The adoption of decentralized architectures for health data management offers benefits including patient data sovereignty and elimination of single points of failure, but introduces questions about network overhead compared to traditional centralized systems. This paper presents a network overhead analysis comparing Firebase Real-Time Database with IPFS-based storage via Pinata for mobile health data transmission. We implemented an Android application that collects physiological data from wearable devices and transmits this information to both backends using REST APIs. Our experimental evaluation across eight transmission scales reveals that Firebase demonstrates lower fixed overhead and latency for small payloads, while Pinata exhibits superior scaling characteristics for larger data volumes. A crossover point occurs around 50 records per payload, beyond which the decentralized architecture transmits less total data than the centralized alternative. The results indicate that neither architecture maintains uniform efficiency across all operational scales, with architectural choice depending on expected transaction patterns in the deployment context.

Mobile health applications integrated with wearable devices enable continuous monitoring for chronic disease management, but the reliability of wireless connectivity between mobile applications and sensors remains a concern for healthcare applications. This paper presents an analysis of Bluetooth Low Energy connection reliability between the TEMPO mobile application and Movesense wearable devices in real-world hemophilia management scenarios. We conducted a 48-hour continuous monitoring experiment simulating typical patient usage patterns, where users operate the system without active connection management. The experimental setup involved a Movesense HR paired with the TEMPO application running on an Android smartphone collecting IMU data for Human Activity Recognition applications. Our analysis demonstrates that BLE connections achieve a 90% automatic reconnection success rate, with the system effectively handling routine connectivity interruptions without requiring user intervention. The system demonstrated resilience during extended disconnection periods and successfully recovered automatically. These findings support the feasibility of BLE-based wearable systems for reliable healthcare monitoring in chronic disease management.

Mobile health applications for chronic disease management require frequent access to health data for continuous monitoring and clinical decision support, creating performance challenges when accessing centralized platforms like Google’s Health Connect (HC). This paper presents an evaluation of content provider mechanisms for optimizing health data access in Human Digital Twin applications. We developed a three-layer Human Digital Twin architecture implementing local data storage through Android’s content provider interface, integrated with TEMPO, a hemophilia management application for continuous physical activity monitoring. Our experimental evaluation compared the HDT content provider approach against direct Health Connect access over continuous monitoring periods, with automated data requests across varying request sizes from single records to large batch operations. Results demonstrate that local data access through content providers can provide performance benefits over direct API access, with implications for healthcare applications requiring frequent data access.

Industry 5.0 emphasizes human-centric manufacturing, where the operator needs to inform system design. While Operator Digital Twins monitor operator states through biometric data, existing approaches have not exploited this information to adapt industrial Human-Machine Interfaces (HMIs), which remain predominantly static and uniform. This paper proposes an architecture that extends Human Digital Twin systems with LLM-driven personalization for web-based HMIs. The framework collects operator-specific data through a mobile application integrating Health Connect and manual inputs. A Large Language Model interprets operator profiles containing biometric signals, permanent characteristics, and preferences, generating customized CSS stylesheets and configuration parameters that adapt visual properties and interaction modalities while preserving safety-critical elements. The architecture’s applicability is illustrated through diverse conceptual scenarios, such as visual impairments, protective equipment usage, elevated stress states, and ergonomic preferences. This demonstrates how a consistent operator profile format could drive varied interface adaptations to meet different operator needs.

The increasing reliance on machine learning in Internet of Things systems demands to evaluate the trade off between computing on the resource constrained devices or offload the computation to more powerful edge devices. Split computing has emerged as a promising paradigm to bridge this gap by partitioning workloads between resource-constrained devices and edge devices in a flexible way. Existing approaches, however, often remain tied to specific model architectures or assume theoretical network and device conditions, hence limiting their applicability in realistic deployments. In this paper, we introduce SCIoT, a framework for Split Computing in the Internet of Things that seeks to address these shortcomings. SCIoT enables flexible and adaptive partitioning across heterogeneous devices, explicitly accounting for resource availability, fluctuating network performance, and data sensitivity. The framework incorporates dynamic policies that balance latency, bandwidth usage, and privacy, moving beyond static or one-size-fits-all strategies. We evaluate SCIoT across representative scenarios, demonstrating its ability to adaptively reconfigure computations while maintaining competitive efficiency. Our results show both the advantages and the current limitations of split computing in practice, contributing a step toward more robust, adaptive, and privacy-aware collaborative inference in IoT ecosystems.

End-to-end latency is a critical metric in interactive and real-time systems, where even short delays can undermine usability, situational awareness, and trust. While most studies focus on network transmission delays, this overlooks other significant sources of latency, including sensor acquisition, processing, and especially visualization. Rendering pipelines are heavily influenced by hardware, visualization strategies, and data complexity, and in visualization-rich domains such as smart cities, these factors can add considerable overhead. Ignoring them leads to overly optimistic performance assessments and missed opportunities for optimization.In this work, we study the last mile of end-to-end latency, explicitly incorporating the visualization stage and bridging the gap between network-level metrics and user-perceived responsiveness. We introduce the Unreal Smart Cities Visualizer (USCV), a framework we have built to support precise latency assessment in realistic, data-intensive urban scenarios. Our contributions include highlighting the limits of network-centric evaluation, presenting a visualization-aware methodology, and demonstrating USCV as a tool for accurate end-to-end latency measurement.Our results show that visualization delays, particularly in latency-critical environments, are not negligible and cannot be underestimated.

This paper presents the Smart City Architecture (SCA), a middleware system built upon the MQTT (Message Queuing Telemetry Transport) protocol and developed within the MASA (Modena Automotive Smart Area) initiative. SCA enables intelligent urban applications by facilitating seamless and scalable communication among heterogeneous entities, including assets, services, and observers. Its structured, topic-based messaging layer supports efficient telemetry exchange, event-driven processing, and dynamic service interaction. The capabilities of SCA are exemplified through two real-world services—Vulnerable Road User (VRU) and GeoPerception—which provide real-time risk detection and localized situational awareness in smart city scenarios.

This paper introduces a low-cost and reproducible framework for measuring glass-to-glass (G2G) latency in real-time video systems. Unlike existing solutions, which are often proprietary, expensive, or poorly documented, our approach combines a photodiode, a microcontroller, and lightweight calibration routines to achieve accurate end-to-end latency measurements. The framework is validated across heterogeneous devices, revealing the impact of hardware tiers and video codecs (e.g., H.264 vs. Motion JPEG) on responsiveness. Beyond smartphones, we demonstrate adaptability to complex pipelines such as remote driving and wearable devices, where latency directly affects safety and user experience. Released as an open-source tool, the framework fills a methodological gap in latency research and offers practical guidelines for optimizing multimedia pipelines in domains including virtual reality, telemedicine, and autonomous mobility.

