Human activity detection in indoor environments is an attractive research field that can assist the elderly and disabled live independently. To detect human activity, various technologies have been proposed, including the use of sensors, cameras, wearables, and contactless radio frequency (RF). RF sensing has the potential to become a universal sensing mechanism due to the omnipresent of electromagnetic singals. This study reports the findings of an experiment to locate activity in an indoor environment utilising USRP. A single subject is observed while sitting, standing, and walking in two directions to collect samples of CSI. The proposed method outperforms current benchmark techniques while combining machine learning and deep learning techniques for improved accuracy and makes use of a K- Nearest Neighbor (KNN) that can identify the location of various activities with a 98.43% accuracy .