• Proposal of Geo-Anchored Datasets (GADs) for vehicular data collection and sync in VANETs. • Design of the State Vector Push Protocol (SVPP) for efficient GAD construction in dense networks. • One-to-many NDN model enabling scalable data exchange in vehicular environments. • Real-time traffic monitoring using GADs for intersection awareness and control. • Adaptive scheduling to reduce packet loss in broadcast-heavy vehicular networks. In this paper, we introduce the concept of Geo-Anchored Datasets (GADs) – named datasets constructed from geographical areas where vehicles tend to aggregate due to traffic rules and density, leveraging Named Data Networking (NDN) applied to Vehicular Ad-Hoc Networks (VANETs). These datasets include information from vehicles that pass through the defined area, remain valid for a controlled period, and, within that brief time, require at least one vehicle to be present in the area at any given moment to retain the information. The focus of this work is to present a mechanism for fast and collision-efficient information synchronization under geographical and temporal restrictions. The proposed State Vector Push Protocol (SVPP) creates Geo-Anchored Datasets using an epidemic push-based model and the state vector technique. We also propose two scheduling mechanisms to mitigate the effects of the resulting broadcast storm by dynamically adjusting transmission delays based on vehicle density, reducing collision occurrences. In our most demanding scenario – 112 vehicles in a 4 lane 140m static segment – SVPP achieves under 10 % packet loss, sub-second convergence, and 1 Mbit/s throughput, showing high efficiency even under extreme conditions.

This research explores the realm of Vehicle Ad-hoc Networks (VANETs) and emphasizes the potential of the field. Applications that enable direct communication between vehicles and Vulnerable Road Users (VRUs) have the potential to significantly improve road safety for all road users. The goal is to let cars know where VRUs are, how fast they are going, and in what direction - all in real-time, taking advantage of the current progress in Vehicle-to-everything (V2X) communications. The implemented solution centers on pedestrian detection using advanced communication protocols with a Road Side Unit (RSU). Acting as a central hub, the RSU collects pedestrian information and disseminates it to vehicles. Equipped with this data, vehicles autonomously execute collision avoidance algorithms, ensuring rapid and informed responses to potential road hazards. The proposed system underwent comprehensive implementation and testing in realistic scenarios using simulation environments. The results align with expectations, validating the effectiveness and feasibility of the proposed solution.

The advent of connected and autonomous vehicles (CAVs) holds immense promise for revolutionizing road safety and efficiency. Communications between the vehicles and other road users or elements (V2X) opens various possibilities that can prevent accidents and fatalities. Vulnerable Road Users (VRUs), like pedestrians and cyclists, are at particular risk since a collision with a vehicle has a high likelihood of inflicting severe injuries. On the other hand, VRUs usually have one or more connected devices, and these can be used to increase their safety by allowing the prediction and prevention of collisions with CAVs. However, for this to be possible, it will be necessary to ensure not only reliable and low latency communication between the VRUs, CAVs and, eventually, Road Side Units (RSUs), but also the development of robust and reliable collision prediction algorithms capable of anticipating potential collisions and enabling timely interventions to prevent accidents. This paper presents a system that can be used to track a vehicle and a VRU, predict possible collisions and alert the VRU about the incoming danger. Additionally, it delves into the intricate aspects of collision prediction algorithms for CAVs, addressing the fundamental requirements for effective collision prediction and presenting a balanced approach that meets the demands of speed, accuracy, robustness, adaptability, and scalability.

The safety factor of ITS is particularly important for VRUs, as they are typically more prone to accidents and fatalities than other road users. The implementation of safety systems for these users is challenging, especially due to their agility and hard to predict intentions. Still, using ML mechanisms on data that is collected from V2X communications, has the potential to implement such systems in an intelligent and automatic way. This paper evaluates the performance of a collision prediction system for VRUs (motorcycles in intersections), by using LSTMs on V2X data - generated using the VEINS simulation framework. Results show that the proposed system is able to prevent at least 74% of the collisions of Scenario A and 69% of Scenario B on the worst case of perception-reaction times; In the best cases, the system is able to prevent 94% of the collisions of Scenario A and 96% of Scenario B.

