Dpto. Teoría de la Señal y Comunicaciones e Ingeniería Telemática - Comunicaciones a congresos, conferencias, etc. https://uvadoc.uva.es/handle/10324/1381 2026-04-19T07:12:19Z 2026-04-19T07:12:19Z Casaseca de la Higuera, Juan Pablo Aguiar Pérez, Javier Manuel Amado Caballero, Patricia Morgado, Antonio J. Silva Moreira, Joana da Luo, Chunbo Wang, Xinheng https://uvadoc.uva.es/handle/10324/78604 2025-12-15T13:51:42Z 2025-01-01T00:00:00Z

This paper introduces an online active deep learning framework tailored for intelligent and sustainable aerial-terrestrial IoT networks. The framework utilizes Long Short-Term Memory (LSTM) models, enhanced with active learning and incremental updates, to address the challenges of dynamic network traffic prediction. Model updates are selectively applied, prioritizing samples with higher prediction errors, enabling the system to adapt dynamically to evolving traffic patterns and improving predictive accuracy. The proposed approach accommodates non-stationary data streams and prioritizes critical information, establishing a robust foundation for effective network management. The framework's performance was validated using the Milan traffic dataset, demonstrating its effectiveness compared to non-weighted online learning methods and static models.

2025-01-01T00:00:00Z Anzola Rojas, Camilo Aguado Manzano, Juan Carlos Durán Barroso, Ramón José Miguel Jiménez, Ignacio de Merayo Álvarez, Noemí Fernández Reguero, Patricia Lorenzo Toledo, Rubén Mateo Abril Domingo, Evaristo José https://uvadoc.uva.es/handle/10324/76304 2025-07-09T19:01:40Z 2023-01-01T00:00:00Z

Connected, cooperative and automated mobility (CCAM) is a growing field as the services required by connected vehicles increase in quantity and complexity. To be able to be connected, vehicles need network infrastructure to communicate with. Roadside Units (RSUs) are communication devices that are placed beside highways and roads and offer connectivity and processing services to the vehicles. The 3rd Generation Partnership Project (3GPP) has developed some standards which specify the characteristics of the services, as well as the spectral bands used to offer vehicle to everything (V2X) connectivity. In this paper, we study the optimal placement of RSUs in a V2X network for different service requirements and channel quality scenarios, considering the 3GPP standards. We formulate an Integer Linear Programming (ILP) model to minimize the number of RSU s needed and solve it for different traffic and channel conditions given by the 3GPP specifications. Results offer the number of required RSU s and their placement for vehicular access networks deployments.

2023-01-01T00:00:00Z Anzola Rojas, Camilo Durán Barroso, Ramón José Miguel Jiménez, Ignacio de Aguado Manzano, Juan Carlos Merayo Álvarez, Noemí Fernández Reguero, Patricia Lorenzo Toledo, Rubén Mateo Abril Domingo, Evaristo José https://uvadoc.uva.es/handle/10324/76301 2025-07-09T19:01:39Z 2023-01-01T00:00:00Z

Multi-Access Edge Computing (MEC) network planning is performed considering a forecast of estimated workload in each coverage zone with the aim of offloading computationally expensive tasks from user's devices to the nearest MEC Data Center (MEC-DC). Nevertheless, in some scenarios, these forecasts are exceeded temporarily due to sudden peaks in demand in a determined MEC-DC, making its planned computing resources (i.e., MEC servers) scarce, and introducing the need of a strategy to face this increment in demand. In this paper, we propose and evaluate an Integer Linear Programming (ILP) model for optimizing the task offloading considering a previously defined MEC network topology. Our model is based on the possibility of offloading some tasks to MEC-DCs different to the initially planned (nearest to the user) one, as long as the latency requirements are met, and the allocated server has enough idle computing power. Results show that the proposed strategy considerably increases the capacity of the network to face sudden workload increments compared to an approach that only assigns the nearest MEC server to every user.

