The deployment of Digital Twins (DTs) in heterogeneous Internet of Things (IoT) environments remains challenging due to protocol diversity, infrastructure complexity, and the absence of standardized methodologies. Existing solutions rely on ad-hoc architectures with manual configuration, limiting scalability and adaptability. This paper introduces DTDeployer, a platform for automated DT deployment and lifecycle management inspired by container orchestration. DTDeployer adopts a declarative configuration model, enabling users to define complex ecosystems through simple specification files, with automated provisioning of ingestion, processing, and output components. Key features include template-driven composability, multi-protocol data ingestion (HTTP, WebSockets, MQTT, AMQP, Kafka), integrated time-series storage, and developer-centric code generation for type-safe processing in Java and Python. By combining requirements analysis with a comprehensive architecture, this work establishes DTDeployer as a foundation for enterprise-grade DT deployments, lowering adoption barriers while supporting complex industrial applications.

Building Automation and Control Systems increasingly rely on BACnet/IP to interconnect heterogeneous field devices and supervisory applications. Although BACnet Secure Connect provides end-to-end protection via TLS, its adoption in smart-building deployments is hindered by the limited capabilities of legacy devices and the additional communication overhead. In this paper, we propose an in-network security approach for BACnet/IP communications based on two P4-programmable boundary switches that transparently provide confidentiality, integrity, and authentication across exposed network segments without requiring modifications to BACnet endpoints. The solution combines AES-based encryption with support for 128-, 192-, and 256-bit keys and SHA-256 HMAC protection. The proposed approach is validated on a virtualized testbed that we developed on top of the NIST Net-Zero Energy Residential Test Facility (NZERTF) HVAC reference scenario. Experimental results show that the proposed in-network approach achieves a mean RTT between 1393μs and 1541μs, thus reducing latency by 22.8%−30.2%with respect to BACnet/SC (1995 μ s), while remaining above plaintext BACnet/IP (951 μs) by 46.5%−62.0%.

Industrial Control Systems (ICS) are increasingly interconnected with enterprise IT and cloud services, yet their communications remain largely unprotected due to the limited adoption of Transport Layer Security (TLS) and other cryptographic standards. Legacy devices often lack the resources to support TLS, and operators face performance constraints and complex certificate management. To address this gap, we present P4ICS, a framework that provides confidentiality, integrity, and replay protection for industrial protocols by shifting security functions from endpoints into P4-programmable switches. P4ICS transparently parses and protects Modbus, DNP3, EtherNet/IP, and MQTT traffic, establishing switch-to-switch encrypted tunnels that secure untrusted network segments while preserving interoperability with legacy equipment. Our evaluation on an ad hoc physical testbed shows that P4ICS introduces only a modest overhead compared to plaintext communication, while consistently outperforming TLS, reducing delays by about 12% for Modbus and DNP3, 43% for EtherNet/IP, and 47% for MQTT. By leveraging in-network computing, P4ICS delivers a practical and deployable security layer for Industry 4.0 communications, narrowing the gap between available secure protocol profiles and their limited use in operational ICS.

The adoption of Digital Twin (DT) technologies in public transport systems, particularly bus networks, is gaining momentum as cities seek smarter, more responsive, and efficient mobility solutions. Enabled by advances in IoT, AI, and Big Data Analytics, DTs offer real-time monitoring, simulation, and optimization of transit operations. However, despite their potential, the application of DTs in bus-based public transport remains relatively underexplored and fragmented across the literature. This study presents a Systematic Literature Review (SLR) aimed at synthesizing current research on DT technologies in this domain. Specifically, it investigates architectural models, technological frameworks, and platform designs; examines how AI and machine learning models are integrated to support operational tasks; and analyzes the role of Human-Computer Interaction (HCI) in the design and usability of such systems. By identifying key trends, challenges, and research gaps, this work provides a structured overview of the current landscape. Furthermore, it outlines directions for future research in DT-enabled public transportation systems.

