ADAPTIVE FILTER EQUALIZER OPTIMIZATION USING HYBRID APPROACH

The LMS (Least Mean square) algorithm was widely utilized in several applications because of its robustness and simplicity. In the practical implementation of the LMS algorithm, a major factor is the phase size. As the step size turns into a small/large, the convergence ratio of the LMS algorithm will be quick and the stable-state MSE (mean square error) will decrease/increase. Consequently, the step size offers a trade-off among the stable-state MSE and the convergence rate of the LMS algorithm. An intuitive method to enhance the efficiency of the LMS algorithm is to do the step size variable instead of static, specifically, during the original convergence of the LMS algorithm select a value of large step size, and utilize the values of minor step size once the system is nearer to its stable state, which leads to consistent VSSLMS (step size Least Mean square) algorithms. Both a small stable-state MSE and a fast convergence rate can be achieved by using a similar method. Though several VSSLMS algorithmic techniques operate properly in specific circumstances, noise could degrade their efficiency and having sensitivity performance throughout the constraint establishing. In this article, a new theory is established to differ the step size which is based on the evolutionary programming (VSSLMSEV) algorithm is portrayed. It has demonstrated that the overall efficiency that is created by this technique is vigorous and does not need any pre-setting of engaged considerations in a solution that is based on the numerical features of the signal. This article establishes an NVSS LMS (new variable step-size LMS) algorithm in adaptive channel equalization. This variation of the VSS LMS algorithm will depend on the weighting variance trade-off/coefficients bias. This article also describes the adaptive equalizer with the NVSS LMS algorithm has had a promising efficiency. Modelling outcomes are offered to assist the recommended execution of the NVSS LMS