# US 2014/0195163 A1 · Processing of skin conductance signals to mitigate noise and detect signal features > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0195163 A1 - Document ID: US20140195163A1 - Status: Published application - Kind: A1 - Application number: 13/734,947 - Filed: 2013-01-05 - Published: 2014-07-10 - Earliest claimed: 2013-01-05 - Assignees: QUALCOMM Incorporated - Applicants: Qualcomm Incorporated - Inventors: Somdeb Majumdar; Aniket A. Vartak; Robert S. Tartz - Web page: https://patentgrep.com/patent/US20140195163A1 ## CPC - G16Z 99/00: Subject matter not provided for in other main groups of this subclass - A61B 5/0531: Measuring skin impedance - A61B 5/7203: {for noise prevention, reduction or removal} - G16Z: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) ## Abstract Methods, systems, and devices are described for identifying noisy regions in a skin conductance signal. The signal is divided into a plurality of windows. Two or more features of the signal within a first window are computed. At least one of the two or more features being in a frequency domain. At least two of the features are combined to obtain at least a first metric. The first metric is compared to a corresponding threshold. The first window is identified as a noisy region of the skin conductance signal based on the comparison. ## Claims ### Claim 1 A method to identify noisy regions in a skin conductance signal, comprising: dividing the signal into a plurality of windows; computing two or more features of the signal within a first window, at least one of the two or more features being in a frequency domain; combining at least two of the features to obtain at least a first metric; comparing the first metric to a corresponding threshold; and identifying the first window as a noisy region of the skin conductance signal based at least in part on the comparison. ### Claim 2 (depends on 1) The method of claim 1, wherein the two or more features of the skin conductance signal comprise at least two of a normalized range, a normalized maximum or minimum, a number of mean crossings, an amount of low frequency energy, or an amount of high frequency energy. ### Claim 3 (depends on 1) The method of claim 1, further comprising: detecting a skin conductance response (SCR) from the skin conductance signal; and ignoring the identified noisy region of the skin conductance signal while detecting the SCR. ### Claim 4 (depends on 1) The method of claim 1, wherein each window comprises a plurality of discrete samples of the skin conductance signal. ### Claim 5 (depends on 4) The method of claim 4, wherein at least one of the plurality of samples of the skin conductance signal is included in both the first window and a second window, such that the second window overlaps the first window. ### Claim 6 (depends on 1) The method of claim 1, wherein at least one of the two or more features is in a time domain. ### Claim 7 (depends on 1) The method of claim 1, wherein combining at least two of the features to obtain at least a first metric comprises: combining a first feature in the frequency domain with a second feature in the frequency domain. ### Claim 8 (depends on 1) The method of claim 1, wherein computing two or more features of the signal within the first window comprises: computing an amount of low frequency energy and an amount of high frequency energy within the first window; and combining the at least two of the features to obtain at least the first metric comprises forming a ratio of the amount of low frequency energy to the amount of high frequency energy. ### Claim 9 (depends on 1) The method of claim 1, further comprising: identifying the corresponding threshold for the first metric from training data. ### Claim 10 (depends on 9) The method of claim 9, wherein the training data is obtained from the skin conductance signal. ### Claim 11 (depends on 1) The method of claim 1, further comprising: comparing a value of at least one of the features to a corresponding at least one additional threshold; and identifying the first window as a noisy region of the skin conductance signal based on a combination of the comparisons. ### Claim 12 (depends on 11) The method of claim 11, wherein comparing a value of at least one of the features to a corresponding at least one additional threshold comprises: comparing a number of signal mean crossings to a corresponding mean crossing threshold; and identifying the first window as a noisy region of the skin conductance signal based on a combination of the comparisons. ### Claim 13 (depends on 1) The method of claim 1, wherein computing two or more features of the signal within the first window comprises: computing an amount of low frequency energy and an amount of high frequency energy within the first window; and forming a ratio of the amount of low frequency energy to the amount of high frequency energy. ### Claim 14 (depends on 1) The method of claim 1, further comprising: acquiring the skin conductance signal from a skin conductance sensor on-board a mobile device. ### Claim 15 A mobile device for identifying noisy regions in a skin conductance signal, comprising: a processor; memory in electronic communication with the processor; and instructions stored in the memory, the instructions being executable by the processor to: divide the signal into a plurality of windows; compute two or more features of the signal within a first window, at least one of the two or more features being in a frequency