National Telecommunications Institute

This study proposes a model of a rotary machine's fault diagnosis system based on vibration signal analysis under Improved eXtreme Gradient Boosting-Recursive Feature Elimination (XGBoost-RFE). A 3D dataset of vibration signals is collected from healthy and faulty induction motors. The Empirical Mode Decomposition (EMD) technique is used to perform signal conditioning. The de-noised signals are obtained to extract the multi-domain features. Finally, multiple classifiers, including K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost), are performed with different kernel settings at the classification step. The results indicate that a hybrid approach that combines time and frequency domain features and classifies them using XGBoost with a Gaussian kernel achieves the highest accuracy (99%) with the lowest error rate <1.3%.

Intelligent Reflecting Surfaces or IRSs are energy-efficient technologies used as “passive” relays to increase coverage. Often fixed, they enable connectivity to users in positions that are difficult for the base station to access, or that are blind. Most studies propose fixed IRS positioning, with the disadvantage of covering only a fixed zone. Therefore, in the vehicular environment, it would be interesting to see the feasibility of placing IRSs following traffic. We, therefore, propose to study the mobile placement of IRSs in a vehicular context first by using an optimizer and then by relying on heuristics. In the first part, we compare fixed and mobile IRS positioning. Then, for the heuristic part, we present different IRS election strategies, which we have compared. Performances are compared between fixed and mobile placement in the first part, and between one and two hops in the second part, all while analyzing the impact of different parameters on these results. We also evaluated the impact of a trajectory predictor and the dataset on these results. • When there are few available IRSs or a small coverage area, mobile placement is more effective than fixed placement. • The selection of IRSs in relation to the population to be covered is crucial. • In addition, incorporating prediction improves these performances.

In the contemporary environment, mobile ad hoc networks (MANETs) are becoming necessary. They are absolutely vital in a variety of situations where setting up a network quickly is required; however, this is infeasible due to low resources. Ad hoc networks have many applications: education, on the front lines of battle, rescue missions, etc. These networks are distinguished by high mobility and constrained compute, storage, and energy capabilities. As a result of a lack of infrastructure, they do not use communication tools related to infrastructure. Instead, these networks rely on one another for routing and communication. Each node in a MANET searches for another node within its communication range and uses it as a hop to relay the message through a subsequent node, and so on. Traditional networks have routers, servers, firewalls, and specialized hardware. In contrast, each node in ad hoc networks has multiple functions. Nodes, for instance, manage the routing operation. Consequently, they are more vulnerable to attacks than traditional networks. This study's main goal is to develop an approach for detecting blackhole attacks using anomaly detection based on Support Vector Machine (SVM). This detection system looks at node activity to scan network traffic for irregularities. In blackhole scenarios, attacking nodes have distinct behavioral characteristics that distinguish them from other nodes. These traits can be efficiently detected by the proposed SVM-based detection system. To evaluate the effectiveness of this approach, traffic under blackhole attack is created using the OMNET++ simulator. Based on the categorization of the traffic into malicious and non-malicious, the malicious node is then identified. The results of the suggested approach show great accuracy in detecting blackhole attacks.

Gastrointestinal cancer is a leading cause of cancer-related deaths globally, prompting significant research into using artificial intelligence (AI) for detection. Researchers have been exploring AI applications in this field since the 1960s, leveraging its ability to handle repetitive tasks and complex computations. In Phase I of this study, various AI models, including basic CNN and more advanced ones like vgg16, Alex, and DenseNet121, were employed to diagnose gastrointestinal cancer using datasets comprising images of benign tumors and malignancies from patients. However, accuracy rates with conventional techniques were found to be insufficient. Thus, Phase II focused on refining the DenseNet121 model, leading to improved accuracy, sensitivity, and specificity. The modified model demonstrated enhanced diagnostic performance, albeit with slightly longer processing times, compared to existing approaches.

Background: Telehealth refers to the delivery of medical care or services from a distance rather than in person. According to Global Market Insights, the global telehealth market was estimated to be worth $56 billion in 2020. Aim: The purpose of this research was to analyze the perceptions of nursing academic staff and care providers about the benefits and barriers to the use of telehealth. Design: Concurrent mixed-method (quantitative-qualitative) design was implemented using an online cross-sectional survey. Sample: A prospective sample (788 nursing academic faculty and 794 nursing care providers) was recruited. Tools: (1): Questionnaires to assess sociodemographic and perception of telehealth. (2): The benefits and barriers to telehealth utilization were qualitatively assessed. Results: The mean scores of participants' perceptions of telehealth were positive (nursing academic faculty group: 54.25±13.78; nursing care providers group: 50.43±15.92). In terms of the usefulness of telehealth in providing health care to rural locations, the views of both groups were 98.4% and 86.2%, respectively. However, the main barrier was the lack of telehealth instruction in academic nursing curricula (75.4%). Telehealth was positively perceived by academic nursing faculty and care providers. Conclusion: Their perceptions on the benefits of telehealth focused on the development of new job descriptions and classifications, the effectiveness of disaster relief, access to healthcare for people living in remote places, and client support through reduced anxiety. On the contrary, the scarcity of personnel from the specialized health team, educational barriers, and technological issues were the main barriers. Recommendations: Incorporating telehealth into the academic program for nursing students and providing relevant techniques for use in capstone projects.

International audience