Abstract Video surveillance has benefited greatly from advances in artificial intelligence, particularly in computer vision, and the number of monitored environments has consequently increased, creating new challenges in extracting relevant information. When multiple cameras cover adjacent areas, overlapping fields of view can cause the same subject to be detected across cameras, introducing duplicate counts. Although the literature offers established solutions, many rely on high-quality recordings and complex, computationally intensive methods, limiting their use on resource-constrained devices. In this work, we address these challenges with a solution that leverages minimal information about target subjects and uses a transductive strategy to detect duplicates based on interactions within overlapping fields of view. Experiments in real-world settings show that our approach suppresses duplicates effectively, making it suitable for deployment on resource-limited hardware. We evaluate state-of-the-art lightweight models with high inference speed, as well as classical re-identification methods, in scenarios with low-quality video and constrained devices. The results underscore the effectiveness of the proposed approach and motivate exploration of re-ID with domain adaptation, as well as anchor-free methods with weak or semi-supervised learning.

Due to limited hardware, consumer-grade surveillance cameras usually rely on cloud-based computer vision models to detect people in video footage. However, this approach introduces a recurring cost, as users must continuously pay for cloud processing. One possible solution is to use mobile devices for person detection, as they can be found in most households. Yet, experimental evaluations on the impact of different computer vision models on mobile device resource usage are limited. This study examines the efficiency of YOLOv8 models on Android devices, assessing detection performance, inference time, memory consumption, and energy efficiency. The models were tested using two machine learning frameworks, LiteRT (formerly TensorFlow Lite) and ONNX Runtime, to determine the most suitable approach for mobile inference. Experimental results confirm that compact models, such as YOLOv8n and YOLOv8s, offer the best trade-off between computational efficiency and detection accuracy, while LiteRT outperforms ONNX in all evaluated metrics.

Smart homes are transforming domestic environments by integrating connected devices and sensors, enabling lighting, temperature, and security automation. While these systems enhance comfort and efficiency, they often rely on predefined settings or manual input due to the absence of adaptive recommendation systems. AI-driven recommendation systems personalize actions by learning from user behavior and environmental data, improving the smart home experience. However, they also introduce cybersecurity risks, particularly data poisoning attacks, where manipulated data disrupts system functionality. This paper exposes and examines vulnerabilities in smart home recommendation systems, categorizing data poisoning attacks and analyzing their impact. Through a literature review and attack vector analysis, we identify key weaknesses and propose mitigation strategies to enhance security. Our goal is to contribute to developing robust smart home technologies that protect user privacy, ensure reliability, and withstand adversarial threats.

ABSTRACT Context Smart home devices have become increasingly popular in modern households, powered by the Internet of Things (IoT) advances. The data generated by smart devices can provide valuable insights into users' behavior and preferences. By analyzing the data, one can understand how people interact with their homes, thus creating a “smart home profile”. To comprehend the complete IoT ecosystem dynamics of an intelligent environment, it is necessary to learn from each IoT device to predict its status in the future time. Nevertheless, dealing with real‐world IoT data structure requires considerable preprocessing tasks and the employment of classifiers that can learn multiple IoT inputs from a single IoT message. Objective Aware of these challenges, this paper proposes a novel methodology to process multi‐label IoT data and provide a comprehensive comparison of multi‐label classifiers for forecasting the status of smart devices, considering their efficiency and accuracy. Method We propose a data transformation method to preprocess the IoT data to be used by multi‐label classifiers. This method is based on real data structure. Results We evaluate our proposal in two real‐world scenarios and various multi‐label classifiers. The promising findings indicate that efficient classifiers can generate many correct predictions for a comprehensive IoT ecosystem in a small fraction of a second. Conclusions Our proposed data transformation can fit the context of prediction to smart homes and work with multi‐label classifiers to understand user behavior.

