The increase in urban traffic has led to significant challenges, primarily congestion and accidents, many of which stem from human errors, both direct and indirect. Connected, Cooperative, Autonomous and Automated Mobility (CCAM) presents a promising approach to improve traffic efficiency and safety. Achieving these goals requires overcoming several obstacles, including the development of robust communication and data processing systems, advancements in vehicle technology. To fully realize the potential of Autonomous Vehicles (AV) in urban and suburban settings, it is essential to tackle infrastructure deficiencies and to provide an alternative way of navigation when Global Navigation Satellite Systems (GNSS) signals are inaccurate or unavailable. This can be achieved through the integration of technologies such as sensor fusion techniques that combine data from multiple sources like LiDAR (Light Detection and Ranging) and cameras. This research evaluates the readiness of the route for autonomous transportation. It specifically examines the reliability and performance speeds of supporting systems such as Cellular Vehicle-to-Everything (C-V2X) and Vehicle-to-Infrastructure (V2I). The findings indicate critical shortcomings in various areas that must be addressed to optimize functionality. The paper will conclude with recommendations for future research and advancements necessary to further optimize AV navigation both on this route and in broader contexts.

Accurate depth perception is vital for autonomous driving and roadside monitoring. Traditional stereo vision methods are cost-effective but often fail under challenging conditions such as low texture, reflections, or complex lighting. This work presents a perception pipeline built around FoundationStereo, a Transformer-based stereo depth estimation model. At low resolutions, FoundationStereo achieves real-time performance (up to 26 FPS) on embedded platforms like NVIDIA Jetson AGX Orin with TensorRT acceleration and power-of-two input sizes, enabling deployment in roadside cameras and in-vehicle systems. For Full HD stereo pairs, the same model delivers dense and precise environmental scans, complementing LiDAR while maintaining a high level of accuracy. YOLO11 object detection and segmentation is deployed in parallel for object extraction. Detected objects are removed from depth maps generated by FoundationStereo prior to point cloud generation, producing cleaner 3D reconstructions of the environment. This approach demonstrates that advanced stereo networks can operate efficiently on embedded hardware. Rather than replacing LiDAR or radar, it complements existing sensors by providing dense depth maps in situations where other sensors may be limited. By improving depth completeness, robustness, and enabling filtered point clouds, the proposed system supports safer navigation, collision avoidance, and scalable roadside infrastructure scanning for autonomous mobility.

Real-time object detection is critical for Autonomous Vehicles (AVs), but balancing accuracy, latency, hardware and software constraints remains challenging. While automotivegrade systems rely on sophisticated 3D neural networks and multi-sensor fusion for orientation and depth estimation, lightweight 2D detectors like You Only Look Once (YOLO) offer potential for cost-effective, real-time performance in specific perception tasks. The study evaluates YOLOv11's real-time capabilities, focusing on its speed, accuracy, and adaptability to diverse automotive environments, highlighting its potential for integration into autonomous systems. While YOLOv11 is a 2D model, the possibility of integrating YOLOv11 into autonomous vehicle systems for specific real-time applications, leveraging its efficiency and accuracy is explored. This paper also evaluates YOLOv11's viability for real-time object detection in AVs using multiple camera types: high-resolution (4K), low-resolution (720p) and fisheye (distorted field of view), simulating diverse automotive hardware in a dynamic road environment to assess its adaptability to diverse situations. YOLOv11 exhibits significant advancements over its predecessors, showcasing enhanced precision and mean Average Precision (m AP) across diverse classes. With its high inference speed and adaptability to challenging environments, YOLOv11 delivers robust real-time detection while maintaining computational efficiency, making it a potent tool for real-world applications.

Autonomous transportation is heavily reliant on precise and reliable localization systems for safe and efficient navigation in dynamic environments. This paper builds upon previous research conducted on the Zilina city and surrounding area, which identified significant challenges in achieving accurate localization. The route encompasses diverse urban landscapes, posing unique challenges. Through a detailed analysis of network performance parameters, this study aims to pinpoint key improvement areas and propose innovative solutions to enhance localization accuracy. Initial research revealed limitations in smartphone-based GPS localization for localization measurements due to fluctuations in accuracy and environmental factors. To overcome these challenges, the second measurement proposed the integration of GNSS and LiDAR technology, complemented by high-resolution land surveyor maps. By leveraging these technologies, the study seeks to bolster the readiness of the route for autonomous transportation. This study sets the stage for an analysis of the challenges and opportunities in achieving precise localization accuracy, highlighting the importance of robust localization systems for successful autonomous vehicles deployment in urban environments.