domain; combine at least two or the features to obtain at least a first metric; compare the first metric to a corresponding threshold; and identify the first window as a noisy region of the skin conductance signal based at least in part on the comparison. ### Claim 16 (depends on 15) The mobile device of claim 15, wherein the two or more features of the skin conductance signal comprise at least two of a normalized range, a normalized maximum or minimum, a number of mean crossings, an amount of low frequency energy, or an amount of high frequency energy. ### Claim 17 (depends on 15) The mobile device of claim 15, wherein the instructions to compute two or more features of the signal within the first window are further executable by the processor to: compute an amount of low frequency energy and an amount of high frequency energy within the first window; and form a ratio of the amount of low frequency energy to the amount of high frequency energy. ### Claim 18 (depends on 15) The mobile device of claim 15, wherein the instructions are further executable by the processor to: acquire the skin conductance signal from one or more skin conductance sensors. ### Claim 19 An apparatus for identifying noisy regions in a skin conductance signal, comprising: means for dividing the signal into a plurality of windows; means for computing two or more features of the signal within a first window, at least one of the two or more features being in a frequency domain; means for combining at least two of the features to obtain at least a first metric; means for comparing the first metric to a corresponding threshold; and means for identifying the first window as a noisy region of the skin conductance signal based at least in part on the comparison. ### Claim 20 (depends on 19) The apparatus of claim 19, wherein the two or more features of the skin conductance signal comprise at least two of a normalized range, a normalized maximum or minimum, a number of mean crossings, an amount of low frequency energy, or an amount of high frequency energy. ### Claim 21 (depends on 19) The apparatus of claim 19, wherein the means for computing two or more features of the signal within the first window comprises: means for computing an amount of low frequency energy and an amount of high frequency energy within the first window. ### Claim 22 (depends on 21) The apparatus of claim 21, wherein the means for combining at least two of the features to obtain at least the first metric comprises: means for forming a ratio of the amount of low frequency energy to the amount of high frequency energy. ### Claim 23 (depends on 19) The apparatus of claim 19, further comprising: means for acquiring the skin conductance signal from a skin conductance sensor on-board the mobile device. ### Claim 24 A computer program product for identifying noisy regions in a skin conductance signal, the computer program product comprising a non-transitory computer-readable medium storing instructions executable by a processor to: divide the signal into a plurality of windows; compute two or more features of the signal within a first window, at least one of the two or more features being in a frequency domain; combine at least two of the features to obtain at least a first metric; compare the first metric to a corresponding threshold; and identify the first window as a noisy region of the skin conductance signal based at least in part on the comparison. ### Claim 25 (depends on 24) The computer program product of claim 24, wherein the two or more features of the skin conductance signal comprise at least two of a normalized range, a normalized maximum or minimum, a number of mean crossings, an amount of low frequency energy, or an amount of high frequency energy. ### Claim 26 (depends on 24) The computer program of claim 24, wherein the instructions to compute two or more features of the signal within the first window are further executable by the processor to compute an amount of low frequency energy and an amount of high frequency energy within the first window. ### Claim 27 (depends on 26) The computer program of claim 26, wherein the instructions to combine at least two of the features to obtain at least the first metric are further executable by the processor to: form a ratio of the amount of low frequency energy to the amount of high frequency energy. ### Claim 28 A method to detect peaks of a skin conductance signal, comprising: identifying a first set of peaks by identifying a first set of one or more parameters of the skin conductance signal, the first set of peaks comprising one or more primary peaks of the skin conductance signal; comparing each parameter of the first set of one or more parameters to a corresponding threshold in a first set of one or more thresholds; computing a second set of one or more parameters of at least one primary peak in the first set of peaks; comparing each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds; and identifying at least one secondary peak associated with the at least one primary peak based at least in part on the comparison of each parameter of the second set of one or more parameters to a corresponding threshold in the second set of one or more thresholds. ### Claim 29 (depends on 28) The method of claim 28, wherein the skin conductance signal comprises a plurality of discrete samples. ### Claim 30 (depends on 29) The method of claim 29, wherein computing the second set of one or more parameters comprises: computing a set of second