Crowd management is the problem of dealing with physical crowds of humans, e.g., for safety reasons or to improve services through crowd-awareness. Information and communication technology can be of great help in supporting the monitoring and management of crowds. Following the “digital twin concept”, a group or crowd of people – as a dynamic physical entity – is amenable to be represented and managed through digital models. Accordingly, in this work we present and develop a general notion of a Crowd Digital Twin (CDT) , intended as the digital twin associated to physical gatherings of humans. We motivate it as a fundamental component of crowd management systems, and investigate the functional and non-functional requirements that such a technical solution should satisfy. Then, we consider what the fundamental components of CDT systems are, and accordingly delineate a reference architecture aimed at flexibly supporting the corresponding physical-digital thread. We also discuss methods and tools that might be useful for supporting implementations of such a CDT concept, and delineate a set of challenges that may serve as a roadmap for future research on the topic.

Crowd management is the problem of dealing with physical crowds of humans, e.g., for safety reasons or to improve services through crowd-awareness. In previous work, the notion of a Crowd Digital Twin (CDT), namely the digital twin associated to physical gatherings of humans, was motivated and proposed as a fundamental component of crowd management systems. In this work, we better characterise the CDT notion, by investigating the functional and non-functional requirements that such a technical solution should satisfy. Then, we consider what are the fundamental components of CDT systems, and accordingly delineate a general architecture aimed at flexibly supporting the corresponding physical-digital thread.

The development of a crowd digital twin offers significant potential for enhancing public safety, urban planning, and event management. A key challenge in creating such a digital twin lies in the efficient and accurate acquisition of crowd-related data, particularly through object detection models deployed on resource-constrained devices. Through a series of experiments, we compare TinyML and Edge approaches in terms of detection accuracy, inferencing time, and resource utilization. Our findings highlight the trade-offs inherent in selecting detection models for crowd digital twin applications, underscoring the importance of aligning model choice with specific deployment needs.

Accurate demand forecasting is essential for optimizing public transportation systems, reducing congestion, and improving service efficiency. Traditional methods often struggle to capture the dynamic nature of urban mobility, leading to increasing reliance on machine learning and deep learning techniques. This paper explores various machine learning models for bus transportation demand forecasting, leveraging real-world data to evaluate their effectiveness. Our findings contribute to the development of intelligent public transport systems, supporting data-driven decision-making for adaptive and sustainable urban mobility solutions.

Adaptive Cruise Control (ACC) is a widely adopted Advanced Driver Assistance System (ADAS) technology that enhances driving comfort and safety. As part of the broader effort to advance autonomous driving, this paper investigates the imple-mentation of ACC systems using two approaches: a standard PID controller, valued for its simplicity, reliability, and widespread use, and a machine learning-based model, designed to offer similarly straightforward usability. The experiments were carried out within the CARLA simulator, a platform that provides a controlled environment for testing and evaluation. By comparing the performance of these two methods, the study aims to evaluate whether machine learning can achieve improved results while retaining the ease of implementation associated with traditional PID controllers. This comparative study highlights the strengths and limitations of both approaches, offering valuable insights into how machine learning can enhance ACC functionality and contribute to safer autonomous driving systems.

The Internet of Things (IoT) is a global network of intelligent interconnected objects, enabling remote recognition, detection, and control. Through virtualization, these devices exchange knowledge in the Cloud, bridging physical and virtual worlds. However, the rapid growth in connected objects and data production necessitates innovative system designs. While virtualization offers scalability and flexibility, it proves insufficient alone. Recent studies recommend integrating social network concepts into IoT, forming the Social Internet of Things (SIoT). SIoT, mirroring human social networks, enhances smart devices with relational capabilities, addressing challenges in managing the surge in connected devices. This paper introduces the “Virtual User” concept-an autonomous entity streamlining user-device interactions in the digital realm. The research proposes, elucidates, and evaluates a Cloud infrastructure integrating the Virtual User into the SIoT paradigm. This novel approach offers a modular, extendable, and secure communication platform, marking the first Cloud solution tailored for the Virtual User concept, leveraging SIoT virtualization through containerization and automated service setup processes.

