Stefanos Bakirtzis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Stefanos Bakirtzis has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Jie Zhang
Chongqing University of Posts and Telecommunications
10 shared publications - Ian J. Wassell
Department of Computer Science and Technology, University of Cambridge, Cambridge, U.K.
9 shared publications - Kehai Qiu
Department of Computer Science and Technology, University of Cambridge, Cambridge, U.K.
5 shared publications - Marco Fiore
IMDEA Networks Institute, Madrid, Spain
4 shared publications Hui Song
Ranplan Wireless Network Design Ltd., Cambridge, U.K.
3 shared publications- André Felipe Zanella
IMDEA Networks Institute, Madrid, Spain
2 shared publications
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25 (9) , 3656-3660
Deep Learning-Based Indoor Path Loss Prediction Model With Spatial Attention
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 25 (9) , 3656-3660
Deep Learning-Based Indoor Path Loss Prediction Model With Spatial Attention
- 9 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Multiscale and Multiphysics Computational Techniques
- Vol. 10, 58-68
Rigorous Indoor Wireless Communication System Simulations With Deep Learning-Based Radio Propagation Models
Recently, there has been a surge in the development of data-driven propagation models. These models aspire to distill knowledge from propagation solvers or measured data and eventually become capable of predicting characteristics related to radiowave propagation. In this paper, we present the functionality of a generalizable and robust data-driven propagation model that enables efficient and reliable simulations of indoor wireless communication systems (IWCSs). In particular, we modify our previously introduced model, EM DeepRay, to consider the impact of antenna directivity, and we present a training and inference strategy that allows the simulation of large-scale and complicated IWCSs. Our data-driven model is trained over a rich data set comprising diverse building geometries, frequency bands, and antenna radiation patterns. Benchmarking its performance with that of a ray-tracer in complicated IWCSs with real-world measured data yields similar results that have a distinct advantage in terms of computational time. Ultimately, our work paves the way for replacing legacy IWCSs simulators, with high-fidelity artificial intelligence-based models.
- 9 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Multiscale and Multiphysics Computational Techniques
- Vol. 10, 58-68
Rigorous Indoor Wireless Communication System Simulations With Deep Learning-Based Radio Propagation Models
Recently, there has been a surge in the development of data-driven propagation models. These models aspire to distill knowledge from propagation solvers or measured data and eventually become capable of predicting characteristics related to radiowave propagation. In this paper, we present the functionality of a generalizable and robust data-driven propagation model that enables efficient and reliable simulations of indoor wireless communication systems (IWCSs). In particular, we modify our previously introduced model, EM DeepRay, to consider the impact of antenna directivity, and we present a training and inference strategy that allows the simulation of large-scale and complicated IWCSs. Our data-driven model is trained over a rich data set comprising diverse building geometries, frequency bands, and antenna radiation patterns. Benchmarking its performance with that of a ray-tracer in complicated IWCSs with real-world measured data yields similar results that have a distinct advantage in terms of computational time. Ultimately, our work paves the way for replacing legacy IWCSs simulators, with high-fidelity artificial intelligence-based models.
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 13 (12) , 3340-3344
Large Language Model-Based Wireless Network Design
In this paper, we present the Large Language Model-based combinatorial optimizer (LMCO) for wireless network optimization and planning tasks, focusing on optimizing the number and the placement of wireless access point placement. The performance and efficiency of LMCO are evaluated and compared with the well-established Ant Colony Optimization (ACO) algorithm. The results indicate that LMCO exhibits superior performance, particularly as the complexity of the problem increases. These findings also underscore the significant potential of LLM-based algorithms in revolutionizing combinatorial optimization across a wide range of applications.
