Electronics

30 pages, 31241 KiB

Open AccessArticle

Coupled Sub-Feedback Hyperchaotic Dynamical System and Its Application in Image Encryption

by Zelong You, Jiaoyang Liu, Tianqi Zhang and Yaoqun Xu

Electronics 2025, 14(10), 1914; https://doi.org/10.3390/electronics14101914 - 8 May 2025

Images serve as significant conduits of information and are extensively utilized in several facets of life. As chaotic encryption evolves, current chaotic key generators have grown increasingly prevalent and susceptible to compromise. We present an advanced chaos architecture that integrates numerous nonlinear functions [...] Read more.

Images serve as significant conduits of information and are extensively utilized in several facets of life. As chaotic encryption evolves, current chaotic key generators have grown increasingly prevalent and susceptible to compromise. We present an advanced chaos architecture that integrates numerous nonlinear functions and incorporates common chaotic maps as perturbation factors. The produced two-dimensional QWT chaotic map exhibits a more stable chaotic state and a broader chaotic range in comparison to existing maps. Simultaneously, we developed a novel roulette scrambling technique that shifts the conventional in-plane scrambling to cross-plane scrambling. Upon evaluation, the encrypted image demonstrates commendable performance regarding information entropy, correlation, and other parameters, while its encryption algorithm exhibits robust security. Full article

(This article belongs to the Section Computer Science & Engineering)

► Show Figures

Figure 1

20 pages, 1154 KiB

Open AccessArticle

DC-TransDPANet: A Transformer-Based Framework Integrating Composite Attention and Polarized Attention for Medical Image Segmentation

by Wenshu Li, Maoli Zhu and Jianping Xie

Electronics 2025, 14(10), 1913; https://doi.org/10.3390/electronics14101913 - 8 May 2025

Abstract

Medical image segmentation is a critical task in image analysis and plays an essential role in computer-aided diagnosis. Despite the promising performance of hybrid models combining U-Net and transformer architectures, these approaches face challenges in extracting local features and optimizing attention mechanisms. To [...] Read more.

Medical image segmentation is a critical task in image analysis and plays an essential role in computer-aided diagnosis. Despite the promising performance of hybrid models combining U-Net and transformer architectures, these approaches face challenges in extracting local features and optimizing attention mechanisms. To address these limitations, we propose the Depthwise Composite Transformer and Depthwise Polarized Attention Network (DC-TransDPANet), a novel framework designed for medical image segmentation. The proposed DC-TransDPANet introduces a Depthwise Composite Attention Module (DW-CAM), which integrates depthwise convolution, and a Composite Attention mechanism to enhance local feature extraction and fuse contextual information. Additionally, a Depthwise Polarized Attention (DPA) block is employed to improve global context representation while preserving high-resolution details, achieving a fine balance between local and global feature extraction. Extensive experiments on benchmark datasets demonstrate that DC-TransDPANet significantly outperforms existing methods in segmentation accuracy. Full article

► Show Figures

Figure 1

16 pages, 4292 KiB

Open AccessArticle

PreEdgeDB: A Lightweight Platform for Energy Prediction on Low-Power Edge Devices

by Woojin Cho, Dongju Kim, Byunghyun Lim and Jaehoi Gu

Electronics 2025, 14(10), 1912; https://doi.org/10.3390/electronics14101912 - 8 May 2025

Abstract

Rising energy costs due to environmental degradation, climate change, global conflicts, and pandemics have prompted the need for efficient energy management. Edge devices are increasingly recognized for improving energy efficiency; however, their role as primary computing units remains underexplored. This study presents PreEdgeDB, [...] Read more.

Rising energy costs due to environmental degradation, climate change, global conflicts, and pandemics have prompted the need for efficient energy management. Edge devices are increasingly recognized for improving energy efficiency; however, their role as primary computing units remains underexplored. This study presents PreEdgeDB, a lightweight platform deployed on low-power edge devices to optimize energy usage in industrial complexes, which consume approximately 57.29% of South Korea’s total energy. The platform integrates real-time data preprocessing, time-series storage, and prediction capabilities, enabling independent operation at individual factories. A low-resource preprocessing module was developed to handle missing and anomalous data. For storage, RocksDB—a lightweight, high-performance key–value database—was optimized for edge environments. For prediction, Light Gradient Boosting Machine (LightGBM) was adopted due to its efficiency and high accuracy on limited-resource systems. The resulting model achieved a coefficient of variation of the root mean squared error (CV(RMSE)) of 14.36% and a prediction score of 0.8240. The total processing time from data collection to prediction was under 300 milliseconds. With memory usage below 150 MB and CPU utilization around 60%, PreEdgeDB enables fully autonomous energy prediction and analysis on edge devices, without relying on centralized servers. Full article

