Pētniecība
Publikācijas
729 publikācijas
- 2025
Slow Sawtooth Voltage in a Stroboscopic Converter of a Georadar Receiver. Effect of Its Nonlinearity on Signal Distortions(atveras jaunā logā)
Vladimirs Aristovs
Automatic Control and Computer Sciences, Sēj. 59 (3), Pleiades Publishing
Kopsavilkums
Abstract: For a stroboscopic converter, which is a receiver of an ultra-wideband pulsed georadar, the effect of the nonlinearity of the “slow” sawtooth voltage on the uniformity of the input signal sampling during its conversion has been considered. It has been shown that due to the exponential nonlinearity of the specified voltage characteristic of the equivalent RC circuit, the order of the sampling pulses of the stroboscopic receiver obeys the law that is based on the Lambert function. The distortions of the converted pulse signal arising due to the nonlinearity of the “slow” sawtooth voltage have been estimated. © 2025 Elsevier B.V., All rights reserved.
- 2025
Implementation of LoRa TDMA-Based Mobile Cell Broadcast Protocol for Vehicular Networks(atveras jaunā logā)
Modris Greitāns, Gatis Gaigals, Aleksandrs Ļevinskis
Information (Switzerland), Sēj. 16 (6), MDPI
Kopsavilkums
<p>With increasing vehicle density and growing demands on transport infrastructure, there is a need for resilient, low-cost communication systems capable of supporting safety-critical applications, especially in situations where primary channels like Wi-Fi or LTE are unavailable. This paper proposes a novel, real-time vehicular network protocol that functions as an emergency fallback communication layer using long-range LoRa modulation and off-the-shelf hardware. The core contribution is a development of Mobile Cell Broadcast Protocol that is implemented using Long-Range modulation and time-division multiple access (TDMA)-based cell broadcast protocol (LoRA TDMA) capable of supporting up to six dynamic clients to connect and exchange lightweight cooperative awareness messages. The system achieves a sub-100 ms node notification latency, meeting key low-latency requirements for safety use cases. Unlike conventional ITS stacks, the focus here is not on full-featured data exchange but on maintaining essential communication under constrained conditions. Protocol has been tested in laboratory to check its ability to ensure real-time data exchange between dynamic network nodes having 14 bytes of payload per data packet and 100 ms network member notification latency. While focused on vehicular safety, the solution is also applicable to autonomous agents (robots, drones) operating in infrastructure-limited environments. © 2025 Elsevier B.V., All rights reserved.</p>
- 2025
Application of differential privacy to sensor data in water quality monitoring task(atveras jaunā logā)
Audris Arzovs, Sergei V. Parshutin, Valts Urbanovics, Janis Rubulis, Sandis Dejus
Ecological Informatics, Sēj. 86, Elsevier B.V.
Kopsavilkums
Although differential privacy (DP) is used to obfuscate local information and avoid data leakage, very little research exists on the neural network model performance with applied DP for datasets from online drinking water sensor monitoring. This study aims to examine the accuracy of four different neural network model architectures with DP applications. To compare the performance of the neural network model performance in total 2 215 906 augmented and experimentally obtained sensor readings were obtained from a drinking-water pilot system. Three types of contaminations at three different concentrations were applied as scenarios for anomalies in drinking water monitoring. The results achieved similar accuracy with all model architectures, with the best result showing only a 0.3% reduction in model accuracy compared with a nonprivate neural network model with 94% and 94.7% accuracy, respectively. Thus, differential privacy can be applied in the field of water quality monitoring with a reasonable decrease in the model performance. © 2025 Elsevier B.V., All rights reserved.
- 2025
Privacy Auditing of Lithium-Ion Battery Ageing Model by Recovering Time-Series Data Using Gradient Inversion Attack in Federated Learning(atveras jaunā logā)
Kaspars Sudars, Ivars Namatēvs, Arturs Nikulins, Kaspars Ozols
Applied Sciences (Switzerland), Sēj. 15 (10), MDPI
Kopsavilkums
The exchange of gradients is a widely used method in modelling systems for machine learning (e.g., distributed training, federated learning) in privacy-sensitive domains. Unfortunately, there are still privacy risks in federated learning, as servers can reconstruct clients‘ training data through gradient inversion attacks. To protect against such attacks, we need to know when and how privacy is being undermined, largely due to the black box nature of deep neural networks. Although gradient inversion has been used to classify images and text data, its use in time-series is still largely unexplored. In this paper, we empirically evaluate the practicality of recovering lithium-ion battery time-series data from the gradients of a transformer-based model, both without and with differential privacy, in a time-series federated learning framework. It is especially significant in the case of electric vehicles (EVs). As shown in this paper, additional protection by differential privacy leads to the saturation of gradient inversion attacks, i.e., the reconstructed signal maintains a certain error level, depending on the applied privacy budget level. With this empirical evaluation, we provide insights into effective gradient perturbation directions, unfairness with respect to privacy, and privacy-preferred model initialisation. © 2025 Elsevier B.V., All rights reserved.