2023-01-01T00:00:00Z Hosseini, Soheil Miguel Jiménez, Ignacio de Casellas, Ramón Abril Domingo, Evaristo José Durán Barroso, Ramón José https://uvadoc.uva.es/handle/10324/76169 2025-07-07T07:58:38Z 2025-01-01T00:00:00Z

The recently proposed band division multiplexing (BDM) technology enables network operators to leverage existing optical fibers more effectively. By expanding from the conventional C-band to the L-band, and potentially to the S and E bands, network capacity improves at least two times. This rate of capacity improvement allows the support of significantly more connections compared to the current elastic optical networks (EONs). However, the increased capacity also means that a network failure might cause huge data loss, making the employment of survivability methods a must. On the other hand, since the required level of protection might vary across different users, the implementation of an effective service level agreement (SLA) mechanism is crucial for enhancing network performance and optimizing costs. Since transceivers are among the most expensive components, in this paper, we aim to modify an SLA-differentiated protection for C+L band networks that we previously proposed, focusing on reducing transceiver usage. The modified SLA-differentiated method is evaluated against the original one in terms of network performance and transceiver consumption. We demonstrate that the introduced variation of the SLA mechanism achieves a significant reduction in the number of utilized transceivers.

2025-01-01T00:00:00Z Masoumi Estahbanati, Maryam Brasca, Fabrizio Bonfaci, Emanuele Rizzi, Gianluca Miguel Jiménez, Ignacio de Durán Barroso, Ramón José https://uvadoc.uva.es/handle/10324/76168 2025-07-07T07:58:37Z 2024-01-01T00:00:00Z

Understanding the unique features of Multi-access Edge Computing (MEC) in the context of 5G is key, as they enable applications such as real-time analytics. This paper explores the integration of 5 G technologies with MEC and the challenges in deploying and maintaining MEC infrastructure, recognizing the impact on network reliability. In particular, the paper considers potential failures in MEC and their broader consequences on overall network reliability, proposing the strategic use of backup resources as a proactive solution to fortify MEC resilience. MEC, evolving as an ecosystem to integrate telecommunication and IT services, establishes a cloud computing platform at the edge of the radio access network (RAN), enhancing storage and computational capabilities with reduced latency for mobile end-users. The analysis covers key technologies enabling MEC integration with 5G, including software-defined networking (SDN), network function virtualization (NFV), and network slicing. The deployment of Network Services (NS) through virtual network functions (VNFs) is explored for latency-sensitive communication services, acknowledging the vulnerability of softwarized and cloud services to failures. Reliability emerges as a major challenge in MEC 5G networks, prompting the proposal of redundancy (backup) as a solution to enhance communication service reliability, setting the stage for further research considerations in MEC reliability.

2024-01-01T00:00:00Z Becattini, Marco Carnevali, Laura Fontani, Giovanni Paroli, Leonardo Scommegna, Leonardo Masoumi Estahbanati, Maryam Miguel Jiménez, Ignacio de Brasca, Fabrizio https://uvadoc.uva.es/handle/10324/76162 2025-07-07T07:58:36Z 2024-01-01T00:00:00Z

The use of innovative technologies in Industry X.0 sce narios, including, but not limited to, Augmented Reality/Virtual Reality (AR/VR), autonomous robotics, and advanced security systems, requires applicative interconnections between a large number of IoT machines and devices. These interconnections must support Ultra-Reliable and Low Latency Communications (URLLC) to optimize usage and per formances of devices related to those new technologies. Notably, the concepts of low latency and reliability are inherently linked; from a device perspective, any service exceeding specific response time thresholds is deemed unresponsive, and thus unreliable. In this paper, we present an innovative approach to quan titatively evaluate reliability in URLLC settings, leveraging the use of Digital Twin Networks (DTN), with a specific focus on Mobile Edge Computing (MEC) and its application to Industry X.0 scenarios. Results obtained so far show the potential for this approach to confer MEC better requests handling capabilities, by providing a near real time re-configuration ability within the MEC itself.