Abstract The Internet of Things (IoT) has been playing an important role in the technology scenario due to its high potential and impact on different society segments. Estimates suggest a trend for an increase in the number of IoT devices connected to the Internet for the next few years. Hence, the volume of data produced by IoT devices will follow this growth perspective, and there will be a demand for systems that can process, store, and promote access to large amounts of data. The data collected from sensors in typical IoT systems is stored and processed in cloud servers. However, some IoT solutions use edge devices to perform specific actions, such as processing, storage, and access, using only local infrastructure for low latency requirements. Fog computing has been used to improve IoT solutions and to transfer some of the complexity from the cloud to the edge of the network, that is, closer to devices, applications and users, working as a kind of “local and private cloud.” The cooperation of devices and applications between edge and cloud creates a need for an interplay to enable data flow among the layers of IoT systems deployed on edge and in the cloud. Thus, IoT data systems should support the data life cycle through its collection, analysis, and use. Performance efficiency is a quality factor of systems and software engineering, which measures “performance relative to the number of resources used under stated conditions.” In particular, in IoT systems that involve a large volume of data, the performance efficiency of data interplay is a relevant requirement. This work proposes a data interplay model of IoT to define and deploy the IoT data life cycle in the collection, analytics, and data use stages. This data interplay proposal aims to improve performance efficiency in IoT data life cycle operations: collection, analytics, and use among devices and applications in edge and cloud infrastructures.

Web of Things (WoT) uses current Web protocols and languages as an integration platform of physical devices on the Internet. Previsions estimate that by 2020, 20 to 50 billion such devices will be connected - producing, collecting, and transmitting data over the Internet. The amount of data associated with such a gigantic number of available devices represents a real challenge in data visualization. This paper proposes a Data Visualization model for WoT, based on Fog Computing concepts, more specifically the Fog of Things paradigm. The aim of our proposal is to enable the creation of different data visualizations that take into account the hierarchical levels in an architecture involving Fog and Cloud. In order to demonstrate our proposal, a prototype of the Web of Things application was developed, which is used to perform usability evaluation through the USE questionnaire technique.

The progress towards proving full deployable Internet of Things solutions for cross-domain data exchange is moving slowly and in todays IoT's challenges the need for integrating data across different software platforms and its use over heterogeneous technology remains open. The use of Cloud computing for data sharing as part of the IoT solution(s) design is provided on the basis that sharing data is enabled by the cloud. The emerging of Fog computing is generating that new scenarios are being re-defined, and in particular cases the need for combining both technologies enabling interoperability is explored/studied. This paper presents an approach of the Data Interplay for the Fog of Things. The need for a more structured way to exchange data seamlessly between edge applications and the cloud is required. The data interplay between Fog and Cloud infrastructures addresses the need for big volume generation and storage of data and the way on how to transfer the data from the edge to the cloud and viceversa.

The fog computing paradigm extends cloud resources near data sources to overcome limitations of cloud-based IoT centralized architectures. Its involve the running of services and applications on the nodes between the IoT devices and the cloud. The prototyping and testing of these distributed software is challenging due to the fact that not only the fog service has to be tested but also its integration with management systems. Despite recent advances in fog platforms, there exists no readily available testbed which can help researchers to design and test real world fog applications. To this purpose, network simulators and cloud middleware are adapted to enable the investigation of fog solutions. This paper presents Fogbed, a framework and toolset integration for rapid prototyping of fog components in virtualized environments. Using a desktop approach, Fogbed enables the deployment of fog nodes as software containers under different network configurations. Its design meets the requirements of low cost, flexible setup and compatibility with real world technologies. Unlike current approaches, the proposed framework allows for the testing of fog components with third-party systems through standard interfaces. A scheme to observe the behavior of the environment is provided. A case study is presented to demonstrate a fog service analysis. In addition, future developments and research directions are discussed.