Urban scenarios present various concerns to micromobility users such as cyclists, which are an important part of the mobility infrastructure. As smarter cities increasingly focus on enhancing citizen well-being and infrastructure efficiency, navigation systems have primarily been designed for car drivers, with a focus on traffic conditions, while disregarding metrics that are important for micromobility users. As a matter of fact, they tend to privilege other aspects of their journey that enhance its comfort, rather than the mere path length. To address this gap, this paper presents a holistic architectural framework for micromobility-oriented navigation systems, incorporating nominal, data-driven and crowdsensed metrics in the loop. The paper presents the definition of metrics and a real world case study for each category of metrics, providing a real implementation. We demonstrate the effectiveness of our approach in a controlled environment before implementing and deploying the complete system in a real-world city, where we provide a detailed analysis of the results.

Crowd-sensing has emerged as a powerful data retrieval model, enabling diverse applications by leveraging active user participation. However, data availability and privacy concerns pose significant challenges. Traditional methods like data encryption and anonymization, while essential, may not fully address these issues. For instance, in sparsely populated areas, anonymized data can still be traced back to individual users. Additionally, the volume of data generated by users can reveal their identities. To develop credible crowd-sensing systems, data must be anonymized, aggregated and separated into uniformly sized chunks. Furthermore, decentralizing the data management process, rather than relying on a single server, can enhance security and trust. This paper proposes a system utilizing smart contracts and blockchain technologies to manage crowd-sensing campaigns. The smart contract handles user subscriptions, data encryption, and decentralized storage, creating a secure data marketplace. Incentive policies within the smart contract encourage user participation and data diversity. Simulation results confirm the system's viability, highlighting the importance of user participation for data credibility and the impact of geographical data scarcity on rewards. This approach aims to balance data origin and reduce cheating risks.

In this work, we investigate the performance and suitability of various mobile frameworks for critical smart city applications. With the increasing reliance on mobile apps for urban services, understanding the capabilities and assessing the limitations of different development platforms is crucial. We evaluate Android OS, Apple iOS, and cross-platform solutions like Flutter, focusing on their ability to deliver real-time information efficiently in mobile applications tailored for smart city environments. In our work, we look at the content delivery latency, scalability, cross-platform capabilities, and developer productivity. We also assess the frameworks' integration with existing smart city infrastructure and their handling of distributed location-dependent data. The study compares native implementations using platform-specific SDKs against cross-platform solutions, with particular attention to map-based applications utilizing both native maps and third-party services like Mapbox. We have tested the platforms within the Modena Automotive Smart Area (MASA), a real-world testbed for innovative mobility solutions in Modena, Italy. This environment provided a realistic setting to evaluate the frameworks' performance in critical smart city scenarios. The outcomes of our work offer insights into the relative strengths and limitations of each mobile framework, considering both technical performance and development efficiency. In this work, we aim to guide developers and city planners in selecting the most appropriate mobile framework for smart city applications, balancing performance requirements with development resources and cross-platform needs.

This study investigates the impact of meteorological variables on air pollutant concentrations, focusing on Carbon Monoxide (CO), Nitrogen Dioxide (NO2), and Ozone (O3). By integrating data from static stations and other services, alongside weather data from online public repositories, we aim to enhance the understanding of air quality dynamics. The research highlights how temperature and solar radiation significantly influence air quality, with wind speed and precipitation aiding in pollutant dispersion. Utilizing the SHAP method, we offer a detailed and interpretable analysis of the factors affecting air quality, emphasizing the crucial role of integrating diverse data sources. Our findings demonstrate that merging various datasets fills critical gaps in environmental monitoring, leading to improved interpretability and reliability in air quality assessments. These insights support more effective environmental management strategies. Future directions include leveraging citizen-generated data to refine pollution modeling and enhance forecast transparency, ultimately contributing to more comprehensive environmental monitoring practices.

Crowd-sensing is considered a robust model for data collection, yet with challenges related to data availability and privacy. Traditional techniques such as data encryption and anonymization may not fully mitigate these issues, since anonymized data can still be traced back to individual users, and the volume of data generated can reveal user identities. This paper introduces a system that employs smart contracts and blockchain technology to manage crowd-sensing campaigns. The smart contract oversees user subscriptions, data encryption, and decentralized storage, creating a secure data marketplace. Incentive mechanisms within the smart contract promote user participation. Simulation results validate the system’s feasibility, emphasizing the importance of user engagement for data credibility and the impact of geographical data scarcity on rewards.

Current health data management relies on centralized architectures that create a single point of failure, limit patient autonomy, and increase vulnerability to data breaches and vendor lock-in. This paper presents a decentralized approach to continuous health monitoring through the integration of wearable devices and distributed file systems. We implemented an Android application that collects physiological data and contextual information from a wearable device, storing it on the IPFS via Pinata API. Additionally, we propose a blockchain architecture for role-based access control. Performance evaluation comparing our IPFS-based implementation against Firebase Real-Time database reveals that the resource requirements remain negligible for modern smartphones while achieving significant benefits, including no single point of failure, enhanced data portability, and patient data sovereignty. The results demonstrate that decentralized health data management is technically feasible on mobile devices, offering an alternative approach to traditional centralized health data architectures.

Industry 5.0 redefines industrial automation by emphasizing human-centricity, sustainability, and resilience. Within this paradigm, the Operator Digital Twin (ODT) has recently emerged as a digital counterpart of the human worker, integrating biometric, contextual, and behavioral data to enable adaptive interactions with machines. However, fully exploiting the potential of ODTs introduces significant challenges, including the dynamic management of AI functionalities, privacy preservation, and context-aware deployment across heterogeneous computing devices. This paper proposes a dynamic, privacy-aware framework for managing AI capabilities within ODTs. The approach supports real-time adaptation of machine learning models based on the operator’s condition and the computational constraints of devices such as smartphones and embedded systems. By leveraging edge processing, the architecture minimizes the exposure of sensitive biometric data while ensuring reliable functionality and compliance with privacy regulations. The framework is validated in a prototypical industrial testbed using real ML models and heterogeneous hardware, demonstrating its effectiveness in enabling context-driven and secure AI orchestration in ODT-enabled industrial environments.

