Search for: All records

Modern vehicles increasingly rely on advanced driver monitoring systems (DMS) to ensure safety and enhance the driving experience. These systems assess driver status to prevent accidents caused by fatigue, inattentiveness, or intoxication. While some DMS applications process video data on vehicle, many rely on edge or cloud-based solutions, raising significant privacy concerns due to the storage of sensor data from vehicles. Existing approaches, such as de-identification and homomorphic encryption, either impose heavy computational overhead on vehicles or insufficiently address privacy. To overcome these limitations, we present the Privacy-preserving Driver Monitoring System (PDMS), a novel framework based on the additive secret sharing theory and privacy-preserving Transformer-based deep learning models. PDMS creates randomized secret shares from driver’s facial video data on vehicle, processes them independently through privacy-preserving Transformer models on edges, and securely aggregates partial results on vehicle, ensuring vehicles’ sensor data and final results remain protected. This approach reduces the computational load on the vehicle, enabling cost-effective and scalable DMS solutions that protect the privacy of the driver both in transit and in processing. Our contributions include the design and optimization of the PDMS system, incorporating privacy-preserving DNN layers that are capable of processing randomized secret shares. Furthermore, we present a practical system that utilizes a vision transformer (ViT)-based gaze estimation model, demonstrating the effectiveness of PDMS through comprehensive experiments. 

As automotive technology advances, ensuring efficient and reliable vehicle storage systems becomes increasingly important to vehicle edges. Environmental factors, like extreme cold or intense heat, can greatly affect how well these critical components function. In this paper, we study the effects of different temperatures on flash-based vehicle storage systems, especially how these conditions impact data storage workloads, machine learning workloads, and vehicle edge computing by analyzing the read and write performance of car flash memory. Our approach combines environmental simulations, performance testing, and data analysis to examine how temperature changes affect the performance and reliability of vehicle storage. By testing conditions from standard room temperatures to extreme heat, this study explores how such environments influence the speed, dependability, and overall functionality of flash memory in automotive systems. The results show detailed relationships between temperature changes and the speed (throughput) and delay (latency) in flash storage, identifying areas where these systems may be vulnerable or where improvements could be made. Understanding these dynamics is essential for improving the durability and flexibility of automotive storage systems and vehicle edges in various environmental conditions. We have refined our conclusions to note that while our findings provide insights into temperature-related performance shifts, they represent one piece of a broader set of design considerations for engineers and manufacturers. Rather than offering definitive guidance for policymakers, our findings primarily help illustrate potential thermal vulnerabilities, informing ongoing work toward more robust and reliable vehicle storage systems. As the automotive industry continues to innovate, this study offers an initial foundation for future developments in vehicle storage technology. 

Vehicular edge computing relies on the computational capabilities of interconnected edge devices to manage incoming requests from vehicles. This offloading process enhances the speed and efficiency of data handling, ultimately boosting the safety, performance, and reliability of connected vehicles. While previous studies have concentrated on processor characteristics, they often overlook the significance of the connecting components. Limited memory and storage resources on edge devices pose challenges, particularly in the context of deep learning, where these limitations can significantly affect performance. The impact of memory contention has not been thoroughly explored, especially regarding perception-based tasks. In our analysis, we identified three distinct behaviors of memory contention, each interacting differently with other resources. Additionally, our investigation of Deep Neural Network (DNN) layers revealed that certain convolutional layers experienced computation time increases exceeding 2849%, while activation layers showed a rise of 1173.34%. Through our characterization efforts, we can model workload behavior on edge devices according to their configuration and the demands of the tasks. This allows us to quantify the effects of memory contention. To our knowledge, this study is the first to characterize the influence of memory on vehicular edge computational workloads, with a strong emphasis on memory dynamics and DNN layers. 

Abstract On June 6, 2023, the Kakhovka Dam in Ukraine experienced a catastrophic breach that led to the loss of life and substantial economic values. Prior to the breach, the supporting structures downstream of the spillway had shown signs of being compromised. Here, we use multi-source satellite data, meteorological reanalysis, and dam design criteria to document the dam’s pre-failure condition. We find that anomalous operation of the Kakhovka Dam began in November 2022, following the destruction of a bridge segment, which led to persistent overtopping from late April 2023 up to the breach, contributing to the erosion of the spillway foundation. Moreover, our findings also highlight safety and risk-reduction measures pivotal in avoiding such scenarios. To help prevent future disasters, we advocate for greater transparency in the design parameters of key water structures to enable risk management, and conclude that remote sensing technology can help ensuring water infrastructure safety.