Internet of Things最新文献

The Internet of Things (IoT) is increasingly integrated into daily life, particularly in healthcare, through the Internet of Medical Things (IoMT). IoMT devices support services like continuous health monitoring but raise significant cybersecurity concerns due to their vulnerability to various attacks. The complexity and data volume of IoMT network traffic requires advanced methods to enhance security and reliability. Machine Learning (ML) offers techniques to detect, prevent, and mitigate cyberattacks. However, existing benchmark datasets lack essential features for robust IoMT security solutions, such as a reduced number of real devices, a limited variety of attacks, and a lack of extensive profiling. We propose a realistic benchmark dataset for IoMT security solutions development and evaluation to address these gaps. We executed 18 attacks on an IoMT testbed with 40 devices (25 real and 15 simulated), using protocols like Wi-Fi, MQTT, and Bluetooth. Supporting technologies, including dedicated network traffic collectors and a Faraday Cage, ensured data quality. The attacks fall into five categories: DDoS, DoS, Recon, MQTT, and spoofing. We aim to establish a baseline that complements existing datasets, aiding researchers in creating secure healthcare systems using ML. Beyond simulating attacks, we capture the lifecycle of IoMT devices from network entry to exit through profiling, allowing classifiers to identify device anomalies. The resulting CICIoMT2024 dataset, published on the CIC dataset page, demonstrates that various methods can classify IoMT cyberattacks. This effort supports new IoMT security solutions and contributes to the broader field of cybersecurity in healthcare, ensuring more reliable IoMT device deployment.

With the popularity of connected electric vehicles, the openness and sharing of charging data between stakeholders allows a more accurate estimation of charging behavior, which is valuable for optimizing energy systems and facilitating travel convenience. However, to enable such an effective mechanism, the challenge of data security and privacy should be addressed. Federated learning in the vehicular network is appealing for utilizing individual vehicle data while preserving data privacy. We propose an improved local differential privacy-based federated learning approach for modeling charging session prediction problems while preserving user privacy against the threat from a honest-but-curious server. In this approach, all vehicles, within the coordination of a cloud server, collaboratively establish a global regression network through parameter exchange. Meanwhile, the servers may belong to third-party model owners and can be semi-honest when inferring private information on the collected model parameters. Hence, local differential privacy is adopted to perturb the parameters. Additionally, a combination of local and global models via elastic synchronization is proposed to improve the accuracy of the learned noisy global model. Through the test on a real data set, the results show the superiority of the proposed algorithm over traditional noisy federated learning methods. Furthermore, the practical value of the proposed method is validated with a real-world charging case. Such an accurate charging session prediction service for electric vehicle drivers facilitates charging and travel convenience in the green transportation world.

The Massive Access Problem of the Internet of Things stands for the access problem of the wireless devices to the Gateway when the device population in the coverage area is excessive. We develop a hybrid model called Data Fusion Integrated Network Forecasting Scheme Classifier (DFI-NFSC) using a Multi-Layer Perceptron (MLP) Decomposition architecture specifically designed to address the Massive Access Problem. We utilize our custom error metric to display throughput and energy consumption results. These results are obtained by emulating the Joint Forecasting–Scheduling (JFS) system on a single IoT Gateway and distinguishing between ARIMA, LSTM, and MLP forecasters of the JFS system. The outcomes indicate that the DFI-NFCS method plays a notable role in improving performance and mitigating challenges arising from the dynamic fluctuations in the diversity of device types within an IoT gateway’s coverage zone.

Intrusion detection systems (IDS) for IoT devices are critical for protecting against a wide range of possible attacks when dealing with Distributed Denial of Service (DDoS) attacks. These attacks have become a primary concern for IoT networks. Intelligent decision-making techniques are required for DDoS attacks, which pose serious threats. The range of devices connected to the IoT ecosystem is growing, and the data traffic they generate is continually changing; the need for models more resistant to new attack types and existing attacks is of research interest. Motivated by this gap, this paper provides an effective IDS powered by deep learning models for IoT networks based on the recently published CICIoT2023 dataset. In this work, we improved the detection and mitigation of potential security threats in IoT networks. To increase performance, we performed preprocessing operations on the dataset, such as random subset selection, feature elimination, duplication removal, and normalization. A two-level IDS using deep-learning models containing binary and multiclass classifiers has been designed to identify DDoS attacks in IoT networks. The effectiveness of several deep-learning models in real-time and detection performance has been evaluated. We trained fully connected, convolutional, and LSTM-based deep learning models for detecting DDoS attacks and sub-classes. According to the results on a partially balanced sub-dataset, two staged models performed better than baseline models such as DNN (Deep Neural Networks), CNN (Convolutional Neural Networks), LSTM (Long Short Term Memory), RNN (Recurrent Neural Network).

