This paper proposes a centralized autonomous intersection management scheme for heterogeneous connected vehicles (HCVs). Contributions of this work are as follows. First, we sustainably classify heterogeneous vehicles with their distinctive safety-related characteristics. Second, we conduct a safe and efficient coordination algorithm with respect to some criteria such as vehicle types, road priorities and right of way rules. Third, we consider the impact of different road conditions, vehicle characteristics, load, and braking technology on the system performance. Forth, we demonstrate the efficiency of the system under various traffic densities with symmetric and asymmetric vehicle distribution. Besides, system performance is to be compared with traffic lights (TLs) scenarios in terms of throughput, average travel time (ATT), intersection busy time (IBT), channel busy rate (CBR), and packet loss rate (PLR) in various road conditions.
It is posited that Low-rate DoS (LDoS) attack against TCP short transfer has not yet been studied. Since LDoS attacks use pulse-shaped attack traffic, the probability of an attack pulse colliding with the targeted traffic is low when the transfer time is short. In this study, we investigated the feasibility of the Shrew LDoS method against short transfers. In the Shrew method, the time difference between the targeted traffic and the attack traffic, i.e., the attack timing skew, has a large impact on the attack effectiveness for short transfers. Therefore, we proposed the First-Attack Pulse Width Expansion Shrew (Fawe-Shrew) method to improve the attack effectiveness in the presence of this skew. We confirmed that the proposed method has increased tolerance of the skew in the attack initiation timing.
Recent advances in the field of deep learning technologies have made it possible to develop practical video analysis systems with embedded platform that are more accurate and faster than prior embedded systems based on pattern analysis technology. In video analysis applications, object detection, face recognition, action recognition and super resolution technology is the most important functions. In this paper, we show action recognition and super resolution on embedded deep learning system. It is discovered that ResNet34 structure with 16 frames analysis is most profitable for speed and accuracy and Attention based super resolution method is enough for real-time processing. Both deep learning models optimized for speed and accuracy are operated on embedded system with Intel NPU at real-time. We introduce main technologies in chapter 1. Proposed method is shown in chapter 2 and the result is shown in chapter 3.
In this paper, a solar tracking system with feedback controller and state estimation filter is designed with consideration of unpredictable disturbance and feedback sensor noise and verified through various computer simulations. It is verified that the designed solar tracking system with PI controller and Kalman filter has the ability to reject disturbance and reduce feedback sensor noise.
The Home IoT device market is expanding rapidly, which has led to the increasing adoption of these devices in the home. However, IoT devices have long been challenged by various vulnerabilities and pose unique security challenges compared to other devices. This paper aims to redefine the process of entering the Home IoT environment by leveraging Decentralized Identifier(DID) authentication, and proposes a comprehensive approach to using Verifiable Credential(VC) and Verifiable Presentation(VP) to control access from outsiders like visitors. Through security analysis, this paper also highlights how these authentication and permission controls address traditional vulnerabilities and provide improved security over traditional IoT authentication methods. The results of this study shed light on the significant impact this methodology can have on the security of the Home IoT environment.
The development of a comprehensive and decisive drone defense integrated control system (DDS) that can provide maximum security is crucial for smart aerial mobility to sustain the emerging drone transportation system (DTS) for priority-based logistics and mobile communication. This study developed a robust drone defense control system that uses blockchain technology to verify and authenticate the legitimacy of a drone operation and verify its package delivery to forestall possible disruption of aerial vehicular mobility, malicious attacks, and invasions. The system accepts inputs about the drone information and attached package in real-time over the cloud and verifies them using a proof of authority algorithm and smart contract and authenticates them using a decentralized application that is deployed on the drone, the ground control station, and the DDS before executing the appropriate neutralization response. To maintain a proper balance between scalable authentication, security, and speed of responsiveness in taking action, the legacy and the proposed blockchain-assisted approach run in parallel, and the disparity in the result is adjudged a perceived illegitimate aerial mobility intention in real-time. With the deployment of the BANDA on three (3) public networks, the Avalanche Fuji achieved a 97.7 ms authentication latency adjudging it an efficient and effective drone defense security approach for ensuring the sustainability of DTS as an autonomous vehicle for mobility.
This paper proposes a hospital assignment scheme for ambulances transporting patients, aiming at efficiently re-matching both parties even in the case of a demand surge. Built on top of the minimum cost maximum flow model, the allocation process adjusts link connectivity in the flow graph according to whether a patient can be cured at a hospital, reach within a deadline, and be reallocated to another hospital even while on the move. The prototype implementation using the Java language shows that the proposed scheme can reallocate patients with stable reassignment overhead, with the total moving distance being managed below 4.3 %.
In this paper, power control is considered to improve the performance of uplink random access in non-orthogonal multiple access (NOMA) employing imperfect successive interference cancellation (SIC). First, the effect of imperfect SIC is modeled exploiting the sigmoid function. Then, in order to improve the performance of the NOMA SIC receiver, we propose an uncoordinated power control method that makes the power received from each terminal different. Based on the simulation results, it is shown that the proposed scheme results in performance improvement compared to the conventional power control method in the orthogonal multiple access (OMA) or other power control schemes in NOMA.