Internet of Things (IoT) devices provide constant, contextual data that can be leveraged to automatically reconfigure and optimize smart environments. Artificial Intelligence (AI) and deep learning techniques are tools of increasing importance for this, as Deep Reinforcement Learning (DRL) can provide a general solution to this problem. However, the heterogeneity of scenarios in which DRL models may be deployed is vast, making the design of universal plug-and-play models extremely difficult. Moreover, the real deployment of DRL models on the Edge, and in the IoT in particular, is limited by two factors: firstly, the computational complexity of the training procedure, and secondly, the need for a relatively long exploration phase, during which the agent proceeds by trial and error. A natural solution to both these issues is to use simulated environments by creating a Digital Twin (DT) of the environment, which can replicate physical entities in the digital domain, providing a standardized interface to the application layer. DTs allow for simulation and testing of models and services in a simulated environment, which may be hosted on more powerful Cloud servers without the need to exchange all the data generated by the real devices. In this paper, we present a novel architecture based on the emerging Web of Things (WoT) standard, which provides a DT of a smart environment and applies DRL techniques on real time data. We discuss the theoretical properties of DRL training using DTs, showcasing our system in an existing real deployment, comparing its performance with a legacy system. Our findings show that the implementation of a DT, specifically for DRL models, allows for faster convergence and finer tuning, as well as reducing the computational and communication demands on the Edge network. The use of multiple DTs with different complexities and data requirements can also help accelerate the training, progressing by steps. • Integration of Web of Things and Digital Twins for seamless DRL training. • Performance evaluation on a real dataset, showing the benefits of digital twins. • Legacy interoperability is demonstrated with off the shelf smart home software.

Location based services (LBS) are leveraged in everyday services and applications, as they can provide contextual and relevant information for the user needs. These services require the location of the user to be sent along with other relevant information, to provide the data in return that is relevant to the sent position. Although this opens up exciting scenarios for users, it has also been studied since it encompasses several potential privacy issues, which range from the re-identification of the user to the discovery of habits and routines. In this work, we present a study on the tradeoff between the information quality obtained from an LBS and the location precision sent by the user. Our results indicate that by sending out queries with imprecise location enhances the privacy of the users, while still providing a satisfactory quality of information.

The concept of Fluid Computing entails a dynamic resource allocation approach, enabling seamless task migration between computing nodes. This paper investigates the fusion of Fluid Computing principles with the Internet of Things (IoT) and introduces the concept of Fluid Digital Twins (FDTs) i.e. cyber-physical entities that bridge the complexities of this integration. FDTs serve as intermediaries, overseeing fluid task migration, optimizing resource use, and simplifying interactions for external digital applications. The paper delves into challenges arising from this fusion, including limited IoT device capabilities, fragmentation, and the necessity of an intelligent intermediary layer. This research article models and presents FDT mechanics, features a prototype with experimental evaluation and concludes by discussing findings and potential future research directions.

The proliferation of Internet of Things (IoT) devices has sparked a growing demand for lightweight and energy-efficient machine learning solutions, leading to the emergence of Tiny Machine Learning (TinyML). This paper presents a thorough evaluation of TinyML, encompassing its performance metrics, challenges, and prospects, focused on the use of Split Computing. Split Computing allows to offload a subset of layers of a neural network to a more powerful Edge server, to achieve a faster computation hence lower inference latency. We evaluate our proposal on a real testbed with ESP32 microcontrollers with different neural network structures, highlighting the benefits of split computing for IoT devices with varying conditions. Our results indicate that split computing on IoT devices is viable and can bring benefits particularly in heavy load scenarios where the network conditions may rapidly change.

Recent studies explore the possibility of detecting events in a room via Wi-Fi Sensing. This practice exploits the interaction between waves carrying Wi-Fi signals and the elements present in an environment. These interactions are called Channel State Information (CSI) and can be analyzed and exploited to infer information about the environment, such as “device-free” Human Activity Recognition, Human Identification, and more. Considering identification, we recently saw an increasing trend in the usage of low-end devices such as ESP32. Being small and low-power, they are cheap and versatile, however, the quality of the collected data is inferior. In this work, we use state-of-the-art tools to perform Human Identification using the ESP32. Software is created to act as an interface between the collected data and the algorithms suitable for Wi-Fi Sensing. To evaluate the final design, we performed a data collection in a controlled environment. The experiments show an accuracy of 95% in distinguishing two users while 74% in distinzuishing three.

Mobile Crowdsensing (MCS) is a paradigm where a crowdsourcer recruits a set of workers through a campaign to collect data using sensors in their mobile device. This process greatly reduces the costs of data collection processes; however, most of the historically proposed systems are centralized. Since this makes the MCS platform a single point of failure, there is an increasing interest in decentralized blockchain-based solutions; regardless, most of the current proposals have a vertical focus and do not account for the heterogeneity of MCS. We propose a decentralized high-level architecture for MCS, based on Distributed Ledger Technology (DLT), that is adaptable to most MCS deployments. We then implement our architecture using the IOTA protocols and evaluate its performance over a real deployment in terms of scalability, showing its advantages over classic blockchains for MCS data.

This paper proposes a groundbreaking architecture that reimagines Industry 5.0, emphasizing human-centric technological integration via the Web of Things (WoT) standard. Our approach innovatively digitizes human operators and machinery, creating a responsive industrial ecosystem attentive to real-time human conditions. Central to this is the Operator Thing (OT), a digital replica representing the human operator's status and needs. This system not only recognizes operator stress and discomfort but intelligently adjusts, ensuring optimal human-machine synergy. Our methodology extends to redefining operational parameters and tasks in response to human states, balancing well-being with production efficiency. The ultimate goal is a transformative, adaptive, and empathetic Industry 5.0 environment, validated through rigorous interdisciplinary evaluation.

The combination of Fluid Computing with the Internet of Things has enabled dynamic orchestration of tasks and functionalities, enhancing performance and responsiveness. Integrating Digital Twins has bridged the cyber-physical gap and the recent introduction of the concept of Fluid Digital Twins (FDTs) opened to the dynamic reconfiguration of functions and simplified augmentation of physical assets’ capabilities. However, introducing new functions or updating existing ones to improve performance or fix bugs poses significant challenges in validating, testing, and deploying these changes in a production environment without disrupting operations. This paper proposes and experimental evaluate an FDT approach for dynamic function management by spawning twin replicas for testing and automatically synchronizing data between production and validation instances.

