A Formal Energy Consumption Analysis to Secure Cluster-Based WSN: A Case Study of Multi-Hop Clustering Algorithm Based on Spectral Classification Using Lightweight Blockchain.

Wireless Sensors Networks are integrating human daily life at a fast rate. Applications cover a wide range of fields, including home security, agriculture, climate change, fire prevention, and so on and so forth. If WSN were initially flat networks, hierarchical, or cluster-based networks have been introduced in order to achieve a better performance in terms of energy efficiency, topology management, delay minimization, load balancing, routing techniques, etc. As cluster-based algorithms proved to be efficient in terms of energy balancing, security has been of less importance in the field. Data shared by nodes in a WSN can be very sensitive depending on the field of application. Therefore, it is important to ensure security at various levels of WSN. This paper proposes a formal modeling of the energy consumed to secure communications in a cluster-based WSN in general. The concept is implemented using the Proof of Authentication (POAh) paradigm of blockchain and applied to a Multi-hop Clustering Algorithm based on spectral classification. The studied metrics are residual energy in the network, the number of alive nodes, first and last dead node.

Automated Ischemic Stroke Classification from MRI Scans: Using a Vision Transformer Approach.

Background: This study evaluates the performance of a vision transformer (ViT) model, ViT-b16, in classifying ischemic stroke cases from Moroccan MRI scans and compares it to the Visual Geometry Group 16 (VGG-16) model used in a prior study. Methods: A dataset of 342 MRI scans, categorized into 'Normal' and 'Stroke' classes, underwent preprocessing using TensorFlow's tf.data API. Results: The ViT-b16 model was trained and evaluated, yielding an impressive accuracy of 97.59%, surpassing the VGG-16 model's 90% accuracy. Conclusions: This research highlights the ViT-b16 model's superior classification capabilities for ischemic stroke diagnosis, contributing to the field of medical image analysis. By showcasing the efficacy of advanced deep learning architectures, particularly in the context of Moroccan MRI scans, this study underscores the potential for real-world clinical applications. Ultimately, our findings emphasize the importance of further exploration into AI-based diagnostic tools for improving healthcare outcomes.

Nonlinear robust sliding mode - Backstepping hybrid control for WECS -theoretical design and experimental evaluation.

This paper proposes a new contribution in the field of optimizing control techniques for wind systems to enhance the quality of the energy produced in the grid. Although the Sliding Mode control technique, whether classical or involving the use of artificial intelligence, has shown interesting results, its main drawback lies in the oscillation phenomenon commonly referred to as "chattering." This phenomenon affects the accuracy and robustness of the system, as well as the parametric variation of the system. In this work, we propose a solution that combines two nonlinear techniques based on the Lyapunov theorem to eliminate the chattering phenomenon. It is a hybrid approach between the Backstepping strategy and the Sliding Mode, aiming to control the active and reactive powers of the doubly fed induction generator (DFIG) connected to the electrical grid by two converters (grid side and machine side). This hybrid technique aims to improve the performance of the wind system in terms of precision errors, stability, as well as active and reactive power. The proposed solution has been validated in the Matlab & Simulink environment to assess the performance and robustness of the proposed model, as well as experimentally validated on a test bench using the DSPACE 1104 card. The obtained results are then compared with other techniques, demonstrating a significant improvement in performance.