38796 research outputs found
Sort by
On the Application of a Sparse Data Observers (SDOs) Outlier Detection Algorithm to Mitigate Poisoning Attacks in UltraWideBand (UWB) Line-of-Sight (LOS)/Non-Line-of-Sight (NLOS) Classification
No full textThe classification of the wireless propagation channel between Line-of-Sight (LOS) or Non-Line-of-Sight (NLOS) is useful in the operation of wireless communication systems. The research community has increasingly investigated the application of machine learning (ML) to LOS/NLOS classification and this paper is part of this trend, but not all the different aspects of ML have been analyzed. In the general ML domain, poisoning and adversarial attacks and the related mitigation techniques are an active area of research. Such attacks aim to hamper the ML classification process by poisoning the data set. Mitigation techniques are designed to counter this threat using different approaches. Poisoning attacks in LOS/NLOS classification have not received significant attention by the wireless communication community and this paper aims to address this gap by proposing the application of a specific mitigation technique based on outlier detection algorithms. The rationale is that poisoned samples can be identified as outliers from legitimate samples. In particular, the study described in this paper proposes a recent outlier detection algorithm, which has low computing complexity: the sparse data observers (SDOs) algorithm. The study proposes a comprehensive analysis of both conventional and novel types of attacks and related mitigation techniques based on outlier detection algorithms for UltraWideBand (UWB) channel classification. The proposed techniques are applied to two data sets: the public eWINE data set with seven different UWB LOS/NLOS different environments and a radar data set with the LOS/NLOS condition. The results show that the SDO algorithm outperforms other outlier detection algorithms for attack detection like the isolation forest (iForest) algorithm and the one-class support vector machine (OCSVM) in most of the scenarios and attacks, and it is quite competitive in the task of increasing the UWB LOS/NLOS classification accuracy through sanitation in comparison to the poisoned model.JRC.E.2 - Space, Connectivity and Economic Securit
A Machine Learning Evaluation of the Impact of Bit-Depth for the Detection and Classification of Wireless Interferences in Global Navigation Satellite Systems
No full textThe performance of the services provided by Global Navigation Satellite Systems (GNSSs) can be seriously degraded by the presence of wireless interferences, and Machine Learning (ML) has been applied to address this problem using the digital artifacts generated by the GNSS receiver. While such an application is not novel in the literature, the analysis of the impact of the bit-depth at which the GNSS signal is recorded has not received significant attention. The type and power level of the wireless interference are also important factors to investigate in this context. This paper addresses this gap by performing an extensive analysis of the impact of these factors on a data set of GNSS signals subject to three different types of wireless interferences with ML and DL algorithms. The analysis is a combination of a pre-processing phase where the Carrier-to-Noise Ratio (CNR) values of different satellites are evaluated, the extraction of relevant features for ML, and the application of a Convolutional Neural Network (CNN) with a multi-head attention layer. The results show that the proposed approach is able to detect the presence of interference with great accuracy (e.g., 99%) but the type of interference and bit-depth can decrease the performance.JRC.E.2 - Space, Connectivity and Economic Securit
Deep cascaded registration and weakly-supervised segmentation of fetal brain MRI
No full textDeformable image registration is a cornerstone of many medical image analysis applications, particularly in the context of fetal brain magnetic resonance imaging (MRI), where precise registration is essential for studying the rapidly evolving fetal brain during pregnancy and potentially identifying neurodevelopmental abnormalities. While deep learning has become the leading approach for medical image registration, traditional convolutional neural networks (CNNs) often fall short in capturing fine image details due to their bias toward low spatial frequencies. To address this challenge, we introduce a deep learning registration framework comprising multiple cascaded convolutional networks. These networks predict a series of incremental deformation fields that transform the moving image at various spatial frequencies, ensuring accurate alignment with the fixed image. This multi-resolution approach allows for a more accurate and detailed registration process, capturing both coarse and fine image structures.We demonstrate the superior