Mu Han’s research while affiliated with Jiangsu University and other places

Accurate obstacle detection is crucial for automated parking systems (APS), yet existing methods often face limitations such as low accuracy in ultrasonic sensors, the inability to capture three-dimensional (3D) information with vision-based techniques, and the high costs and complexity associated with multi-sensor fusion. This study introduces GridReconNet, a novel 3D obstacle detection model for APS, which combines grid-structured light projection with deep learning techniques. The model integrates recurrent residual convolutional units, deformable convolutions, and attention mechanisms, significantly enhancing feature extraction, reducing noise, and enabling precise 3D depth reconstruction from two-dimensional (2D) images. Experimental results demonstrate that GridReconNet achieves a 122.0% improvement in Structural Similarity Index (SSIM) and a 15.6% improvement in Peak Signal-to-Noise Ratio (PSNR) compared to the baseline UNet, outperforming R2U-Net and DUNet models. While introducing a moderate increase in computational complexity, the approach offers significant improvements in accuracy and robustness, providing an efficient and practical solution for 3D obstacle detection in automated parking environments. Future research will focus on optimizing the model for lightweight applications and enhancing its generalization in dynamic and complex parking scenarios.

The correct use of high beams is essential when driving at night, and improper use can lead to glare or even temporary blindness for oncoming drivers, increasing the risk of accidents. Although adaptive driving beam system can mitigate this issue, shortcomings remain in environmental perception and beam control accuracy. To address these deficiencies, this paper proposes the following improvements. First, we propose an improved Yolov5s-CBAM-Pruning visual detection model and a two-level fusion method based on camera and millimeter-wave radar for “target-decision” fusion. By adding the CBAM attention module into the Yolov5s backbone network and applying channel pruning, the model’s detection performance is enhanced. The Global Nearest Neighbor algorithm enables target-level fusion between the camera and radar, followed by decision-level fusion for unmatched targets. Second, we proposes a light attenuation model based on the “distance-angle” framework, achieving precise beam control by turning off light sources that exceed the glare threshold. The experiments show that the improved Yolov5s-CBAM-Pruning model achieves a 3.2% increase in mAP and a 34.81% reduction in complexity compared to Yolov5s. The two-level fusion detection accuracy improves by 10.4% compared to single-vision methods. An adaptive LED shutdown control was implemented with Arduino UNO, validating the strategy’s effectiveness.

Vehicular ad-hoc networks (VANET) are crucial for improving road safety and traffic management in Intelligent Transportation Systems (ITS). However, these networks face significant security and privacy challenges due to their dynamic and decentralized nature. Traditional authentication methods, such as Public Key Infrastructure (PKI) and centralized systems, struggle with scalability, single points of failure, and privacy issues. To address these issues, this paper introduces DBVA, a Double-Layered Blockchain Architecture that integrates private and consortium blockchains to create a robust and scalable authentication framework for VANET. The DBVA framework segregates public transactions, such as traffic data, from private transactions, such as identity and location information, into separate blockchain layers, preserving privacy and enhancing security. Additionally, DBVA introduces strict access control smart contracts for the decentralized revocation of unauthorized vehicle privileges, minimizing communication risks and enhancing system resilience. A dynamic pseudonym identity generation mechanism with periodic updates further strengthens privacy by segregating real and pseudonymous identities into separate blockchain layers. Comprehensive performance evaluations reveal that DBVA significantly enhances computational efficiency, reducing the computational cost to 18.34 ms, lowering communication overhead to 992 bits per message, and minimizing storage requirements to just 50 units, making it competitive among contemporary schemes. Extensive security analysis and formal proof confirm that DBVA effectively meets all essential privacy and security requirements, making it a robust, reliable, and scalable solution for enhancing the security, privacy, and resilience of VANET.

According to research, the vast majority of road accidents (90%) are the result of human error, with only a small percentage (2%) being caused by malfunctions in the vehicle. Smart vehicles have gained significant attention as potential solutions to address such issues. In the future of transportation, travel comfort and road safety will be ensured while also offering several value-added services. The automotive industry has undergone a significant transformation through the use of emerging technologies and wireless communication channels, resulting in vehicles becoming more interconnected, intelligent, and safe. However, these technologies and communication systems are susceptible to numerous security attacks. The objective of this paper is to present a comprehensive overview of the smart vehicle’s architecture, encompassing emerging technologies and security challenges and solutions associated with smart vehicles. There has been a significant surge in the utilization of machine learning techniques in smart vehicles. We categorically discuss common security measures, including machine learning and deep learning based solutions that have been mentioned in the literature and implemented against security threats on smart vehicles. This paper has also been titled a tutorial due to its layout, which begins with covering preliminary knowledge, terminologies, and encompassing technologies required to comprehend smart vehicles. Following this, the paper addresses the overall challenges associated with smart vehicles and then focuses on security issues. In terms of solutions, the paper discusses overall solutions to security issues in smart vehicles before delving into a specific solution based on machine learning and deep learning.

