NLMS adaptive filter

Normalised LMS: divides the step by the instantaneous input power x_seg' x_seg + eps, giving robust convergence over a wider range of input scales.

Usage

morie_dsp_nlms(x, d, order = 16L, mu = 0.5, eps = 1e-08)

Arguments

x

Input (reference) vector.

d

Desired vector, same length as x.

order

Filter order (taps). Default 16.

mu

Normalised step size (typically 0 < mu < 2). Default 0.5.

eps

Power-floor for division. Default 1e-8.

Value

List with y, e.

References

Rangayyan & Krishnan (2015), Ch. 3, sec. 3.6.

Examples

if (requireNamespace("ranger", quietly = TRUE)) {
  set.seed(1L)
  x <- rnorm(4000)
  d <- x + 0.01 * rnorm(4000)
  out <- morie_dsp_nlms(x, d, order = 8L, mu = 1.0)
  mean(out$e[3001:4000]^2)
}
#> [1] 2.906927