The SARS-CoV-2 virus is a big threat to the global population in the recent century. However, by utilizing current computer approaches such as artificial intelligence (AI), the Internet of Things (IoT), and distributed cloud computing, the COVID-19 pandemic may be effectively controlled. In order to assess the state of the body’s health, the wearable sensors connect to an edge node in the cloud-based IOT system. Additionally, the technology automatically warns the pertinent medical authorities about any breach of quarantine for exposure to contagious patients by tracking patients’ real-time Global Positioning System (GPS) data. Users will be able to recognize serious medical disorders early and take the appropriate action thanks to AI. The implementation of AI of things has the ability to substantially decrease the rate of death by enabling early detection of high- risk patients, methods to limit the spread, monitoring the spread of the disease, mortality risk prediction by analyzing patient’s health history, remote or in-home therapies to minimize hospital availability, and other techniques to remarkably prevent it from spreading and treating the patients successfully. The aim of this work is to review the different IoT based-wearable devices for the detection of SARS-CoV-2 and Zika viruses.
Hybrid Wireless Network-on-Chip (HWNoC) architecture has been introduced as a promising communication infrastructure for multicore systems. HWNoC-based multicore systems encounter extremely dynamic application workloads that are submitted at run-time. Mapping and scheduling of these applications are critical for system performance, especially for real-time applications. The existing resource allocation approaches either ignore the use of wireless links in task allocation on cores or ignore the timing characteristic of tasks. In this paper, we propose a new deadline-aware and energy-efficient dynamic task mapping and scheduling approach for the HWNoC-based multicore system. By using of core utilization threshold and tasks laxity time, the proposed approach aims to minimize communication energy consumption and satisfy the deadline of the real-time applications tasks. Through cycle-accurate simulation, the performance of the proposed approach has been compared with state-of-the-art approaches in terms of communication energy consumption, deadline violation rate, communication latency, and runtime overhead. The experimental results confirmed that the proposed approach is a very competitive approach among the alternative approaches.
1 January 2019 in Journal of Parallel and Distributed Computing