Lastest News
October 2026 Mejor Conectados program by Telefónica dedicated a spot to our work of collaboration with the Museo Nacional del Prado and the C2 Project, see https://mejorconectados.com/proyecto-aracne-museo-del-prado/ and the video there in or at https://www.youtube.com/watch?v=9M_p0c1mu78&t=8s. We deeply thank the team of Mejor Conectados for this great job describing the scope of our work and contributions.
June 2026 We were granted a project “Rapid detection of unmanned aircraft with limited resources” DGP_PIDI_2024_0 by EIDIA de la Junta de Andalucía, Spain. In this project we will develop for three years new tools to detect non authorized drones based on radiofrequency and visible images.
May 2026 We were granted a project by Fundación Española para la Ciencia y Tecnología (FECYT), a public organization to foster the public and social value of science and innovation, as well as their contribution to democratic quality and social well-being. We presented a proposal named “Textile Analysis for the Improvement of Study and Conservation Policies in Historical Heritage: Advisory Services and Knowledge Transfer: ARACNE 2.0” FCT-25-23070 to improve, during two years, the Aracne Software as it needed a profound review to cope with new versions of the programming languages. We also aim at improving performance by using artificial intelligence.
February 2026 We started an exiting collaboration in place with the C2 project (https://www.c2luxor.es/) at Luxor, becoming part of the C2 team. We shared a one week with the C2 team at tomb MMA1103. We selected, processed and picture about 50 textiles now being analyzed with the Aracne software.
September 2025 A year ago, Laura Alba at the Museo Nacional del Prado and our group started a collaboration with the team C2 (https://www.c2luxor.es/), led by José Ramón Pérez Accino at the Universidad Complutense de Madrid. They proposed that we apply Aracne to old Egyptian textiles. We presented preliminary results last year at the ‘Hide & Seek’ workshop in Madrid, then our first steps adapting the software to these types of fabrics at Eusipco 2025, last September in Palermo.
September 2025 We delivered a talk about general-purpose artificial intelligence (GPAI) at the Universidad Pablo Olavida as part of an international workshop on data and intelligence within the criminal proceedings, at the invitation of Prof. Ignacio Colomer Hernández. With the most interesting talk by Andrej Karpathy as the starting point, we dived into the principles of ChatGPT, Gemini, or Copilot: learning, fine-tuning and reinforce learning.
June 2025 We are happy to share a new result on the thread counting for old Canvases: https://rdcu.be/eo0kd coauthored with Antonio Delgado Bejarano and Laura Alba Carcelén. This work uses deep learning to tackle this problem, allowing the matching of patterns of densities of threads along the paintings. Previous frequency domain approaches failed to establish this matching for some canvases in the Museo Nacional del Prado collection. In this proposal, we directly compute the density value with IA instead of locating the crossing points as in our previous approach https://lnkd.in/dxs5pXhY, with reduced error. Besides, we get better results than the previous machine learning tool by Laurens van der Maaten and Robert G Erdmann, where we needed to label part of the masterpiece at hand. We are grateful to Catherine Higgitt at the The National Gallery, for facilitating part of this research. One of the outcomes of this work was the change of authorship for “An Artillery General” by Rizzi and now attributed to Herrera El Mozo, as we found the same pattern.
April 2025 We were recently interviewed about the Aracne Software on TV7 (in Spanish), see https://lnkd.in/daUGvxwK. It will be open for one week. Also available in two parts at https://lnkd.in/dxKzuN72 and https://lnkd.in/dqtj_BHd. Also my colleague al the MNP, Laura Alba Carcelén, was interviewed by the Observer at https://lnkd.in/dMhBna9m.
April 2025 Las thrusday our member Prof. Murillo was the recipient of the Oscar for the best lecturer in the master in Telecommunications Engineering at ETSi – Escuela Técnica Superior de Ingeniería de la Universidad de Sevilla. I have a pretty collection already :- ). Thanks to my students, a formidable group this year that not only had remarkable academic achievements but also showed extraordinary generosity in welcoming foreign students. I shared this Oscar with my colleagues at the master, we are jointly making it to succeed, with an increase of 40% in students this year. Finally, I had a few words for Corsino at our Centro de Cálculo, retiring this year after decades at our school, Lourdes Pizarraya was secretary at the direction and now moving to another service at the US after 17 years with us, and Alejandro García Lomas retiring after being in charge of our infraestructures for 27 years at ETSi. Oscars are for lecturers, but Corsino, Lourdes, and Alejandro are those making things happen. They always had a “yes” for the ETSi, no matter when, how, or what. Their capacity and availability have been paramount for the success of the ETSi over the last decades. We will really miss them around.