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 13 (12) , 3340-3344
Large Language Model-Based Wireless Network Design
In this paper, we present the Large Language Model-based combinatorial optimizer (LMCO) for wireless network optimization and planning tasks, focusing on optimizing the number and the placement of wireless access point placement. The performance and efficiency of LMCO are evaluated and compared with the well-established Ant Colony Optimization (ACO) algorithm. The results indicate that LMCO exhibits superior performance, particularly as the complexity of the problem increases. These findings also underscore the significant potential of LLM-based algorithms in revolutionizing combinatorial optimization across a wide range of applications.
- 22 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 42 (11) , 3232-3245
A Joint Optimization Approach for Power-Efficient Heterogeneous OFDMA Radio Access Networks
Heterogeneous networks have emerged as a popular solution for accommodating the growing number of connected devices and increasing traffic demands in cellular networks. While offering broader coverage, higher capacity, and lower latency, the escalating energy consumption poses sustainability challenges. In this paper a novel optimization approach for orthogonal heterogeneous networks is proposed to minimize transmission power while respecting individual users’ throughput constraints. The problem is formulated as a mixed integer geometric program, and optimizes at once multiple system variables such as user association, working bandwidth, and base stations transmission powers. Crucially, the proposed approach becomes a convex optimization problem when user-base station associations are provided. Evaluations in multiple realistic scenarios from the production mobile network of a major European operator and based on precise channel gains and throughput requirements from measured data validate the effectiveness of the proposed approach. Overall, our original solution paves the road for greener connectivity by reducing the energy footprint of heterogeneous mobile networks, hence fostering more sustainable communication systems.
- 22 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 42 (11) , 3232-3245
A Joint Optimization Approach for Power-Efficient Heterogeneous OFDMA Radio Access Networks
Heterogeneous networks have emerged as a popular solution for accommodating the growing number of connected devices and increasing traffic demands in cellular networks. While offering broader coverage, higher capacity, and lower latency, the escalating energy consumption poses sustainability challenges. In this paper a novel optimization approach for orthogonal heterogeneous networks is proposed to minimize transmission power while respecting individual users’ throughput constraints. The problem is formulated as a mixed integer geometric program, and optimizes at once multiple system variables such as user association, working bandwidth, and base stations transmission powers. Crucially, the proposed approach becomes a convex optimization problem when user-base station associations are provided. Evaluations in multiple realistic scenarios from the production mobile network of a major European operator and based on precise channel gains and throughput requirements from measured data validate the effectiveness of the proposed approach. Overall, our original solution paves the road for greener connectivity by reducing the energy footprint of heterogeneous mobile networks, hence fostering more sustainable communication systems.
- 6 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (6) , 451-458
AI-Assisted Indoor Wireless Network Planning With Data-Driven Propagation Models
Propelled by rapid advances in artificial intelligence (AI), the design and operation of 5G and beyond networks are anticipated to be radically different from those of legacy communication systems. Indeed, AI can be exploited to automate and optimize various essential functionalities of the wireless ecosystem, such as resource allocation, channel modeling, or network planning. This article explores how AI-driven propagation models can be leveraged for the automated and expedient deployment of small cells in indoor environments. To this end, we couple a generalizable data-driven propagation model with an AI-based optimizer to determine the optimal network topology with respect to a target key performance indicator. Our approach reduces the computational time of indoor wireless network design by two to three orders of magnitude, thus enabling accurate planning that would be extremely expensive to conduct using conventional indoor propagation tools and yielding significant gains in the resulting indoor planning quality and performance.
- 6 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (6) , 451-458
AI-Assisted Indoor Wireless Network Planning With Data-Driven Propagation Models
Propelled by rapid advances in artificial intelligence (AI), the design and operation of 5G and beyond networks are anticipated to be radically different from those of legacy communication systems. Indeed, AI can be exploited to automate and optimize various essential functionalities of the wireless ecosystem, such as resource allocation, channel modeling, or network planning. This article explores how AI-driven propagation models can be leveraged for the automated and expedient deployment of small cells in indoor environments. To this end, we couple a generalizable data-driven propagation model with an AI-based optimizer to determine the optimal network topology with respect to a target key performance indicator. Our approach reduces the computational time of indoor wireless network design by two to three orders of magnitude, thus enabling accurate planning that would be extremely expensive to conduct using conventional indoor propagation tools and yielding significant gains in the resulting indoor planning quality and performance.