In the era of ubiquitous computing, the collection of users' geographical location is increasingly widespread. This represents an enabling technology, capable of creating new type of services but at the same time represents a new digital asset that needs to be protected in order to safeguard the users' privacy. In fact, exploiting everyday movements, it is possible for a threat actor to gather sensible information about the victims that can be leveraged afterwards. In this preliminary paper, we reproduced some major results in the field of re-identification of users' trajectories, validating them under scenarios where different countermeasures for geographical data are in place. Specifically, we tested generalization of spatial data using geohashing and K-means clustering. The results were obtained using a dataset that collects users from all over the world, allowing the clustering methods to range on very different scales. Results shows that, even if a strong data generalization is applied, users' trajectories keep their uniqueness, showing high re-identification ratios. Nevertheless, the usability issues typical of these techniques are still present, having only few tens of points for covering the entire globe which cannot be considered a general solution for every possible use case of such data.

The rapid advancement of Internet of Things (IoT) devices requires innovative approaches to implement machine learning (ML) in resource-constrained environments. This paper explores the integration of Tiny Machine Learning (TinyML) with split computing, focusing on classification using an ESP32 microcontroller connected to a Raspberry Pi edge server. We conduct a series of experiments to measure the time required for image capture and classification, rather than focusing solely on model accuracy. Our findings indicate that while local processing on the ESP32 is limited by its computational capabilities, the split computing approach significantly reduces the processing time by leveraging the Raspberry Pi's computing resources. We then highlight the benefits of our approach considering a dynamic scenario, in which networking changes hence the possibilities to perform split computing vary over time.

In the fast-paced urban landscapes of today, the demand for advanced route planning solutions that cater to personalized navigation and environmental consciousness is more pressing than ever. In this work we present the Urban Route Planner, a state-of-the-art framework poised to revolutionize urban navigation by providing tailored route recommendations that not only consider individual preferences but also prioritize environmental factors. Our work delves into the conceptual framework and essential design elements of our proposal, high-lighting its potential to redefine urban navigation by delivering personalized route recommendations that promote environmental awareness and user satisfaction. With a focus on enhancing both the functionality and sustainability of urban transportation, this framework represents a significant step towards a more intelligent and user-centric approach to city navigation. Our results on different cities highlight the viability of our approach, and pave the way for future contributions to this field.

The rise of wearable devices offers numerous opportunities for monitoring human activities and behaviors, even outside hospital settings. Human Activity Recognition techniques utilize sensor data from wearables and smartphones to extract patterns and determine the activities performed. This study focuses on Human Activity Recognition for Haemophilia patients, to identify the optimal sensor positions for accurate activity detection. We have collected data from 5 key activities using multiple wearable devices, to determine the most informative features and device positions. Our results indicate that placing the devices on the ankle, closer to the source of movements, achieves the highest performance. Using such device, we are able to recognize these activities with F1 scores close to 1.

Road surface quality is a major concern for bicycle riders and plays an important role in the mobility infrastructure. In the era where smarter cities aim to increase the well-being of citizens and the efficiency of infrastructures, navigation systems relying on Mobile Crowdsensing (MCS) are mostly designed for car drivers, and account for road traffic conditions. To cover the gap, in this paper, we propose a full architectural pipeline of an MCS-based navigation system for bicycle riders that accounts for the road surface quality. The MCS paradigm leverages the sensor data produced by the personal devices of participating citizens to describe phenomena of common interest. Our system classifies road segments using inertial sensor data gathered by users, using a combination of supervised and unsupervised methods, as human labeling in this context is impractical and too subjective. We prove the efficacy of our method in a controlled environment, and then we implement and deploy the full system in a real city, finally reporting on its results.

In recent years, there has been a growth in the development of numerous software algorithms dedicated to pedometers (or step counters). This surge has subsequently spurred the creation of various context-aware smartphone applications for sports, healthcare, and other fields. Most works that compare commercial offerings do not adopt a sound and rigorous method, as human testers are asked to stick to a defined set of constraints, and experiments are carried out within controlled environments. However, each application is still tested separately, with no guarantee that the conditions are really the same, plus these conditions cannot resemble the real environment where pedometers are going to be used. Our proposal features a software solution that records the sensor readings of human testers and inject the exact same sensor values into different pedometer applications to produce a sound result by using the same testing conditions. We implement our solution and perform with it a comparison study.

The recognition of the activity of texting while driving is an open problem in literature and it is crucial for the security within the scope of automotive. This can bring to life new insurance policies and increase the overall safety on the roads. Many works in literature leverage smartphone sensors for this purpose, however it is shown that these methods take a considerable amount of time to perform a recognition with sufficient confidence. In this paper we propose to leverage the smartphone front camera to perform an image classification and recognize whether the subject is seated in the driver position or in the passenger position. We first applied standalone Convolutional Neural Networks with poor results, then we focused on object detection-based algorithms to detect the presence and the position of discriminant objects (i.e. the security belts and the car win-dow). We then applied the model over short videos by classifying frame by frame until reaching a satisfactory confidence. Results show that we are able to reach around 90 % accuracy in only few seconds of the video, demonstrating the applicability of our method in the real world.

Pervasive devices are now part of daily lives for a multitude of human beings, due to their ability to perform simple to more complex tasks. Scenarios like Industry 4.0 and drone delivery are only few of the several ones which benefit from autonomous and modern smart devices. Due to their tasks, almost all of these devices are battery powered, with some of them for which it is hard to preventively maintain it. Most of the works which tackles this problem rely on processes which could be unpractical in the real world due to complexity, time or cost constraints. In this paper we propose a novel methodology which leverages data obtained from normal charge and discharge cycles to diagnose the current battery for power fade faults and possibly perform maintenance before service interruption occurs. Tests performed on a real dataset demonstrate the feasibility of our approach.

Mobile Crowdsensing (MCS) is a paradigm involving a crowd of participants, called workers, into sensor data gathering campaigns through their personal devices. Some campaigns require workers to contribute with small amounts of geolocalized data at a constant rate, while being not directly aware of the global conditions of the system. In the scope of this reduced awareness, it is crucial to consider the privacy preservation of single workers at design time, as the disclosure of their exact location may lead to severe privacy issues. In this paper we design a privacy by design MCS framework that leverages variable rewards for workers willing to submit their location with an higher precision than others. Privacy is ensured through a negotiation phase that estimates the reward of the workers for different levels of location precision. This way, it helps them decide autonomously the spatial granularity of their data in order to preserve their privacy, yet obtaining a reward for their data. We design a metric based on k-anonymity to evaluate the level of privacy achieved, and validate the proposed framework over a real dataset. Our results show the efficacy of the framework as well as interesting effects caused by the topology of the environment.

