Signal processing method and system for noise removal and signal extraction
First Claim
1. A signal processing method comprising:
- receiving a signal corrupted with noise;
performing an n-level decomposition of said signal using a discrete wavelet transform to produce a smooth component and a rough component for each decomposition level;
inputting the nth level smooth component into a corresponding neural network pre-trained to recognize and extract substantially-noiseless ideal signal feature patterns from the nth level smooth component exclusively;
performing an inverse discrete wavelet transform on the substantially-noiseless ideal signal feature patterns extracted by the neural network to recover a clean signal, andoutputting the de-noised clean signal.
4 Assignments
0 Petitions
Accused Products
Abstract
A signal processing method and system combining smooth level wavelet pre-processing together with artificial neural networks all in the wavelet domain for signal denoising and extraction. Upon receiving a signal corrupted with noise, an n-level decomposition of the signal is performed using a discrete wavelet transform to produce a smooth component and a rough component for each decomposition level. The nth level smooth component is then inputted into a corresponding neural network pre-trained to filter out noise in that component by pattern recognition in the wavelet domain. Additional rough components, beginning at the highest level, may also be retained and inputted into corresponding neural networks pre-trained to filter out noise in those components also by pattern recognition in the wavelet domain. In any case, an inverse discrete wavelet transform is performed on the combined output from all the neural networks to recover a clean signal back in the time domain.
118 Citations
13 Claims
-
1. A signal processing method comprising:
-
receiving a signal corrupted with noise; performing an n-level decomposition of said signal using a discrete wavelet transform to produce a smooth component and a rough component for each decomposition level; inputting the nth level smooth component into a corresponding neural network pre-trained to recognize and extract substantially-noiseless ideal signal feature patterns from the nth level smooth component exclusively; performing an inverse discrete wavelet transform on the substantially-noiseless ideal signal feature patterns extracted by the neural network to recover a clean signal, and outputting the de-noised clean signal. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A signal processing method comprising:
-
receiving a signal corrupted with noise; decomposing the signal into a plurality of signal components using a predetermined transform; inputting each of the plurality of decomposed signal components into a corresponding neural network pre-trained to recognize and extract substantially-noiseless ideal signal feature patterns from the corresponding one of the decomposed signal components exclusively, in the transform domain; performing an inverse transform on the substantially-noiseless ideal signal feature patterns extracted by all the neural networks to recover a clean signal in the time domain, and outputting the de-noised clean signal. - View Dependent Claims (11)
-
-
12. A system comprising:
-
a discrete wavelet transformer for iteratively decomposing a signal into a plurality of decomposition levels each having a smooth component and a rough component; at least one neural network(s) corresponding in number to a set of pre-selected components comprising a highest-level smooth component and a predetermined number of rough components, each neural network operatively coupled to the discrete wavelet transformer to receive a corresponding one of the pre-selected components, and pre-trained to recognize and extract substantially-noiseless ideal signal feature patterns from the corresponding one of the pre-selected components exclusively; and an inverse discrete wavelet transformer capable of recovering a clean signal in the time domain from the substantially-noiseless ideal signal feature patterns extracted by the plurality of neural networks. - View Dependent Claims (13)
-
Specification