Vehicular Ad hoc Networks (VANETs) are the underlying network for all the communications between Intelligent Transportation System entities as, for example, Road Side Units, On-Board Units, etc. These networks are growing in size and interest from the automaker's industry, software companies and scientific researchers. The intrinsic technological nature of VANETs (based on a wireless shared medium) with no predefined structure of architecture (high mobility of their users) makes them very susceptible to security attacks, which may impact deeply on road users. There are several works on VANET security, usually involving cryptography. But, these can only prevent some types of attacks. Others, mainly the ones carried by authentic entities, cannot be prevented applying this strategy. Nonetheless, attacks can be detected triggering responses minimizing the repercussions. This can be accomplished by using an Intrusion Detection System (IDS) which can be enhanced with the help of Machine Learning (ML). But, one of the main difficulties for an effective implementation of a ML-based IDSs is the (un)availability of valid public datasets. The IDSs need a large collection of VANET messages to be able to recognize types of behaviors. However, most of the studies in the literature discussing solutions and their results do not make the datasets publicly available for scrutiny by third parties. As such, the results are difficult to properly verify and validate. The goal of this work is to provide well documented and publicly available datasets to be used by researchers on future R&D projects, enabling a fairer comparison of their results.

This paper proposes a reliable step counting system for pedestrians based on pattern recognition, more specifically, on autocorrelation. It uses the accelerometer signal obtained through the accelerometer sensor that is usually present in almost all smartphones and was designed to be easily integrated as a system component in an Indoor Positioning System. The system has been thoroughly tested using specific planned scenarios always with more than one hundred steps. The ground truth was carefully obtained by counting the real number of steps. The same path was travelled many times, by different users, carrying the smartphone in 5 distinct positions: in hand, in front pocket, in back pocket, all-mixed and running. Experimental results show that on average only one step is miscounted using the proposed method, regardless of the position of the smartphone.

Wi-Fi fingerprinting is a popular technique for smartphone-based indoor positioning. However, well-known RF propagation issues create signal fluctuations that translate into large positioning errors. Large errors limit the usage of Wi-Fi fingerprinting in industrial environments, where the reliability of position estimates is a key requirement. One successful approach to deal with signal fluctuations is to average the signals collected simultaneously through independent Wi-Fi interfaces. Another successful approach is to average the estimates provided by models built on independent radio maps. This paper explores multiple combinations of both approaches and determines the procedure to select the best model based on them through a simulated environment. The evaluation of the proposed model in a real-world industrial scenario shows that the positioning error (according to different metrics including the 95thand 99thpercentiles) is highly improved with respect to the traditional fingerprint.

A Vehicular Adhoc Network (VANET) is a conceptualisation that can be applied to the communications domain of an Intelligent Transportation System (ITS), where vehicles use On Board Units (OBU)to communicate with other vehicles or the road side infrastructure. The heterogeneity of the interconnected devices, the large number of devices inside and outside the vehicles and different degrees of device density, makes interoperability a challenging problem, making way to many machine-2-machine (M2M)autonomous communications. This paper discusses two different M2M architectures that may be used in ITS context and presents test results obtained from implementations of these M2M architectures on real OBU hardware, in the context of ETSI compliant safety applications. The experimental results were taken using two different oneM2M Vehicular Architectures Types in order to analyse their suitability to be used in a platooning car-following use case. For the car-following use case, OBUs interface the vehicle's Controller Area Network (CAN)bus and use Dedicated Short Range Communications (DSRC)for M2M communications to convey dynamic data to enable controlling the vehicle's speed and longitudinal distance. The experiments and corresponding result analysis has shown that the oneM2M architectures are suitable to be used for such car-following application, which requires latencies below 100 ms.

