HP Labs Technical Reports
Click here for full text:
A General Technique for Filtering Random Noise
Natarajan, Balas K.
Abstract: We present a novel technique for the design of filters for random noise, leading to a class of filters called Occam filters. The essence of the technique is that when a lossy data compression algorithm is applied to a noisy signal with the allowed loss set equal to the noise strength, the loss and the noise tend to cancel rather than add. We give two illustrative applications of the technique to univariate signals. We observe that an Occam filter can outperform the Wiener filter, but, unlike the Wiener filter, does not require prior knowledge of the spectral properties of the noise-free signal. We also prove asymptotic convergence bounds on the effectiveness of Occam filters.
Back to Index