Least Mean Square Algorithm Based on Bessel Function and Multiscale Wavelet Transform

The development of adaptive technology is inseparable from the theory of adaptive filtering and its algorithms. The least mean square (LMS) algorithm occupies an important position as one of the most important algorithms in adaptive filtering and is widely used due to its simplicity, ease of implementation, and good convergence in stationary environments. In this paper, for the LMS algorithm, the LMS algorithm based on Bessel function and the multiscale wavelet transform algorithm are studied respectively. On this basis, a new LMS algorithm based on Bessel function and multiscale wavelet transform is proposed, using MATLAB computer simulation software to conduct experiments, the time domain waveform and convergence effect of the algorithm have been significantly improved, verifying the feasibility of the algorithm. The simulation results show that the newly algorithm has faster convergence speed, higher processing speed, strong denoising ability, and its steady state offset is small.