Albert Cabellos-Aparicio
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Albert Cabellos-Aparicio has a total of 119 co-authors. Here are the top co-authors by number of shared publications:
- Sergi Abadal
Universitat Politecnica de Catalunya, Barcelona, Spain
14 shared publications - Eduard Alarcón
Universitat Politecnica de Catalunya, Barcelona, Spain
12 shared publications - Pere Barlet-Ros
Universitat Politecnica de Catalunya, Barcelona, Spain
12 shared publications - Hamidreza Taghvaee
University of Surrey, Guildford
5 shared publications - Jordi Paillisse
Barcelona Neural Networking Center, Universitat Politcnica de Catalunya, Barcelona, Catalonia, Spain
5 shared publications - José Suárez-Varela
Telefónica Research, Madrid, Spain
5 shared publications
Latest Publications
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 2060-2066
Towards Traffic Modelling of Multi-Agent Systems: The Role of Coordination Topology
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 2060-2066
Open AccessTowards Traffic Modelling of Multi-Agent Systems: The Role of Coordination Topology
- 25 March 2026
- Published by Elsevier in Computer Networks
- Vol. 282
From simulation to deep learning: Survey on network performance modeling approaches
- 25 March 2026
- Published by Elsevier in Computer Networks
- Vol. 282
From simulation to deep learning: Survey on network performance modeling approaches
- Open Access
- 9 October 2025
- Published by Elsevier in Computer Networks
- Vol. 273
BGP anomaly detection using the raw internet topology
- 9 October 2025
- Published by Elsevier in Computer Networks
- Vol. 273
Open AccessBGP anomaly detection using the raw internet topology
- Open Access
- 9 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 14-19
Wavelet-Enhanced Graph Neural Networks: Towards Non-Parametric Network Traffic Modeling
Network modeling is crucial for the design, management, and optimization of modern telecommunications and data networks. Recent advancements in Machine Learning (ML), specifically Graph Neural Networks (GNNs), offer promising solutions but rely on parameterized traffic representations, necessitating retraining for new traffic patterns. This paper explores integrating the Discrete Wavelet Transform (DWT) with GNNs to enhance network traffic modeling. By leveraging wavelets, which decompose signals into both time and frequency components, we aim to encode traffic patterns without assuming specific distributions, improving model adaptability and accuracy. We modify the state-of-the-art RouteNet-Fermi model to incorporate wavelet-based traffic encoding and evaluate its performance across different synthetic and real traffic scenarios. Our findings show that wavelet-based encoding handles unseen traffic distributions with minimal impact on performance, unlike traditional parameter-based approaches. This work represents a step forward in bridging the gap between non-parametric traffic representation and advanced network modeling, offering a promising solution for dynamic and complex network environments.
- 9 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 14-19
Open AccessWavelet-Enhanced Graph Neural Networks: Towards Non-Parametric Network Traffic Modeling
Network modeling is crucial for the design, management, and optimization of modern telecommunications and data networks. Recent advancements in Machine Learning (ML), specifically Graph Neural Networks (GNNs), offer promising solutions but rely on parameterized traffic representations, necessitating retraining for new traffic patterns. This paper explores integrating the Discrete Wavelet Transform (DWT) with GNNs to enhance network traffic modeling. By leveraging wavelets, which decompose signals into both time and frequency components, we aim to encode traffic patterns without assuming specific distributions, improving model adaptability and accuracy. We modify the state-of-the-art RouteNet-Fermi model to incorporate wavelet-based traffic encoding and evaluate its performance across different synthetic and real traffic scenarios. Our findings show that wavelet-based encoding handles unseen traffic distributions with minimal impact on performance, unlike traditional parameter-based approaches. This work represents a step forward in bridging the gap between non-parametric traffic representation and advanced network modeling, offering a promising solution for dynamic and complex network environments.