Mobile Crowdsensing has become an important paradigm in the last decade for on-demand monitoring scenarios in Smart Cities and vehicular networks, when the deployment of a dedicated sensor network is no longer affordable. To foster the participation of a large user base, it is common to reward them on top of the amount and the quality of data provided. Regardless of the MCS policy adopted, this requires the crowdsourcer to keep track of the participants. Since the contributed data inherently carries sensitive spatio-temporal information, privacy problems arise if a malicious entity gains access to it; still, in some cases, the spatio-temporal precision is crucial for the benefit of the application and cannot be distorted. In this paper we propose a privacy preserving framework for opportunistic MCS scenarios that includes data collection and rewarding phases. The framework both retains the precision of spatio-temporal information and limits the sensitivity of information disclosed through an algorithm that clusters the data points into low correlated sets. The framework is agnostic about how correlation is calculated, and we propose three exemplary correlation functions. We evaluate our framework against six real world datasets, assessing its efficacy and envisioning its implementation in practical deployments.

The Internet of Things and more recently the Web of Things are changing how we interact with devices. The possibilities and novel services they provide enables the users to perform automatic operations and to monitor data of interest. Although many operations are performed autonomously by devices, there is still the need for the user to understand the data provided, and to configure their own services according to it. In this work we explore the possibility for devices to autonomously organize and understand the effects of the actions on the scenario, and provide a better status of the system. We do so by presenting a novel architecture, and developing a Q-learning algorithm which learns from the different statuses in which the system is. Our results indicate that devices with no prior knowledge of each other may eventually collaborate to provide a novel service to the end user, without any human intervention, and eventually achieve a better system status.

The Internet of Things is permeating our everyday life and the number of sensors and actuators around us is increasing at an exponential pace. Data generated by such heterogeneous devices is hard to organize, therefore, in pervasive scenarios like Smart Cities, there is an increasing need for service infrastructures that play the role of intermediary between citizen and things. Often, end users call for customized services that are tailored to their specific need rather than general-purpose ones. For this reason, in this paper we propose a service architecture based on End-User Service Composition (EUSC), through which individuals can aggregate primary sources of data to compose services. Furthermore, we investigate the requirements for service reusability and inherently leverage a hierarchical paradigm by introducing a specific class of composition languages. Finally, we show our Proof-of-Concept (PoC) middleware implementation, namely SenSquare, to show how this is achievable in a real deployment through visual programming, specifically illustrating how hierarchization is achieved.

Wearable sensors and Internet of Things technologies are enabling automated health monitoring applications, where signals captured by sensors are analyzed in real-time by algorithms detecting health issues and conditions. However, continuous clinical-level monitoring of patients in everyday settings often requires computation, storage and connectivity capabilities beyond those possessed by wearable sensors. While edge computing partially resolves this issue by connecting the sensors to compute-capable devices positioned at the network edge, the wireless links connecting the sensors to the edge servers may not have sufficient capacity to transfer the information-rich data that characterize these applications. A possible solution is to compress the signal to be transferred, accepting the tradeoff between compression gain and detection accuracy. In this paper, we propose SIC-EDGE: a "semantic compression" framework whose goal is to dynamically optimize the resolution of an electrocardiogram (ECG) signal transferred from a wearable sensor to an edge server to perform real-time detection of heart diseases. The core idea is to establish a collaborative control loop between the sensor and the edge server to iteratively build a semantic representation that is: (i) ECG-cycle specific; (ii) personalized, and (iii) targeted to support the classification task rather than signal reconstruction. The core of SIC-EDGE is a Sequential Hypothesis Testing (SHT) algorithm that analyzes partial representations along the iterations to determine which and how many representation layers (wavelet coefficients in our implementation) are requested. Our results on established datasets demonstrates the need for adaptive "semantic" compression, and illustrate the dynamic compression strategy realized by SIC-EDGE. We show that SIC-EDGE leads to an increase in terms of recall and F1 score of up to 35% and 26% respectively compared to an optimized but static wavelet compression for a given maximum channel usage.

Sensory data generated around us in the context of IoT is huge and heterogeneous. To fully unleash the potential of IoT Open Data there is a need for service infrastructures that facilitate the interaction of users with such data, especially when they are able to customize such services to fit their needs. In this paper we propose to use our tool SenSquare for IoT End-User Service Composition, by presenting its main features and its recent advances towards providing a data historian and importing other services, as well as evaluating the performance of its implementation in parallel.

With the ever-increasing popularity of fitness trackers, data on the time and location of popular walking, running, and bicycling routes is expansive and growing rapidly. This data is currently used primarily for route discovery and personal fitness tracking, but it may also be leveraged to build ad-hoc transportation flows. We present a novel model that creates delivery networks from these zero-emission transportation flows, and we evaluate the model using data from two popular datasets. Our results indicate that such networks are indeed possible, and can help reduce traffic, emissions, and delivery times. Moreover, we demonstrate how our results can be consistently reproduced in different cities with different subsets of carriers.

Mobile crowdsensing (MCS) has become a popular paradigm for data collection in urban environments. In MCS systems, a crowd supplies sensing information for monitoring phenomena through mobile devices. Typically, a large number of participants is required to make a sensing campaign successful. For such a reason, it is often not practical for researchers to build and deploy large testbeds to assess the performance of frameworks and algorithms for data collection, user recruitment, and evaluating the quality of information. Simulations offer a valid alternative. In this paper, we present CrowdSenSim 2.0, a significant extension of the popular CrowdSenSim simulation platform. CrowdSenSim 2.0 features a stateful approach to support algorithms where the chronological order of events matters, extensions of the architectural modules, including an additional system to model urban environments, code refactoring, and parallel execution of algorithms. All these improvements boost the performances of the simulator and make the runtime execution and memory utilization significantly lower, also enabling the support for larger simulation scenarios. We demonstrate retro-compatibility with the older platform and evaluate as a case study a stateful data collection algorithm.