- Open Access
- 24 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 645-659
Characterizing Mobile Service Demands at Indoor Cellular Networks
- 24 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 645-659
Open AccessCharacterizing Mobile Service Demands at Indoor Cellular Networks
- Open Access
- 2 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
An Expedient, Generalizable and Realistic Data-Driven Indoor Propagation Model
Efficient and realistic indoor radio propagation modelling tools are inextricably intertwined with the design and operation of next generation wireless networks. Machine learning (ML)-based radio propagation models can be trained with simulated or real-world data to provide accurate estimates of the wireless channel characteristics in a computationally efficient way. However, most of the existing research on ML-based propagation models focuses on outdoor propagation modelling, while indoor data-driven propagation models remain site-specific with limited scalability. In this paper we present an efficient and credible ML-based radio propagation modelling framework for indoor environments. Specifically, we demonstrate how a convolutional encoder-decoder can be trained to replicate the results of a ray-tracer, by encoding physics-based information of an indoor environment, such as the permittivity of the walls, and decode it as the path-loss (PL) heatmap for an environment of interest. Our model is trained over multiple indoor geometries and frequency bands, and it can eventually predict the PL for unknown indoor geometries and frequency bands within a few milliseconds. Additionally, we illustrate how the concept of transfer learning can be leveraged to calibrate our model by adjusting its pre-estimate weights, allowing it to make predictions that are consistent with measurement data.
- 2 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessAn Expedient, Generalizable and Realistic Data-Driven Indoor Propagation Model
Efficient and realistic indoor radio propagation modelling tools are inextricably intertwined with the design and operation of next generation wireless networks. Machine learning (ML)-based radio propagation models can be trained with simulated or real-world data to provide accurate estimates of the wireless channel characteristics in a computationally efficient way. However, most of the existing research on ML-based propagation models focuses on outdoor propagation modelling, while indoor data-driven propagation models remain site-specific with limited scalability. In this paper we present an efficient and credible ML-based radio propagation modelling framework for indoor environments. Specifically, we demonstrate how a convolutional encoder-decoder can be trained to replicate the results of a ray-tracer, by encoding physics-based information of an indoor environment, such as the permittivity of the walls, and decode it as the path-loss (PL) heatmap for an environment of interest. Our model is trained over multiple indoor geometries and frequency bands, and it can eventually predict the PL for unknown indoor geometries and frequency bands within a few milliseconds. Additionally, we illustrate how the concept of transfer learning can be leveraged to calibrate our model by adjusting its pre-estimate weights, allowing it to make predictions that are consistent with measurement data.
- Open Access
- 28 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
AI-assisted Indoor Wireless Network Planning with Data-Driven Propagation Models
Propelled by rapid advances in artificial intelligence (AI), the design and operation of 5G and beyond networks are anticipated to be radically different from those of legacy communication systems. Indeed, AI can be exploited to automate and optimize various functionalities that are essential for the wireless ecosystem, such as resource allocation, channel modeling, or network planning. This article explores how data-driven propagation models can be leveraged for the automated and expedient deployment of small cells in indoor environments. In particular, a generalizable data-driven propagation model is entwined with AI-based optimizers aiming at determining the optimal network topology with respect to some target key performance indicators. The data-driven model combines the accuracy of a high-performance propagation solver with the computational efficiency of deep neural networks, inferring the signal level spatial distribution based on physics-based information of the indoor geometry. The capability of the data-driven model to conduct multiple simulations within a few seconds is vital for the optimization procedure, enabling accurate network planning that would be extremely expensive to produce with a conventional indoor propagation tool, especially for ultra-dense networks where macro-base stations coexist with numerous small cells.