March 2025 We are happy to announce that today we published in open access the software Aracne. More than 10 years of collaboration of the ETSi – Escuela Técnica Superior de Ingeniería de la Universidad de Sevilla with Ph. D. Alba-Carcelen, curator at the Museo Nacional del Prado, led to this software. Aracne is a tool to process images of fabrics of canvases that goes further than the handmade and hence limited study of cloths. Aracne provides insightful information to support curators in making decisions on the date, authorship, or integrity, among others, of paintings. A short description of the software can be enjoyed in the following video (Spanish): https://lnkd.in/dnnVkeu4. See https://lnkd.in/dRJPSSM6 for a short description in instagram (https://lnkd.in/d6cX8FJ5 in facebook) or go to https://lnkd.in/dZZNkMvC for the full information. Also in https://lnkd.in/dqT_6jsj.
October 2023 Our work on AI applied to the segmentation of crossing points of threads in X-rays of canvases, a joint work with the Museo Nacional del Prado, has been published, https://www.sciencedirect.com/science/article/pii/S0952197623012848 see https://lnkd.in/dw7Hitvh at us.es. In this work, a collaboration with Museo Nacional del Prado, we locate crossing points between horizontal and vertical threads to estimate thread densities later. We obtain a good estimation of the separation between threads at every point of the painting. This novel approach has been successfully applied to canvases by Velázquez, where the SOTA, the Fourier transform, fails. This is part of the PhD Thesis of Antonio Delgado Bejarano.
March 2023. We successfully designed a new thread counting algorithm based on deep learning and semi-supervised regression, outperforming previous approaches. See https://arxiv.org/abs/2303.15999 for the full manuscript, sent to the Neural Networks journal, and https://github.com/gapsc-us/DL4ART for code.
July 2023 Appointed as vice-dean for international relations and mobility at the Escuela Técnica Superior de Ingeniería: moving students in and out around the globe.
October 2022. We published a new book entitled Laboratorio de Comunicaciones Digitales con Python, in the editorial Universidad de Sevilla, under the collection “Publicaciones de la Escuela Técnica Superior de Ingeniería”. See this url and the code at github.
April 2022. We participated in the Computational approaches for technical imaging in cultural heritage (7th IP4AI meeting, https://art-ict.github.io/artict/Conference.html) at the National Gallery, London, with the following works:
- María del Mar Velasco Montero, Juan José Murillo-Fuentes, Laura Alba-Carcelén, “Complex Twill Fabrics Pattern Recognition in Canvases”,
- Antonio D. Bejarano, Juan José Murillo-Fuentes, Laura Alba-Carcelén, “Crossings Segmentation in Plain Weaves for X-Rays of Canvases with Deep Learning”.
- Antonio D. Bejarano, Juan José Murillo-Fuentes, Laura Alba-Carcelén, “Crossings Segmentation in Plain Weaves for X-Rays of Canvases with Deep Learning”.
October 2021. We start a new project with El Museo Nacional del Prado: Artificial Intelligence Applied to the Characterization and Comparison of Fabrics of Canvases in Art Masterpieces (ATENEA). In this project we face the analysis of fabrics of old canvases by means of artificial intelligence (AI). See the web page of the ATENEA project for further information.
September 2022. We successfully designed a new thread counting algorithm based on deep learning and segmentation. See https://arxiv.org/abs/2302.11924 for the full manuscript, sent to the Engineering Applications of Artificial Intelligence journal.
January 2021. Spanish Newspaper “EL PAIS” reports on our advances in the automatic analysis of fabrics with Aracne. See the piece of news. (you might need to register to read it). See also the public piece of news at the local Seville edition of the “ABC”.
December 21th 2020. A new MATLAB software tool has been publicly published at GitHub including the codes to perform iterative detection and decoding based on Expectation Propagation (EP), using LDPC, for equalisation and MIMO detection See link at GitHub. In particular, it generates the code used in the papers, including the figures:
- Murillo-Fuentes JJ, Santos I, Aradillas JC, Sánchez Fernández M. A Low-Complexity Double EP-based Detector for Iterative Detection and Decoding in MIMO. IEEE Trans Commun. Accepted 2021. See link at IEEE.
- I. Santos, J. J. Murillo-Fuentes, J. C. Aradillas and E. Arias-De-Reyna, “Channel Equalization With Expectation Propagation at Smoothing Level,” in IEEE Transactions on Communications, vol. 68, no. 5, pp. 2740-2747, May 2020, doi: 10.1109/TCOMM.2020.2975624. See https://ieeexplore.ieee.org/document/9006952 to get citation
- I. Santos and J. J. Murillo-Fuentes, “Self and Turbo Iterations for MIMO Receivers and Large-Scale Systems,” in IEEE Wireless Communications Letters, vol. 8, no. 4, pp. 1095-1098, Aug. 2019, doi: 10.1109/LWC.2019.2907941. See https://ieeexplore.ieee.org/document/8675457 to get a citation.
November 28th 2020. A new work reporting advances on the application of Expectation Propagation (EP) to MIMO and massive MIMO has been accepted for publication at the IEEE Transactions on Communications. See link at IEEE.
November 7th 2020. A new Python software tool has been publicly published at GitHub to accomplish an elementary laboratory on digital comunica See link at GitHub. In particular, it includes a file labcomdig.py, with several functions to simulate PAM, PSK and QAM transmitter and receivers for AWGN channels. Several examples using jupyter notebooks are also included. See the manual of the laboratory to follow this examples.