2024-01-01T00:00:00Z Hosseini, Soheil Durán Barroso, Ramón José Miguel Jiménez, Ignacio de González de Dios, Óscar Merayo Álvarez, Noemí Aguado Manzano, Juan Carlos https://uvadoc.uva.es/handle/10324/76160 2025-07-07T07:58:34Z 2024-01-01T00:00:00Z

The use of multi-band elastic optical networks (MB-EONs) over C+L bands is considered as a realistic mediumterm solution to extend the capacity of current C-band optical networks. In this kind of networks, different levels of survivability can be offered, as we showed in a previous work in which we analysed the impact of these techniques in fully upgraded C+L networks. However, postponing the high CAPEX cost of activating the L-band for all the optical fibers within a network has shifted the attention of the network operators to partially upgraded networks. Thus, we now focus on partially upgraded networks and demonstrate that it is necessary to modify the protection method that we previously proposed for fully upgraded networks in order to efficiently work in these environments. The performance of the modified method is then evaluated in terms of request blocking ratio in both partially and fully upgraded networks, and we show that the introduced variation outperforms the original method.

2024-01-01T00:00:00Z Hosseini, Soheil Miguel Jiménez, Ignacio de González de Dios, Óscar Fernández-Palacios, Juan Pedro Durán Barroso, Ramón José https://uvadoc.uva.es/handle/10324/76159 2025-07-07T07:58:33Z 2024-01-01T00:00:00Z

Operators have two alternatives to acquire new transceivers when migrating from current C-band optical networks to C+L multi-band optical networks: i) multi-band C+L transceivers, or ii) separate single-band C and single-band L transceivers. The deployment of these new transceivers, along with the other costly components required, delays the completion of a fully upgraded network on which the L-band is active on all network links. Therefore, the concept of partial migration during the network planning phase has been proposed. This paper attempts to shed light on the question of which type of transceivers should the industry focus to help operators in the migration of their networks from the C-band to the C+L-bands. Simulation results demonstrate that the employment of multi-band transceivers does not lead to a significant reduction in the number of required additional transceivers that should be installed for different levels of partial migration. Additionally, we present a techno-economic study about this issue, demonstrating that, in fact, the use of single-band transceivers leads to lower costs.

2024-01-01T00:00:00Z Ferens Michalek, Mieszko Jan Hortelano Haro, Diego Miguel Jiménez, Ignacio de Durán Barroso, Ramón José Kosta, Sokol https://uvadoc.uva.es/handle/10324/76157 2025-07-07T07:58:32Z 2024-01-01T00:00:00Z

Large scale deployment of Internet-of-Things (IoT) devices is projected to grow in the coming years. These devices are expected to be low-cost while supporting applications with growing computational demands. To enable the necessary computations, offloading of computational tasks to Edge and Cloud nodes is a fundamental technology. However, orchestration for such networks is a complex problem which affects both the network design and the decision system. To aid in solving this problem, simulation tools are essential for predicting the performance of networks in different conditions and under different orchestration policies. In this paper, we propose STEROCEN, a Cloud-Edge network resource orchestration simulation and training tool which allows for different configurations of up-to a four-layer network composed of: (i) end-device, (ii) Close Edge, (iii) Far Edge, and (iv) Cloud layers. Our tool collects delay metrics for flexibly defined applications, especially in regard to computation in the network nodes and including uncertainty in processing times. Additionally, the tool only needs the initial configuration and an independently defined orchestrator, allowing for testing of many strategies. As an example, we provide results of testing some Deep Reinforcement Learning (DRL) algorithms using the same training and simulation environment.

