Sub-band structures for adaptive line enhancement

A new structure for adaptive line enhancement (ALE) is proposed. Separating the input signal in complementary sub-bands gives an increasing of the signal-to-noise ratio within each sub-band. So, short length filters can be used reducing the misadjustment and the eigenvalue spread, which is important when the LMS algorithm is used to adapt the weight vector. The steady-state behavior of the adaptive structure is analyzed for an input signal containing multiple sinusoids in white noise. It is shown that the sub-band processing improves the transient and the steady-state performance of the adaptive system.