This paper addresses the critical challenge of privacy in Online Social Networks (OSNs), where centralized designs compromise user privacy. We propose a novel privacy-preservation framework that integrates blockchain technology with deep learning to overcome these vulnerabilities. Our methodology employs a two-tier architecture: the first tier uses an elitism-enhanced Particle Swarm Optimization and Gravitational Search Algorithm (ePSOGSA) for optimizing feature selection, while the second tier employs an enhanced Non-symmetric Deep Autoencoder (e-NDAE) for anomaly detection. Additionally, a blockchain network secures users’ data via smart contracts, ensuring robust data protection. When tested on the NSL-KDD dataset, our framework achieves 98.79% accuracy, a 10% false alarm rate, and a 98.99% detection rate, surpassing existing methods. The integration of blockchain and deep learning not only enhances privacy protection in OSNs but also offers a scalable model for other applications requiring robust security measures.

The automotive sector faces escalating security risks due to advances in wireless communication technology. Expanding on our previous research using a sensor pairing technique and machine learning models to evaluate IoT sensor data reliability, this study broadens its scope to address security concerns in Connected and Autonomous Vehicles (CAVs). The objectives of this research include identifying and mitigating specific security vulnerabilities related to CAVs, thereby establishing a comprehensive understanding of the risks these vehicles face. Additionally, our study introduces two innovative pairing approaches. The first approach focuses on pairing Electronic Control Units (ECUs) within individual vehicles, while the second extends to pairing entire vehicles, termed as vehicle pairing. Rigorous preprocessing of the dataset was carried out to ensure its readiness for subsequent model training. Leveraging Support Vector Machine (SVM) and TinyML methods for data validation and attack detection, we have been able to achieve an impressive accuracy rate of 97.2%. The proposed security approach notably contributes to the security of CAVs against potential cyber threats. The experimental setup demonstrates the practical application and effectiveness of TinyML in embedded systems within CAVs. Importantly, our proposed solution ensures that these security enhancements do not impose additional memory or network loads on the ECUs. This is accomplished by delegating the intensive cross-validation to the central module or Roadside Units (RSUs). This novel approach not only contributes to mitigating various security loopholes, but paves the way for scalable, efficient solutions for resource-constrained automotive systems.

The application of convolutional neural network (CNN) has greatly broaden the application range of intelligent fault diagnosis, which is beneficial to the stability of industrial production. However, traditional CNN based fault diagnosis methods cannot capture the global features. To address the problem, this paper proposed a method called PSO-Conformer to extract features integrating global and local information. First, convolutional layers are uesd for local feature extraction, and then using Encode module of Transformer to extract global features. The diagnostic performance of the model can be significantly improved and the model hyperparameters are optimized using the PSO algorithm. The proposed method is applied to the fault diagnosis of pneumatic brake systems of commercial vehicles and compared with several methods based on deep learning. And this paper use the t-Distributed Stochastic Neighbor Embdedding (TSNE) to visualize the output features of models in two dimensional plane and four evaluation indexes based on confusion matrix are used to evaluate the model. The results show that the diagnostic performance of the proposed method is better than the existing method.

Outsourcing storage and computation to cloud servers have become a trend. Although searchable symmetric encryption (SSE) had handled the data privacy issue caused by honest‐but‐curious servers, a semi‐honest server may return incomplete or incorrect results when users search for their encrypted data. To against such servers, scholars have recently used blockchain/Ethereum‐based SSE schemes which utilize the public, that is, active nodes, to verify the search process. However, the search operation in existing schemes is very expansive in terms of fee and time. In this paper, we propose a new blockchain‐based searchable encryption framework with search optimized, that is, free of charge, quicker, and more private, at the cost of some extra storage. Besides, we design a general and efficient verification algorithm for our framework, which makes the search verifiable. In addition, we deploy an instance of our framework on an official Ethereum test network, and the experimental results and evaluations demonstrate the advantage of our framework.

With the rapid development of vehicle intelligence, the in-vehicle network is no longer a traditional closed network. External devices can be connected through Bluetooth, WiFi or OBD interfaces, so that attackers can remotely attack vehicles through these channels. Hence we create one-time pads to protect the in-vehicle network. Intelligent connected vehicle (ICV) is an information physical system, thus finding a suitable entropy source from its physical properties to extract true random numbers as a one-time pad can well ensure the security of ICV. During the driving process of ICV, the driving decision will change in real time, and these changes will directly act on the generator of the vehicle’s power system, causing the voltage to change in real time. Therefore, we observe that the on-board power voltage of ICV is a very useful source of entropy. We propose a scheme to extract random numbers from the voltage entropy source. First, we filter the weak periodicity in the voltage signal using wavelet variations. After obtaining the non-periodic voltage signal, we fuse the high voltage time interval with it as a second entropy source to improve the extraction efficiency of the random numbers. Secondly, we build Markov chains by analysing the partial autocorrelation coefficient of the quantized bits of one trace. Finally, we extract perfect random numbers from the Markov chain by using cascaded XOR and hash function. Extensive realistic experiments are conducted to validate our scheme.