(This article belongs to the Section Computer Science & Engineering)

► Show Figures

Figure 1

26 pages, 469 KiB

Open AccessArticle

Research on Offloading and Resource Allocation for MEC with Energy Harvesting Based on Deep Reinforcement Learning

by Jun Chen, Junyu Mi, Chen Guo, Qing Fu, Weidong Tang, Wenlang Luo and Qing Zhu

Electronics 2025, 14(10), 1911; https://doi.org/10.3390/electronics14101911 - 8 May 2025

Abstract

Mobile edge computing (MEC) systems empowered by energy harvesting (EH) significantly enhance sustainable computing capabilities for mobile devices (MDs). This paper investigates a multi-user multi-server MEC network, in which energy-constrained users dynamically harvest ambient energy to flexibly allocate resources among local computation, task [...] Read more.

Mobile edge computing (MEC) systems empowered by energy harvesting (EH) significantly enhance sustainable computing capabilities for mobile devices (MDs). This paper investigates a multi-user multi-server MEC network, in which energy-constrained users dynamically harvest ambient energy to flexibly allocate resources among local computation, task offloading, or intentional task discarding. We formulate a stochastic optimization problem aiming to minimize the time-averaged weighted sum of execution delay, energy consumption, and task discard penalty. To address the energy causality constraints and temporal coupling effects, we develop a Lyapunov optimization-based drift-plus-penalty framework that decomposes the long-term optimization into sequential per-time-slot subproblems. Furthermore, to overcome the curse of dimensionality in high-dimensional action, we propose hierarchical deep reinforcement learning (DRL) solutions incorporating both Q-learning with experience replay and asynchronous advantage actor–critic (A3C) architectures. Extensive simulations demonstrate that our DRL-driven approach achieves lower costs compared with conventional model predictive control methods, while maintaining robust performance under stochastic energy arrivals and channel variations. Full article

► Show Figures

Figure 1

23 pages, 6496 KiB

Open AccessArticle

Research on Accurate Fault Location of Multi-Terminal DC Distribution Network

by Zhuolin Chen and Qing Liu

Electronics 2025, 14(10), 1910; https://doi.org/10.3390/electronics14101910 - 8 May 2025

Abstract

The rise of direct current (DC) distribution networks, driven by distributed energy storage and large-scale photovoltaic integration, has significantly altered distribution network configurations. In DC networks, short-circuit faults cause a sharp drop in voltage and a rapid increase in current, negatively impacting system [...] Read more.

The rise of direct current (DC) distribution networks, driven by distributed energy storage and large-scale photovoltaic integration, has significantly altered distribution network configurations. In DC networks, short-circuit faults cause a sharp drop in voltage and a rapid increase in current, negatively impacting system stability. To solve this problem, we used an improved red fox optimization (IRFO) algorithm to calculate the distance to failure of the protection device. The algorithm shows higher convergence and accuracy compared to conventional methods. The isolated forest algorithm rejects anomalous data, while an adjustable feedback factor and genetic crossover operator further improve performance. Adaptive interpolation is employed to address low sampling frequency issues, enhancing fault localization precision. Simulations performed in Simulink show that the method is highly resistant to interference with minimal localization error. It is also resistant to changes in system parameters, highlighting its robustness and usefulness in fault localization. Full article

(This article belongs to the Special Issue Advanced Online Monitoring and Fault Diagnosis of Power Equipment)

► Show Figures

Figure 1

20 pages, 5632 KiB

Open AccessArticle

Filtering Unintentional Hand Gestures to Enhance the Understanding of Multimodal Navigational Commands in an Intelligent Wheelchair

by Kodikarage Sahan Priyanayana, A. G. Buddhika P. Jayasekara and Ruwan Gopura

Electronics 2025, 14(10), 1909; https://doi.org/10.3390/electronics14101909 - 8 May 2025

Abstract

Natural human–human communication consists of multiple modalities interacting together. When an intelligent robot or wheelchair is being developed, it is important to consider this aspect. One of the most common modality pairs in multimodal human–human communication is speech–hand gesture interaction. However, not all [...] Read more.