The fog computing model integrates cloud services into the network on a widely distributed level to overcome challengers of cloud-based IoT platforms. The deployment of IoT solutions employing a fog architecture requires a decentralized and scalable computing infrastructure, which places networking, compute and storage resources in a hierarchy of levels arranged between the data source and the cloud. Despite recent advances in fog platforms, there exists no readily available testbed which can help researchers to design and evaluate fog applications on a truly IoT scale. The current fog prototyping tools are adapted from cloud middleware or network simulators to enable the evaluation of fog solutions in limited environments. This paper presents a framework and toolset integration that uses the Fogbed emulator to enable fog distributed testbeds in virtualized environments. Unlike current approaches, the proposed framework allows for the deployment and testing of fog components in a scalable way. Its design is compatible with real world technologies and meets the requirements of low cost, flexible setup and supports third-party systems through standard interfaces. A case study is presented to demonstrate the emulation of fog distributed components and services using a cluster approach. In addition, future developments and research directions are discussed.

The Internet of Things (IoT) is a paradigm in which physical objects embedded with sensors and actuators are connected by networks to the Internet. IoT allows the creation of solutions for various areas of human life. However, predictions show that the number of these objects connected to the Internet tend to increase more and more, thus necessitating new techniques of management, control and maintenance. These techniques should allow IoT platforms to be highly available and coverage. Thus, this work presents M2-FoT, a proposal for monitoring and management of Fog of Things (FoT) platforms. Unlike other works, M2-FoT also presents a method for retrieving device configuration. M2-FoT was implanted on a Fog platform for IoT: the SOFT-IoT. Finally, some experiments were performed to verify the performance of M2-FoT, in order to evaluate its impact on the SOFT-IoT platform.

The Internet of Things (IoT) allows for the communication of a large number of physical or virtual objects through different technologies, protocols and patterns. Several organizations have tried to predict the number of IoT devices that will be connected to the Internet by 2020. Although they have not yet to agree on an exact `magical' number of such devices (in billions), we can expect this huge number of devices to be active which may pose several challenges regarding connectivity and information exchange over network infrastructures. And despite the fact that Cloud and Fog computing addresses some of these challenges, it can also create other problems due to the high demand for storage as well as high network traffic at the edge level. In this paper, we model the IoT data traffic in order to execute several experiments related to the demand for storage and network traffic. As a consequence, the results of our experiments, which have been validated in a modeling based on the Fog of Things paradigm, can be used to evaluate the impact that variation in IoT data traffic patterns has on IoT/Fog environments.

The Internet of Things (IoT) involves the connectivity of a number of different physical and virtual devices, allowing the development of new services and applications. These intelligent objects, along with their tasks, constitute domain-specific applications (vertical markets), while ubiquitous and analytical services form domain-independent services (horizontal markets). The development of these applications and services in these markets brings challenges such as deployment, scalability, integration, interoperability, mobility and performance. Recent researches indicates that Microservices have been adopted in several domains such as IoT. This paper provides an overview of the current state of the art (studies selected by May 2018) regarding the adoption of Microservices in the development of IoT applications by means of the mapping study methodology. In this context, we present the eighteen selected studies, the existing contributions and the future research perspectives.

This work aimed to propose LDoW-PaN, a Linked Data presentation and navigation model focused on the average user. The LDoW-PaN model is an extension of the Dexter Hypertext Reference Model. Through the LDoW-PaN model, ordinary people—who have no experience with technologies that involve the Linked Data environment—can interact with the Web of Data (RDF) more closely related to how they interact with the Web of Documents (HTML). To evaluate the proposal, some tools were developed, including the following: (i) a Web Service, which implements the lower-level layers of the LDoW-PaN model; (ii) a client-side script library, which implements the presentation and navigation layer; and (iii) a browser extension, which uses these tools to provide Linked Data presentation and navigation to users browsing the Web. The browser extension was developed using user interface approaches that are well known, well accepted, and evaluated by the Web research community, such as faceted navigation and presentation through tooltips. Therefore, the prototype evaluation included: usability evaluation through two classical techniques; computational complexity measures; and an analysis of the performance of the operations provided by the proposed model.