- 2025
Prediction of energy resolution in the JUNO experiment(atveras jaunā logā)
Ints Mednieks, Ángel Abusleme, Thomas Adam, Kai Adamowicz, Shakeel Ahmad, Rizwan Ahmed, Sebastiano Aiello, Fengpeng An, Qi An, Giuseppe Andronico, Nikolay V. Anfimov and Antonelli
Chinese Physics C, Sēj. 49 (1), Institute of Physics
Kopsavilkums
This paper presents an energy resolution study of the JUNO experiment, incorporating the latest knowledge acquired during the detector construction phase. The determination of neutrino mass ordering in JUNO requires an exceptional energy resolution better than 3% at 1 MeV. To achieve this ambitious goal, significant efforts have been undertaken in the design and production of the key components of the JUNO detector. Various factors affecting the detection of inverse beta decay signals have an impact on the energy resolution, extending beyond the statistical fluctuations of the detected number of photons, such as the properties of the liquid scintillator, performance of photomultiplier tubes, and the energy reconstruction algorithm. To account for these effects, a full JUNO simulation and reconstruction approach is employed. This enables the modeling of all relevant effects and the evaluation of associated inputs to accurately estimate the energy resolution. The results of this study reveal an energy resolution of 2.95% at 1 MeV. Furthermore, this study assesses the contribution of major effects to the overall energy resolution budget. This analysis serves as a reference for interpreting future measurements of energy resolution during JUNO data collection. Moreover, it provides a guideline for comprehending the energy resolution characteristics of liquid scintillator-based detectors. © 2025 Elsevier B.V., All rights reserved.
- 2025
JUNO sensitivity to invisible decay modes of neutrons(atveras jaunā logā)
Ints Mednieks, Ángel Abusleme, Thomas Adam, Kai Adamowicz, Shakeel Ahmad, Rizwan Ahmed, Sebastiano Aiello, Fengpeng An, Qi An, Giuseppe Andronico, Nikolay V. Anfimov and Antonelli
European Physical Journal C, Sēj. 85 (1), Springer Nature
Kopsavilkums
We explore the decay of bound neutrons in the JUNO liquid scintillator detector into invisible particles (e.g., n→3ν or nn→2ν), which do not produce an observable signal. The invisible decay includes two decay modes: n→inv and nn→inv. The invisible decays of s-shell neutrons in 12C will leave a highly excited residual nucleus. Subsequently, some de-excitation modes of the excited residual nuclei can produce a time- and space-correlated triple coincidence signal in the JUNO detector. Based on a full Monte Carlo simulation informed with the latest available data, we estimate all backgrounds, including inverse beta decay events of the reactor antineutrino ν¯e, natural radioactivity, cosmogenic isotopes and neutral current interactions of atmospheric neutrinos. Pulse shape discrimination and multivariate analysis techniques are employed to further suppress backgrounds. With two years of exposure, JUNO is expected to give an order of magnitude improvement compared to the current best limits. After 10 years of data taking, the JUNO expected sensitivities at a 90% confidence level are τ/B(n→inv)>5.0×1031years and τ/B(nn→inv)>1.4×1032years. © 2025 Elsevier B.V., All rights reserved.
- 2025
Evaluation of Subcutaneous and Intermuscular Adipose Tissues by Application of Pattern Recognition and Neural Networks to Ultrasonic Data: A Model Study(atveras jaunā logā)
Alexey Tatarinov, Aleksandrs Sisojevs, Vladislavs Agarkovs and Jegors Lukjanovs
Bioengineering, Sēj. 12, lpp. 1373, MDPI
Kopsavilkums
Distinguishing subcutaneous adipose tissue (SAT) from intermuscular adipose tissue (IMAT) is clinically important because IMAT infiltration is strongly associated with age-related functional decline, sarcopenia, diabetes, cardiovascular disease, and obesity. Current assessments rely on MRI or CT, which are stationary, costly, and labor-intensive. Portable ultrasound-based solutions could enable broader, proactive screening. This model study investigated the feasibility of differentially assessing SAT and IMAT using features extracted from propagating ultrasound signals. Twenty-five phantoms were constructed using gelatin as a muscle-mimicking matrix and oil as the SAT and IMAT compartments, arranged to provide gradual variations in fat fractions ranging from 0% to 50%. Ultrasound measurements were collected at 0.8 MHz and 2.2 MHz, and multiple evaluation criteria were computed, including ultrasound velocity and parameters derived from the signal intensity. Classification domains were then generated from intersecting decision rules associated with these criteria. In parallel, artificial neural networks (ANN/LSTM) were trained and tested on identical phantom subsets to evaluate data-driven classification performance. Both the rule-based and ANN/LSTM approaches achieved diagnostically meaningful separation of SAT and IMAT. The aim of this work was to perform an experimental proof-of-concept study on idealized tissue models to demonstrate that ultrasound measurements can reliably differentiate SAT and IMAT, supporting the development of future screening devices.
