# US 2014/0195201 A1 · Signal Source Separation Partially Based on Non-Sensor Information > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0195201 A1 - Document ID: US20140195201A1 - Status: Published application - Kind: A1 - Application number: 13/933,060 - Filed: 2013-07-01 - Published: 2014-07-10 - Earliest claimed: 2012-06-29 - Assignees: Speech Technology & Applied Research Corporation - Applicants: Speech Technology & Applied Research Corporation - Inventors: Richard Goldhor; Keith Gilbert; Joel MacAuslan; Karen Payton - Web page: https://patentgrep.com/patent/US20140195201A1 ## CPC - G01N 29/4472: {Mathematical theories or simulation} - G01H 3/12: by electric means (G01H3/14 takes precedence) - G01N 29/14: using acoustic emission techniques {(echo of particles G01N29/046; measuring mechanical vibrations or acoustic waves in solids in general G01H1/00)} - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) - G01H: MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES ## Abstract In environments (such as acoustic and bioelectrical environments) characterized by multiple simultaneous sources, effective blind source separation from sensor response mixtures becomes difficult as the number of sources increases—especially when the true number of sources is both unknown and changing over time. However, in some environments, non-sensor information can provide useful hypotheses for some sources. Embodiments of the present invention provide an adaptive filtering architecture for validating such source hypotheses, extracting an estimated representation of source signals corresponding to valid hypotheses, and improving the separation of the remaining “hidden” source signals from the sensor response mixtures. ## Claims ### Claim 1 A method comprising: (A) receiving at least one sensor response signal; (B) receiving at least one source hypothesis signal; and (C) generating at least one estimated source signal corresponding to the at least one sensor response signal, based on the at least one sensor response signal and the at least one source hypothesis signal. ### Claim 2 (depends on 1) The method of claim 1, wherein the at least one source hypothesis signal comprises a first source hypothesis signal that is coherent with the response signal of a first sensor, wherein (A) comprises receiving a first one of the at least one sensor response signal from the first sensor. ### Claim 3 (depends on 2) The method of claim 2, wherein the at least one source hypothesis signal further comprises a second source hypothesis signal that is coherent with the response signal of a second sensor, wherein (A) further comprises receiving a second one of the at least one sensor response signal from the second sensor. ### Claim 4 (depends on 1) The method of claim 1, wherein the at least one sensor response signal is generated by at least one sensor, wherein the at least one source hypothesis signal comprises a first source hypothesis signal that is provided as an input to a first one of at least one source, and wherein the first one of the at least one source generates a first source output signal that differs from the first source hypothesis signal. ### Claim 5 (depends on 1) The method of claim 1, wherein the at least one source hypothesis signal is generated by a source hypothesis generator from a source hypothesis description. ### Claim 6 A non-transitory computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions are executable by a computer processor to perform a method, the method comprising: (A) receiving at least one sensor response signal; (B) receiving at least one source hypothesis signal; and (C) generating at least one estimated source signal corresponding to the at least one sensor response signal, based on the at least one sensor response signal and the at least one source hypothesis signal. ### Claim 7 (depends on 6) The non-transitory computer-readable medium of claim 6, wherein the at least one source hypothesis signal comprises a first source hypothesis signal that is coherent with the response signal of a first sensor, wherein (A) comprises receiving a first one of the at least one sensor response signal from the first sensor. ### Claim 8 (depends on 7) The non-transitory computer-readable medium of claim 7, wherein the at least one source hypothesis signal further comprises a second source hypothesis signal that is coherent with the response signal of a second sensor, wherein (A) further comprises receiving a second one of the at least one sensor response signal from the second sensor. ### Claim 9 (depends on 6) The non-transitory computer-readable medium of claim 6, wherein the at least one sensor response signal is generated by at least one sensor, wherein the at least one source hypothesis signal comprises a first source hypothesis signal that is provided as an input to a first one of at least one source, and wherein the first one of the at least one source generates a first source output signal that differs from the first source hypothesis signal. ### Claim 10 (depends on 6) The non-transitory computer-readable medium of claim 6, wherein the at least one source hypothesis signal is generated by a source hypothesis generator from a source hypothesis description. ### Claim 11 A method comprising: (A) receiving at least one sensor response signal; (B) receiving at least one source hypothesis signal; and (C) generating at least one estimated source signal corresponding to the at least one source hypothesis signal, based on the at least one sensor response signal and the at least one source hypothesis signal. ### Claim 12 (depends on 6) The method of claim 6, wherein the number of the at least one estimated source signals is less than the number of the at least one source hypothesis signal. ### Claim 13 A non-transitory computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions are executable by a computer processor to perform a method, the method comprising: (A) receiving at least one sensor response signal; (B) receiving at least one source hypothesis signal; and (C) generating at least one estimated source signal corresponding to the at least one source hypothesis signal, based on the at least one sensor response signal and the at least one source hypothesis signal. ### Claim 14 (depends on 13) The non-transitory computer-readable medium of claim 13, wherein the number of the at least one estimated source signals is less than the number of the at least one source hypothesis signal. ### Claim 15 A method comprising: (A) receiving a first source hypothesis signal; (B) receiving at least one initial sensor response signal including a first sensor response signal having a component that is coherent with the first source hypothesis signal; (C) removing the coherent component from the first sensor response signal based on the first source hypothesis signal. ### Claim 16 (depends on 15) The method of claim 15, wherein the first sensor response signal represents a mixture of the first filtered source signal and a second filtered source signal. ### Claim 17 (depends on 15) The method of claim 15, further comprising: (D) receiving a second source hypothesis signal that is coherent with the second filtered source signal; and wherein the at least one initial sensor response signal further includes a second sensor response signal having a component that is coherent