September 16th 2020. Invited talk to the Optical Algorithm Workshop organized by Huawei in Moscow. A review of the latest advances in Expectation Propagation for equalization will be presented. See presentation.
September 8th 2020. New work on automatic handwritting recognition using recurrent deep neural networks, Improving offline HTR in small databases by purgin Unreliable Labels, presented at the 17th Int. Conf. on Frontiers in Handwriting Recognition
February 24-26th. 2020. Prof. Murillo-Fuentes teaching digital communications, antennas and propagation at the Master’s Degree in Remotely Piloted Aircraft Systems (RPAS) at the Universidad de Huelva.
February 10-11th. 2020. Prof. Simois and Murillo-Fuentes introduced high-school students to the Signal Processing labs.
February 11th. 2020. A new work on how the Kalman smoother can be approximated for discrete signals has been published. We reported significant improvements compared to our previous paper by exploiting expectation propagation (EP) at smoothing level. With this work we push the Gaussian approximation toward the optimal for channels with larger memories and higher order modulations. See I. Santos, J. J. Murillo-Fuentes, José C. Aradillas and E. Arias-de-Reyna, “Channel Equalization with Expectation Propagation at Smoothing Level,” in IEEE Transactions on Communications, see also personal.us.es/papers/IEEETCOM_KSEP2020.pdf
December 19th. 2019. A new work on how the expectation propagation (EP) can be used twice, within the equalizer and at the turbo equalization was published! I. Santos, J. J. Murillo-Fuentes and E. Arias-de-Reyna, “A Double EP-Based Proposal for Turbo Equalization,” in IEEE Signal Processing Letters, vol. 27, pp. 121-125, 2020. See DOI: 10.1109/LSP.2019.2959900, see also personal.us.es/papers/IEEESPL_DEP2020.pdf
November 26th. 2019. Prof. Murillo was keynote speaker at the international congress on the penal process and the limits in the use of the data https://www.diariodesevilla.es/juzgado_de_guardia/actualidad/Jornadas-uso-informacion-procesos-penales_0_1411959052.html
October 25th. 2019. Rafael Boloix Tortosa became profesor titular (associate professor) at the Department of Signal Theory and Communications at the Universidad de Sevilla. Congratulations!!!
October 15th. 2019. The joint work with prof. Sotos (Univ. of Edinburgh) has been published: by Boloix-Tortosa, J.J. Murillo-Fuentes S.A. Tsaftaris. “The Generalized Complex Kernel Least-Mean-Square Algorithm” IEEE Transactions on Signal Processing. DOI 10.1109/TSP.2019.2937289, https://arxiv.org/abs/1902.08692
March 1st. 2019. Published paper by Santos, J.J. Murillo-Fuentes. (2019). “Self and Turbo Iterations for MIMO Receivers and Large-Scale Systems” IEEE Wireless Communications Letters. DOI 10.1109/LWC.2019.2907941, https://arxiv.org/abs/1805.05065
December 10th. The joint work with Petar Djuric has been accepted for publication: Irene Santos, Juan José Murillo-Fuentes, Petar M. Djurić (2019) “Recursive Estimation of Dynamic RSS Fields Based on Crowdsourcing and Gaussian Processes” To appear in IEEE Transactions on Signal Processing. Paper in arxiv.org.
November 22nd. The work “Turbo EP-Based Equalization: A Filter-Type Implementation” and its first author, Irene Santos, are the recipients of the best scientific paper of the quarter at the ETSI.
November 16th. Irene Santos successfully defended her PhD with cum laudem. Congrats!
November 9th 2018. Invited talk at the 3rd Workshop on Communication Networks and Communication Systems. See presentation (pdf). Watch talk (starts at minute 5:40)
October 26th 2018. Invited Talk at the VII Arte y Ciencia Congress, at Museo Nacional del Prado
October 19th 2018. Eva Arias de Reyna passed the exam to become associate professor. Congrats!
November 1st 2018. Published work: R. Boloix-Tortosa, J. J. Murillo-Fuentes, F. J. Payán-Somet and F. Pérez-Cruz, “Complex Gaussian Processes for Regression,” in IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 11, pp. 5499-5511, Nov. 2018. At ieeexplore
September 5th 2018. Paper accepted at NIPS, in collaboration with the University of Cambridge, see paper
September 1st 2018. Published work in collaboration with the Stoony Brooks University, University of Bologna and the Northwestern University: E. Arias-de-Reyna, P. Closas, D. Dardari and P. M. Djuric, “Crowd-Based Learning of Spatial Fields for the Internet of Things: From Harvesting of Data to Inference,” in IEEE Signal Processing Magazine, vol. 35, no. 5, pp. 130-139, Sept. 2018. At ieeeplore
September 1st 2018. Pubished work: Santos, J. J. Murillo-Fuentes, E. Arias-de-Reyna and P. M. Olmos, “Turbo EP-Based Equalization: A Filter-Type Implementation,” in IEEE Transactions on Communications, vol. 66, no. 9, pp. 4259-4270, Sept. 2018. At ieeexplore