- 2025
Depth-Aware RGB and IR Image Alignment Using ToF-Based Homography Interpolation
Roberts Treize
2025 International Conference on Networking, Sensing and Control (ICNSC), IEEE
Kopsavilkums
This paper presents a depth-aware method for pixelwise registration of RGB and infrared (IR) images using Time-of-Flight (ToF) sensing. Unlike fixed homography models that fail under varying scene depths, our approach dynamically adapts the transformation matrix based on per-pixel depth information. We perform a two-depth calibration using a 6×7 checkerboard target, obtaining homographies between the ToF and both RGB and IR image planes. These matrices are linearly interpolated to generate a continuous, depth-dependent transformation model. Each pixel's depth is retrieved from the ToF camera and used to evaluate a tailored homography, which maps coordinates between modalities. To handle missing or uncertain depth data, we apply percentile-based outlier removal and edge-aware interpolation. Final alignment is performed via bilinear resampling. The method is scalable to high-resolution input, and achieves precise multimodal overlay even under non-uniform depth conditions.
Indeksēts Scopus / WoS
- 2025
Real-time accelerator design for lossless compression of 1D data(atveras jaunā logā)
Marko Melnyk, Rihards Novickis
2025 International Conference on Networking, Sensing and Control (ICNSC), lpp. 8, IEEE
Kopsavilkums
The increasing number of Internet of Things devices puts a strain on the available wireless communication bandwidth and consequent power consumption. However, the modular System-on-Chip architecture is well-suited for on-the-fly data processing. Subsequently, this technical work proposes an accelerator-based solution for real-time compression and decompression of one-dimensional data, which consists of an encoder, decoder, predictor, variable-to-fixed and fixed-to-variable converter digital circuits. The solution is tested, validated and characterised with multiple real and synthetic datasets. The experiments demonstrate that using even the simplest predictor type can improve communication bandwidth by up to 49% in a serialised communication scenario.
Indeksēts Scopus / WoS
- 2025
An Approach to Hardware-Accelerated Quantum Sensing Based on NV Centers in Diamonds(atveras jaunā logā)
Andrejs Cvetkovs. Gatis Supols. Rihards Novickis.
2025 International Conference on Networking, Sensing and Control (ICNSC)
Kopsavilkums
Magnetometers based on the nitrogen-vacancy defect centers in diamonds are currently an active research topic due to their high sensitivity and relative ease of use compared to other similar solutions. Despite already being used by numerous research institutions, their widespread adoption is still hindered by high cost and complexity of the required associated equipment. This paper presents a hardware-accelerated measurement system implemented in FPGA chip, as well as the supporting analog electronics and software application. The developed system is capable of ODMR spectrum measurement, lock-in detection and multipoint resonance frequency tracking.
Indeksēts Scopus / WoS
- 2024
Energy Citizenship in Energy Transition: The Case of the Baltic States(atveras jaunā logā)
Ikstena R., Lagzdiņa Ē., Brizga J., Kudrenickis I., Ernšteins R.