The paper considers the delay performance of underwater acoustic networks. The focus is on the network delay when both the queueing delay and the propagation delay are considered. The queueing delay is evaluated for the case of general queues. General queues are of interest since the traffic in underwater acoustic networks cannot necessarily be modeled by a Poisson distribution. The propagation delay is of interest due to the low speed at which sound propagates underwater. Numerical examples are presented that illustrate the delay performance for various distances between the network nodes.
For the decision-making problem involving the interaction of an autonomous vehicle with other participating vehicles during the process of autonomous valet parking (AVP), an interactive decision-making method is proposed based on non-cooperative complete information static game theory. Firstly, the conflict types that occur between participating vehicles during the AVP process are analyzed, specifically in the intersection scenario and the front of the parking spot scenario within the parking lot. And all conflict modes are systematically summarized based on the inductive method. Then, the non-cooperative complete information static game is adopted to model the vehicle-vehicle interaction process, additionally, an interactive game cost function is designed by integrating driving style, driving safety, and traffic efficiency. Finally, the optimal driving strategy is obtained by solving the Nash equilibrium solution of the non-cooperative complete information static game. Through simulating various diverse vehicle-vehicle interaction scenarios in the parking lot, the experimental results demonstrate that this method dynamically determines correct and reasonable driving strategies and the effectiveness of the proposed method is verified.
The problem of route planning is important for setting up communication base station based on unmanned aerial vehicle (UAV). This paper proposes a route planning algorithm by creating control points and route patterns in the resource-limited scenario of maritime communication. The algorithm could realize the route planning under coupling constraints of the UAV communication base stations’ total number, transmission distance, flight speed and turning radius. The simulation result shows a significant improvement than existing methods. The minimum on-line rate of the users obtained by the algorithm proposed in this paper is up to 33% higher than the former algorithm, and up to 54% higher than distribute the UAVs evenly. The increase of the average on-line rate also gets a slightly improvement up to 16% and 27% respectively.
The Internet of Things (IoT) is paving the way for the development of Cyber-Physical Systems (CPS), the next step of the Internet evolution, which will allow the development of several new systems and applications. Likewise, urged by the adoption of 5G and Beyond networks, the massive, ubiquitous spread of interconnected IoT devices has increasingly exposed the vulnerability of data and related applications in an unprecedented way. If the security of any component in such systems gets compromised, affecting its trust with respect to others, an associated data leak may cause serious threats to privacy, material losses, and even put people’s lives at risk. In this paper, we present IoT devices’ traffic characterization to provide trust values to enable secure communications among such devices. We develop experiments using a real IoT dataset to demonstrate the feasibility and the effectiveness of our proposal. Considering that complementary features between blockchain technology and information theory triggers a great potential for research and innovation, the key idea of the contribution consists in modeling trust using a two-level approach, which is based on a distributed-ledger (at the high level), and a relative entropy measure (at the low level). The results show the feasibility of our approach.
The increasing popularity of Internet of Things (IoT) devices has brought significant security challenges to IoT networks. However, most deep learning-based anomaly detection solutions often require high computation performance so that it is difficult to be implanted on low-end IoT devices with limited power and memory capacity. In this paper, we propose a low-complexity network anomaly detection method based on feature selection using the Shapley value for the Isolation Forest algorithm. The proposed feature selection method using the Shapley value can reduce the dimension of input data, thereby improving the performance with reduced computational complexity. We provide simulation results to demonstrate the effectiveness of the proposed method. The results show that the proposed method based on Isolation Forest achieves comparable performance to the deep learning method based on neural networks while using fewer dimensions than the deep learning method.
This paper proposes a technique to translate Topcon to Eidon fundus photography using a multimodal image-to-image translation technique based on the BicycleGAN model. The proposed approach aims to address the limitations of Topcon-type fundus photography and leverage the advantages of Eidon-type fundus photography, which captures high-resolution images of the retina using a confocal scanning technique. To this end, we constructed a dataset comprising 475 pairs of Topcon and Eidon fundus images from the same subjects, and we used it to train the proposed model. We evaluated the generated Eidon-type fundus images qualitatively and quantitatively using various image quality metrics, including SSIM, MSE, PSNR and FID. Our results demonstrate that the proposed technique can effectively translate Topcon-type fundus photography to high-quality Eidon-type fundus photography. This technique could potentially enhance the diagnosis and treatment of retinal diseases by providing high-resolution images that capture fine details of the retina.
Wireless technology has faced technical challenges that have been unresolved or only partially addressed. Issues such as modeling the wireless channel and selecting the optimum signal This paper proposes using Artificial Intelligence (AI) to tackle these concerns. Machine Learning (ML) can estimate wireless channel states based on available data. Regression and classification techniques have been used to improve communication and meet 5G standards. The effectiveness of ML and Deep Learning techniques were compared to achieve the best accuracy. This paper shows how AI can revolutionize the design of 5G-NR and future generations with an accurate prediction of 99.99%.
This paper presents a soft-error injection system on the FPGA platform using an ARM Cortex-M3 processor. The system consists of an error injector that generates random numbers using a Fibonacci linear feedback shift register to introduce errors into an application program status register. The efficacy of the soft-error injection system is assessed and validated on an FPGA platform employing Xilinx XCKU115. Additionally, a user interface is created to configure and test the error-injection system. The system is capable of simulating both single-event transient and single-event upset errors, and it can determine the error duration before returning to the previous error-free state. The soft-error injection system is a valuable tool for evaluating the dependability of embedded systems used in industries such as automotive, industrial control, and consumer electronics.
