ADAPTIVE NOISE CANCELLATION USING WEINER FILTER

Abstract

This paper defines the concept of adaptive noise cancelling, an alternative method of estimating signals corrupted
by additive no`q ise or interference. The method uses a primary input containing the corrupted Signal and a reference input
containing noise correlated in some unknown way with the primary noise. The reference input is adaptively filtered and
subtracted from the primary input to obtain the signal estimate. Wiener solutions are developed to describe asymptotic
adaptive performance and output signal-to-noise ratio for stationary stochastic inputs, including single and multiple
reference inputs. These solutions show that when the reference input is free of signal and certain other conditions are met
noise in the primary input can be essentially eliminated without signal distortion. It is further shown that in wiener filter with
narrow bandwidth, infinite null, and the capability of treating periodic interference the adaptive noise canceller acts as
tracking the exact frequency of the interference; in this case the canceller behaves as a linear, time-invariant system, with the
adaptive filter results presented that illustrate the usefulness of the adaptive noise cancelling on a dynamic rather than a
static solution. In the first stage the narrowband noise is reduced by using the ALE technique. The second stage gets a signal
with ideally only the wideband noise which is reduced using the NLMS technique. The reference input is adaptively filtered
and subtracted from primary input to obtain signal estimate.