derivatives for a set of discrete samples located along a first edge of a first primary peak; and determining, from the set of second derivatives, a curvedness factor of the first edge. ### Claim 31 (depends on 30) The method of claim 30, wherein determining the curvedness factor of the first edge comprises: determining a range of the set of second derivatives; and identifying a magnitude of the range as the curvedness factor. ### Claim 32 (depends on 30) The method of claim 30, wherein comparing each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds comprises: comparing the curvedness factor of the first edge to a curvedness threshold. ### Claim 33 (depends on 30) The method of claim 30, further comprising: identifying a location of a secondary peak associated with the first primary peak based on a change in sign in the set of second derivatives. ### Claim 34 (depends on 28) The method of claim 28, wherein the first set of one or more parameters of the skin conductance signal comprise at least one of a slope, a rise time, or an amplitude. ### Claim 35 (depends on 28) The method of claim 28, further comprising: acquiring the skin conductance signal from a skin conductance sensor on-board a mobile device. ### Claim 36 A mobile device for detecting peaks of a skin conductance signal, comprising: a processor; memory in electronic communication with the processor; and instructions stored in the memory, the instructions being executable by the processor to: identify a first set of peaks by identifying a first set of one or more parameters of the skin conductance signal, the first set of peaks comprising one or more primary peaks of the skin conductance signal; comparing each parameter of the first set of one or more parameters to a corresponding threshold in a first set of one or more thresholds; compute a second set of one or more parameters of at least one primary peak in the first set of peaks; compare each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds; and identify at least one secondary peak associated with the at least one primary peak based at least in part on the comparison of each parameter of the second set of one or more parameters to a corresponding threshold in the second set of one or more thresholds. ### Claim 37 (depends on 36) The mobile device of claim 36, wherein the skin conductance signal comprises a plurality of discrete samples. ### Claim 38 (depends on 37) The mobile device of claim 37, wherein the instructions to compute the second set of one or more parameters are further executable by the processor to: compute a set of second derivatives for a set of discrete samples located along a first edge of a first primary peak; and determine, from the set of second derivatives, a curvedness factor of the first edge. ### Claim 39 (depends on 38) The mobile device of claim 38, wherein the instructions to determine the curvedness factor of the first edge are further executable by the processor to: determine a range of the set of second derivatives; and identify a magnitude of the range as the curvedness factor. ### Claim 40 (depends on 36) The mobile device of claim 36, wherein the instructions are further executable by the processor to: identify a location of a secondary peak associated with the first primary peak based on a change in sign in the set of second derivatives. ### Claim 41 An apparatus for detecting peaks of a skin conductance signal, comprising: means for identifying a first set of peaks by identifying a first set of one or more parameters of the skin conductance signal, the first set of peaks comprising one or more primary peaks of the skin conductance signal; means for comparing each parameter of the first set of one or more parameters to a corresponding threshold in a first set of one or more thresholds; means for computing a second set of one or more parameters of at least one primary peak in the first set of peaks; means for comparing each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds; and means for identifying at least one secondary peak associated with the at least one primary peak based at least in part on the comparison of each parameter of the second set of one or more parameters to a corresponding threshold in the second set of one or more thresholds. ### Claim 42 (depends on 41) The apparatus of claim 41, wherein the skin conductance signal comprises a plurality of discrete samples. ### Claim 43 (depends on 42) The apparatus of claim 42, wherein the means for computing the second set of one or more parameters comprises: means for computing a set of second derivatives for a set of discrete samples located along a first edge of a first primary peak; and means for determining, from the set of second derivatives, a curvedness factor of the first edge. ### Claim 44 (depends on 43) The apparatus of claim 43, wherein the means for determining the curvedness factor of the first edge comprises: means for determining a range of the set of second derivatives; and means for identifying a magnitude of the range as the curvedness factor. ### Claim 45 (depends on 41) The apparatus of claim 41, further comprising: means for identifying a location of a secondary peak associated with the first primary peak based on a change in sign in the set of second derivatives. ### Claim 46 A computer program product for detecting peaks of a skin conductance signal, the computer program product comprising a non-transitory computer-readable medium storing instructions executable by a processor