Velocity and Distance Estimation of Multiple Objects Using YOLO-Based OFDM Radar System

In this study, we propose a method for simultaneously estimating the speed and distance of multiple objects using a YOLO model within an OFDM radar system. The proposed technique inputs a 2D spectrogram image generated from OFDM radar signal processing into the YOLO model to predict the speed and distance of objects. Simulation results demonstrate that the proposed method performs well in recognizing multiple objects. Notably, the speed and distance estimation accuracy improves as the number of OFDM symbols increases and the number of objects decreases. The proposed technology is applicable not only to 5G but also to 6G OFDM signals.