2026-03-14
Néstor Leyva-López, Ramón Zatarain-Cabada, María Lucía Barrón-Estrada, Hugo Jair Escalante-Balderas
Automatic personality recognition has become relevant due to the advancement of artificial intelligence and the widespread use of mobile devices. This study proposes predicting personality traits according to the OCEAN model by integrating mobile sensor data and self-reports using deep neural networks. Through a mobile application, sensor data and daily surveys related to device usage were collected. Four model architectures (MLP, CNN, LSTM and Transformer) were evaluated, finding that CNN is most effective with raw data, while the Transformer excels at including temporal and frequential attributes. These results represent a breakthrough in customized and empathic technologies in mobile data-based personality recognition.
2026-03-14
Iván Alfonso Reyes-Portillo, Saúl Rolando Méndez-Elizondo, Jorge Alberto Morales-Saldaña, Jorge Ulises Muñoz-Minjares, Claudia Angelica Rivera-Romero, Dora Luz Castro-López
The use of photovoltaic (PV) systems has experienced rapid development as part of renewable energy sources. These systems require DC/DC converters with wide transformation ranges to provide regulated output voltages. In PV applications isolated from the electrical grid, output voltage levels of 24 V and 48 V have commonly been reported. Due to the inherent low efficiency of PV modules, it is essential that the converter performs highly efficient power processing in order to properly utilize the energy generated. This paper proposes the analysis of a boost converter based on the concept of Partial Power Processing (PPP) as an alternative for photovoltaic applications. The analysis and evaluation of PPP are presented through modeling in the Volt-Ampere area, as well as through the study of the dynamic effects that this type of processing introduces into the system. Additionally, the switched and linear models of the converter are developed, along with the analysis of PPP through the buffer element. Finally, the obtained results through simulations and experimental measurements are presented, demonstrating a 4.21% increase in the overall efficiency of the system compared to the conventional boost converter.
2026-03-14
Murugesan S, Ramjethmalani C H, Navin Sam K, Venkadesan A, Sadheesh Kumar S J
The increasing integration of advanced information and communication technologies in power systems has increased their vulnerability to cyberattacks. False Data Injection Attacks (FDIAs) are a concerning and widely encountered cyberattack. FDIAs pose a critical threat to the security and reliability of modern power systems by manipulating measurement data. Traditional state estimation techniques often fail to detect stealthy FDIAs, particularly in large-scale grids. This paper proposes a Self-Attention-enabled Conditional Generative Adversarial Network with Gradient Penalty (SACGAN-GP) for effective FDIA detection. The framework leverages a self-attention mechanism to capture long-range dependencies among state variables, enhancing feature representation and improving detection performance. The gradient penalty term ensures training stability and mitigates mode collapse, a common issue in standard GANs. SACGAN-GP can effectively utilize limited labelled attack data and learn robust representations of normal and anomalous patterns, without relying heavily on large, labelled datasets as required by supervised models. Experimental validation on IEEE 14-bus and 118-bus test systems demonstrates that the proposed model performs better than other methods. The proposed method achieves an accuracy of 97.06%, F1-score of 98.28%, and an AUC of 100% on the IEEE 14-bus system, while attaining 95.59% accuracy, 97.39% F1-score, and 100% AUC on the IEEE 118-bus system. Detection times remain under 0.52 seconds, confirming the method’s applicability for real-time scenarios. Furthermore, attention heatmaps generated by the model provide interpretable insights into the localized impacts of FDIAs, offering a promising direction for intelligent, secure power grid monitoring.
2026-03-14
Daniel Ulises Campos Delgado
2026-03-14
Maria de Lourdes Angulo Dominguez, Pedro Mejia Alvarez, Rolando Menchaca Mendez, Arturo Yee Rendon
Urban traffic congestion remains a critical challenge for modern cities, impacting travel efficiency, environmental sustainability, and quality of life. This paper introduces the Collective Optimization Scheme (COS), a collaborative routing framework that integrates Best Response Dynamics with Dijkstra’s algorithm to promote cooperative decision-making among drivers. Unlike traditional navigation systems that optimize routes individually, COS computes routes that account for prevailing congestion conditions and aim to minimize the total travel time across all trips. The proposed approach is evaluated through extensive simulations on real-world urban maps, demonstrating substantial reductions in travel time particularly under low to moderate congestion levels. These results highlight COS as a scalable and effective strategy for sustainable congestion management and improved urban mobility.
2026-03-14
Dieu Nguyen-Khanh, Dat Nguyen-Tien, Nguyen Tran, Thai Nguyen-Dinh, Niamat Hussain, Hung Tran-Huy
This paper presents a method to design a high-gain dual circularly polarized (CP) antenna with compact size based on high-order mode patch structure. For high-gain radiation, a square patch is excited to operate in high-order TM03 mode. To suppress grating lobes and reduce the overall size, four open slots are etched on the edges of the patch. For dual-CP realization, a 90 ◦ hybrid coupler is employed as the feeding network. By changing the feeding port, either right-hand CP (RHCP) or left-hand CP (LHCP) with high isolation can be produced. An antenna prototype with overall dimensions of 1 . 49 λ × 1 . 49 λ × 0 . 04 λ at 5.6 GHz is fabricated and measured for validation. The measurements demonstrate that the proposed design achieves good dual-CP performance around 5.6 GHz with a peak realized gain of approximately 12 dBi. In comparison with the high-gain dual-CP antennas using array or Fabry-Perot structures, the proposed approach has the advantage of achieving high gain with a compact and simple configuration.
2026-03-14
Carolina Quintero, Alexander Andrade, Erith Alexander Muñoz
This study investigates the potential of structural break detection in stock price time series as a tool for investment decision-making in emerging markets. Operating under the hypothesis that structural breaks reflect shifts in underlying price trends, we conduct an empirical analysis of investment performance in the Quito Stock Exchange (QSE) using monthly average prices from 2013 to 2022 for the ten most actively traded companies, selected for their transaction volume and sectoral representativeness. Importantly, this period coincided with the COVID-19 pandemic, providing a natural context to explore how structural breaks behave under heightened market volatility. Two algorithms—CUSUM and BFAST—are applied and compared in terms of their ability to identify actionable breakpoints and generate profitable buy/sell signals. Results show that BFAST, originally developed for remote sensing applications, consistently outperforms CUSUM: it detects a higher proportion of successful signals, yields stronger average returns over a six-month evaluation window (+17% in the financial sector and +18.75% in the productive/commercial sector), and achieves superior risk-adjusted performance as measured by Sharpe ratios. Statistical validation using the Wilcoxon signed-rank test confirms the significance of BFAST’s advantage (p = 0.004). Taken together, these findings position BFAST as a robust and economically relevant tool for financial time-series analysis, extending its utility beyond traditional domains and offering in- vestors a methodologically sound framework for decision-making in volatile market environments.