As urbanization accelerates, the demand for efficient parking surveillance solutions has increased. However, existing solutions often face challenges related to energy consumption, scalability, and reliability. This paper introduces a smart hybrid parking surveillance system integrating wireless sensor networks (WSNs) with vision based solution at the edge for resource constrained IoT devices to address these challenges. The solution leverages WSNs for periodic readings of parking space occupancy and introduces a low power sleep mode in the network for energy efficiency, along with optical verification strategies using computer vision models like R-CNN and Faster R-CNN FPN on ResNet50 and MobileNetv2 backbones for distinguishing between true and false positives in the WSN data for a greater accuracy in parking space occupancy. The system utilizes edge for computing on edge servers resulting in increased responsiveness of the system, reduced data transmission and real time processing of data. The proposed solution is formulated in such a way that it automatically switches between WSN and vision based sensing resulting in less energy consumption and longer lifespan of the system without compromising on accuracy. Through experimental results it is observed that models trained on the MobileNetv2 backbone demonstrated at least twice faster for both processing the images and training compared to those models trained on the ResNet backbone. On the other hand, both Faster R-CNN FPN (input resolution: 1440) and R-CNN (input resolution: 128) models trained on the MobileNetv2 backbone have slightly lower accuracies than the same models trained on the ResNet50 backbone.

Data is becoming increasingly pivotal and foundational in the development of smart agriculture, underscoring the importance of efficient methods for obtaining high-value data. Data sensing methods have become the key technologies and methods to realize the agricultural Internet of Things (IoT). However, in the face of the new agricultural paradigm driven by big data, traditional agricultural IoT confronts numerous challenges at the data sensing level. This article, therefore, adopts a data sensing perspective and, based on the agricultural IoT, explores the evolution of data sensing technology in the agricultural domain. Initially, it introduces a data sensing framework for the agricultural Internet of Things, which integrates cloud and edge computing. Subsequently, it reviews the sensors commonly deployed in agricultural scenarios. Then, common methods for collaborative sensing of agricultural data were discussed from three aspects: intra-node, multiple nodes, and cross-domain. At the same time, the issues of data security and privacy in data collaborative sensing were discussed. Next, we integrate multi-dimensional technology to construct an application case for data sensing in the agricultural IoT. Finally, it discusses the challenges that Collaborative sensing technology encounters within the agricultural IoT.

The Internet of Things (IoT) has emerged as one of the growing fields in digital technology over the past decade. A primary goal of IoT is to connect physical objects to the Internet to provide various services. Due to the vast number and diversity of these objects, referred to as devices, IoT must tackle both traditional and novel theoretical and practical network problems. Among these, multi-interface problems are well-known and have been extensively studied.

This research focuses on one of the newest multi-interface models that fits well within the IoT context. It is known as the Coverage in the budget-constrained multi-interface problem, where the budget represents the total amount of energy in the network, and coverage refers to the model’s goal of activating all required communications among IoT devices. Since most IoT devices are battery-powered, energy consumption must be considered to extend the network’s lifespan. This means selecting the most energy-efficient interface configuration that allows all desired connections to function. To achieve this, both global energy consumption and the local number of active interfaces are limited. Moreover, this model also incentivize devices to turn on the available interfaces to create a more performant network. Finally, this model also takes into account the performance of the networks assigning a profit to devices that activate interfaces and realize connections.

This problem can be represented using an undirected graph where each vertex represents a device, and each edge represents a desired connection. Every device is equipped with a set of available interfaces that can be used to facilitate transmission among the devices. The final goal is to activate a subset of the available interfaces that maximize the total profit, while not violating the constraints.