Texting while Driving has been reported as one of the major sources of inattention by car drivers, leading to an increased probability of severe road accidents. In fact, notifications, messages and other interactions with mobile devices may make the driver unaware of road and traffic events. To prevent or mitigate this issue, solutions have been proposed that either block the smartphone when inside the vehicle or recognize the activity to issue monetary fines at a later time. This paper proposes a classification framework capable to identify the location of a device within the vehicle using data from integrated sensors. This allow more selective countermeasures targeted specifically to mobile devices used by the driver, rather than by any person inside the vehicle. The framework extracts sensor data from the smartphone, computes ad-hoc features and feeds them to a neural network. Different from prior work, we demonstrate that accurate detection can be achieved even using only one device by combining subsequent turns of the vehicle.

Context-aware computing uses the wide range of information produced by mobile platforms to optimize the parameters of applications providing important services. However, as some of the data are either publicly exposed or can be acquired by malicious parties, a privacy issue arises. Recent studies extend this concept to context prediction, that is, the ability to forecast the future user context from current or past data. Whereas privacy in context-aware computing has been widely studied, the issue of impairing the ability to predict user context remains largely unexplored. This paper presents a framework based on Markov process theory to reduce the accuracy of predictions made by a malicious observer. Rather than attempting to hide current context, which is often purposely exposed by the user, the proposed methodology injects manufactured, and temporary, context updates to impair prediction. Numerical results from FourSquare databases demonstrate the privacy increase granted by the proposed technique and illustrate the tradeoff between user privacy and noise injection.

Internet of Things (IoTs)-based monitoring applications usually involve large-scale deployments of battery-enabled sensor nodes providing measurements at regular intervals. In order to guarantee the service continuity over time, the energy-efficiency of the networked system should be maximized. In this paper, we address such issue via a combination of novel hardware/software solutions including new classes of Wake-up radio IoT Nodes (WuNs) and novel data- and hardware-driven network management algorithms. Three main contributions are provided. First, we present the design and prototype implementation of WuN nodes able to support two different energy-saving modes; such modes can be configured via software, and hence dynamically tuned. Second, we show by experimental measurements that the optimal policy strictly depends on the application requirements. Third, we move from the node design to the network design, and we devise proper orchestration algorithms which select both the optimal set of WuN to wake-up and the proper energy-saving mode for each WuN, so that the application lifetime is maximized, while the redundancy of correlated measurements is minimized. The proposed solutions are extensively evaluated via OMNeT++ simulations under different IoT scenarios and requirements of the monitoring applications.

Nowadays, Location Based Services (LBS) are fore- seen as a fundamental building block of modern mobile applications and services. Important examples of LBS concerns indoor environments in which GPS technology cannot be used. On the other hand, the great diffusion of pervasive Mobile Devices (MDs) as smartphones and tablets has enabled many positioning techniques, such as WiFi FingerPrinting (FP), that exploits all the MD's embedded sensors. This paper proposes and investigates the performance of a method exploiting a WiFi FP algorithm for indoor localization fed with information from the barometer to estimate the floor in which the MD is located. Our results, carried out in indoor areas at the University of Genoa (UniGE) and at the University of Bologna (UniBO), show that when more than 5 Access Points (APs) are used the proposed 3D positioning system is able to accurately localize the user with an error below 2 and 1.2 and meters for the UniBO and UniGE case, respectively.

Upcoming mobile network technologies developed in the context of 5G and DSRC are expected to finally legitimize direct data transfers among vehicles as a standard communication paradigm. We investigate fundamental properties of the topology of vehicular networks built on top of these emerging vehicle-to-vehicle communication technologies. Our study yields multiple elements of originality: ( i) it addresses temporal connectivity, which has been poorly investigated despite a high relevance for vehicular network operations; (ii) it introduces exact but computationally efficient models of the temporal connectivity of vehicular networks; (iii) it evaluates the proposed models in urban settings that exhibit an unprecedented combination of dependability, scale and generality of vehicular mobility. This approach lets us unveil an apparent scale-and city-invariant law of temporal reachability in vehicular networks. Finally, we open our original scenarios to the research community, so as to ensure reproducibility of our results and foster further investigations of vehicular network performance.

The collaborative Internet of Things (C-IoT) is an emerging paradigm that involves many communities with the idea of cooperating in data gathering and service sharing. Many fields of application, such as smart cities and environmental monitoring, use the concept of crowdsensing in order to produce the amount of data that such Internet of Things (IoT) scenarios need in order to be pervasive. In this paper we introduce an architecture, namely SenSquare, able to handle both the heterogeneous data sources coming from open IoT platform and crowdsensing campaigns, and display a unified access to users. We inspect all the facets of such a complex system, spanning over issues of different nature: we deal with heterogeneous data classification, mobile crowdsensing management for environmental data, information representation, and unification, IoT service composition and deployment. We detail our proposed solution in dealing with such tasks and present possible methods for meeting open challenges. Finally, we demonstrate the capabilities of SenSquare through both a mobile and a desktop client.

The Internet of Things (IoT) environments are no more a vision as they are already surrounding the everyday life of citizens. Its pervasive nature brings IoT ecosystems closer and closer to the end users, facilitating their domestic lives through home automation appliances and platforms. Several M2M communication technologies and data representation techniques have been standardized and often established by the manufacturers. For such reasons, many Home Automation Systems (HAS) are irreconcilable due to the incompatibility of their communication technologies. In order to take a step towards HAS interoperability, in this paper we propose RouteX, an experimental cross-technology platform able to manage home devices belonging to different networks and using different technologies. It leverages the potential of Service Oriented Architectures (SOA), providing the user with an abstraction layer over a multitude of home sensors and actuators. We also present a practical user interface, through which the user is able to administrate efficiently all his or her home appliances no matter which technology they use.

The integration of automotive communications in 5G systems must build on a clear understanding of the performance of services for connected vehicles in today's LTE deployments. In this paper, we carry out a simulation-based performance evaluation of automotive communications in LTE, with particular attention to realism: to that end, we investigate the impact of different road traffic models, employ a state-of-the-art commercial LTE tool, and study a practical service use case. Our results demonstrate that unrealistic road traffic datasets can bias network simulations in urban vehicular environments, and provide insights on the limitations of the current radio access architecture, when confronted to connected vehicles.