2024-01-01T00:00:00Z Hosseini, Soheil Miguel Jiménez, Ignacio de González de Dios, Óscar Merayo Álvarez, Noemí Fernández Palacios, Fernando Durán Barroso, Ramón José https://uvadoc.uva.es/handle/10324/76156 2025-07-07T07:58:31Z 2024-01-01T00:00:00Z

The attraction of new services and applications over the Internet makes it critical for the network operators to increase the capacity of optical networks. To this end, multi-band elastic optical networks (MB-EON) are at the forefront of this capacity improvement for the short and medium term. These networks employ additional spectrum bands from the optical fiber other than the C (conventional) band. Due to the current availability of L-band erbium doped fiber amplifiers (EDFAs), lighting up the L-band of already installed fibers is considered as a pragmatic approach. However, network operators envision a soft migration from current C-band to fully upgraded C+L-bands networks in order to distribute the high cost of equipping all fibers and nodes of a network with multi-band devices over several years. Therefore, it is essential to propose solutions for scenarios where just a subset of network elements has been upgraded, i.e., to consider a hybrid C/C+L optical network rather than a fully upgraded one. As using the L-band besides the C-band doubles the network capacity, the management of the potential fiber failures is an important factor. In this work, the focus is set on the evaluation of the network performance, when considering different levels of partial migration, and comparing the use of dynamic restoration versus dedicated protection. Simulation results demonstrate that the usage of dynamic restoration, mainly in the middle stages of migration towards a fully upgraded network can remarkably boost the supported traffic load compared to the use of dedicated protection.

2024-01-01T00:00:00Z Amado Caballero, Patricia Garmendia Leiza, José Ramón Aguilar García, M. D. Martínez Fernández de Septiem, C. San José Revuelta, Luis Miguel García Ruano, A. Alberola López, Carlos Casaseca de la Higuera, Juan Pablo https://uvadoc.uva.es/handle/10324/72906 2025-02-20T13:53:03Z 2024-01-01T00:00:00Z

This study explores the feasibility of employing eXplainable Artificial Intelligence XAI methodologies for the analysis of cough patterns in respiratory diseases. A cohort of 20 adult patients, all presenting persistent cough as a symptom of respiratory disease, was monitored for 24 hours using a smartphone. The audio signals underwent frequency domain transformation to yield 1-second spectrograms, subsequently processed by a CNN to detect cough events. Quantitative analysis of spectrogram regions relevant for cough detection highlighted by occlusion maps, revealed significant differences between patient groups. Notably, distinctions were observed between the Chronic Obstructive Pulmonary Disease (COPD) patient group and groups with other respiratory pathologies, both chronic and non-chronic. In conclusion, interpretability analysis methods applied to neural networks offer insights into cough-related distinctions among patients with varying respiratory conditions.

2024-01-01T00:00:00Z Jiménez Galindo, Daniel Casaseca de la Higuera, Juan Pablo San José Revuelta, Luis Miguel https://uvadoc.uva.es/handle/10324/71907 2025-03-26T19:10:05Z 2019-01-01T00:00:00Z

The design of both FIR and IIR digital filters is a multi-variable optimization problem, where traditional algorithms fail to obtain optimal solutions. A modified Shuffled Flog Leaping Algorithm (SFLA) is here proposed for the design of FIR and IIR discrete-time filters as close as possible to the desired filter frequency response. This algorithm can be considered a type of memetic algorithm. In this paper, simulations prove the obtained filters outperform those designed using the traditional bilinear Z transform (BZT) method with elliptic approximation. Besides, results are close to, and even slightly better, than those reported in recent bio-inspired approaches using algorithms such as particle swarm optimization (PSO), differential evolution (DE) and regularized global optimization (RGA).

