In this study, we propose a method for simultaneously estimating the number, velocity, and distance of multiple targets in an Orthogonal Frequency Division Multiplexing (OFDM) radar environment using YOLO (You Only Look Once). The proposed approach employs Doppler-range two-dimensional (2D) signals as the input to the YOLO model, enabling it to learn and predict target characteristics. Since YOLO performs object detection in a single forward pass, it achieves higher computational efficiency compared to conventional CNN-based methods, making it suitable for multi-target estimation tasks. To validate the performance of the proposed method, we conducted simulations under various signal-to-noise ratio (SNR) conditions ranging from –10 dB to 20 dB and considered scenarios with one to five targets. The results show that, with 32 OFDM symbols, the YOLO-based model achieved an average velocity estimation error of 1.34 km/h and an average distance estimation error of 0.71 m. These results represent improvements of 0.56 km/h and 0.89 m, respectively, over conventional CNN-based single-target estimation models, demonstrating the precision of the proposed method. Such performance indicates its potential for effective application in next-generation joint communication and sensing systems.
Jeon et al. (Tue,) studied this question.