Smart house, is a set of devices and smart applications designed to provide a comfortable and safe environment for its residents. Home devices have the ability to communicate with each other. Through a network of networks. This network of devices is called internet of things (IoT). Our contribution is to propose an architecture based on an autonomic and intelligent object using multi-Agent systems. We have created an environment that simulates a smart home with a range of devices and sensors. The agent manages the built-in device. Through user orders and information captured by sensors. The home is controlled remotely via the Internet. Through the smartphone.
30 August 2017 in International Journal of Software Engineering and Its Applications
Internet of things is a network of objects mainly supported by electronic devices and electronic components such as sensors and electronic cards. These objects can be physical and virtual devices, sensors or actuators, are autonomous and have their own intelligence characteristics. On the other hand, smart environments are those in which sensors and actuators have been integrated, to react to events and to adapt to those present. The environment acquires intelligence through its intelligent components, or through the intelligence resulting from its interaction with other components. Our contribution is a proposal of Cognitive IoT (CIoT) devices structure by adding an agent layer to the device. Such layer provides the device with agent characteristics (intelligence, autonomy, cooperation and organization).
Electroencephalography (EEG) signals present unique challenges for classification tasks due to their non-stationary and high-dimensional nature. In this paper, we propose a novel method that combines Riemannian geometry with deep learning to classify multi-class EEG data. Specifically, we compute covariance matrices of EEG signals and map them onto the tangent space of the Symmetric Positive Definite (SPD) manifold. A deep neural network architecture, termed NeuroSPDNet, is designed to effectively learn from these tangent space features. The method incorporates cross-validation for robust performance evaluation and utilizes Receiver Operating Characteristic (ROC) and Precision-Recall (PR) curves to assess classification effectiveness across four classes. Experimental results demonstrate that the proposed approach achieves an overall accuracy of 96.23% with high precision and recall, significantly outperforming traditional approaches in EEG signal classification.