performance of our method by a significant margin compared to other state-of-the-art methods, including other multi-resolution techniques. Our method outperforms existing state-of-the-art techniques, including other multi-resolution strategies, by a substantial margin. Additionally, we integrate this cascaded registration framework into a multi-atlas segmentation pipeline, demonstrating competitive performance against the nnU-Net while using a significantly smaller set of annotated images as atlases. This approach is particularly valuable in the context of fetal brain MRI, where annotated datasets are limited. Our pipeline for registration and multi-subject segmentation is publicly available at \url{https://github.com/ValBcn/CasReg}.JRC.F.7 - Digital Healt
A comprehensive analysis of midcap enterprises in the EU business economy
Get PDFThis report presents a comprehensive analysis of the role and contribution of small and large midcap enterprises (midcaps) to the EU business economy. Using a sample of micro-data extracted from Orbis and supplemented with indicators from the World Bank Enterprise Survey (WBES), we provide detailed estimates of midcaps' impact on the EU business economy, and of their key characteristics in terms of financial indicators, innovation, and others. In line with previous research, we find that small and large midcaps provide a substantial contribution to the EU business economy in terms of both employment and turnover, particularly in specific industrial ecosystems, such as electronics, energy renewables, and energy intensive industries. When it comes to the financial indicators, some associations with class size are detected. Interpreting these trends, however, requires caution, as the analysis did not control for differences in sector concentration between classes. We also find that small midcaps are more likely to engage in innovation activities, certain operational and management practices, and have also a greater international presence as compared to Small and Medium Enterprises. Results provide up to date understanding of this specific subset of the enterprise population, providing valuable information for policymaking.JRC.S.3 - Science for Modelling, Monitoring and Evaluatio
Robust Precise On Board Orbit Determination exploiting T-RAIM for LEO-PNT
No full textThe emergence of Low Earth Orbit Position, Navigation, and Timing (LEO-PNT) satellite constellations is set to transform the landscape of PNT services, providing Global Navigation Satellite System (GNSS) users with supplementary signals from an extensive network of LEO satellites. These additional signals are intended to significantly enhance both performance and reliability. Numerous initiatives are taking shape globally, with some leveraging existing broadband LEO constellations through a fused approach, while others are creating dedicated LEO infrastructures [1]. Regardless of the approach, all LEO-PNT endeavours offer cost-effective solutions by adopting multi-tiered architectural designs. Within these architectures, the LEO-PNT services depend on spaceborne GNSS receivers. These receivers function as autonomous systems for Orbit Determination and Time Synchronization (ODTS), capable of generating on-board ephemeris and other necessary corrections that are disseminated to the final LEO-PNT end-users. The driving technology behind this opportunity is the innovative onboard Precise Orbit Determination (P2OD) capability, bolstered by Precise Point Positioning (PPP)-like correction signals in space, similar to those offered by Galileo High-Accuracy Services (HAS). The goal is to achieve accurate real time decimeter-level reconstruction of spacecraft orbits and, more critically, to provide unbiased nanosecond-level time synchronization by using GNSS Pulse Per Second (PPS) events. In the LEO-PNT multi-tier solution, the spaceborne GNSS PPS is used to maintain the clock stability and discipline the transmitter generating the navigation signals. This synchronization mechanism significantly affects the precision of the corresponding LEO-PNT observables, such as pseudorange and carrier phase. This framework creates a strong dependency between the Medium Earth Orbit (MEO) constellations and the LEO layer, meaning that any faults in the upper layer can propagate to the lower one generating a cascade effect detrimental for the final end-user. This phenomenon is thoroughly explained in [1], which also suggests that adopting Receiver Autonomous Integrity Monitoring (RAIM)-like capabilities within the spaceborne receivers could be the best recovery. This paper investigates the possibility to extend to spaceborne receivers T-RAIM solutions generally proposed for ground and envisage the possibility to integrate it with the state of art of P2OD HAS based techniques ( [2], [3] ). This work builds upon and enhances the ongoing European Commission (EC) activities [4] , which