The Coronavirus or COVID-19 is a highly infectious disease which is caused by SARS-CoV-2 virus. Therefore, it is necessary to identify Covid symptoms at an early stage and get the adequate treatment. Wireless Body Area Networks (WBAN) and Internet of Things (IoT) combined with machine learning can be used to detect Covid from biomedical sensors such as body temperature, blood pressure, and oxygen saturation. In this paper, an energy efficient IoT-based WBAN to control corona virus outbreak is proposed. The IoT framework uses non-invasive sensors interfaced with Arduino Uno for collection of real-time data which is sent to the receiver comprising of Espressif Systems32 (ESP32) using Long-Range Radio (LoRa) Technology gateway. LoRa helps connect large number of nodes and offers low power, adaptive data rate algorithm which maximizes nodes battery life and long-range transmission which makes this application energy efficient as compared to other WBANs’ that carry more data and thus, consume more energy. The collected data is then stored and visualized using the ThingSpeak cloud platform. The obtained dataset is labeled using rule-based decision making and given as input to three supervised machine learning algorithms namely Classic Logistic Regression, Support Vector Machine, and Naïve Bayes to classify whether a person has COVID-19 symptoms or not. The model performance parameters such as accuracy, precision, recall, and F1 score are calculated and compared.