In IoT (the Internet of Things) platforms, sensors generate a huge amount of data for a particular application. This application might be remote surveillance and monitoring of different diseases, such as to observe a change in the health of a person or to detect the change in the state of a sleep-disordered subject’s breathing during the night for paramedics to take the appropriate action. Some other challenging applications might be the prediction of faults to avoid any disturbance/discontinuity of various services (broadband, landline and television) using real-time user data sent back by Wi-Fi routers for better customer experience. The nature of data in such applications can be speech, images, videos, multidimensional time series representing different measures, and other types of sensor data such as ECG/EEG/EMG signals.
The objective of this Special Issue is to showcase computationally efficient intelligent systems for IoT-based applications using statistical models and machine learning. The huge amount of data collected by sensors is not possible to process, as it makes the IoT platform painfully slow. Unnecessary data should be analyzed and rectified at the point of collection or before transmission using Edge/Fog computing with efficient intelligent systems. Furthermore, to avoid breach of sensitive information, the sensor data should not be stored/transmit in the original form. This Special Issue includes but is not limited to the following topics:
Real-time analysis to eliminate unwanted data in IoT-based applications;
Feature engineering for identification of patterns in data;
Computationally efficient Intelligent systems for Edge/Fog computing;
New approaches for privacy and security of sensitive information;
Innovative methods to examine integrity of data for reliable decision making.
Dr. Zulfiqar Ali
Prof. Dr. Sally McClean
Dr. Muhammad Imran