focus on the development of innovative algorithms and technologies aiming to achieve unparalleled accuracy and reliability in the LEO Space Service Volume (SSV). Although the challenges of on-board RAIM in space extend to both positioning and timing accuracy, our research intentionally begins with the temporal aspect. Within the LEO PNT multi-tier architecture, ephemeris errors can be mitigated by projecting the most recent valid ephemeris, thus affording a grace period for users. However, timing errors cannot be similarly compensated, as they immediately result in undetectable delays in LEO-PNT signal generation. We contend that to ensure terrestrial integrity when using LEO-PNT observations, the timing requirements for P2OD should serve as a mission-critical threshold for the T-RAIM algorithm.JRC.E.2 - Space, Connectivity and Economic Securit
Unlocking the full potential of behavioural insights for policy
Get PDFThis report examines the transformative role of behavioural insights (BI) in EU policymaking, advocating for integrating BI early in the policy cycle to enhance policy effectiveness. It challenges the misconception that BI is limited to designing behavioural interventions with marginal impacts and demonstrates its potential to guide the development of both traditional policy instruments and behavioural interventions. The report underscores the importance of BI in identifying synergies and conflicts between policies across different areas, thereby improving policy coherence. It advocates for the use of BI in combination with systems analysis to achieve systemic changes. The policy relevance of this work lies in its timely contribution to evidence-based approaches, particularly in areas where the human dimension is key to policy success.JRC.S.1 - EU Policy Lab: Foresight, Design & Behavioural Insight
The effect of incorporating Cs, Sr and Eu nitrates on the matrix development of Fe-rich polymers
No full textRadionuclides like 137Cs, 90Sr and 152+154Eu need to be immobilised from liquid radioactive waste to a suitable final encapsulation matrix. Alkali-activated Materials (AAMs) have the potential to be more effective in immobilising Cs+ and Sr2+ than Portland cement because they can produce stable phases and incorporate them into their structure. Less explored in AAMs is their capacity to immobilise Eu-ions. Nanoparticles are investigated for extracting radionuclides from liquid radioactive waste. CeO2 nanoparticles have exhibited great potential in
their ability to sorb Eu3+ but after several adsorption/desorption cycles also these need to be immobilised into a final encapsulation matrix. In this work, AAMs were prepared from synthetic Fe-rich slag. Two sodium silicate ratios were examined and the AAMs were doped with various mixtures of CsNO3, Sr(NO3)2, Eu(NO3)3, and CeO2 nanoparticles. Contaminants added to the AAM matrices can change the properties of the encapsulation system. To understand the impact on the AAM structure and to determine whether the effects are derived from the simulated radioactive Cs+, Sr2+, and Eu3+, the presence of the CeO2 nanoparticles or the presence of the nitrate ions, samples were examined during their matrix development using isothermal calorimetry, and investigations were made on their microstructural and physicochemical characteristics. The introduction of Cs+ to the matrix showed no notable impact on the activation kinetics, but Eu3+ seems to form Eu(OH)3 similarly to Sr2+ which forms Sr(OH)2 reducing the available hydroxides during the activation and ultimately hindering the polymerisation.JRC.G.I.2 - Nuclear Material Researc
Measuring sustainable and inclusive wellbeing: a multidimensional dashboard approach
Get PDFAnnounced in the 2023 Strategic Foresight Report of the Commission, the sustainable and inclusive wellbeing initiative recognizes the usefulness of GDP but also the need for complimentary indicators to fully capture all aspects of the quality of life, inclusiveness, and sustainability.
One of its main objectives is to develop a multidimensional dashboard, which integrates existing tools and frameworks into a set of indicators that provide a holistic view of the wellbeing of people and the planet. This development involved a rigorous process in an inter-service working group, narrowing down over a thousand potential measures to a comprehensive dashboard of 140 and eventually 50 indicators.
Besides documenting the process, this report presents some preliminary analyses based on the dashboard of 50 indicators and corresponding synthetic indices. The analysis shows that the state of wellbeing and its components in the European Union varies across Member States, presenting important examples of a decoupling of wellbeing from income. While there is a general correlation between economic prosperity and wellbeing, there are notable exceptions and trade-offs between different aspects of wellbeing.