2019-01-01T00:00:00Z González Ortega, David Díaz Pernas, Francisco Javier Martínez Zarzuela, Mario Antón Rodríguez, Miriam https://uvadoc.uva.es/handle/10324/65443 2024-01-31T20:02:03Z 2019-01-01T00:00:00Z

In this paper, we present a custom-developed driving simulation scenario and a vehicle model developed to evaluate driving efficiency and safety. The scenario includes different road sections, traffic conditions, and events, including a through road similar to a real road section in the city of Valladolid (Spain). The modeled vehicle is an ICE vehicle with manual or automatic gear shift. During a simulation, following a guided or a free route, the speed, rpm, gear, consumption, and traffic offences are showed in real time and stored in files for further processing. Six people drove in the scenario twice, one time with automatic gear shift and another with manual gear shift. An analysis of the results has been carried out to know the factors with influence on driving efficiency and safety. A significant relation between efficient and safe driving was found. People that took part in the experiments ranked the simulation scenario positively regarding ease of interaction, realistic experience, usefulness for driving learning, and entertainment capacity.

2019-01-01T00:00:00Z González Ortega, David Díaz Pernas, Francisco Javier Martínez Zarzuela, Mario Antón Rodríguez, Miriam https://uvadoc.uva.es/handle/10324/65439 2024-02-01T09:20:47Z 2018-01-01T00:00:00Z

In this paper, we present two 3D driving simulation scenarios and a vehicle model developed to analyze driving efficiency and safety. The scenarios include different roads, traffic, and events so that drivers can acquire driving competences and their performance can be tested. They were developed with Unity game engine, which allows to create 2D and 3D games and applications and to export them to many different platforms. The first scenario is urban with varied roads, crossroads controlled by traffic lights, and roundabouts. The second scenario is interurban with different road sections connected by roundabouts. The vehicle engine has been modeled setting the dependency between engine torque and rpm and the dependency of each gear and speed on fuel consumption. Automatic or manual gear shifting can be selected. During a simulation, the speed, rpm, gear, consumption, and traffic offences are showed in real time and stored in files for further processing. Seven people drove in the scenarios and their stored data enabled us to analyze differences in driving performance between them. The simulator was ranked positively by all the participants regarding ease of interaction, similarity to real driving, and usefulness for driving learning.

2018-01-01T00:00:00Z Andrés, Jesús Ángel https://uvadoc.uva.es/handle/10324/62381 2023-10-26T19:01:07Z 2023-01-01T00:00:00Z

Los laboratorios virtuales y remotos resultan de gran utilidad en ámbitos de la inge-niería, especialmente en entornos universitarios, debido a que el acceso a los labora-torios físicos es limitado, costoso y requieren un espacio físico. En particular, en el campo de las redes y comunicaciones ópticas, esto es especialmente cierto. En este artículo se presenta el diseño e implementación de un sistema remoto y virtual utili-zando LabVIEW, para aplicar al análisis de la capa física en redes de Acceso Ópti-cas Pasivas Gigabit (GPON, Gigabit Passive Optical Networks), de especial relevan-cia por su nivel de penetración en el mercado mundial. El sistema permite interac-tuar con el laboratorio real a larga distancia y de manera activa, lo que proporciona una experiencia cercana a situaciones reales que se pueden encontrar en el ámbito profesional.

2023-01-01T00:00:00Z Pérez Pons, María Eugenia Parra Domínguez, Javier Anzola Rojas, Camilo Durán Barroso, Ramón José Miguel Jiménez, Ignacio de Queiroz, Jonás Leitão, Paulo https://uvadoc.uva.es/handle/10324/62375 2025-02-18T11:29:27Z 2021-01-01T00:00:00Z

Agents are understood by definition as entities that interact with their environment and also with other agents. Thus, when two or more agents are able to work together in order to solve a common problem, they form a multi-agent system (MAS). MAS are systems that integrate a set of agents that interact, communicate and coordinate to achieve the established objectives. MAS are designed to meet a set of objectives according to a set of rules and standards. This article briefly summarizes different MAS classifications and methodologies that have been extensively used in the literature.

2021-01-01T00:00:00Z