- Open Access
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 25-30
Detecting Contextual Network Anomalies with Graph Neural Networks
Detecting anomalies on network traffic is a complex task due to the massive amount of traffic flows in today's networks, as well as the highly-dynamic nature of traffic over time. In this paper, we propose the use of Graph Neural Networks (GNN) for network traffic anomaly detection. We formulate the problem as contextual anomaly detection on network traffic measurements, and propose a custom GNN-based solution that detects traffic anomalies on origin-destination flows. In our evaluation, we use real-world data from Abilene (6 months), and make a comparison with other widely used methods for the same task (PCA, EWMA, RNN). The results show that the anomalies detected by our solution are quite complementary to those captured by the baselines (with a max. of 36.33% overlapping anomalies for PCA). Moreover, we manually inspect the anomalies detected by our method, and find that a large portion of them can be visually validated by a network expert (64% with high confidence, 18% with mid confidence, 18% normal traffic). Lastly, we analyze the characteristics of the anomalies through two paradigmatic cases that are quite representative of the bulk of anomalies.
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 25-30
Open AccessDetecting Contextual Network Anomalies with Graph Neural Networks
Detecting anomalies on network traffic is a complex task due to the massive amount of traffic flows in today's networks, as well as the highly-dynamic nature of traffic over time. In this paper, we propose the use of Graph Neural Networks (GNN) for network traffic anomaly detection. We formulate the problem as contextual anomaly detection on network traffic measurements, and propose a custom GNN-based solution that detects traffic anomalies on origin-destination flows. In our evaluation, we use real-world data from Abilene (6 months), and make a comparison with other widely used methods for the same task (PCA, EWMA, RNN). The results show that the anomalies detected by our solution are quite complementary to those captured by the baselines (with a max. of 36.33% overlapping anomalies for PCA). Moreover, we manually inspect the anomalies detected by our method, and find that a large portion of them can be visually validated by a network expert (64% with high confidence, 18% with mid confidence, 18% normal traffic). Lastly, we analyze the characteristics of the anomalies through two paradigmatic cases that are quite representative of the bulk of anomalies.
- Open Access
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Atom: Neural Traffic Compression with Spatio-Temporal Graph Neural Networks
Storing network traffic data is key to efficient network management; however, it is becoming more challenging and costly due to the ever-increasing data transmission rates, traffic volumes, and connected devices. In this paper, we explore the use of neural architectures for network traffic compression. Specifically, we consider a network scenario with multiple measurement points in a network topology. Such measurements can be interpreted as multiple time series that exhibit spatial and temporal correlations induced by network topology, routing, or user behavior. We present Atom, a neural traffic compression method that leverages spatial and temporal correlations present in network traffic. Atom implements a customized spatio-temporal graph neural network design that effectively exploits both types of correlations simultaneously. The experimental results show that Atom can outperform GZIP's compression ratios by 50%--65% on three real-world networks.
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessAtom: Neural Traffic Compression with Spatio-Temporal Graph Neural Networks
Storing network traffic data is key to efficient network management; however, it is becoming more challenging and costly due to the ever-increasing data transmission rates, traffic volumes, and connected devices. In this paper, we explore the use of neural architectures for network traffic compression. Specifically, we consider a network scenario with multiple measurement points in a network topology. Such measurements can be interpreted as multiple time series that exhibit spatial and temporal correlations induced by network topology, routing, or user behavior. We present Atom, a neural traffic compression method that leverages spatial and temporal correlations present in network traffic. Atom implements a customized spatio-temporal graph neural network design that effectively exploits both types of correlations simultaneously. The experimental results show that Atom can outperform GZIP's compression ratios by 50%--65% on three real-world networks.
- Open Access
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 7-12
Topological Network Traffic Compression
The surge in computer network traffic, fueled by emerging applications and technology advancements, is creating a pressing need for efficient techniques to store and analyze massive traffic traces. This paper introduces a novel approach to address this challenge by employing Topological Deep Learning (TDL) for lossy traffic compression. Unlike Graph Neural Networks (GNNs), which rely on the binary interactions and local neighborhoods defined by graph representations, TDL methods can naturally accommodate higher-order relations among arbitrary network elements, thus offering more expressive representations beyond the graph domain. In particular, our proposed traffic compression framework aims to detect higher-order correlated structures within traffic data and employs Topological Neural Networks to generate compressed representations within these sets. This work assesses if this approach can outperform conventional Machine Learning (ML) compression architectures by better exploiting multi-datapoint interactions, potentially capturing correlations between distant network elements. We evaluate our method on two real-world networking datasets, comparing it against GNN-based architectures and a Multi-Layer Perceptron autoencoder designed for this task. The results demonstrate significant improvements w.r.t. ML baselines --from 30% up to 90% better reconstruction errors across all scenarios--, establishing our topological framework as a strong baseline for lossy neural traffic compression.
