The convergence of wireless communications and transportation technologies has propelled the evolution of Intelligent Transportation Systems (ITS), enhancing connectivity, safety, and efficiency in vehicular networks. As electric vehicle (EV) adoption grows within connected ecosystems, the need for intuitive charging management solutions has surged. This paper presents a task-oriented, bilingual chatbot to optimize EV charging control via an intelligent human-machine interface (HMI). Using Natural Language Processing (NLP) and Artificial Intelligence (AI), the system enables seamless interaction with charging infrastructure through wireless channels, supporting English and Spanish to address linguistic diversity. Integrated with ITS frameworks, it facilitates real-time monitoring and control, aligning with cooperative sensing and user experience goals. Two approaches are assessed: Voiceflow, for fast prototyping, and BERT (Bidirectional Encoder Representations from Transformers)-based models, implemented as a multilingual system and separate monolingual variants. Results show monolingual BERT models outperform the multilingual model in terms of intent classification accuracy, reinforcing reliability in dynamic settings. By linking wireless connectivity with AI-driven HMI, our proposal enhances EV charging scalability and accessibility within ITS, advancing sustainable transportation and efficient vehicular networking.

Vehicular networks have transformed Intelligent Transportation Systems (ITS) by enabling Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications, enhancing road safety and traffic efficiency. However, these technologies face significant challenges in rural and mountainous environments, where communication infrastructure is limited. In this context, LoRaWAN emerges as a viable solution to extend coverage in vehicular networks, providing connectivity in remote areas where traditional solutions, such as LTE-V2X and 5G-V2X, are insufficient due to their reliance on dense infrastructure.This paper evaluates the use of LoRaWAN in V2I communications in remote areas using FresSim, a simulation tool based on Fresnel Zone analysis and geospatial data. In particular, LoRaWAN coverage has been evaluated considering the strategic placement of gateways, achieving connectivity extension up to 140 km with only five gateways. The results confirm that LoRaWAN enables the transmission of emergency alerts and vehicular data in hard-to-reach environments with minimal infrastructure, optimizing resource utilization and providing an efficient alternative to enhance V2I connectivity on rural roads and highways.

This article introduces the LoRaWAN Fresnel Zone Simulator (FresSim), a tool designed to evaluate the feasibility of LoRaWAN deployments by analyzing the Fresnel Zone and terrain topography, allowing for the early prediction of coverage between nodes and the gateway. Furthermore, unlike other tools, FresSim can integrate real data from platforms such as The Things Network (TTN) and ChirpStack, providing an accurate assessment of potential coverage and signal quality in real deployments. Comparisons between FresSim and LoRaSim, a widely used signal propagation simulator, reveal that FresSim offers superior accuracy in scenarios with complex topography, achieving a 100% success rate in predicting coverage, doubling the 50% accuracy rate of LoRaSim. These results demonstrate that FresSim is a valuable tool for optimizing LoRaWAN networks in open environments, significantly contributing to the improvement of LoRaWAN deployment planning.

Device registration in LoRaWAN systems can be a tedious process, particularly when managing a large number of devices. To automate this task, we previously developed the AutoLoraConfig protocol, although it includes vulnerabilities that attackers can exploit. To solve this, in this work we introduce the SecureAutoLoRa, a protocol designed to enhance security in the automated device registration process for LoRaWAN networks. SecureAutoLoRa blocks abuse of the guest DevEUI used in automated joins by obfuscating registration messages and enforcing a configurable positional security code that dictates where the real DevEUI appears across packets, markedly hindering unauthorized registrations.Results show that SecureAutoLoRa sharply reduces attackers’ success. Compared with AES-128 and SPECK, it complements standard encryption to provide a practical, layered defense suitable for large-scale LoRaWAN onboarding.

The rapid adoption of electric vehicles (EVs) demands advanced residential charging solutions, where user behavior varies widely due to factors like electricity tariffs and weather conditions. Such conditions make charging predictions particularly complex. This study addresses this challenge by predicting the time an EV remains connected to a residential charger using a dataset of 106,260 sessions. To this end, we propose the Connection Time Neural Model (CTNM), a deep neural network designed to model the complex dynamics of domestic charging, and we introduce the Weighted Error Metric (WEM), a novel metric that penalizes overestimations and under-estimations differently to reflect their real-world impacts on both grid management and user experience. Utilizing bidirectional charger data, we focus solely on connection time, bypassing energy prediction. CTNM is benchmarked against state-of-the-art methods (i.e., Random Forest, Dense Neural Network, XGBoost, and Support Vector Regression) using Mean Absolute Error, Root Mean Squared Error, and WEM as performance metrics. Results demonstrate CTNM’s superiority, reducing average error by 18%, and weighted error by 31% compared to Random Forest, thanks to its deep architecture and integration of contextual features, like variable tariffs and weather.

LoRaWAN (Long Range Wide Area Network) is a communications protocol stack based on LoRa which provides long-range connectivity and low power consumption, making it a strong candidate for Internet of Things (IoT) device connectivity. However, unlike other technologies like WiFi or Bluetooth, LoRaWAN lacks an automatic mechanism to facilitate the connection of new devices to a network, which is undoubtedly a limitation for widespread device deployment. This paper presents AutoLoRaConfig, a rule-based system designed to automate the configuration and registration of devices in LoRaWAN networks, reducing the manual intervention required, and therefore minimizing potential human errors. The proposal aims to simplify the work required by users regarding the device configuration and registration process, enabling agile and efficient large-scale deployments. Results indicate that, for a set of 100 devices, automated deployment is 136.6 times faster than the manual procedure.

Long Range Wide Area Network (LoRaWAN) is a network protocol specifically designed to leverage the advantages of Long Range technology. LoRaWAN is employed to connect Internet of Things devices through a long-range network infrastructure, providing a secure and efficient communication layer that facilitates connectivity for a vast number of devices in a LoRa network. The aim of this study is to assess the real-life performance of LoRaWAN in different topographical environments. In particular, two scenarios (one flat and one mountainous) are compared, with an analysis of the success rate in data packet reception. The purpose is to examine how topography influences communication between LoRa devices and gateways, as well as to explore alternatives for overcoming challenges in rugged terrain environments. Therefore, in the mountainous environment, the study also investigates whether the use of drones could enhance communications. The findings of this study reveal that the topography of the terrain significantly impacts LoRa communications, resulting in a 58.63% decrease in the number of successfully received packets compared to a flat environment. Moreover, one-third of the nodes failed to establish communication with the ground gateway, and a partial improvement in connectivity was achieved by using gateways deployed on drones.