An improved Multi-harmonic Sine Fitting Algorithm based on Tabu Search

• The proposed algorithm introduces the Tabu Search.
• It can converge to the global minimum with initial frequencies obtained by IpDFT.
• The iteration process is improved by adjusting the first iteration step-size.

In the Multi-harmonic Fitting Algorithm based on four parameters sine fitting, the problem is that the global convergence highly depends on the initial estimation frequency. To overcome this problem, an improved Multi-harmonic Sine Fitting Algorithm based on Tabu Search (TSMSFA) is introduced. Combining Tabu Search theory, TSMSFA algorithm can converge to the global minimum with the initial frequencies obtained by Interpolation Discrete Fourier Transform (IpDFT) while multiharmonic signals are disturbed by additive noise, phase noise and jitter. The iteration process of TSMSFA is improved by adjusting the first iteration step-size. Compared with the execution speed of the Classical Multi-harmonic Sine Fitting Algorithm (CMSFA), TSMSFA has some improvement.