Accurate extraction of multicomponent linear frequency modulation (LFM) signal parameters, such as onset frequency, linear modulation frequency, amplitude, and initial phase, is of great importance in the fields of ISAR, cognitive radio, electronic countermeasures, and star-ground communications. However, the task of accurately extracting the characteristic parameters of a signal is challenging when it has an extraordinarily large bandwidth as well as cross or neighboring components in the time-frequency domain. In this paper, we first review the main current methods used for multicomponent LFM signal decomposition and their challenges, and then propose a novel multi-parameter feature parameter extraction algorithm. The algorithmrealizes the direct and accurate extraction of thefeature parameters of multicomponent LFM signals at ultra-low sub-Nyquist sampling rate for the first time. Moreover, the algorithm is optimized for the computational complexity and anti-noise problems in practical applications, so that it has high accuracy, high efficiency and good noise robustness. We also compare the algorithm with innovative and existing methods, and the results show that the algorithm has excellent performance in feature parameter extraction accuracy, noise immunity and computational speed.