Signal separation method and apparatus for restoring original signal from observed data

A signal separation method comprising:

• estimating and restoring an original signal from observed data obtained by observing multiple mixed signals, which include said original signal, including the steps of:
• introducing, for said observed data, a cost function based on a function having a monotonously increasing characteristic;
• estimating a separation matrix using an adaptive filter that optimizes said cost function;
• and
• estimating and restoring said original signal by multiplying said separation matrix by said observed data.

A signal separation method comprising:

• separating and extracting an original signal from observed data obtained by observing multiple mixed signals, which include said original signal, including the steps of:
• reading observed signals;
• subtracting the average of said observed signals and performing zero averaging for said observed signals;
• whitening the observed signals obtained by zero averaging;
• separating said whitened observed signals based on a cost function that has a monotonously increasing characteristic;
• and
• performing, as a post processing, inverse whitening for the obtained observed signals.

A signal processing apparatus comprising:

• input means, for receiving observed data obtained by observing multiple mixed signals, which include an original signal;
• separation matrix estimation means, for estimating, for said observed data, a separation matrix using adaptive filtering for suppressing the H-infinity norm concerning said separation matrix until the H-infinity norm is to equal to or smaller than a provided scalar value;
• and
• estimation/restoration means, for estimating and restoring said original signal by multiplying said separation matrix by said observed data.

A signal processing apparatus comprising:

• input means, for receiving observed data obtained by observing multiple mixed signals, which include an original signal;
• selection means, for employing, for said observed data, the MinMax strategy in game theory to select, from separation matrixes, a specific separation matrix;
• and
• estimation/restoration means, for estimating and restoring an original signal by multiplying said separation matrix by said observed data.

A signal processing apparatus comprising:

• input means, for receiving observed data obtained by observing multiple mixed signals;
• separation matrix estimation means, for estimating, for said observed data received from said input means, a separation matrix by using an adaptive filter with optimizing a cost function that is based on a function having a monotonously increasing characteristic;
• and
• estimation/restoration means, for estimating and restoring an original signal by multiplying said separation matrix by said observed data.

A signal processing apparatus comprising:

• a non-linear function unit, for performing a non-linear function for an input observed signal and a separation matrix estimated during a previous cycle;
• an error signal calculator, for calculating an error signal based on the value obtained by said non-linear function unit, said separation matrix estimated during the previous cycle, and said observed signal at a present time;
• and
• a separation matrix update unit, for updating said separation matrix estimated at said time based on said error signal, so that error evaluation is weighted by said cost function having the monotonously increasing characteristic, for outputting and/or separating an original signal from an input observed signal.

A signal processing apparatus comprising:

• input means, for receiving mixed speech data obtained by observing multiple mixed speech signals;
• separation matrix estimation means, for estimating a separation matrix, for said mixed speech data, using an adaptive filter with optimizing a cost function that is based on a function having a monotonously increasing characteristic;
• and
• separation/extraction means, for separating and extracting said speech signals from said mixed speech data by multiplying said separation matrix by said mixed speech data.

A signal processing apparatus for separating an artifact from an observed bio-signal, said apparatus comprising:

• input means, for receiving observed data containing said artifact in said observed bio-signal;
• separation matrix estimation means, for estimating a separation matrix for said observed data, using an adaptive filter with optimizing a cost function that is based on a function having a monotonously increasing characteristic;
• and
• separation/extraction means for separating and extracting said observed bio-signal from said observed data by multiplying said separation matrix by said observed data.

A signal processing apparatus for extracting from economic statistical data, a fluctuation element that is hidden during an observation, comprising:

• input means, for receiving economic statistical data;
• separation matrix estimation means, for estimating a separation matrix for said economic statistical data using an adaptive filter with optimizing a cost function that is based on a function having a monotonously increasing characteristic;
• and
• separation/extraction means, for separating and extracting said fluctuation element from said economic statistical data by multiplying said separation matrix by said economic statistical data.

A mobile terminal device, for receiving, from a base station for code division multiple access, observed data that include the spread information to other users, and for extracting a local user signal from said observed data, comprising:

• input means, for receiving observed data from said base station;
• separation matrix estimation means, for estimating a separation matrix for said observed data using an adaptive filter with optimizing a cost function that is further based on a function having a monotonously increasing characteristic;
• and
• separation/extraction means, for separating and extracting a user signal from said observed data by multiplying said separation matrix by said observed data.