Procedia Computer Science最新文献

In this work, we classified the wireless internet of things (IoT) traffic of the IoT Health intensive care unit (IHI) dataset which belongs to three general classes: patient monitoring, environment monitoring, and network attack. We trained and tested 7 machine learning (ML) models with Orange 3 including kNN, Decision Tree (tree), SVM, Random Forest (RF), Neural Network (NN), Gradient Boosting (GB), and AdaBoost (AB). With the original dataset, 5 ML models performed perfect classification. After pruning the dataset columns by keeping the features with the highest correlations with the label in the dataset, good classifications were obtained with only 4 TCP/IP features by the Gradient Boosting, kNN, and RF models with MSEs in the range 0.008-0.011, and R²s in the range 0.978-0.984. With only 6 MQTT features, Gradient Boosting, RF, Tree, and NN were the top classifiers with MSEs in the range 0.073-0.074, and R²s in the range 0.856-0.859. This work demonstrates the effectiveness of guiding the feature pruning process by the values of the correlation coefficients in order to minimize the long training times of ML models while achieving good accuracies.

The magnetic coupler is a key component of the wireless power transfer (WPT) system, which greatly affects the performance of the WPT. This paper proposes a novel magnetic coupler with ferrite PQI cores for the WPT of small drones, which can enhance the magnetic coupling and improve power transfer efficiency. This magnetic coupler includes the transmitting (T_X) side and receiving (R_X) side. The T_X side is made of a helical coil and a ferrite PQ core; and the R_X side is made of a helical coil and a ferrite I core plate. Finite element simulations are used to investigate the performance of the proposed magnetic coupler with PQI cores, and compare it with a conventional planar magnetic coupler with two I core plates. In addition, an experimental platform is built to prove the validity of the proposed magnetic coupler with PQI cores. The results show that the coupling coefficient can reach 0.97, and it can exceed 0.446 even under the worst coupling conditions. Meanwhile, power transfer efficiency increases by 10.3% using the magnetic coupler with PQI cores.

This study discusses the surge in energy consumption in the Caribbean region over the past decade, notably in Trinidad, where per capita consumption exceeds 6500 kWh. In response to rising electricity tariffs, energy sector entities are implementing conservation initiatives. The study focuses on a Direct Reduced Iron (DRI) plant in Trinidad's Point Lisas Industrial Estate, specifically examining alternatives to conventional air conditioning in a DRI processing laboratory. Various insulating materials were simulated using CHVAC software and evaluated against a 5-ton air-conditioning unit using the CLTD method. Analysis reveals standalone use of insulating systems in the laboratory is impractical due to orientation, location, and internal heat loads, necessitating a hybrid cooling system. Economically viable configurations involve a 3-ton A/C unit paired with PVC, foam, or fibreglass walls and ceilings. Despite higher initial costs, configurations with a 3-ton unit offer savings in maintenance and electricity. Findings extend beyond the laboratory, potentially influencing passive cooling material adoption and active cooling load reduction in other contexts, thus promoting sustainable energy practices.

Fog computing has emerged as an essential technology for enabling real-time, low-latency responses in cloud-based applications within Internet of Things (IoT) systems. This study explored the impact of the number of fog devices (NFDs) on the performance of IoT-fog-cloud systems. Through comparative analysis, two scheduling algorithms—First Come First Serve (FCFS) and Priority—were evaluated across different scales of system deployment. The results indicated that the FCFS algorithm was optimal for systems with fewer NFDs, whereas the Priority algorithm proved advantageous in larger settings. These findings not only establish guidelines for selecting appropriate scheduling strategies based on the scale of fog device deployment but also provide strategic insights for businesses on the optimal distribution of computational resources. This research aids companies in deciding whether to centralize resources in fewer branches with more employees or to decentralize into more branches with fewer employees, thereby optimizing operational efficiency and responsiveness.

There currently exists a need for a platform that provides users with the ability to perform extensive, repeatable and meaningful simulation and testing for the hardware and software which compose vehicle/autonomous vehicle systems whilst being broadly accessible, widely supported and provides robust features and development tools. The contemporary implementations of similar systems are either financially exorbitant or highly contained. The system reflected in this paper aims to fill a gap in the industry of vehicle/autonomous vehicle development by extending on currently existing open-source software to provide a highly streamlined platform to support the production of general road vehicle and autonomous vehicle driving systems. The software tools and hardware components chosen for the system will be discussed, followed by the features constructed throughout the development process. The end result of the system is a platform that allows for quick, repeatable, accurate, and nearly endless testing of a digital twin of real life vehicles. This system will allow users to gain valuable simulation and testing data of hardware and software components in a manner which is not always feasible using the traditional methods of autonomous vehicle testing.

This paper addresses the multi- UAV patrolling problem, a NP-hard optimization problem that is focused on minimizing idleness, which is defined as the time between consecutive visits to specific locations. We propose the Discrete Grey Wolf Optimizer (D-GWO), which is specifically developed to handle the discrete aspects of UAV patrolling routes. This new algorithm is enhanced with a 2-opt local search strategy, which integrates the global search capabilities of D-GWO with the precision of local optimization to effectively refine solutions. Comparative experimental results show that our algorithm outperforms established methods such as ant colony optimization and simulated annealing in terms of reducing global worst idleness and overall exploration time. Our findings suggest that the D-GWO algorithm is particularly effective for complex multi-UAV patrolling tasks, significantly enhancing efficiency in security and disaster response missions.