Nowadays, people usually connect to the Internet through a multitude of different devices. Video streaming takes the lion's share of the bandwidth, and represents the real challenge for the service providers and for the research community. At the same time, most of the connections come from indoor, where Wi-Fi already experiences congestion and coverage holes, directly translating into a poor experience for the user. A possible relief comes from the TV white space (TVWS) networks, which can enhance the communication range thanks to sub-GHz frequencies and favorable propagation characteristics, but offer slower datarates compared with other 802.11 protocols. In this paper, we show the benefits that TVWS networks can bring to the end user, and we present CABA, a connection aware balancing algorithm able to exploit multiple radio connections in the favor of a better user experience. Our experimental results indicate that the TVWS network can effectively provide a wider communication range, but a load balancing middleware between the available connections on the device must be used to achieve better performance. We conclude this paper by presenting real data coming from field trials in which we streamed an MPEG dynamic adaptive streaming over HTTP video over TVWS and Wi-Fi. Practical quantitative results on the achievable quality of experience for the end user are then reported. Our results show that balancing the load between Wi-Fi and TVWS can provide a higher playback quality (up to 15% of average quality index) in scenarios in which the Wi-Fi is received at a low strength.

In this article, we assess the viability of underlay sensor networks in frequencies used by an incumbent digital TV broadcasting system, that is, in the so-called TV grayspaces (TVGS). Grayspace operations are particularly interesting when other unlicensed bands are overcrowded, for example, due to high-volume WiFi operations. We simulate the operational characteristics of the recent IEEE 802.15.4m standard for low-rate wireless personal area networks to evaluate the performance degradation of an incumbent Digital Video Broadcasting - Terrestrial (DVB-T) system if a secondary network of low-power low-rate devices are co-deployed in the same frequency bands. Our results show that short sensor messages will not disrupt the DVB-T service due to the existing error-correction capabilities. Furthermore, if sufficient separation distances to primary transmitters are maintained, transmit powers are sufficient to achieve reasonable connectivity levels of the secondary network. In order to obtain realistic figures on the predicted feasibility of grayspace sensor networks, we study the deployment constraints of a hypothetical secondary network co-located with the TV broadcasting network of Germany. Our analysis shows that if we aim to support a minimum sensor-sensor distance, no universal coverage can be maintained in this country. While our quantitative results are specific to Germany, we deem them indicative for the expected results also in other potential deployments. We found that while a secondary wireless sensor network in TVGS is technically possible, the necessary constraints on operational parameters and service levels for TVGS co-existence will significantly limit its practical viability.

Modern smartphones are nowadays equipped with a multitude of sensors, which extend their capabilities paving the way for a multitude of services. Among these, the ability to locate the device is exploited by many. While outdoor the GPS provides good accuracy, indoor localization is challenging to be performed with it, as buildings shadow the satellite signal. In particular, the barometric pressure sensor is often used to determine the altitude of the device from the ground floor, particularly for safety applications and indoor navigation. However, pressure changes during the day, and thus it is challenging to bind a static value to a specific altitude. In this work, we propose a self-adapting algorithm able to determine the height at which the device is in a building, by exploiting the barometric pressure. We implemented and tested our algorithm on an Android application, and we compared it against other techniques. We tested our proposal for three specific use-cases, and our results show the benefit of our proposal.

It is our great pleasure to welcome you to the Fourth IEEE International Workshop on Emerging COgnitive Radio Applications and aLgorithms (CORAL 2016), held in Coimbra, Portugal on June 21, 2016, in conjunction with the IEEE WoWMoM 2016 Conference.

Abstract In this paper, we study the problem of how to detect the current transportation mode of the user from the smartphone sensors data, because this issue is considered crucial for the deployment of a multitude of mobility‐aware systems, ranging from trace collectors to health monitoring and urban sensing systems. Although some feasibility studies have been performed in the literature, most of the proposed systems rely on the utilization of the GPS and on computational expensive algorithms that do not take into account the limited resources of mobile phones. On the opposite, this paper focuses on the design and implementation of a feasible and efficient detection system that takes into account both the issues of accuracy of classification and of energy consumption. To this purpose, we propose the utilization of embedded sensor data (accelerometer/gyroscope) with a novel meta‐classifier based on a cascading technique, and we show that our combined approach can provide similar performance than a GPS‐based classifier, but introducing also the possibility to control the computational load based on requested confidence. We describe the implementation of the proposed system into an Android framework that can be leveraged by third‐part mobile applications to access context‐aware information in a transparent way. Copyright © 2016 John Wiley & Sons, Ltd.

Spectrum scarcity demands for additional bandwidth where new services can be deployed on. However, today's spectrum allocation leaves almost no bands unallocated. Thus, Cognitive Radio has been studied to bring relief to the lack of spectrum, moving towards a more efficient and dynamic spectrum access. In this domain TV White Space have been proposed as a possible solution to bring new, valuable spectrum for opportunistic services. However, their availability is quite low in highly populated areas, and thus their viability is limited. This is mainly because the availability of TV White Space is typically considered at the rooftop, through two-dimensional propagation models which do not account for possible spectrum re-utilization policies within a building, or in a small-scale area. In this paper, we show that much more communication opportunities can be found when we consider also the third-dimension, i.e. the height from the terrain, and novel per-floor allocation policies. We propose three main contributions in this paper. First, we describe an analytical model through which we derive the number of available spectrum resources for indoor secondary networks, considering PU protection policies in the same building, and in surrounding buildings. Second, we estimate the number of TV Gray Space (TVGS) over realistic scenarios in candidate cities, considering realistic street topology and buildings locations, and we show that this value can be much higher than what reported in the spectrum database. Finally, we investigate co-existence of secondary networks on TVWS, when novel per-floor spectrum sharing models are used.

Guaranteeing network connectivity in post-disaster scenarios is challenging yet crucial to save human lives and to coordinate the operations of first responders. In this paper, we investigate the utilization of low-altitude aerial mesh networks composed by Small Unmanned Aerial Vehicles (SUAVs) in order to re-enstablish connectivity among isolated end-user (EU) devices located on the ground. Aerial ad-hoc networks provide the advantage to be deployable also on critical scenarios where terrestrial mobile devices might not operate, however their implementation is challenging from the point of view of mobility management and of coverage lifetime. In this paper, we address both these issues with three novel research contributions. First, we propose a distributed mobility algorithm, based on the virtual spring model, through which the SUAV-based mesh node-called also Repairing Units (RUs) in this study- can self-organize into a mesh structure by guaranteeing Quality of Service (QoS) over the aerial link, and connecting the maximum number of EU devices. Second, we evaluate our scheme on a realistic 3D environment with buildings, and we demonstrate the effectiveness of the aerial deployment compared to a terrestrial one, in terms of coverage and wireless link reliability. Third, we address the problem of energy lifetime, and we propose a distributed charging scheduling scheme, through which a persistent coverage of RUs can be guaranteed over the emergency scenario.