In times of renewed discussions around the need to boost EU’s competitiveness, the SIWB dashboard can be a central monitoring tool to make sure that reigniting Europe’s economic engine does not become an end in itself but rather a means for delivering wellbeing to all people of the current and future generations, and to the planet.JRC.B.1 - Economic and Financial Resilienc
MITICA MonItoring Transport Infrastructures with Connected and Automated Vehicles
Get PDFThis technical report is produced as part of the deliverables of the exploratory research project MITICA (European Commission Joint Research Centre) which investigates innovative methods for the indirect monitoring of European bridges in line with the EU priority of "A Europe fit for the digital age". This is achieved by exploiting the technological advances in vehicles and sensors with the scope of addressing the ageing problem in the European infrastructure. In this respect, an experimental drive-by monitoring campaign is performed under laboratory-controlled conditions using a full-scale bridge-like structure of 9-meter long and a lightweight vehicle equipped with sensors. Additional sensors are installed onto the bridge specimen to provide a controlled experimental environment and offer a conventional monitoring solution for verification purposes. The signal acquisition is based on a power-autonomous wireless sensor network, enabling the maximum flexibility within the roving monitoring experimental testing.
In this report, a detailed description of the MITICA experimental campaign is presented,
It is experimentally demonstrated that the iSHM methods are case-specific, depending on the dynamic properties of the two systems involved. The obtained research findings highlight the limitations and constraints within iSHM methods as experimentally identified, and show the direction for future work.JRC.E.3 - Built Environmen
Le programme de subvention des intrants agricoles au Sénégal [The agricultural input subsidy program in Senegal]
No full textDepuis maintenant une quinzaine d’années, le Sénégal a mis en place un programme de subvention d’intrants agricoles à destination de ses producteurs agricoles visant à leur fournir engrais, semences et matériel agricole. Cependant, très peu d’études ont été réalisées pour en évaluer les impacts. Ce rapport présente les résultats détaillés d’une étude visant à évaluer les effets de ce programme dans deux régions, la vallée du fleuve Sénégal et le bassin arachidier, au travers de plusieurs dispositifs de collecte de données, et notamment d’une enquête auprès des producteurs agricoles. Les résultats mettent en lumière plusieurs faiblesses du programme. La mise en œuvre du programme diffère d’une région à l’autre et les critères de sélection des bénéficiaires souffrent d’un certain flou. Les bénéficiaires du programme tendent à être sélectionnés parmi les producteurs les plus grands, mieux éduqués et bénéficiant d’un capital social plus important. L’estimation des effets du programme en prenant en compte ce biais de sélection montre bien un effet positif sur l’utilisation d’engrais inorganique, et dans le cas du bassin arachidier, également sur le recours aux semence certifiées. Cependant, aucun effet ni sur les rendements des principales cultures ni sur les performances économiques des bénéficiaires n’est observé dans les deux régions de l’étude. Cela s’explique notamment par les dysfonctionnements du programme, tels que les retards de livraison des intrants ou leur mauvaise qualité, mais aussi par un effet d’éviction sur les achats d’engrais commerciaux, notamment dans la vallée du fleuve Sénégal, qui résulte d’un ciblage inefficace des bénéficiaires. Au final, le programme de subvention des intrants agricoles ne permet pas d’augmenter la production ou le revenu des producteurs bénéficiaires. Le rapport conclut sur des recommandations pour améliorer l’efficacité de cet outil essentiel de la politique agricole Sénégalaise.
[For the past fifteen years, Senegal has implemented a program to subsidize agricultural inputs for its producers with the aim of providing them with fertilizers, seeds, and agricultural equipment. However, very few studies have been conducted to evaluate its impacts. This report presents detailed results of a study aimed at assessing the effects of this program in the Senegal River Valley and the Bassin Arachidier, using several data collection mechanisms, including a survey of agricultural producers. The results highlight several weaknesses of the program. The implementation of the program varies from one region to another, and the selection criteria lack clarity. Program beneficiaries tend to be selected among the larger producers who are better educated and have greater social capital. Estimating the effects of the program while accounting for this selection bias does show a positive effect on the use of inorganic fertilizer and, in the case of the Bassin Arachidier, also on the use of certified seeds. However, no effect on the yields of the main crops nor on economic performance of farms is observed in either region. This can be explained by the program's deficiences, such as delays in input delivery or input poor quality, but also by a crowding-out effect on the purchases of commercial fertilizers, particularly in the Senegal River Valley, resulting from inefficient targeting of beneficiaries. In the end, the agricultural input subsidy program does not increase production nor the income of beneficiary producers.]JRC.D.4 - Economics of the Food Syste