Driving involves numerous factors demanding a driver's attention. To enhance road safety, there has been a significant increase in the development and implementation of vehicle sensors. These sensors, in conjunction with mobile applications, can assess a driver's emotional state and focus, providing feedback to enhance their attention. This paper explores the monitoring of driver behavior, emphasizing the effects of distractions and emotions on driving performance. Stemming from the European initiative, NextPerception, this research focuses on advancing perception sensors and refining distributed intelligence models in various domains, including automotive. The initiative aims to develop a range of sensors, from obstacle detection tools to those monitoring a driver's eye movements and vital signs. A key goal is to define a “fitness-to-drive” metric, representing the driver's attentiveness level. The project also seeks to use gamification to emphasize the importance of this metric, increasing drivers' awareness of their driving skills. The ultimate aim is to create a system that determines fitness-to-drive based on distractions and emotions, integrating this into a prototype simulating sensor data, and introducing a web-based application to display this data for a community of drivers.

The term Digital Twin (DT) refers to an emerging concept and technology extending the value of physical assets through the services supported by a bidirectional connection with their digital counterparts. Since its very introduction, the DT concept has been applied to empower several kinds of physical assets across multiple domains, including Industry 4.0, infrastructures, autonomous vehicles, e-health, to name a few. The rise of socio-technical systems and cyber-physical-human systems, which feature humans as key system components, has led to the extension of the DT concept to humans themselves, bringing in the notion of "human digital twin", supporting human-system integration. However, in several contexts, humans are not just isolated entities, but rather form groups or crowds. The peculiar dynamics of such groups have to be carefully considered in various scenarios like emergency management or logistics scenarios. Motivated also by recent trends in social Internet of Things and collective computing, in this paper, we propose the notion of a "crowd digital twin", meant to support services considering the crowd and for the crowd itself. Supporting a bidirectional connection with a physical crowd, this notion is shown to present peculiar requirements, opportunities, and challenges, including low-latency synchronisation, environment-driven prediction, actuation, and validation of emergent behaviour.

With the integration of virtualization technologies, the Internet of Things (IoT) is expanding its capabilities and quickly becoming a complex ecosystem of networked devices. The Social Internet of Things (SIoT), where intelligent things include social properties that improve functioning and user engagement, is the result of this progress. The SIoT still has issues with scalability, data management, and user-centric operations, despite tremendous progress. In order to overcome these obstacles, a strong architecture is needed that can handle the enormous number of IoT devices while simultaneously streamlining the user interface. This study provides a unique architecture for the IoT that uses containerization to efficiently deploy and manage services while integrating Virtual Users (VUs) and Social Virtual Objects (SVOs) into a scalable Cloud/Edge infrastructure. These innovative aspects collectively advance previous works presented in literature and focused on novel SIoT architectures and implementations, by addressing key challenges in scalability, efficiency, and automation within the SIoT. The proposed method presents an extensible, modular architecture that lets VUs self-manage IoT services, making user administration easier and improving system security and scalability. Important parts of the design include a host controller for container orchestration, a deployer for automated service deployment, and user clusters for aggregating VUs, SVOs, and apps to provide secured and efficient data sharing. We show through experimental assessment that the architecture can manage high-volume installations and operating needs, exceeding the conventional platform based on Google App Engine in terms of system overhead and deployment timeframes. The obtained results highlight how our suggested architecture, which provides an easy-to-use, scalable, and secure foundation for IoT deployments, has the potential to advance the SIoT landscape.

Several elements can affect the drivers' behaviour while they are performing driving activities. Ranging from visual to cognitive distractions, emotions and other drivers' conditions (that could emerge from biometric data, such as temperature, heartbeat, pressure, etc.) can play a significant role, performing as a factor that can increase drivers' response time. This could be crucial in avoiding dangerous situations and in deciding and performing actions that could influence the happening of car accidents. This paper introduces the concept of the “Fitness-to-Drive” index and aims to evaluate how the arousal effects can influence the drivers' status. The paper presents some experimental evaluations we have conducted on a driver simulator, discussing the obtained results.

Recently in the world of boats we hear about “smart boats”, or 3.0 connected boats. For many years, technology has entered the world of boats, especially as regards safety, some systems have become mandatory. With the spread of IoT systems, boats could be considered as sources of information for other boats in the same sea area or for the construction of coastal IoT services. In the marine context, measurements of environmental values such as sea water temperature, wave period and height or direction of currents, are carried out using buoys of considerable size and at a great distance from the coast. In this paper we want to present a Social Internet of Things system that allows the distribution of information between boaters and the integration with fixed IoT networks consisting of buoys and coastal stations. A preliminary LoRa mesh communication test is presented showing a stable mesh network in which each node is able to communicate in the sea with a radius of 5 Km.