With an increasing demand for monitoring energy consumption at granularity levels down to single household appliances, it is necessary to develop new means to collect sensor measurements in a robust and cost-efficient manner. The smart grid paradigm foresees using wireless links for data transfer, albeit no dedicated spectrum bands have been designated for this purpose. In this paper we study the feasibility of opportunistic spectrum access for smart grids, and focus on underlay spectrum sharing over occupied TV channels. These frequency bands, which are commonly denoted as TV gray spaces, provide superior propagation characteristics, but are locally used by high-power (mostly DTV) broadcasting transmitters. For selected reference geometries of intra-meter and meter-to-operator communications, we study the smart meter performance (in terms of achievable throughput and transmission range), and the necessary power limits. We compare our results from a small-scale measurement campaigns against existing wireless technologies for low-power communications in other adjacent bands. Our results show that wall shielding and fading in indoor to outdoor propagation channels sufficiently protects the primary system from the interference introduced by gray space meter-to-meter communications, but that the required transmit powers to send operations data from indoor meters to outdoor collection point severely limit the applicability of TV gray spaces for such network topologies.

Nowadays several mobile applications connect to the internet through 2G/3G/LTE, which are becoming more crowded. Cognitive wireless networks have been proposed as a possible solution to supply additional bandwidth, and more recently TV White Spaces (TVWS) have been investigated as one candidate. TVWS devices should contact a remote spectrum database, which will reply with the channels available to use. It is not specified how devices should contact the remote spectrum database, so in this work we focus on the usage of a cellular connection, where however the number of the queries could rapidly grow and occupy considerable bandwidth. In this paper we present the idea of Femto-Databases, i.e. devices which act as distributed mobile databases able to satisfy the spectrum requests by opportunistic devices. Extensive simulations through the Omnet++ platform show that our approach can effectively reduce the load on the cellular infrastructure, and improve the latency of the query communication to the remote spectrum database.

The spectrum scarcity is a known problem for a multitude of services. Several bands have been licensed, and nowadays it is difficult to find unused spectrum. Cognitive radio networks have been proposed as a possible solution to contrast the experienced spectrum scarcity. One case of particular interest come from the scarce utilization of TV frequencies, which form the so-called TV White Spaces. In this paper we investigate the utilization of occupied frequencies by secondary devices for indoor communication. We conduct spectrum measurements to quantify the availability of spectrum, and study how indoor communications could impact the DTV receiver. We show that this portions of spectrum, called gray spaces, can be utilized under certain circumstances, for example in highly populated areas, which is the scenario in which it is harder to find TV White Spaces. Simulation studies show the impact gray spaces can have on the available spectrum for opportunistic use.

Machine-to-Machine communications is envisioned to become one of the fundamental pillars of the future Internet of Things paradigm, enabling platoons of devices to be seamlessly connected and to cooperate over smart spaces. Among the possible application scenarios, smart metering represents an already existing technology that might take benefit from the capability of autonomous configuration and setup of M2M networks. At present, smart meters communicate over the 2G/3G network, however the utilization of the cellular technology poses several problems, such as low coverage and spectrum shortage over dense areas. To overcome these issues, in this paper we investigate the application of cognitive radio principles over TV White Spaces to M2M communication for the smart metering scenario. Following the recent regulations of FCC and Ofcom, that foresees the presence of a spectrum database for TV white spaces detection, we study the trade-off between protection of licensees and energy consumption in a cluster of smart meters. We provide three novel research contributions: (i) an analytical model to estimate the lifetime of a cluster of smart meters; (ii) centralized and distributed algorithms to determine the schedule operations of Master/Slave devices foreseen by the spectrum regulations; (iii) performance evaluation of the proposed framework through extensive Omnet++ simulations.

In this paper, we address the problem of re-establishing the network connectivity in post-disaster scenarios, where the original wireless infrastructure has been partitioned into multiple network fragments (called islands), operating on different frequencies. To this purpose, we propose the utilization of swarms of dedicated repairing units, called Stem-Nodes (SNs). SNs are provided with Cognitive Radio (CR) and self-positioning capabilities, in order to offer maximum reconfigurability in terms of mobility and wireless technologies supported. Moreover, swarms of SNs can self-organize into STEM-Mesh structure, that works as a dynamic backbone to connect heterogeneous islands using different technologies (e.g. Wi-Fi, Wi-MAX, etc). In this paper, we present three contributions pertaining to STEM-Mesh: (i) we describe a distributed motion control scheme (based on virtual springs approach) that enables SNs to self-organize into dynamic STEM-Mesh structures, (ii) we introduce a discovery scheme, through which SNs can explore the scenario in both spatial and frequency domains, and possibly connect the islands to the STEM-Mesh backbone and (iii) we validate the correctness of the proposed scheme, by verifying the optimal placements of the SNs composing the STEM-Mesh on a simplified scenario (e.g. chain topology). Finally, we evaluate through Omnet++ simulations the ability of STEM-Mesh to maximally re-establish connectivity on partitioned network scenarios.

The Internet of Energy (IoE) for Electric Mobility is an European research project that aims at deploying a communication infrastructure to facilitate and support the operations of Electric Vehicles (EVs). In this paper, we present three research contributions of IoE. First, we describe a software architecture to support the deployment of mobile and smart services over an Electric Mobility (EM) scenario. The proposed architecture relies on an ontology-based data representation, on a shared repository of information (Service Information Broker), and on software modules (called Knowledge Processors -KPs) for standardized data access/management. As a result, information sharing among the different stakeholders of the EM scenario (i.e. EVs, EVSEs, City Services, etc) is enabled, and the interoperability of smart services offered by heterogeneous providers is guaranteed by the common ontology. Second, we rely on the proposed architecture to develop a remote charging reservation system, that runs on top of mobile smarthphones, and allows drivers to monitor the current state-of-charge of their EV, and to reserve a charging slot at a specific EVSE. Finally, we validate our architecture through a benchmark framework, that supports the embedding of mobile EV applications and of real KPs into a simulated vehicular scenario, including realistic traffic, wireless communication and battery models. Evaluation results confirm the scalability of our architecture, and the ability to support EVs charging operations on a large-scale scenario (i.e. the downtown of Bologna).