- 5 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 7-12
Open AccessTopological Network Traffic Compression
The surge in computer network traffic, fueled by emerging applications and technology advancements, is creating a pressing need for efficient techniques to store and analyze massive traffic traces. This paper introduces a novel approach to address this challenge by employing Topological Deep Learning (TDL) for lossy traffic compression. Unlike Graph Neural Networks (GNNs), which rely on the binary interactions and local neighborhoods defined by graph representations, TDL methods can naturally accommodate higher-order relations among arbitrary network elements, thus offering more expressive representations beyond the graph domain. In particular, our proposed traffic compression framework aims to detect higher-order correlated structures within traffic data and employs Topological Neural Networks to generate compressed representations within these sets. This work assesses if this approach can outperform conventional Machine Learning (ML) compression architectures by better exploiting multi-datapoint interactions, potentially capturing correlations between distant network elements. We evaluate our method on two real-world networking datasets, comparing it against GNN-based architectures and a Multi-Layer Perceptron autoencoder designed for this task. The results demonstrate significant improvements w.r.t. ML baselines --from 30% up to 90% better reconstruction errors across all scenarios--, establishing our topological framework as a strong baseline for lossy neural traffic compression.
- Open Access
- 10 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 14-20
Enhancing 5G Radio Planning with Graph Representations and Deep Learning
The roll out of new mobile network generations poses hard challenges due to various factors such as cost-benefit tradeoffs, existing infrastructure, and new technology aspects. In particular, one of the main challenges for the 5G deployment lies in optimal 5G radio coverage while accounting for diverse service performance metrics. This paper introduces a Deep Learning-based approach to assist in 5G radio planning by utilizing data from previous-generation cells. Our solution relies on a custom graph representation to leverage the information available from existing cells, and employs a Graph Neural Network (GNN) model to process such data efficiently. In our evaluation, we test its potential to model the transition from 4G to 5G NSA using real-world data from a UK mobile network operator. The experimental results show that our solution achieves high accuracy in predicting key performance indicators in new 5G cells, with a Mean Absolute Percentage Error (MAPE) <17% when evaluated on samples from the same area where it was trained. Moreover, we test its generalization capability over various geographical areas not included in the training, achieving a MAPE <19%. This suggests beneficial properties for achieving robust solutions applicable to 5G planning in new areas without the need of retraining.
- 10 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 14-20
Open AccessEnhancing 5G Radio Planning with Graph Representations and Deep Learning
The roll out of new mobile network generations poses hard challenges due to various factors such as cost-benefit tradeoffs, existing infrastructure, and new technology aspects. In particular, one of the main challenges for the 5G deployment lies in optimal 5G radio coverage while accounting for diverse service performance metrics. This paper introduces a Deep Learning-based approach to assist in 5G radio planning by utilizing data from previous-generation cells. Our solution relies on a custom graph representation to leverage the information available from existing cells, and employs a Graph Neural Network (GNN) model to process such data efficiently. In our evaluation, we test its potential to model the transition from 4G to 5G NSA using real-world data from a UK mobile network operator. The experimental results show that our solution achieves high accuracy in predicting key performance indicators in new 5G cells, with a Mean Absolute Percentage Error (MAPE) <17% when evaluated on samples from the same area where it was trained. Moreover, we test its generalization capability over various geographical areas not included in the training, achieving a MAPE <19%. This suggests beneficial properties for achieving robust solutions applicable to 5G planning in new areas without the need of retraining.
