Alvaro Alonso
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Alvaro Alonso has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Joaquín Salvachúa
Universidad Politcnica de Madrid, Madrid, Spain
12 shared publications - Sonsoles López-Pernas
School of Computing, University of Eastern Finland, Joensuu, Finland
11 shared publications - Alejandro Pozo
Universidad Politcnica de Madrid, Madrid, Spain
10 shared publications - Andres Munoz-Arcentales
Universidad Politcnica de Madrid, Madrid, Spain
9 shared publications - Javier Conde
Universidad Politécnica de Madrid
8 shared publications - Gabriel Huecas
Universidad Politcnica de Madrid, Madrid, Spain
7 shared publications
Latest Publications
- Open Access
- 18 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
Real-time Spatial Retrieval Augmented Generation for Urban Environments
The proliferation of Generative Artificial In t elligence (AI), especially Large Language Models, presents transformative opportunities for urban applications through Urban Foundation Models. However, base models face limitations, as they only contain the knowledge available at the time of training, and updating them is both time-consuming and costly. Retrieval Augmented Generation (RAG) has emerged in the literature as the preferred approach for injecting contextual information into Foundation Models. It prevails over techniques such as fine-tuning, which are less effective in dynamic, real-time scenarios like those found in urban environments. However, traditional RAG architectures, based on semantic databases, knowledge graphs, structured data, or AI-powered web searches, do not fully meet the demands of urban contexts. Urban environments are complex systems characterized by large volumes of interconnected data, frequent updates, real-time processing requirements, security needs, and strong links to the physical world. This work proposes a real-time spatial RAG architecture that defines the necessary components for the effective integration of generative AI into cities, leveraging temporal and spatial filtering capabilities through linked data. The proposed architecture is implemented using FIWARE, an ecosystem of software components to develop smart city solutions and digital twins. The design and implementation are demonstrated through the use case s of a tourism assistant and for the management of traffic lights and luminaires in the city of Madrid. The use case s serve to validate the correct integration of Foundation Models through the proposed RAG architecture. They also enable the analysis of current model limitations, such as their inability to handle large volumes of information, even when it fits within their context window, and the high latency of Large Language Models caused by transformer-based architectures, which generate output token by token.
- 18 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
Open AccessReal-time Spatial Retrieval Augmented Generation for Urban Environments
The proliferation of Generative Artificial In t elligence (AI), especially Large Language Models, presents transformative opportunities for urban applications through Urban Foundation Models. However, base models face limitations, as they only contain the knowledge available at the time of training, and updating them is both time-consuming and costly. Retrieval Augmented Generation (RAG) has emerged in the literature as the preferred approach for injecting contextual information into Foundation Models. It prevails over techniques such as fine-tuning, which are less effective in dynamic, real-time scenarios like those found in urban environments. However, traditional RAG architectures, based on semantic databases, knowledge graphs, structured data, or AI-powered web searches, do not fully meet the demands of urban contexts. Urban environments are complex systems characterized by large volumes of interconnected data, frequent updates, real-time processing requirements, security needs, and strong links to the physical world. This work proposes a real-time spatial RAG architecture that defines the necessary components for the effective integration of generative AI into cities, leveraging temporal and spatial filtering capabilities through linked data. The proposed architecture is implemented using FIWARE, an ecosystem of software components to develop smart city solutions and digital twins. The design and implementation are demonstrated through the use case s of a tourism assistant and for the management of traffic lights and luminaires in the city of Madrid. The use case s serve to validate the correct integration of Foundation Models through the proposed RAG architecture. They also enable the analysis of current model limitations, such as their inability to handle large volumes of information, even when it fits within their context window, and the high latency of Large Language Models caused by transformer-based architectures, which generate output token by token.