- 28 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessAI-assisted Indoor Wireless Network Planning with Data-Driven Propagation Models
Propelled by rapid advances in artificial intelligence (AI), the design and operation of 5G and beyond networks are anticipated to be radically different from those of legacy communication systems. Indeed, AI can be exploited to automate and optimize various functionalities that are essential for the wireless ecosystem, such as resource allocation, channel modeling, or network planning. This article explores how data-driven propagation models can be leveraged for the automated and expedient deployment of small cells in indoor environments. In particular, a generalizable data-driven propagation model is entwined with AI-based optimizers aiming at determining the optimal network topology with respect to some target key performance indicators. The data-driven model combines the accuracy of a high-performance propagation solver with the computational efficiency of deep neural networks, inferring the signal level spatial distribution based on physics-based information of the indoor geometry. The capability of the data-driven model to conduct multiple simulations within a few seconds is vital for the optimization procedure, enabling accurate network planning that would be extremely expensive to produce with a conventional indoor propagation tool, especially for ultra-dense networks where macro-base stations coexist with numerous small cells.
- 13 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 9 (23) , 24529-24540
Deep-Learning-Based Multivariate Time-Series Classification for Indoor/Outdoor Detection
Recently, the topic of indoor outdoor detection (IOD) has seen its popularity increase, as IOD models can be leveraged to augment the performance of numerous Internet of Things and other applications. IOD aims at distinguishing in an efficient manner whether a user resides in an indoor or an outdoor environment, by inspecting the cellular phone sensor recordings. Legacy IOD models attempt to determine a user’s environment by comparing the sensor measurements to some threshold values. However, as we also observe in our experiments, such models exhibit limited scalability, and their accuracy can be poor. Machine learning (ML)-based IOD models aim at removing this limitation, by utilizing a large volume of measurements to train ML algorithms to classify a user’s environment. Yet, in most of the existing research, the temporal dimension of the problem is disregarded. In this article, we propose treating IOD as a multivariate time-series classification (TSC) problem, and we explore the performance of various deep learning (DL) models. We demonstrate that a multivariate TSC approach can be used to monitor a user’s environment, and predict changes in its state, with greater accuracy compared to conventional approaches that ignore the feature variation over time. Additionally, we introduce a new DL model for multivariate TSC, exploiting the concept of self-attention and atrous spatial pyramid pooling. The proposed DL multivariate TSC framework exploits only low power consumption sensors to infer a user’s environment, and it outperforms state-of-the-art models, yielding a higher accuracy combined with a smaller computational cost.
- 13 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 9 (23) , 24529-24540
Deep-Learning-Based Multivariate Time-Series Classification for Indoor/Outdoor Detection
Recently, the topic of indoor outdoor detection (IOD) has seen its popularity increase, as IOD models can be leveraged to augment the performance of numerous Internet of Things and other applications. IOD aims at distinguishing in an efficient manner whether a user resides in an indoor or an outdoor environment, by inspecting the cellular phone sensor recordings. Legacy IOD models attempt to determine a user’s environment by comparing the sensor measurements to some threshold values. However, as we also observe in our experiments, such models exhibit limited scalability, and their accuracy can be poor. Machine learning (ML)-based IOD models aim at removing this limitation, by utilizing a large volume of measurements to train ML algorithms to classify a user’s environment. Yet, in most of the existing research, the temporal dimension of the problem is disregarded. In this article, we propose treating IOD as a multivariate time-series classification (TSC) problem, and we explore the performance of various deep learning (DL) models. We demonstrate that a multivariate TSC approach can be used to monitor a user’s environment, and predict changes in its state, with greater accuracy compared to conventional approaches that ignore the feature variation over time. Additionally, we introduce a new DL model for multivariate TSC, exploiting the concept of self-attention and atrous spatial pyramid pooling. The proposed DL multivariate TSC framework exploits only low power consumption sensors to infer a user’s environment, and it outperforms state-of-the-art models, yielding a higher accuracy combined with a smaller computational cost.