- Open Access
- 29 April 2023
- Published by Springer Nature in Scientometrics
- Vol. 128 (6) , 3313-3335
Bayesian inference of spatial and temporal relations in AI patents for EU countries
In the paper, we propose two models of Artificial Intelligence (AI) patents in European Union (EU) countries addressing spatial and temporal behaviour. In particular, the models can quantitatively describe the interaction between countries or explain the rapidly growing trends in AI patents. For spatial analysis Poisson regression is used to explain collaboration between a pair of countries measured by the number of common patents. Through Bayesian inference, we estimated the strengths of interactions between countries in the EU and the rest of the world. In particular, a significant lack of cooperation has been identified for some pairs of countries. Alternatively, an inhomogeneous Poisson process combined with the logistic curve growth accurately models the temporal behaviour by an accurate trend line. Bayesian analysis in the time domain revealed an upcoming slowdown in patenting intensity.
- 29 April 2023
- Published by Springer Nature in Scientometrics
- Vol. 128 (6) , 3313-3335
Open AccessBayesian inference of spatial and temporal relations in AI patents for EU countries
In the paper, we propose two models of Artificial Intelligence (AI) patents in European Union (EU) countries addressing spatial and temporal behaviour. In particular, the models can quantitatively describe the interaction between countries or explain the rapidly growing trends in AI patents. For spatial analysis Poisson regression is used to explain collaboration between a pair of countries measured by the number of common patents. Through Bayesian inference, we estimated the strengths of interactions between countries in the EU and the rest of the world. In particular, a significant lack of cooperation has been identified for some pairs of countries. Alternatively, an inhomogeneous Poisson process combined with the logistic curve growth accurately models the temporal behaviour by an accurate trend line. Bayesian analysis in the time domain revealed an upcoming slowdown in patenting intensity.
- Open Access
Proximal Policy Optimization with Graph Neural Networks for Optimal Power Flow
Optimal Power Flow (OPF) is a very traditional research area within the power systems field that seeks for the optimal operation point of electric power plants, and which needs to be solved every few minutes in real-world scenarios. However, due to the nonconvexities that arise in power generation systems, there is not yet a fast, robust solution technique for the full Alternating Current Optimal Power Flow (ACOPF). In the last decades, power grids have evolved into a typical dynamic, non-linear and large-scale control system, known as the power system, so searching for better and faster ACOPF solutions is becoming crucial. Appearance of Graph Neural Networks (GNN) has allowed the natural use of Machine Learning (ML) algorithms on graph data, such as power networks. On the other hand, Deep Reinforcement Learning (DRL) is known for its powerful capability to solve complex decision-making problems. Although solutions that use these two methods separately are beginning to appear in the literature, none has yet combined the advantages of both. We propose a novel architecture based on the Proximal Policy Optimization algorithm with Graph Neural Networks to solve the Optimal Power Flow. The objective is to design an architecture that learns how to solve the optimization problem and that is at the same time able to generalize to unseen scenarios. We compare our solution with the DCOPF in terms of cost after having trained our DRL agent on IEEE 30 bus system and then computing the OPF on that base network with topology changes
Open AccessProximal Policy Optimization with Graph Neural Networks for Optimal Power Flow
Optimal Power Flow (OPF) is a very traditional research area within the power systems field that seeks for the optimal operation point of electric power plants, and which needs to be solved every few minutes in real-world scenarios. However, due to the nonconvexities that arise in power generation systems, there is not yet a fast, robust solution technique for the full Alternating Current Optimal Power Flow (ACOPF). In the last decades, power grids have evolved into a typical dynamic, non-linear and large-scale control system, known as the power system, so searching for better and faster ACOPF solutions is becoming crucial. Appearance of Graph Neural Networks (GNN) has allowed the natural use of Machine Learning (ML) algorithms on graph data, such as power networks. On the other hand, Deep Reinforcement Learning (DRL) is known for its powerful capability to solve complex decision-making problems. Although solutions that use these two methods separately are beginning to appear in the literature, none has yet combined the advantages of both. We propose a novel architecture based on the Proximal Policy Optimization algorithm with Graph Neural Networks to solve the Optimal Power Flow. The objective is to design an architecture that learns how to solve the optimization problem and that is at the same time able to generalize to unseen scenarios. We compare our solution with the DCOPF in terms of cost after having trained our DRL agent on IEEE 30 bus system and then computing the OPF on that base network with topology changes