to: identify a first set of peaks by identifying a first set of one or more parameters of the skin conductance signal, the first set of peaks comprising one or more primary peaks of the skin conductance signal; compare each parameter of the first set of one or more parameters to a corresponding threshold in a first set of one or more thresholds compute a second set of one or more parameters of at least one primary peak in the first set of peaks; compare each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds; and identify at least one secondary peak associated with the at least one primary peak based at least in part on the comparison of each parameter of the second set of one or more parameters to a corresponding threshold in a second set of one or more thresholds. ### Claim 47 (depends on 46) The computer program product of claim 46, wherein the skin conductance signal comprises a plurality of discrete samples. ### Claim 48 (depends on 47) The computer program product of claim 47, wherein the instructions to compute the second set of one or more parameters are further executable by the processor to: compute a set of second derivatives for a set of discrete samples located along a first edge of a first primary peak; and determine, from the set of second derivatives, a curvedness factor of the first edge. ### Claim 49 (depends on 48) The computer program product of claim 48, wherein the instructions to determine the curvedness factor of the first edge are further executable to: determine a range of the set of second derivatives; and identify a magnitude of the range as the curvedness factor. ### Claim 50 (depends on 46) The computer program product of claim 46, wherein the instructions are further executable by the processor to: identify a location of a secondary peak associated with the first primary peak based on a change in sign in the set of second derivatives. ### Claim 51 A method to convert a skin conductance signal from a time domain to an energy domain, comprising: dividing the skin conductance signal in the time domain into a plurality of windows; computing a function of the skin conductance signal in a first window to obtain a first metric of energy in the energy domain; and computing a second signal in the energy domain, the second signal being a function of the metric of energy. ### Claim 52 (depends on 51) The method of claim 51, wherein each window comprises a plurality of discrete samples of the skin conductance signal. ### Claim 53 (depends on 52) The method of claim 52, wherein at least one of the plurality of discrete samples of the skin conductance signal is included in overlapping windows of the plurality of windows. ### Claim 54 (depends on 51) The method of claim 51, wherein computing the function of the skin conductance signal in the first window to obtain a first metric of energy in the energy domain comprises: computing an area defined at least in part by the first window and the skin conductance signal. ### Claim 55 (depends on 54) The method of claim 54, further comprising: comparing the second signal to a corresponding threshold; and identifying a peak of the skin conductance signal in the energy domain based on the comparison. ### Claim 56 (depends on 51) The method of claim 51, further comprising: computing the function of the skin conductance signal in a second window to obtain a second metric of energy in the energy domain. ### Claim 57 (depends on 56) The method of claim 56, wherein dividing the skin conductance signal in the time domain into a plurality of windows comprises: adjusting a width of the second window to mitigate a variance between the first metric of energy and the second metric of energy. ### Claim 58 (depends on 51) The method of claim 51, further comprising: acquiring the skin conductance signal from a skin conductance sensor on-board a mobile device. ### Claim 59 A mobile device for converting a skin conductance signal from a time domain to an energy domain, comprising: a processor; memory in electronic communication with the processor; and instructions stored in the memory, the instructions being executable by the processor to: divide the skin conductance signal in the time domain into a plurality of windows; compute a function of the skin conductance signal in a first window to obtain a first metric of energy in the energy domain; and compute a second signal in the energy domain, the second signal being a function of the metric of energy. ### Claim 60 An apparatus for converting a skin conductance signal from a time domain to an energy domain, comprising: means for dividing the skin conductance signal in the time domain into a plurality of windows; means for computing a function of the skin conductance signal in a first window to obtain a first metric of energy in the energy domain; and means for computing a second signal in the energy domain, the second signal being a function of the metric of energy. ### Claim 61 A computer program product for converting a skin conductance signal from a time domain to an energy domain, the computer program product comprising a non-transitory computer-readable medium storing instructions executable by a processor to: divide the skin conductance signal in the time domain into a plurality of windows; compute a function of the skin conductance signal in a first window to obtain a first metric of energy in the energy domain; and compute a second signal in the energy domain, the second signal being a function of the metric of energy.