The recent trend for vehicles to be connected to unspecified devices, vehicles, and infrastructure increases the potential for external threats to vehicle cybersecurity. Thus, intrusion detection is a key network security function in vehicles with open connectivity, such as self-driving and connected cars. Specifically, when a vehicle is connected to an external device through a smartphone inside the vehicle or when a vehicle communicates with external infrastructure, security technology is required to protect the software network inside the vehicle. Existing technology with this function includes vehicle gateways and intrusion detection systems. However, it is difficult to block malicious code based on application behaviors. In this study, we propose a machine learning-based data analysis method to accurately detect abnormal behaviors due to malware in large-scale network traffic in real time. First, we define a detection architecture, which is required by the intrusion detection module to detect and block malware attempting to affect the vehicle via a smartphone. Then, we propose an efficient algorithm for detecting malicious behaviors in a network environment and conduct experiments to verify algorithm accuracy and cost through comparisons with other algorithms.

The communication and connectivity functions of vehicles increase their vulnerability to hackers. The unintended failure and malfunction of in-vehicle systems caused by external factors threaten the security and safety of passengers. As the controller area network alone cannot protect vehicles from external attacks, techniques to analyze and detect external attacks are required. Therefore, we propose a multi-labeled hierarchical classification (MLHC) intrusion detection model that analyzes and detects external attacks caused by message injection. This model quickly determines the occurrence of attacks and classifies the attack using only existing classified attack data. We evaluated the performance of the model by analyzing its learning space. We further verified the model by comparing its accuracy, F1 score and data learning and evaluation times with the two layers multi-class detection (TLMD) and single-layer multi-class classification (SLMC) models. The simulation results show that the MLHC model has the highest F1 score of 0.9995 and is 87.30% and 99.92% faster than the SLMC and TLMD models in terms of detection time, respectively. Consequently, the proposed model can classify both the type and existence or absence of attacks with high accuracy and can be used in interior communication environments of high-speed vehicles with a high throughput.

On a connected car, the performance of Internet access will significantly affect the user experience. For electric cars that use vehicle-to-grid (V2G) communication to interact with the Internet during charging, the charge cable quality poses a challenge to the V2G communication. Specifically, the performance of Transmission Control Protocol (TCP), the transport protocol that most Internet applications use, may suffer due to the high noise and consequent errors that the charge cable presents. Currently, TCP NewReno is the TCP implementation that ISO 15118 standard stipulates for the V2G communication. However, its congestion control algorithm has been designed for the general Internet environment where congestion, not link errors, account for most of packet losses. Indeed, we confirm that the throughput of TCP NewReno rapidly degrades as the error rate increases on the charge cable. Specifically, we show that other TCP variants such as TCP Illinois far exceeds TCP NewReno in both lossy and non-lossy link environments. Finally, we propose how to configure TCP NewReno parameters to make it achieve the throughput comparable to other TCP variants on V2G communication environments, regardless of the link quality presented by the charging cable.

Humans possess the remarkable capacity to comprehend narratives presented in text and subsequently conjure associated mental images through their imagination. This cognitive ability enhances their grasp of the content and augments their overall enjoyment. Consequently, the development of an automated system aimed at producing visually faithful images based on textual descriptions, often referred to as the text-to-image task, stands as a profoundly meaningful endeavor. For this reason, a variety of text-to-image generating artificial intelligences (AIs) have been devised until now. Nevertheless, the generative AIs introduced thus far encounter an issue wherein they struggle to uphold the coherence of input sentences, particularly when multiple sentences are provided. Within this paper, we present a remedy to this challenge through the application of prompt editing. Furthermore, our experimental results substantiate that our proposed solution more effectively preserves contextual coherence among the generated images in comparison to other preexisting generative artificial intelligence models. The experimental results demonstrate that the proposed scheme improves performance by at least 30 percent in terms of the similarity of the generated image and by 130 percent in terms of ROUGE_recall .