In this paper, we propose a novel federated deep unrolling method for increasing the accuracy of the Lidar Super resolution. The proposed framework not only offers notable improvements in Lidar-based SLAM methodologies but also provides a solution to the significant cost associated with high-resolution Lidar sensors. Particularly, our method can be adopted by a number of vehicles coordinated with a server towards learning a regularizer - a neural network - for capturing the dependencies of the Lidar data. To tackle this adaptive federated optimization problem effectively, we initially propose a deep unrolling framework, converting our solution into a well-justified deep learning architecture. The learnable parameters of this architecture are directly derived from the solution of the proposed optimization problem, thus resulting in an explainable architecture. Further, we extend the capabilities of our deep unrolling technique by incorporating a federated learning strategy. Our federated deep unrolling model employs an innovative Adapt-then-Combine strategy, where each vehicle optimizes its model and, subsequently, their learnable regularizers are combined to formulate a robust global regularizer, equipped to handle diverse environmental conditions. Through extensive numerical evaluations on real-world Lidar based SLAM applications, our proposed model demonstrates superior performance along with a significant reduction in trainable parameters, with 99.75% fewer parameters compared to state of the art lidar super-resolution deep neural networks. Essentially, this study is the first initiative to combine deep unrolling with federated learning, showcasing an efficient, and data-secure approach to automotive Lidar super-resolution SLAM applications. The source code can be found at: https://github.com/alexandrosgk/Federated-Deep-Unrolling-Lidar-Super-resolution-SLAM.git .

To overcome the security vulnerabilities caused by weak passwords, thus bridge the gap between user friendly interfaces and advanced security features, the Fast IDentity Online (FIDO) alliance defined a number of authentication protocols. The existing literature leverages all versions of the FIDO protocols, without indicating the reasons behind the choice of each individual FIDO protocol (i.e., U2F, UAF, FIDO2). Inevitably, the question “which protocol is more suitable per case” becomes significant. To provide an answer to the previous question, this article performs a thorough comparative analysis on the different protocol specifications and their technological and market support, to identify whether any protocol has become obsolete. To reach to a conclusion, the proposed approach (i) explores the existing literature, (ii) analyses the specifications released by the FIDO Alliance, elaborating on the security characteristics, (iii) inspects the technical adoption by the industry and (iv) investigates the compliance of the FIDO with standards, regulations and other identity verification protocols. Our results indicate that FIDO2 is the most widely adopted solution; however, U2F remains supported by numerous web services as a two-factor authentication (2FA) choice, while UAF continues to be utilised in mobile clients seeking to offer the Transaction Confirmation feature.

The growth of IoT networks necessitates robust and adaptive trust management (TM) systems to ensure secure and reliable interactions between devices. This paper introduces a novel TM framework for IoT devices, leveraging a statistical Markov chain model to calculate dynamic trust scores. Our approach integrates a Multi-Attribute Decision-Making (MADM) methodology to rank devices based on trustworthiness, providing a resource-efficient alternative to machine learning (ML)-based models. In contrast to ML approaches, which often require extensive data and are vulnerable to adversarial attacks, our statistical model provides a resilient and computationally efficient solution suitable for environments with limited data availability. The system architecture combines a Trust Management Server (TMS), an Intrusion Detection System (IDS), and Distributed Ledger Technology (DLT) to secure data integrity and enable real-time trust assessment. Performance evaluations confirm the model’s capacity to manage diverse security threats within IoT ecosystems, while future work will focus on enhancing the system’s adaptability to novel threats, such as zero-day attacks, and exploring alternative decision-making models to improve resilience under uncertainty. This TM approach advances IoT network security by offering a scalable, lightweight solution adaptable to varied IoT environments.

The fifth generation (5G) of mobile networks is designed to mark the beginning of the hyper-connected society through a broad set of novel features and disruptive characteristics, delivering massive connectivity, coverage and availability paired with unprecedented speed, throughput and capacity. Such a highly capable networking paradigm, facilitated by its integrated segments and available subsystems, will propel numerous cutting-edge, innovative and versatile services, spanning every possible business vertical. Augmented, response-capable healthcare services have already been identified as one of the prime objectives of both vendors and customers; therefore, addressing controversies and shortcomings related to the specific field is considered a priority for all stakeholders. The scope of this paper is to present the architectural elements of 5G which enable efficient, remote healthcare services along with emergency health monitoring and response capability. In addition, we propose a holistic scheme based on technical enablers such as Internet-of-Things (IoT) and Fog Computing, for mitigating common issues and current limitations which may compromise the proclaimed service delivery.

In this article, the issue of multi-lane merging control for a group of connected vehicles modeled as third-order nonlinear dynamics in two-dimension (2-D) plane is investigated, wherein uncertain dynamics, external disturbances, and performance constraints are taken into account. To eliminate the unnecessary heading control response, a virtual vehicle strategy is first introduced, and a new tracking model for both spacing and angle is further constructed. Then, based on two pairs of novel finite-time performance functions (FnTPFs), an improved prescribed performance control (PPC) method is proposed, which enables the boundaries of the FnTPFs to be actively adjusted according to the magnitude of the tracking errors, thus playing a positive role in mitigating singularity and accelerating the convergence rate. In what follows, a new control scheme, by the virtue of composite finite-time sliding mode surface, is developed. It is proved that the control algorithm designed in this paper is capable of driving the errors (namely spacing error and heading error) approach to a predefined region with a given time with explicit upper bound, that is the finite-time individual vehicle stability can be guaranteed, and meanwhile, the finite-time string stability and the reachability of finite-time prescribed performance are also realized. In the end, the validity of the proposed scheme is demonstrated by constructing serious of simulations and experiments based on ITS-NEU platform.