with the second source hypothesis signal; and wherein the method further comprises: (E) removing the coherent component from the second sensor response signal based on the second source hypothesis signal. ### Claim 18 A non-transitory computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions are executable by a computer processor to perform a method, the method comprising: (A) receiving a first source hypothesis signal; (B) receiving at least one initial sensor response signal including a first sensor response signal having a component that is coherent with the first source hypothesis signal; (C) removing the coherent component from the first sensor response signal based on the first source hypothesis signal. ### Claim 19 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the first sensor response signal represents a mixture of the first filtered source signal and a second filtered source signal. ### Claim 20 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the method further comprises: (D) receiving a second source hypothesis signal that is coherent with the second filtered source signal; and wherein the at least one initial sensor response signal further includes a second sensor response signal having a component that is coherent with the second source hypothesis signal; and wherein the method further comprises: (E) removing the coherent component from the second sensor response signal based on the second source hypothesis signal. ### Claim 21 A method comprising: (A) receiving at least one source hypothesis signal; (B) receiving at least one initial sensor response signal including a first sensor response signal representing a mixture consisting of components that are coherent with the received source hypothesis signals; and (C) removing the first sensor response signal from the at least one initial sensor response signal based on the at least one source hypothesis signal, thereby producing a reduced sensor response signal set. ### Claim 22 (depends on 21) The method of claim 21, further comprising: (D) generating, based on the reduced sensor response signal set and the at least one source hypothesis signal, at least one estimated source signal corresponding to the reduced sensor response signal set. ### Claim 23 A non-transitory computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions are executable by a computer processor to perform a method, the method comprising: (A) receiving at least one source hypothesis signal; (B) receiving at least one initial sensor response signal including a first sensor response signal representing a mixture consisting of components that are coherent with the received source hypothesis signals; and (C) removing the first sensor response signal from the at least one initial sensor response signal based on the at least one source hypothesis signal, thereby producing a reduced sensor response signal set. ### Claim 24 (depends on 23) The non-transitory computer-readable medium of claim 23, wherein the method further comprises: (D) generating, based on the reduced sensor response signal set and the at least one source hypothesis signal, at least one estimated source signal corresponding to the reduced sensor response signal set. ### Claim 25 A method comprising: (A) receiving a plurality of sensor response signals; (B) receiving a plurality of source hypothesis signals; (C) determining, based on the plurality of sensor response signals, that a first subset of the plurality of source hypothesis signals are currently valid and that a second subset of the plurality of source hypothesis signals are currently invalid; (D) generating, for each particular source hypothesis signal in the first and second subsets, an output signal indicating whether the particular source hypothesis signal is currently valid or currently invalid, based on the determinations in (C). ### Claim 26 (depends on 25) The method of claim 25, wherein (C) comprises: (C)(1) determining, for each particular signal in the plurality of the source hypothesis signals, whether the particular signal is invalid, comprising: (C)(1) determining whether the particular signal is incoherent with all of the plurality of sensor response signals; and (C)(2) determining that the particular signal is invalid if the particular signal is determined to be incoherent with all of the plurality of sensor response signals. ### Claim 27 (depends on 26) The method of claim 26, further comprising: (E) estimating, for each valid source hypothesis signal in the plurality of source hypothesis signals, a traceable source signal based on the valid source hypothesis signal and the plurality of sensor response signals. ### Claim 28 (depends on 25) The method of claim 25, wherein (C) comprises: (C)(1) determining, for each particular signal in the plurality of source hypothesis signals, whether the particular signal is valid, comprising: (C)(1) determining whether the particular signal is coherent with at least one of the plurality of sensor response signals; and (C)(2) determining that the particular signal is valid if the particular signal is determined to be coherent with at least one of the plurality of sensor response signals. ### Claim 29 A non-transitory computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions are executable by a computer processor to perform a method, the method comprising: (A) receiving a plurality of sensor response signals; (B) receiving a plurality of source hypothesis signals; (C) determining, based on the plurality of sensor response signals, that a first subset of the plurality of source hypothesis signals are currently valid and that a second subset of the plurality of source hypothesis signals are currently invalid; (D) generating, for each particular source hypothesis signal in the first and second subsets, an output signal indicating whether the particular source hypothesis signal is currently valid or currently invalid, based on the determinations in (C). ### Claim 30 (depends on 29) The non-transitory computer-readable medium of claim 29, wherein (C) comprises: (C)(1) determining, for each particular signal in the plurality of the source hypothesis signals, whether the particular signal is invalid, comprising: (C)(1) determining whether the particular signal is incoherent with all of the plurality of sensor response signals; and (C)(2) determining that the particular signal is invalid if the particular signal is determined to be incoherent with all of the plurality of sensor response signals. ### Claim 31 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the method further comprises: (E) estimating, for each valid source hypothesis signal in the plurality of source hypothesis signals, a traceable source signal based on the valid source hypothesis signal and the plurality of sensor response signals. ### Claim 32 (depends on 29) The non-transitory computer-readable medium of claim 29, wherein (C) comprises: (C)(1) determining, for each particular signal in the plurality of source hypothesis signals, whether the particular signal is valid, comprising: (C)(1) determining whether the particular signal is coherent with at least one of the plurality of sensor response signals; and (C)(2) determining that the particular signal is valid if the particular signal is determined to be coherent with at least one of the plurality of sensor response signals.