Jitter is one of the important reasons to limit the upper limit of circuit system performance, and it is necessary to decompose the various components of jitter. The traditional time delay correlation jitter decomposition algorithm is insufficient in the accuracy and stability of the decomposition results and the timeliness of the operation. In this paper, we propose a fitting method combined with least squares method, which can effectively extract the jitter component information from the TLC function affected by the stochastic jitter component, thus avoiding the influence of stochastic data on the system of equations and enhancing the stability of the algorithm. At the same time, the FFT algorithm is introduced, so that the TLC algorithm can decompose the frequency points of the jitter component more accurately and efficiently, which reduces the running time of the algorithm. To verify the effectiveness of the improved algorithm, this paper compares the proposed improved algorithm with the existing improved algorithm. The experimental results show that the jitter parameters decomposed by the improved TLC algorithm are more accurate, the stability is improved by 41%~54%, and the timeliness is improved by 52%~68%.