- 23 May 2026
- Published by Springer Nature
- p. 229-254
Perturbation-Based Error Tolerance for Large-Scale Networks
- 23 May 2026
- Published by Springer Nature
- p. 229-254
Perturbation-Based Error Tolerance for Large-Scale Networks
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 75, 4332-4346
Dependability and Error Detection for the Stable Diffusion Generative Model
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 75, 4332-4346
Dependability and Error Detection for the Stable Diffusion Generative Model
- Open Access
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 29 (5) , 17-24
Assessing Latency in ASR Systems: A Methodological Perspective for Real-Time Use
Automatic Speech Recognition (ASR) systems generate real-time transcriptions, but often miss nuances that human interpreters capture. While ASR is useful in many contexts, interpreters - who already use ASR tools such as Dragon - add critical value, especially in sensitive settings such as diplomatic meetings where subtle language is key. Human interpreters not only perceive these nuances but can adjust in real-time, improving accuracy, while ASR handles basic transcription tasks. However, ASR systems introduce a delay that does not align with real-time interpretation needs. The user-perceived latency (UPL) of ASR systems differs from that of interpretation because it measures the time between speech and transcription delivery. To address this, we propose a new approach to measuring delay in ASR systems and validate if they are usable in live interpretation scenarios.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 29 (5) , 17-24
Open AccessAssessing Latency in ASR Systems: A Methodological Perspective for Real-Time Use
Automatic Speech Recognition (ASR) systems generate real-time transcriptions, but often miss nuances that human interpreters capture. While ASR is useful in many contexts, interpreters - who already use ASR tools such as Dragon - add critical value, especially in sensitive settings such as diplomatic meetings where subtle language is key. Human interpreters not only perceive these nuances but can adjust in real-time, improving accuracy, while ASR handles basic transcription tasks. However, ASR systems introduce a delay that does not align with real-time interpretation needs. The user-perceived latency (UPL) of ASR systems differs from that of interpretation because it measures the time between speech and transcription delivery. To address this, we propose a new approach to measuring delay in ASR systems and validate if they are usable in live interpretation scenarios.
- Open Access
- 10 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 159098-159108
Tree-Based Architecture for Enabling Live Interaction in Massive Online Events
Massive online and hybrid events have become important during and after the COVID-19 pandemic. These events aim to replicate live experiences by maintaining interaction between the speakers and the audience. However, existing streaming and videoconferencing solutions fail to provide sufficient scalability and real-time interactions. Streaming technologies lack support for live interaction, whereas videoconferencing technologies scale enough. Events such as panel debates, conference presentations, and sessions with live interpretation require both the ability to support a large number of participants and real-time voice interactions. In this paper, we propose a new tree-based architecture that meets both requirements using videoconferencing technologies. This approach combines the scalability of streaming architectures with the low latency of videoconferencing technologies. The objective was to increase the maximum number of participants that videoconference providers can accept while maintaining live interactions. A prototype implementation of the proposed architecture was developed to test and validate it. Finally, this study provides valuable information for implementing and adapting the proposed architecture to various production environments.
- 10 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 159098-159108
Open AccessTree-Based Architecture for Enabling Live Interaction in Massive Online Events
Massive online and hybrid events have become important during and after the COVID-19 pandemic. These events aim to replicate live experiences by maintaining interaction between the speakers and the audience. However, existing streaming and videoconferencing solutions fail to provide sufficient scalability and real-time interactions. Streaming technologies lack support for live interaction, whereas videoconferencing technologies scale enough. Events such as panel debates, conference presentations, and sessions with live interpretation require both the ability to support a large number of participants and real-time voice interactions. In this paper, we propose a new tree-based architecture that meets both requirements using videoconferencing technologies. This approach combines the scalability of streaming architectures with the low latency of videoconferencing technologies. The objective was to increase the maximum number of participants that videoconference providers can accept while maintaining live interactions. A prototype implementation of the proposed architecture was developed to test and validate it. Finally, this study provides valuable information for implementing and adapting the proposed architecture to various production environments.