In the last years, the smart city (SC) paradigm has been deeply studied to support sustainable mobility and to improve human living conditions. In this context, a new SC based on the Social Internet of Things paradigm is presented in this article. Starting from the tracking of all vehicles (that is, private and public) and pedestrians, integrated with air quality measurements (that is, in real time by mobile and fixed sensors), the system aims to improve the viability of the city, both for pedestrian and vehicular users. A monitoring network based on sensors and devices hosted on board in local public transport allows real-time monitoring of the most sensitive areas both from traffic congestion and from an environmental point of view. The proposed solution is equipped with an appropriate intelligence that takes into account instantaneous speed, type of traffic, and instantaneous pollution data, allowing to evaluate the congestion and pollution condition in a specific moment. Moreover, specific tools support the decisions of public administration facilitating the identification of the most appropriate actions for the implementation of effective policies relating to mobility. All collected data are elaborated in real time to improve traffic viability suggesting new directions and information to citizens to better organize how to live in the city.

The idea of equipping smart devices with the capacity to form social connections with their peers is gaining popularity. Indeed, it is a feature with a lot of promise for fostering device collaboration, speeding up and improving service discovery, and evaluating the reliability of devices by utilizing appropriate features based on peer evaluations in social networks. While numerous architectural solutions have been offered along with various concepts for creating the devices’ social network, the question of how to build and sustain the devices’ social connections has been neglected. This paper proposes a possible implementation of a Social Virtual Object as a constituent entity of the Social Internet of Things (SIoT) based on the concept of virtualization. In the experimental evaluation, the performances in the management of the relationships that originate from the proximity among objects and that represent the greatest difficulties of realization are analyzed. The field trials show the effectiveness of the proposed system in creating relationships and highlight some problems that should be addressed in the formalization of the social relationship in the SIoT.

The climatic conditions of places with a high tourist vocation greatly affect the flow of visitors. In the coastal context, measurements of environmental values such as sea water temperature, wave period and height or direction of currents, are carried out using buoys of considerable size and at a great distance from the coast. This results in a lack of information to satisfy the preferences of tourists and guarantee adequate safety in certain sea conditions. In this paper, we show the implementation of a small buoy that can be positioned near the beaches and that, through a Social Internet of Things platform, can share information with tourists. The first results of a real installation are also shown and what challenges need to be addressed for an efficient use of such a system.

As the world becomes increasingly interconnected, emerging and innovative sensing technologies are shaping the future of agriculture, with a special focus on sustainability-related issues. In this context, we envision the possibility to exploit Social Internet of Things for sensing of environmental conditions (solar radiation, humidity, air temperature, and soil moisture) and communications, deep learning for plant disease detection, and crowdsourcing for images collection and classification, engaging farmers and community garden owners and experts. Through, data fusion and deep learning, the designed system can exploit the collected data and predict when a plant would (or not) get a disease, with a specific degree of precision, with the final purpose to render agriculture more sustainable. We here present the architecture, the deep learning model, and the responsive Web app. Finally, some experimental evaluations and usability/engagement tests are reported and discussed, together with final remarks, limitations, and future work.

Coast erosion is a process that degrades a coastal profile and is mainly due to natural factors (e.g., related to climate change) and overcrowding (e.g., urbanization and massive tourism). While the first cause can be considered as a slow process, the growing presence of humans is leading to rapid aging of coasts. The Mediterranean Sea authorities are focusing on the necessity for a systematic and comprehensive approach to the management of littoral areas. In this context, Italy is searching for a promising solution to safeguard coasts but, at the same time, to manage in an intelligent and “green” way the big amount of tourists. Research communities all over the world indicated the Internet of Things (IoT) as a valid technology to develop solutions in order to try solving or mitigating the coastal erosion problem. IoT-based techniques allow to manage heterogeneous and massive data for real-time monitoring and decision making and can be used for coastal environment and crowd level monitoring. This article presents a monitoring system based on the Social IoT (SIoT), a new paradigm that defines a network where every node is an object capable of establishing the social relationships with other things in an autonomous way according to specific rules. Thanks to social relationships, all involved devices in the monitoring system (i.e., sensors, cameras, and smartphones) are able to collect and exchange information. The proposed system, developed and installed in Cagliari (Italy), is able to evaluate the occupational state of a beach considering environmental and crowding data collected by devices and feedback sent by users.