The surge in urban traffic has been linked to various challenges, most notably congestion and accidents-many of which result from human errors, both direct and indirect. CCAM (Connected, Cooperative, Autonomous and Automated Mobility) stands as a potential solution to enhance traffic efficiency, safety, and overall user comfort. Realizing these objectives necessitates the resolution of numerous challenges, such as communication and data processing infrastructure, vehicle technology, protection of personal data, and cybersecurity. This study contributes to the assessment of the readiness of the University of Zilina - Vranie Village route for autonomous transport, focusing on the reliability and speeds of the supporting C- V2X (Cellular vehicle-to-everything), V2I (Vehicle-to-Infrastructure) systems. The results reveal crucial deficiencies in various aspects, necessitating improvements for optimal functionality. While some services meet the requirements at high percentages, significant issues persist, especially in vehicle location accuracy and latency. Notably, non-compliance occurrences suggest inadequacies in the 5G network, emphasizing the need for enhanced network infrastructure. Addressing the deficiencies in network infrastructure, predictive models and stability of the network, while also conducting further research, will be crucial for realizing the potential of autonomous vehicles in both urban and suburban areas.

Metamaterial engineering has become established as an essential design tool in silicon photonics. The utilization of state-of-the-art semiconductor manufacturing methods to create metamaterials within optical waveguides has provided unparalleled control over the manipulation of light propagation in silicon photonic chips [1–6]. In this invited presentation, we will review recent breakthroughs in this rapidly advancing field. Additionally, we will introduce a nascent research area of resonant integrated photonics, leveraging Mie resonances in dielectrics for on-chip guidance of optical waves, as exemplified by the recent demonstration of the first Huygens’ metawaveguide [7].

This paper describes design, theoretical analysis, and experimental evaluation of a π-Phase-Shifted Fiber Bragg Grating (π-PSFBG) inscribed in the standard telecom fiber for slow light generation. At first, the grating was designed for its use in the reflection mode with a central wavelength of 1552 nm and a pass band width of less than 100 pm. The impact of fabrication imperfections was experimentally investigated and compared to model predictions. The optical spectra obtained experimentally show that the spectral region used for slow light generation is narrower (less than 10 pm), thus allowing for too-low levels of slow light optical-output power. In the next step, the optimization of the grating design was conducted to account for fabrication errors, to improve the grating’s spectral behavior and its temporal performance, and to widen the spectral interval for slow light generation in the grating’s transmission mode. The targeted central wavelength was 1553 nm. The π-PSFBG was then commercially fabricated, and the achieved parameters were experimentally investigated. For the region of (1551–1554) nm, a 15-fold increase in the grating’s pass band width was achieved. We have shown that a pair of retarded optical pulses were generated. The measured group delay was found to be ~10.5 ps (compared to 19 ps predicted by the model). The π-PSFBG operating in its transmission mode has the potential to operate as tunable delay line for applications in RF photonics, ultra-fast signal processing, and optical communications, where tunable high precision delay lines are highly desirable. The π-PSFBG can be designed and used for the generation of variable group delays from tens to hundreds of ps, depending on application needs.

The surge in urban traffic has spurred the investigation and development of Connected, Cooperative, Autonomous, and Automated Mobility (CCAM) solutions to mitigate challenges such as congestion and accidents. Localization accuracy is one of the most critical components of CCAM systems, ensuring precise positioning. Building upon the insights gained from previous research, which uncovered significant challenges in achieving precise location along the Zilina city and surrounding region, this study focuses on implementing solutions to address these shortcomings. Our previous investigations highlighted fluctuations in location accuracy, attributed to environmental factors and technological limitations. In response, this research aims to enhance localization accuracy through the integration of GNSS and LiDAR technology while utilizing the previously acquired high-resolution land surveyor maps for detailed analysis. By leveraging these approaches, we aim to bolster the readiness of the route for autonomous transportation.

We report on a numerical model to manage the impact of polarization mode dispersion using in-band channel reservation approach. The model is feeded by experimental data based 20 km long fiber link by measuring its polarization parameters. The model relies on evaluating the link quality using eye opening penalty. We predict the link operation, while reducing potential link outage from 25% down to 5% using two back-up reservation channels. This way, stochastic fiber degradation can be potentially mitigated by using spectrally effective channel reservation, which can be a low-cost and reliable solution for high-speed optical systems.

