Post-surgical fall detection by exploiting the 5 G C-Band technology for eHealth paradigm

Abstract

Internet of Thing (IoT) in the wireless sensing utilizing 5G C-Band is the modern technology and taking 5G C-Band wireless sensing to the higher paradigms. The healthcare sector utilizing 5G C-Band technology will increase the efficiency of monitoring the fall and other body motions with maximum capacity and extensive spectrum range. We have proposed the unique system operating at 4.8 GHz frequency for measuring the wireless channel information (WCI) of post-surgical falls and other related human activities. The overall system exploit the low cost simple wireless devices including NIC card, RF signal generator, Omni directional antenna and a desktop PC. The fall and other human activities are classified by using machine-learning algorithms like Support Vector Machine (SVM), Naïve Bayes and Decision Tree. The three main kernel functions RBF, Linear and polynomial were used for the classification along with the 10 time domain features. The performance of the classifiers increases by using the maximum number of features. We have concluded that the SVM proves to be the best classifier for our wireless data by measuring the performance metrics like accuracy (above 90 %), F-measure, Precession, Specificity, Recall, and Kappa coefficient. The error rate for SVM proved to be the less as compared with the other two classifiers. The accuracy increases with the rate of 4 to 5 %. Our approach shows more feasibility for detecting the post-surgical fall in hospitals and more reliable with maximum accuracy.