- Open Access
- 19 May 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 193, 110305
Activity recognition in patients with tremor: Integrating time–length windows for enhanced detection
- 19 May 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 193, 110305
Open AccessActivity recognition in patients with tremor: Integrating time–length windows for enhanced detection
- Open Access
- 15 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26 (5) , 55-61
Enhanced FIWARE-Based Architecture for Cyberphysical Systems With Tiny Machine Learning and Machine Learning Operations: A Case Study on Urban Mobility Systems
The rise of AI and the Internet of Things is accelerating the digital transformation of society. Mobility computing presents specific barriers due to its real-time requirements, decentralization, and connectivity through wireless networks. New research on edge computing and tiny machine learning (tinyML) explores the execution of AI models on low-performance devices to address these issues. However, there are not many studies proposing agnostic architectures that manage the entire lifecycle of intelligent cyberphysical systems. This article extends a previous architecture based on FIWARE software components to implement the machine learning operations flow, enabling the management of the entire tinyML lifecycle in cyberphysical systems. We also provide a use case to showcase how to implement the FIWARE architecture through a complete example of a smart traffic system. We conclude that the FIWARE ecosystem constitutes a real reference option for developing tinyML and edge computing in cyberphysical systems.
- 15 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26 (5) , 55-61
Open AccessEnhanced FIWARE-Based Architecture for Cyberphysical Systems With Tiny Machine Learning and Machine Learning Operations: A Case Study on Urban Mobility Systems
The rise of AI and the Internet of Things is accelerating the digital transformation of society. Mobility computing presents specific barriers due to its real-time requirements, decentralization, and connectivity through wireless networks. New research on edge computing and tiny machine learning (tinyML) explores the execution of AI models on low-performance devices to address these issues. However, there are not many studies proposing agnostic architectures that manage the entire lifecycle of intelligent cyberphysical systems. This article extends a previous architecture based on FIWARE software components to implement the machine learning operations flow, enabling the management of the entire tinyML lifecycle in cyberphysical systems. We also provide a use case to showcase how to implement the FIWARE architecture through a complete example of a smart traffic system. We conclude that the FIWARE ecosystem constitutes a real reference option for developing tinyML and edge computing in cyberphysical systems.
- Open Access
- 8 August 2024
- Published by River Publishers in Journal of Web Engineering
- Vol. 23 (4) , 595-610
Integration of an Open Source Identity Management System in Educational Platforms
Making research advances available to the community in the shape of open source software has the potential to introduce cutting-edge innovations from early on, foster collaborative development, and revolutionize industrial applications. However, including open source software resulting from a research project as part of a production system poses some risks and must be evaluated in detail, considering all pros and cons. This is especially delicate when that piece of software is in charge of authentication and authorization. This article reports on an experience of integrating open source identity and access management (IAM) software that is the result of multiple research projects, the FIWARE Keyrock IAM, into three educational web-based platforms: two learning object repositories and a course management platform. We intend to draw the lessons learned from this experience so they can guide software practitioners when deciding if they should integrate open source software developed in research projects.
- 8 August 2024
- Published by River Publishers in Journal of Web Engineering
- Vol. 23 (4) , 595-610
Open AccessIntegration of an Open Source Identity Management System in Educational Platforms
Making research advances available to the community in the shape of open source software has the potential to introduce cutting-edge innovations from early on, foster collaborative development, and revolutionize industrial applications. However, including open source software resulting from a research project as part of a production system poses some risks and must be evaluated in detail, considering all pros and cons. This is especially delicate when that piece of software is in charge of authentication and authorization. This article reports on an experience of integrating open source identity and access management (IAM) software that is the result of multiple research projects, the FIWARE Keyrock IAM, into three educational web-based platforms: two learning object repositories and a course management platform. We intend to draw the lessons learned from this experience so they can guide software practitioners when deciding if they should integrate open source software developed in research projects.
- Open Access
- 6 February 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (28) , 71951-71971
An autonomous low-cost studio to record production-ready instructional videos
Producing high-quality educational videos usually requires a large budget as it involves the use of expensive recording studios, the presence of a technician during the entire recording session and often post-production tasks. The high costs associated with video production represent a major hindrance for many educational institutions and, thus, many teachers regard high-quality video recording as inaccessible. As a remedy to this situation, this article presents SAGA (Autonomous Advanced Recording Studio in its Spanish acronym), a low-cost autonomous recording set that allows teachers to produce educational content in video format in an agile way and without the need for post-production. The article provides an overview of SAGA, including a description of its hardware and software so that anyone with basic technical knowledge can replicate and operate the system. SAGA has been used to record more than 1,500 videos including the contents of six MOOCs hosted on the MiriadaX platform, as well as four courses at UPM. SAGA has been evaluated in two ways: (1) from the video producers’ perspective, it was evaluated with a questionnaire based on the Technology Acceptance Model, and (2) from the video consumers’ perspective, a questionnaire was conducted among MOOC participants to assess the perceived technical quality of the videos recorded with SAGA. The results show a very positive general opinion of the SAGA system, the recorded videos and the technical features thereof. Thus, SAGA represents a good opportunity for all those educational institutions and teachers interested in producing high-quality educational videos at a low cost.