This problem has been recognized as NP-hard, which is why we decided to investigate the decision version from the perspective of fixed-parameter tractability (FPT) theory. FPT is an advanced area of complexity theory that aims to identify the core complexity of a combinatorial problem by incorporating parameters into the time complexity domain.

We provide two fixed-parameter tractability results, each describing an FPT algorithm. One algorithm is based on the well-known pathwidth parameter, the number of available interfaces, and the maximum available energy. The other algorithm considers pathwidth, the number of available interfaces, and an upper bound on the optimal profit. Finally, we show that these two algorithms can be applied to the maximization version of the problem.

Amid rising concerns about campus safety in Taiwan, particularly with the global trend towards smart cities, integrating the Internet of Things (IoT) into institutional security frameworks has become pivotal. The paper discusses the implementation of using iBeacons and Long Range (LoRa) technology to locating the student position and ensure his safety in the school campus. It uses Internet of Things (IoT) approach in real time to monitor and locate the student presence in the school compound. This paper unveils an innovative design for a campus security system that harnesses the LoRa technology. In the system, the students are equipped with devices containing Bluetooth Low Energy (BLE) beacons to capture and transmit real-time location data. The system response time to locating student in abnormal locations such as cornered and concealed areas is about one second. By extending this system to cover all individuals on campus, a closely monitored environment and areas is enabled that significantly bolstering the security measures. This not only furnishes a dynamic protective layer for educational institutions but also serves as a proactive deterrent against potential security breaches. Ultimately, this research underscores the transformative potential of merging IoT with campus security to ushering in a new era of student safety. LoRa technology offers advantages in battery life, cost-effectiveness, deployment flexibility, and network coverage etc. Therefore, this paper ultimately provides a method of how to utilize the LoRa technology to develop a campus security system. Unlike artificial intelligence (AI)-based image recognition, which raises concerns about privacy and human rights; the features of LoRa’s long-range communication and low power consumption make it a more suitable choice.

Wireless Sensor Networks (WSN) play a vital role in the Internet of Things (IoT) and show great potential in monitoring applications. However, due to harsh environmental conditions and unreliable communication links, WSN often encounter partial data loss during data collection, which inevitably affects the quality of service. To address this challenge, researchers have employed matrix completion techniques to recover missing data by exploiting the low-rank features in the data, but its accuracy is not satisfactory. This paper argues that the spatiotemporal characteristics of the data underlie its low-rank nature, enabling a more accurate capture of the intrinsic patterns within the data. Drawing on this insight, we propose a missing data recovery algorithm based on Temporal Smoothness and Time-Varying Similarity (TSTVS). Unlike traditional low-rank methods, the TSTVS algorithm directly utilizes the structural features of data in the spatiotemporal domain to establish a missing data completion model. Subsequently, the model is converted into an unconstrained optimization problem using the penalty function method, and the gradient descent method is applied to solve it, reconstructing the complete data matrix. Finally, simulation experiments were conducted on three real-world monitoring datasets, comparing the TSTVS with three low-rank methods, Efficient Data Collection Approach (EDCA), Matrix factorization with Smoothness constraints (MFS) and Data Recovery Based on Low Rank and Short-Term Stability(DRLRSS). The experimental results indicate that the proposed TSTVS algorithm consistently outperforms the three low-rank based algorithms in terms of recovery accuracy across different datasets and missing rate scenarios.

Reliable storage and high-speed data networks enable individuals to access high-quality Internet of Things (IoT) data for scientific research through global transactions. Blockchain technology provides transparency for institutions to securely store and manage IoT data, while cross-chain transaction mechanisms facilitate the flow of IoT data. However, fairness issues may arise when it comes to cross-chain transactions of IoT data. This paper proposes a mechanism for multi-attribute data transactions to support cross-chain. The solution utilizes Vickrey–Clarke–Groves (VCG) auction, Paillier, Intel SGX, and other technologies to design a secure and equitable data seller selection scheme. The scheme ensures that the selection process for data sellers is both informed and private. Additionally, we generate a key pair for each attribute in the dataset to produce the corresponding attribute data signature. The dataset’s legitimacy is verified through batch verification to ensure that the data seller’s purchased attributes align with their requirements. The exchange of crypto assets and private keys between data sellers and buyers is designed to achieve fair payment. Our research suggests that the scheme meets the necessary security standards, and simulation results confirm its feasibility and effectiveness.