Range Estimation of Radar Targets Using a Temporal Convolutional Network
David Chuong, Zachary Poole, Webert Montlouis · 2022
This paper investigates the radar range estimation problem. Using classical signal processing techniques, current radar systems are generally able to accurately detect targets and estimate their ranges. However, radar systems that need to directly sample signals at RF are too computationally intensive to obtain an accurate range estimate in the required time window. In addition, as the amount of data increases the number of computations will also increase. This paper will demonstrate that by using a trained neural network to perform sequence-to-sequence classification on a radar signal, it is possible to detect targets that are within the radar system’s field of vision and estimate their ranges with an overall accuracy of 99% in a fraction of time as compared to conventional radar signal processing.