An Experiment of Human Presence and Movement using Impulse Radar and Machine Learning
Young‐Jin Park, Hui‐Sup Cho · 2019
Research and product development using machine-learning technology is ongoing worldwide. It is especially active in the field of predicting a person's activity. In this paper, a person's presence and movement are determined through machine learning by using radar signals. The preprocessing stage of receiving and processing radar signals produces a basic dataset. Next, we create a machine-learning model and conduct a prediction experiment with four test-sets that is created for this paper. Experiments conducted in this study showed a high level of classification accuracy. However, data sets, pre-treated only by classification of signals, seemed to be quite sensitive to the quality of the data when used as a feature of machine learning.