult to deploy them on low-power or resource-constrained devices, leading to long inference times and high energy consumption. To address these challenges, we propose FlexTrain, a framework that accommodates the diverse storage and computational resources available on different devices during the training phase. FlexTrain enables efficient deployment of deep learning models, while respecting device constraints, minimizing communication costs, and ensuring seamless integration with diverse devices. We demonstrate the effectiveness of FlexTrain on the CIFAR-100 dataset, where a single global model trained with FlexTrain can be easily deployed on heterogeneous devices, saving training time and energy consumption. We also extend FlexTrain to the federated learning setting, showing that our approach outperforms standard federated learning benchmarks on both CIFAR-10 and CIFAR-100 datasets.
Open AccessFlexTrain: A Dynamic Training Framework for Heterogeneous Devices Environments
As deep learning models become increasingly large, they pose significant challenges in heterogeneous devices environments. The size of deep learning models makes it difficult to deploy them on low-power or resource-constrained devices, leading to long inference times and high energy consumption. To address these challenges, we propose FlexTrain, a framework that accommodates the diverse storage and computational resources available on different devices during the training phase. FlexTrain enables efficient deployment of deep learning models, while respecting device constraints, minimizing communication costs, and ensuring seamless integration with diverse devices. We demonstrate the effectiveness of FlexTrain on the CIFAR-100 dataset, where a single global model trained with FlexTrain can be easily deployed on heterogeneous devices, saving training time and energy consumption. We also extend FlexTrain to the federated learning setting, showing that our approach outperforms standard federated learning benchmarks on both CIFAR-10 and CIFAR-100 datasets.