Sustainability, Sēj. 16 (22), lpp. 9665
- 2024
Assessment of Various Factors Affecting Economic Indicators in Prosumer and Consumer Energy Communities: A Case Study in Latvia(atveras jaunā logā)
R. Lazdins and A. Mutule
Latvian Journal of Physics and Technical Sciences, lpp. 23-42
- 2024
Comparative Assessment of Prosumer Support Scheme Adoption for Energy Communities in Latvia(atveras jaunā logā)
R. Lazdins, A. Mutule, L. Petrichenko
2024 IEEE 65th International Scientific Conference on Power and Electrical Engineering of Riga Technical University (RTUCON), lpp. 1-5
- 2024
Assessing the Role of Electricity Sharing in Meeting the Prerequisites for Receiving Renewable Support in Latvia(atveras jaunā logā)
L. Petrichenko, A. Mutule, I. Zalitis, R. Lazdins, J. Kozadajevs and D. Mihaila
Sustainability, Sēj. 16 (23), lpp. 10495
- 2024
Automatization of CT Annotation: Combining AI Efficiency with Expert Precision(atveras jaunā logā)
Edgars Edelmers, Dzintra Kazoka, Katrina Bolocko, Kaspars Sudars, Mara Pilmane
Diagnostics, Sēj. 14 (2), lpp. 185, MDPI
Kopsavilkums
The integration of artificial intelligence (AI), particularly through machine learning (ML) and deep learning (DL) algorithms, marks a transformative progression in medical imaging diagnostics. This technical note elucidates a novel methodology for semantic segmentation of the vertebral column in CT scans, exemplified by a dataset of 250 patients from Riga East Clinical University Hospital. Our approach centers on the accurate identification and labeling of individual vertebrae, ranging from C1 to the sacrum–coccyx complex. Patient selection was meticulously conducted, ensuring demographic balance in age and sex, and excluding scans with significant vertebral abnormalities to reduce confounding variables. This strategic selection bolstered the representativeness of our sample, thereby enhancing the external validity of our findings. Our workflow streamlined the segmentation process by eliminating the need for volume stitching, aligning seamlessly with the methodology we present. By leveraging AI, we have introduced a semi-automated annotation system that enables initial data labeling even by individuals without medical expertise. This phase is complemented by thorough manual validation against established anatomical standards, significantly reducing the time traditionally required for segmentation. This dual approach not only conserves resources but also expedites project timelines. While this method significantly advances radiological data annotation, it is not devoid of challenges, such as the necessity for manual validation by anatomically skilled personnel and reliance on specialized GPU hardware. Nonetheless, our methodology represents a substantial leap forward in medical data semantic segmentation, highlighting the potential of AI-driven approaches to revolutionize clinical and research practices in radiology.
Indeksēts Scopus / WoS · Scopus procentile: 60
- 2024
Distributed Learning in the IoT–Edge–Cloud Continuum(atveras jaunā logā)
Arzovs, A.; Judvaitis, J.; Nesenbergs, K.; Selavo, L
Machine Learning and Knowledge Extraction, Sēj. 6 (1), lpp. 283-315., MDPI
Kopsavilkums
The goal of the IoT–Edge–Cloud Continuum approach is to distribute computation and data loads across multiple types of devices taking advantage of the different strengths of each, such as proximity to the data source, data access, or computing power, while mitigating potential weaknesses. Most current machine learning operations are currently concentrated on remote high-performance computing devices, such as the cloud, which leads to challenges related to latency, privacy, and other inefficiencies. Distributed learning approaches can address these issues by enabling the distribution of machine learning operations throughout the IoT–Edge–Cloud Continuum by incorporating Edge and even IoT layers into machine learning operations more directly. Approaches like transfer learning could help to transfer the knowledge from more performant IoT–Edge–Cloud Continuum layers to more resource-constrained devices, e.g., IoT. The implementation of these methods in machine learning operations, including the related data handling security and privacy approaches, is challenging and actively being researched. In this article the distributed learning and transfer learning domains are researched, focusing on security, robustness, and privacy aspects, and their potential usage in the IoT–Edge–Cloud Continuum, including research on tools to use for implementing these methods. To achieve this, we have reviewed 145 sources and described the relevant methods as well as their relevant attack vectors and provided suggestions on mitigation.
Indeksēts Scopus / WoS · Scopus procentile: 91
- 2024
Spatial Transformation Accelerator with Parallel Data Access Scheme for Sample Reconstruction(atveras jaunā logā)
Rihards Novickis, Edgars Lielāmurs, Daniels Jānis Justs, Andrejs Cvetkovs, Kaspars Ozols
Electronics, Sēj. 13 (5), lpp. 922, MDPI
Kopsavilkums
Spatial image transformation is a commonly used component in many image processing pipelines. It enables the correction of optical distortions, image registration onto a common reference plane, electronic image stabilisation, digital zoom, video mosaicking, etc. With the growing tendency to embed image processing in low-power devices, attaining an efficient transformation solution becomes increasingly decisive. Furthermore, interpolation is the key operation in achieving the high quality of the transformed data from the original data. Fortunately, different implementations have already seen several efficiency improvements in recent years. However, interpolation relies on sampling a set of neighbouring points from memory, which has yet to be addressed efficiently for smaller computational platforms with limited memory resources. In this work, we derive a generic mathematical model and circuit design principles for the spatial transformation accelerator design for N-dimensional data. Furthermore, we present an efficient simultaneous access scheme for high-quality signal reconstruction. Finally, the introduced ideas are verified in field programmable gate arrays using one-dimensional and two-dimensional data transformation use cases. The presented solution is able to transform images with sizes ranging from 256 × 256 to 8192 × 8192 and achieves a transfer rate of 275 frames per second with 512 × 512 images.
