Statistical efficiency of adaptive algorithms
Neural Networks - 2003 Special issue: Advances in neural networks research — IJCNN'03
Linear & non-linear channel equalization using Chebyshev artificial neural network
Proceedings of the International Conference on Advances in Computing, Communication and Control
Improvement of esophageal speech by adaptive line enhancement with bias model
Proceedings of the 2nd International Convention on Rehabilitation Engineering & Assistive Technology
| Hi-index | 35.68 |
The least mean squares adaptive line enhancer (LMS ALE) has been widely used for the enhancement of coherent sinusoids in additive wideband noise. This paper studies the behavior of the LMS ALE when applied to the enhancement of sinusoids that have been corrupted by both colored multiplicative and white additive noise. The multiplicative noise decorrelates the sinusoid, spreads its power spectrum, and acts as an additional corrupting noise. Closed-form expressions are derived for the optimum (Wiener filter) ALE output SNR as a function of the residual coherent sine wave power, the noncoherent sine wave power spectrum, and the background additive white noise. When the coherent to noncoherent sine wave power ratio is sufficiently small, it is shown that a nonlinear (e.g., square law) transformation of the ALE input results in a larger optimum ALE output SNR