Natural human–human communication consists of multiple modalities interacting together. When an intelligent robot or wheelchair is being developed, it is important to consider this aspect. One of the most common modality pairs in multimodal human–human communication is speech–hand gesture interaction. However, not all the hand gestures that can be identified in this type of interaction are useful. Some hand movements can be misinterpreted as useful hand gestures or intentional hand gestures. Failing to filter out these unintentional gestures could lead to severe faulty identifications of important hand gestures. When speech–hand gesture multimodal systems are designed for disabled/elderly users, the above-mentioned issue could result in grave consequences in terms of safety. Gesture identification systems developed for speech–hand gesture systems commonly use hand features and other gesture parameters. Hence, similar gesture features could result in the misidentification of an unintentional gesture as a known gesture. Therefore, in this paper, we have proposed an intelligent system to filter out these unnecessary gestures or unintentional gestures before the gesture identification process in multimodal navigational commands. Timeline parameters such as time lag, gesture range, gesture speed, etc., are used in this filtering system. They are calculated by comparing the vocal command timeline and gesture timeline. For the filtering algorithm, a combination of the Locally Weighted Naive Bayes (LWNB) and K-Nearest Neighbor Distance Weighting (KNNDW) classifiers is proposed. The filtering system performed with an overall accuracy of 94%, sensitivity of 97%, and specificity of 90%, and it had a Cohen’s Kappa value of 88%. Full article

► Show Figures

Figure 1

24 pages, 8128 KiB

Open AccessArticle

Model Adaptive Kalman Filter for Maneuvering Target Tracking Based on Variational Inference

by Junxiang Wang, Xin Wang, Yingying Chen, Mengting Yan and Hua Lan

Electronics 2025, 14(10), 1908; https://doi.org/10.3390/electronics14101908 - 8 May 2025

Abstract

This study introduces a new variational Bayesian adaptive estimator that enhances traditional interactive multiple model (IMM) frameworks for maneuvering target tracking. Conventional IMM algorithms struggle with rapid maneuvers due to model-switching delays and fixed structures. Our method uses Bayesian inference to update change-point [...] Read more.

This study introduces a new variational Bayesian adaptive estimator that enhances traditional interactive multiple model (IMM) frameworks for maneuvering target tracking. Conventional IMM algorithms struggle with rapid maneuvers due to model-switching delays and fixed structures. Our method uses Bayesian inference to update change-point statistics in real-time for quick model switching. Variational Bayesian inference approximates the complex posterior distribution, transforming target state estimation and model identification into an optimization task to maximize the evidence lower bound (ELBO). A closed-loop iterative mechanism jointly optimizes the target state and model posterior. Experiments in six simulated and two real-world scenarios show our method outperforms current algorithms, especially in high maneuverability contexts. Full article

(This article belongs to the Special Issue New Insights in Radar Signal Processing and Target Recognition)

► Show Figures

Figure 1

19 pages, 2439 KiB

Open AccessArticle

Mind Mapping Prompt Injection: Visual Prompt Injection Attacks in Modern Large Language Models

by Seyong Lee, Jaebeom Kim and Wooguil Pak

Electronics 2025, 14(10), 1907; https://doi.org/10.3390/electronics14101907 - 8 May 2025

Abstract

Large language models (LLMs) have made significant strides in generating coherent and contextually relevant responses across diverse domains. However, these advancements have also led to an increase in adversarial attacks, such as prompt injection, where attackers embed malicious instructions within prompts to bypass [...] Read more.

Large language models (LLMs) have made significant strides in generating coherent and contextually relevant responses across diverse domains. However, these advancements have also led to an increase in adversarial attacks, such as prompt injection, where attackers embed malicious instructions within prompts to bypass security filters and manipulate LLM outputs. Various injection techniques, including masking and encoding sensitive words, have been employed to circumvent security measures. While LLMs continuously enhance their security protocols, they remain vulnerable, particularly in multimodal contexts. This study introduces a novel method for bypassing LLM security policies by embedding malicious instructions within a mind map image. The attack leverages the intentional incompleteness of the mind map structure, specifically the absence of explanatory details. When the LLM processes the image and fills in the missing sections, it inadvertently generates unauthorized outputs, violating its intended security constraints. This approach applies to any LLM capable of extracting and interpreting text from images. Compared to the best-performing baseline method, which achieved an ASR of 30.5%, our method reaches an ASR of 90%, yielding an approximately threefold-higher attack success. Understanding this vulnerability is crucial for strengthening security policies in state-of-the-art LLMs. Full article

(This article belongs to the Special Issue AI in Cybersecurity, 2nd Edition)

► Show Figures

Figure 1

18 pages, 8552 KiB

Open AccessArticle

PID-NET: A Novel Parallel Image-Dehazing Network

by Wei Liu, Yi Zhou, Dehua Zhang and Yi Qin

Electronics 2025, 14(10), 1906; https://doi.org/10.3390/electronics14101906 - 8 May 2025

Abstract

Image dehazing is a critical task in image restoration, aiming to retrieve clear images from hazy scenes. This process is vital for various applications, including machine recognition, security monitoring, and aerial photography. Current dehazing algorithms often encounter challenges in multi-scale feature extraction, detail [...] Read more.