Indeksēts Scopus / WoS · Scopus procentile: 68 · SNIP: 1.045
- 2024
Hierarchical Integration of UAS and Sentinel-2 Imagery for Spruce Bark Beetle Grey-Attack Detection by Vegetation Index Thresholding Approach(atveras jaunā logā)
Goldbergs, G., & Upenieks, E. M.
Forests, Sēj. 15 (4), lpp. 644, MDPI
Kopsavilkums
This study aimed to examine the efficiency of the vegetation index (VI) thresholding approach for mapping deadwood caused by spruce bark beetle outbreak. For this, the study used upscaling from individual dead spruce detection by unmanned aerial (UAS) imagery as reference data for continuous spruce deadwood mapping at a stand/landscape level by VI thresholding binary masks calculated from satellite Sentinel-2 imagery. The study found that the Normalized Difference Vegetation Index (NDVI) was most effective for distinguishing dead spruce from healthy trees, with an accuracy of 97% using UAS imagery. Based on accuracy assessment, the summer leaf-on period (June–July) was found to be the most appropriate for spruce deadwood mapping by S2 imagery with an accuracy of 85% and a deadwood detection rate of 83% in dense, close-canopy mixed conifer forests.
Indeksēts Scopus / WoS · Scopus procentile: 78
- 2024
Deep Learning for Wind and Solar Energy Forecasting in Hydrogen Production(atveras jaunā logā)
Arturs Nikulins, Kaspars Sudars, Edgars Edelmers, Ivars Namatevs, Kaspars Ozols, Vitalijs Komasilovs, Aleksejs Zacepins, Armands Kviesis, Andreas Reinhardt
Energies, Sēj. 17 (5), lpp. 12, MDPI
Kopsavilkums
This research delineates a pivotal advancement in the domain of sustainable energy systems, with a focused emphasis on the integration of renewable energy sources—predominantly wind and solar power—into the hydrogen production paradigm. At the core of this scientific endeavor is the formulation and implementation of a deep-learning-based framework for short-term localized weather forecasting, specifically designed to enhance the efficiency of hydrogen production derived from renewable energy sources. The study presents a comprehensive evaluation of the efficacy of fully connected neural networks (FCNs) and convolutional neural networks (CNNs) within the realm of deep learning, aimed at refining the accuracy of renewable energy forecasts. These methodologies have demonstrated remarkable proficiency in navigating the inherent complexities and variabilities associated with renewable energy systems, thereby significantly improving the reliability and precision of predictions pertaining to energy output. The cornerstone of this investigation is the deployment of an artificial intelligence (AI)-driven weather forecasting system, which meticulously analyzes data procured from 25 distinct weather monitoring stations across Latvia. This system is specifically tailored to deliver short-term (1 h ahead) forecasts, employing a comprehensive sensor fusion approach to accurately predicting wind and solar power outputs. A major finding of this research is the achievement of a mean squared error (MSE) of 1.36 in the forecasting model, underscoring the potential of this approach in optimizing renewable energy utilization for hydrogen production. Furthermore, the paper elucidates the construction of the forecasting model, revealing that the integration of sensor fusion significantly enhances the model’s predictive capabilities by leveraging data from multiple sources to generate a more accurate and robust forecast. The entire codebase developed during this research endeavor has been made available on an open access GIT server.
Indeksēts Scopus / WoS · Scopus procentile: 80
- 2024
Wearable Device Bluetooth/BLE Physical Layer Dataset(atveras jaunā logā)
Rusins, A.; Tiscenko, D.; Dobelis, E.; Blumbergs, E.; Nesenbergs, K.; Paikens, P.
Data, Sēj. 9 (4), lpp. 53, MDPI
Kopsavilkums
Wearable devices, such as headsets and activity trackers, rely heavily on the Bluetooth and/or the Bluetooth Low Energy wireless communication standard to exchange data with smartphones or other peripherals. Since these devices collect personal health and activity data, ensuring the privacy and security of the transmitted data is crucial. Therefore, we present a dataset that captures complete Bluetooth communications—including advertising, connection, data exchange, and disconnection—in an RF isolated environment using software-defined radio. We were able to successfully decode the captured Bluetooth packets using existing tools. This dataset provides researchers with the ability to fully analyze Bluetooth traffic and gain insight into communication patterns and potential security vulnerabilities.
Indeksēts Scopus / WoS · Scopus procentile: 60