In recent years identifier/locator splitting (ILS) has been proposed as a promising future Internet solution to the so-called “semantic overload of addresses” problem. It is currently under debate whether the ILS approach, which entails identifier-to-locator resolution procedures, could ensure scalability, session continuity, and mobility across heterogeneous multiaccess networks as demanded by Internet of Things (IoT). In this paper, it is proposed to exploit the Social IoT concept to address the above issues. More specifically, a scheme is introduced that browses the social graph of devices to find information about the locator of the intended destination. Results show that the proposed solution outperforms the alternative ILS approaches in terms of number of hops and latency incurred to accomplish the resolution procedure, at the cost of a slight increase in the storage demands to track social relationships.

The Social Internet of Things (SIoT) is a novel communication paradigm according to which the objects connected to the Internet create a dynamic social network that is mostly used to: route information and service requests, disseminate data, and evaluate the trust level of each member of the network. In this paper, the SIoT paradigm is applied to a scenario where geolocated sensing tasks are assigned to fixed and mobile devices, providing the following major contributions. The SIoT model is adopted to find the objects that can contribute to the IoT application by crawling the social network through the nodes profile and trust level. A new algorithm to address the resource management issue is proposed so that sensing tasks are fairly assigned to the objects in the SIoT. To this, an energy consumption profile is created per device and task, and shared among nodes of the same category through the SIoT. The resulting solution is also implemented in the SIoT-based Lysis platform. Emulations have been performed, which showed an extension of the time needed to completely deplete the battery of the first device of more than 40% with respect to alternative approaches.

This paper presents Lysis, which is a cloud-based platform for the deployment of Internet of Things (IoT) applications. The major features that have been followed in its design are the following: each object is an autonomous social agent; the platform as a service (PaaS) model is fully exploited; reusability at different layers is considered; the data is under control of the users. The first feature has been introduced by adopting the social IoT concept, according to which objects are capable of establishing social relationships in an autonomous way with respect to their owners with the benefits of improving the network scalability and information discovery efficiency. The major components of PaaS services are used for an easy management and development of applications by both users and programmers. The reusability allows the programmers to generate templates of objects and services available to the whole Lysis community. The data generated by the devices is stored at the object owners cloud spaces. This paper also presents a use-case that illustrates the implementation choices and the use of the Lysis features.

The integration of social networking concepts into the Internet of things has led to the Social Internet of Things (SIoT) paradigm, according to which objects are capable of establishing social relationships in an autonomous way with respect to their owners with the benefits of improving the network scalability in information/service discovery. Within this scenario, we focus on the problem of understanding how the information provided by members of the social IoT has to be processed so as to build a reliable system on the basis of the behavior of the objects. We define two models for trustworthiness management starting from the solutions proposed for P2P and social networks. In the subjective model each node computes the trustworthiness of its friends on the basis of its own experience and on the opinion of the friends in common with the potential service providers. In the objective model, the information about each node is distributed and stored making use of a distributed hash table structure so that any node can make use of the same information. Simulations show how the proposed models can effectively isolate almost any malicious nodes in the network at the expenses of an increase in the network traffic for feedback exchange.

The integration of social networking concepts into the Internet of Things (IoT) has led to the so called Social Internet of Things (SIoT) paradigm, according to which the objects are capable of establishing social relationships in an autonomous way with respect to their owners. The benefits are those of improving scalability in information/service discovery when the SIoT is made of huge numbers of heterogeneous nodes, similarly to what happens with social networks among humans. In this paper we focus on the problem of understanding how the information provided by the other members of the SIoT has to be processed so as to build a reliable system on the basis of the behavior of the objects. We define a subjective model for the management of trustworthiness which builds upon the solutions proposed for P2P networks. Each node computes the trustworthiness of its friends on the basis of its own experience and on the opinion of the common friends with the potential service providers. We employ a feedback system and we combine the credibility and centrality of the nodes to evaluate the trust level. Preliminary simulations show the benefits of the proposed model towards the isolation of almost any malicious node in the network.