Design and implementation of low complexity LMS adaptive filter

An adaptive filter is a real-time computational device that iteratively simulates the relationship between a filter’s input and output signals. It is based on an adaptive algorithm that iteratively selfadjusts the linear filter coefficients to decrease the power of e. (n). The LMS method is one of the most widely used adaptive algorithms for adjusting the coefficients of adaptive filters, among others. The error-computation block and the weight-update block, which determine the filter’s efficiency, are the two key computing blocks of the direct-form LMS adaptive filter. In this paper, adaptive filter is implemented in two different architectures namely, zero adaptation delay adaptive filter and two adaptations delay adaptive filter which results in low power consumption and less area complexity. Zero adaptation delay adaptive filter provides nearly 52% savings in the area and the delay decreases by 26% in two adaptations delay adaptive filter over the conventional adaptive filter. Hence based on the required speed and area for the application, any one of the proposed structures can be used.