Connected and Autonomous Vehicles (CAV) technology is a key technology that can improve safety, reliability and efficiency in transportation. Autonomous driving uses information from a variety of sources: internal sensors, messages coming from other vehicles, traffic signs and centralized control systems. This information is processed to provide a variety of services. Some of these services require broadband communication with low latency. Complex vehicle dynamics and a constantly changing environment require high-performing communication technology to support CAV communications. Fifth generation (5G) networks are becoming a key technology for supporting CAV communications due to their high throughput and low latency communication. This paper presents an experimental measurement of Quality of Service and Quality of Experience parameters of 4G and 5G networks and their comparison. The measurements were performed on a selected route in the city of Žilina,, Slovakia. The aim of the research is to reveal the limitations of 5G networks with non-standalone architecture in real conditions, especially in terms of latency.

The rapid integration of Connected and Automated Vehicles (CAVs) into modern transportation systems necessitates a robust and systematic approach to assess the quality of the underlying digital infrastructure. In the presented work, we propose a methodology and evaluation of framework that can be used to assess digital infrastructure segments based on their readiness for the deployment of CAVs. The methodology encompasses a comprehensive framework that collects, processes, and evaluates diverse data sources, including real-time traffic, communication, and environmental data. The proposed framework is developed based on experimental data and provides a systematic approach to assess infrastructure readiness for CAVs. The proposed methodology is applied in a system for detecting the readiness status of digital infrastructure from a Cooperative, Connected, and Automated Mobility (CCAM) perspective. The system can determine the percentage of non-compliance of technical service requirements in terms of latency, bandwidth, and localization accuracy. Thanks to this, we can determine in advance in which state the current digital infrastructure is and which services can be currently operated, and thus locate the segments of the route in which the telecommunication systems need to be supported.

Internet of Things (IoT) becomes indispensable for transport and automotive industry to advance functions in on-road traffic monitoring. Indeed, smart management tools and machine learning concepts are inevitable in vehicle categorization systems. However, to date, existing systems for vehicle classification are exclusively based on singular technological platforms only. This not only limits their long-term use and future scaling, but also sets restrictions to obtain high classification accuracies with modern machine learning tools feed with diversified big volume data. In this work, we design a novel convolutional neural network (CNN) that substantially improves the on-road vehicle classification. In particular, we experimentally harness, to the best of our knowledge for the first time, two different datasets from separated technological platforms based on close-circuit television (CCTV) and fiber Bragg grating (FBG) sensors, respectively. The hybrid CNN classification system, with individual CCTV and FBG datasets, substantially improves detection levels, reaching in-class accuracy up to 90% - 97%. Moreover, this classification concept includes an intrinsic back-up verification with respect to each platform compensating the shortcomings of individual technologies. Our demonstration can make key advances towards near-unity accuracy in vehicle classifications for IoT systems, capitalizing on cost-effective and well-established platforms.

Vehicle-to-infrastructure communications can inform an intersection controller about the location and speed of connected vehicles. Recently, the design of adaptive intersection control algorithms that utilize this information received substantial research attention. These studies typically assume perfect communications. This study explores the possible effects of a temporal decrease in the reliability of the communication channel, on the intersection throughput. Road traffic and DSRC-VANET communications are modelled by integrating traffic and communication simulation tools (Vissim and OMNeT++, respectively). Simulations of scenarios with challenging, but realistic communication distortions conditions show significantly larger average delays to vehicles compared to scenarios with perfect communication conditions. These additional delays are largely independent of whether all or only some of the intersection approaches are affected by the communication distortions. Furthermore, delays do not increase uniformly on all signal groups. They may even decrease for some, which causes unfair allocation of green times. The control may be corrected for the lost communications using the information received in previous time intervals and simple assumptions about the vehicle movements. This correction decreases delays in all scenarios both for isolated and connected intersections. It performs similarly to the case with perfect communications when the communication distortions are distributed uniformly among all intersection approaches. Overall, the results demonstrate that the impact of the communication distortions should be considered in the design of the adaptive intersection control algorithms.

In this paper, we present results of numerical analysis of phase-shifted fiber Bragg gratings aimed at slowing down the group velocity of light propagating through these structures. Using coupled-mode theory and transfer matrix method we model the impact of several parameters such as length of the grating, refractive index modulation depth and phase shift in the periodicity of Brag gratings to calculate transmission spectral properties and to maximize the final group delay. By introducing irregularity into a periodic structure of refractive index, Falmy-Perot like cavity formed in the grating results in a favorable narrow dip in resonance spectra characteristics. Correspondingly, we observe a significant spike in group delay at the same wavelength. Simulations results obtained by numerical approach show a strong need for parameter optimization in process of tailoring gratings behavior. Considerable attention must be put on width of spectral region suitable for large group delay realization and suitable interrogation schemes need to be implemented when adopting studied structures for sensors applications.