- 6 February 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (28) , 71951-71971
Open AccessAn autonomous low-cost studio to record production-ready instructional videos
Producing high-quality educational videos usually requires a large budget as it involves the use of expensive recording studios, the presence of a technician during the entire recording session and often post-production tasks. The high costs associated with video production represent a major hindrance for many educational institutions and, thus, many teachers regard high-quality video recording as inaccessible. As a remedy to this situation, this article presents SAGA (Autonomous Advanced Recording Studio in its Spanish acronym), a low-cost autonomous recording set that allows teachers to produce educational content in video format in an agile way and without the need for post-production. The article provides an overview of SAGA, including a description of its hardware and software so that anyone with basic technical knowledge can replicate and operate the system. SAGA has been used to record more than 1,500 videos including the contents of six MOOCs hosted on the MiriadaX platform, as well as four courses at UPM. SAGA has been evaluated in two ways: (1) from the video producers’ perspective, it was evaluated with a questionnaire based on the Technology Acceptance Model, and (2) from the video consumers’ perspective, a questionnaire was conducted among MOOC participants to assess the perceived technical quality of the videos recorded with SAGA. The results show a very positive general opinion of the SAGA system, the recorded videos and the technical features thereof. Thus, SAGA represents a good opportunity for all those educational institutions and teachers interested in producing high-quality educational videos at a low cost.
- 31 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 73 (4) , 1782-1791
Concurrent Classifier Error Detection (CCED) in Large Scale Machine Learning Systems
The complexity of machine learning (ML) systems increases each year. As these systems are widely utilized, ensuring their reliable operation is becoming a design requirement. Traditional error detection mechanisms introduce circuit or time redundancy that significantly impacts system performance. An alternative is the use of concurrent error detection (CED) schemes that operate in parallel with the system and exploit their properties to detect errors. CED is attractive for large ML systems because it can potentially reduce the cost of error detection. In this article, we introduce concurrent classifier error detection (CCED), a scheme to implement CED in ML systems using a concurrent ML classifier to detect errors. CCED identifies a set of check signals in the main ML system and feed them to the concurrent ML classifier that is trained to detect errors. The proposed CCED scheme has been implemented and evaluated on two widely used large-scale ML models: Contrastive language-image pretraining (CLIP) used for image classification and bidirectional encoder representations from transformers (BERT) used for natural language applications. The results show that more than 95% of the errors are detected when using a simple Random Forest classifier that is orders of magnitude simpler than CLIP or BERT.
- 31 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 73 (4) , 1782-1791
Concurrent Classifier Error Detection (CCED) in Large Scale Machine Learning Systems
The complexity of machine learning (ML) systems increases each year. As these systems are widely utilized, ensuring their reliable operation is becoming a design requirement. Traditional error detection mechanisms introduce circuit or time redundancy that significantly impacts system performance. An alternative is the use of concurrent error detection (CED) schemes that operate in parallel with the system and exploit their properties to detect errors. CED is attractive for large ML systems because it can potentially reduce the cost of error detection. In this article, we introduce concurrent classifier error detection (CCED), a scheme to implement CED in ML systems using a concurrent ML classifier to detect errors. CCED identifies a set of check signals in the main ML system and feed them to the concurrent ML classifier that is trained to detect errors. The proposed CCED scheme has been implemented and evaluated on two widely used large-scale ML models: Contrastive language-image pretraining (CLIP) used for image classification and bidirectional encoder representations from transformers (BERT) used for natural language applications. The results show that more than 95% of the errors are detected when using a simple Random Forest classifier that is orders of magnitude simpler than CLIP or BERT.