Image dehazing is a critical task in image restoration, aiming to retrieve clear images from hazy scenes. This process is vital for various applications, including machine recognition, security monitoring, and aerial photography. Current dehazing algorithms often encounter challenges in multi-scale feature extraction, detail preservation, effective haze removal, and maintaining color fidelity. To address these limitations, this paper introduces a novel Parallel Image-Dehazing Network (PID-Net). PID-Net uniquely combines a Convolutional Neural Network (CNN) for precise local feature extraction and a Vision Transformer (ViT) to capture global contextual information, overcoming the shortcomings of methods relying solely on either local or global features. A multi-scale CNN branch effectively extracts diverse local details through varying receptive fields, thereby enhancing the restoration of fine textures and details. To optimize the ViT component, a lightweight attention mechanism with CNN compensation is integrated, maintaining performance while minimizing the parameter count. Furthermore, a Redundant Feature Filtering Module is incorporated to filter out noise and haze-related artifacts, promoting the learning of subtle details. Our extensive experiments on public datasets demonstrated PID-Net’s significant superiority over state-of-the-art dehazing algorithms in both quantitative metrics and visual quality. Full article

► Show Figures

Figure 1

13 pages, 1027 KiB

Open AccessArticle

Vision Transformers for Efficient Indoor Pathloss Radio Map Prediction

by Rafayel Mkrtchyan, Edvard Ghukasyan, Khoren Petrosyan, Hrant Khachatrian and Theofanis P. Raptis

Electronics 2025, 14(10), 1905; https://doi.org/10.3390/electronics14101905 - 8 May 2025

Abstract

Indoor pathloss prediction is a fundamental task in wireless network planning, yet it remains challenging due to environmental complexity and data scarcity. In this work, we propose a deep learning-based approach utilizing a vision transformer (ViT) architecture with DINO-v2 pretrained weights to model [...] Read more.

Indoor pathloss prediction is a fundamental task in wireless network planning, yet it remains challenging due to environmental complexity and data scarcity. In this work, we propose a deep learning-based approach utilizing a vision transformer (ViT) architecture with DINO-v2 pretrained weights to model indoor radio propagation. Our method processes a floor map with additional features of the walls to generate indoor pathloss maps. We systematically evaluate the effects of architectural choices, data augmentation strategies, and feature engineering techniques. Our findings indicate that extensive augmentation significantly improves generalization, while feature engineering is crucial in low-data regimes. Through comprehensive experiments, we demonstrate the robustness of our model across different generalization scenarios. Full article

(This article belongs to the Special Issue Localization, Navigation, and Channel Modeling: AI Solutions for Complex Indoor and Urban Environments)

► Show Figures

Figure 1

14 pages, 4752 KiB

Open AccessArticle

An Ultra-Wideband Low-Noise Amplifier with a New Cross-Coupling Noise-Canceling Technique for 28 nm CMOS Technology

by Yuanping Cui, Kaixue Ma and Kejie Hu

Electronics 2025, 14(10), 1904; https://doi.org/10.3390/electronics14101904 - 8 May 2025

Abstract

This paper presents an ultra-wideband low-noise amplifier (LNA) with a new cross-coupling noise-canceling technique for 28 nm CMOS technology. The entire LNA contains two stages. The first stage employs inductively coupled Gm-boosted technology, while the second stage is a novel asymmetric cross-coupling noise-canceling [...] Read more.

This paper presents an ultra-wideband low-noise amplifier (LNA) with a new cross-coupling noise-canceling technique for 28 nm CMOS technology. The entire LNA contains two stages. The first stage employs inductively coupled Gm-boosted technology, while the second stage is a novel asymmetric cross-coupling noise-canceling structure (ACCNCS). Through the introduction of these two key techniques, the LNA achieves balanced performance across a relative bandwidth of 56%. Input/output/inter-stage impedance matching uses a transformer-based network with series-parallel combinations of inductors and capacitors. The LNA is designed in a 28 nm CMOS process with a chip core area of 335 × 665 µm². The operating frequency range is 26–46 GHz. Post-layout simulation results show that the peak gain of the LNA is 12.6 dB, and the noise figure is between 2.9 and 4.2 dB across the wideband range. At a center frequency of 36 GHz with a supply voltage (V_DD) of 0.9 V, the input 1 dB compression point (IP_1dB) is −7.6 dBm, while the power consumption is 22 mW. Full article

(This article belongs to the Section Microelectronics)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Electronics, Volume 14, Issue 10 (May-2 2025) – 11 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI