The drone industry is developing rapidly, and the presence of drones is covering many different fields and areas. The detection, classification, and control of drones are becoming urgent, especially in airports, factories, and important areas in the inner city, etc. In the urban scenario, with a limited line of sight due to obstructions, the method of detecting and identifying drones using radio frequency (RF) signals has many advantages over other methods. In this study, we propose a method for classifying drones using RF signal analysis. The performance of the classifier is evaluated under various scenarios using real-world datasets. Specifically, the model is tested with additive white Gaussian noise (AWGN) and multipath fading conditions with assumed Doppler frequencies. Additionally, scenarios involving the simultaneous presence of multiple drones are thoroughly examined. The evaluation results demonstrate classification accuracy exceeding 98% across all tested conditions.