In this paper, we present an assessment framework that can be used to score segments of physical and digital infrastructure based on their features and readiness to expedite the deployment of Connected and Automated Vehicles (CAVs). We discuss the equipment and methodology applied for the collection and analysis of required data to score the infrastructure segments in an automated way. Moreover, we demonstrate how the proposed framework can be applied using data collected on a public transport route in the city of Zilina, Slovakia. We use two types of data to demonstrate the methodology of the assessment-connectivity and positioning data to assess the connectivity and localization performance provided by the infrastructure and image data for road signage detection using a Convolutional Neural Network (CNN). The core of the research is a dataset that can be used for further research work. We collected and analyzed data in two settings-an urban and suburban area. Despite the fact that the connectivity and positioning data were collected in different days and times, we found highly underserved areas along the investigated route. The main problem from the point of view of communication in the investigated area is the latency, which is an issue associated with infrastructure segments mainly located at intersections with heavy traffic or near various points of interest. The low accuracy of localization has been observed mainly in dense areas with large buildings and trees, which decrease the number of visible localization satellites. To address the problem of automated assessment of the traffic sign recognition precision, we proposed a CNN that achieved 99.7% precision.

Ensuring quick and fluent transit of an emergency response vehicle (ERV) can be a particularly challenging task in areas with increased traffic density. Sites where big cultural or sports events are organized, e.g., stadiums, concert halls, etc., impose one of the major challenges for emergency vehicles during peak hours before or after the event takes place. During a short time interval, many vehicles enter the road segments around the area, which can increase the emergency vehicle’s driving time significantly.In this study, we examined the impact of various levels of ICT deployment on the driving time of an ERV under specific traffic conditions near a site where a large cultural or sports event takes place. Driving time of the ERV is evaluated using a realistic simulation of both traffic scenarios as well as communication technologies deployment. The impact of the communication on emergency vehicle’s driving time is evaluated for varying penetration of connected vehicles.For the information exchange between Variable Message Sign (VMS) and Traffic Management Center (TMC), we designed Vehicle-To-Everything (V2X) application protocol which can be universally used both for the communication between TMC and VMS, as well as between VMS and vehicles.

From the past years vehicular Ad-Hoc networks (VANET) are being increasingly researched. This special communication network ensures maximum safety to transportation and supports develop of autonomously controlled vehicles. In this network are transported a lot of data packets from most of which are produced by cars in high motion. Therefore, data routing is one from main key of this network because data routing has affect to security, power and last but not least robustness to this network. This work aims to provide a properties of topology based data routing protocols used in vehicular Ad-Hoc networks to present the features related to this field to help researchers and developers to understand the main features of data routing in VANET technology. We mainly focus on OLSR, DSDV, DSR, AODV and DYMO routing protocols. The paper also represents the main key of data routing in VANET network and its division.

Optical photodetectors are at the forefront of photonic research since the rise of integrated optics. Photodetectors are fundamental building blocks for chip-scale optoelectronics, enabling conversion of light into an electrical signal. Such devices play a key role in many surging applications from communication and computation to sensing, biomedicine and health monitoring, to name a few. However, chip integration of optical photodetectors with improved performances is an on-going challenge for mainstream optical communications at near-infrared wavelengths. Here, we present recent advances in heterostructured silicon-germanium-silicon p-i-n photodetectors, enabling high-speed detection on a foundry-compatible monolithic platform.

The fast signal transmission is critical long-haul communication systems. They represent the key advancements, shaping information-communication technologies. Fiber-optic transmission suffers from many degradation effects, and of particular concern are stochastic fiber impairments represented by polarization mode dispersion (PMD). The PMD is critical as it limits link operation at data rates higher than 10 Gbps. In this work, we report on experimental measurements and theoretical analysis characterization for PMD-based propagation effects in optical fibers influenced by wind gusts. The study was performed on fiber-optic link that runs through 111-km-long optical power ground wire cables. Measured maximum of DGD was up to 10 ps for a wind speed of 20 m/s. This wind condition, the optical link maintained a reliable operation only for established 10 Gbps, while considerable link degradation was seen for data rates of between 40 and 100 Gbps.