Position falsification attacks are one of the most dangerous internal attacks in vehicular networks. Several Machine Learning-based Misbehavior Detection Systems (ML-based MDSs) have recently been proposed to detect these attacks and mitigate their impact. However, existing ML-based MDSs require numerous features, which increases the computational time needed to detect attacks. In this context, this paper introduces a novel ML-based MDS for the early detection of position falsification attacks. Based only on received positions, our system provides real-time and accurate predictions. Our system is intensively trained and tested using a publicly available data set, while its validation is done by simulation. Six conventional classification algorithms are applied to estimate and construct the best model based on supervised learning. The results show that the proposed system can detect position falsification attacks with almost 100% accuracy.
In this present work, we have established exact solutions for (2+1) and (3+1) dimensional extended shallow-water wave equations involving parameters by applying the improved (G'/G) -expansion method. Abundant traveling wave solutions with arbitrary parameter are successfully obtained by this method, and these wave solutions are expressed in terms of hyperbolic, trigonometric, and rational functions. The improved (G'/G) -expansion method is simple and powerful mathematical technique for constructing traveling wave, solitary wave, and periodic wave solutions of the nonlinear evaluation equations which arise from application in engineering and any other applied sciences. We also present the 3D graphical description of the obtained solutions for different cases with the aid of MAPLE 17.
Environmental perception is a key element of autonomous driving because the information received from the perception module influences core driving decisions. An outstanding challenge in real-time perception for autonomous driving lies in finding the best trade-off between detection quality and latency. Major constraints on both computation and power have to be taken into account for real-time perception in autonomous vehicles. Larger object detection models tend to produce the best results, but are also slower at runtime. Since the most accurate detectors cannot run in real-time locally, we investigate the possibility of offloading computation to edge and cloud platforms, which are less resource-constrained. We create a synthetic dataset to train object detection models and evaluate different offloading strategies. Using real hardware and network simulations, we compare different trade-offs between prediction quality and end-to-end delay. Since sending raw frames over the network implies additional transmission delays, we also explore the use of JPEG and H.265 compression at varying qualities and measure their impact on prediction metrics. We show that models with adequate compression can be run in real-time on the cloud while outperforming local detection performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.