Adaptive signal processing
Theory and design of adaptive filters
Theory and design of adaptive filters
Hi-index | 0.00 |
A block adaptive filter accumulates an average error over L samples before a coefficient update is completed. Analysis of such filters can predict learning rate and variance at convergence. This paper studies the performance advantages of the time domain block LMS algorithm (BLMS) relative to the single update LMS algorithm. An exact relationship of weight update between BLMS and LMS is presented. Based upon this relationship an equation of equivalent µB for µ is derived. Also the performance of BLMS in terms of convergence rate, misadjustment, signal to noise ratio and stability bound are analyzed and compared to LMS in terms of the equivalent µB. Simulation results support the analysis.