<p>Fourth-order cumulants (FOC) are widely adopted in direction of arrival (DOA) estimation due to their capability of mitigating Gaussian noise and constructing virtual arrays. However, the prerequisite of knowing the source number becomes a significant bottleneck when a nested array (NA) is utilized, limiting its practical applications. To address this issue, we propose an improved propagator-based method that eliminates the need for prior knowledge of the source number. By introducing a small-value factor, the proposed method generates an estimator equivalent to the classical multiple signal classification (MUSIC) estimator while retaining the benefits of NA, such as high degrees of freedom and low mutual coupling. Numerical simulations demonstrate the superior performance of the proposed estimator, particularly in scenarios with a large number of signals relative to the number of sensors. Furthermore, experimental results validate the effectiveness of the proposed method in practical situations, showcasing its robustness and accuracy. This approach not only overcomes the limitation of source number estimation but also provides a reliable solution for underdetermined DOA estimation, making it highly applicable in fields such as wireless communications, radar systems, and medical electronics.</p>