The accuracy of fingerprinting-based positioning methods accuracy is limited by the fluctuations in the radio signal intensity mainly due to reflections, refractions, and multipath interference, among other factors. We consider that the fluctuations (often modelled as a Gaussian process for simplification purposes) can be minimized by exploiting the richness of multiple signals collected simultaneously through independent network interfaces. This paper introduces an analysis of Wi-Fi signals' statistics using simultaneous measurements which shows that RSSI values obtained from independent devices are not highly correlated. The low correlation between Wi-Fi interfaces might be exploited to improve the positioning accuracy. The validation of the proposed fingerprinting approach in a real scenario shows that the mean and maximum error in positioning can be reduced by more than 40% when five Wi-Fi interfaces are simultaneously used for fingerprinting.

Due to the prohibitive cost of deploying a real vehicular network, most VANETs applications are evaluated using simulation. Considering that one important group of vehicular applications is related to cooperative traffic efficiency, this work evaluates the influence of road traffic warning message dissemination strategies on vehicular traffic and communication metrics. This study uses the prototype implementation of a simple adaptive traffic warning system deployed on the Veins simulation framework. The implemented scenario considers five distinct road incidents to evaluate the effect of three warning message dissemination strategies, based on: i) distance between the sender and the receiver; ii) maximum number of hops that a message is allowed to travel; and iii) probabilistic rebroadcasting. Results show that these strategies affect differently important metrics of road traffic (travel time, CO2emission) and wireless communications (packet loss, duplicate packets).

Tracking the position of people or vehicles in large indoor settings with high accuracy is still a challenge despite the significant progress observed in indoor positioning technology in the last decade. To date, there is not a clearly dominant indoor positioning solution for general use, and challenges related to seamless indoor-outdoor positioning, reliable floor estimation and indoor maps are still needing more research. In this context, the IPIN 2016 conference is promoting a competition to evaluate a set of competing indoor positioning solutions in a realistic scenario. This paper describes the proposal of the UMINHO team and some of the obtained results.

Research and development around indoor positioning and navigation is capturing the attention of an increasing number of research groups and labs around the world. Among the several techniques being proposed for indoor positioning, solutions based on Wi-Fi fingerprinting are the most popular since they exploit existing WLAN infrastructures to support software-only positioning, tracking and navigation applications. Despite the enormous research efforts in this domain, and despite the existence of some commercial products based on Wi-Fi fingerprinting, it is still difficult to compare the performance, in the real world, of the several existing solutions. The EvAAL competition, hosted by the IPIN 2015 conference, contributed to fill this gap. This paper describes the experience of the RTLS@UM team in participating in track 3 of that competition.

Due to its general usage in modern buildings, WiFi has been studied as a promising good low-cost solution to build localization services, but many other approaches are possible. Both industry and academy are putting considerable efforts in studying affordable and accurate alternatives, and no de facto standard exists yet. While technological diversity can be good in terms of flexibility, allowing distinct solutions to be used in distinct contexts according to accuracy needs or cost constraints, it is also a big obstacle to the goal of building a global positioning system. So one important requirement that has yet to be addressed is this diversity, allowing for multiple localization systems to be easily plugged into a standard and global localization service. In this paper, a solution to build a global localization service able to integrate various heterogeneous localization systems is presented and discussed. A prototype implementation built on top of DNS is proposed. The DNS subtree structure, as well as the Resource Records able to store the localization information, are presented and justified. The results obtained from prototype testing have proven that DNS is a valid and good candidate to support a global Localization Service.

Human mobility has been widely studied for a variety of purposes, from urban planning to the study of spread of diseases. These studies depend heavily on large datasets, and recent advances in collaborative sensing and WiFi infrastructures have created new opportunities for generating that data. However, these methods and procedures require the participation of a significant community of users through extended periods of time. In this paper, we address the problem of how to engage people to participate in the data collection process. We have conducted a user study on the utilisation of a mobile collaborative sensing application. We have found that users react positively to campaigns, but it is difficult to keep them participating for long periods of time. We also hypothesise that one must close the loop, rewarding the participants with services based on the collected data, eventually showing that there is added value obtainable from crowd sourcing.