# US 2018/0020953 A1 · Method and device for identifying human movement state > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0020953 A1 - Document ID: US20180020953A1 - Status: Published application - Kind: A1 - Application number: 15/541,313 - Filed: 2016-06-23 - Published: 2018-01-25 - Earliest claimed: 2015-12-28 - Assignees: Goertek Inc. - Applicants: Goertek Inc. - Inventors: Bo LI; Na LI - Web page: https://patentgrep.com/patent/US20180020953A1 ## CPC - A61B 5/1123: {Discriminating type of movement, e.g. walking or running (A61B5/1116, A61B5/112 take precedence)} - A61B 5/0205: Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition - A61B 5/4809: {Sleep detection, i.e. determining whether a subject is asleep or not} - A61B 5/4812: {Detecting sleep stages or cycles} - A61B 5/6801: {specially adapted to be attached to or worn on the body surface} - A61B 5/7225: {Details of analogue processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation (input circuits for detecting, measuring, or recording bioelectric or biomagnetic signals A61B5/30; specific diagnostic methods using bioelectric or biomagnetic signals A61B5/316)} - A61B 5/7278: {Artificial waveform generation or derivation, e.g. synthesizing signals from measured signals} - G01C 22/006: {Pedometers} - A61B 5/024: Measuring pulse rate or heart rate - A61B 5/1116: {Determining posture transitions} - A61B 5/681: {Wristwatch-type devices} - A61B 5/6843: {Monitoring or controlling sensor contact pressure} - A61B 5/7203: {for noise prevention, reduction or removal} - A61B 5/7246: {using correlation, e.g. template matching or determination of similarity} - A61B 2562/0219: Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) ## Abstract A method and device for identifying a human movement state that includes: determining according to acceleration signals that are collected by a three-axis acceleration sensor that a human is in a walking state, calculating a walking step number of the human, and calculating a walking step frequency according to the number; calculating a corresponding physical sign frequency during the walking process according to a physical sign signal that is collected; and comparing the walking step frequency and the physical sign frequency that are obtained by calculating respectively with a step frequency threshold and a physical sign frequency threshold, and if the walking step frequency is greater than the step frequency threshold, and the physical sign frequency is greater than the physical sign frequency threshold, determining that the human movement state is a running state, and recording the calculated walking step number to be a running step number. ## Claims ### Claim 1 A method for identifying a human movement state, wherein the method comprises: providing a three-axis acceleration sensor and a human physical sign sensor in a wearable device; determining according to acceleration signals that are collected by the three-axis acceleration sensor that a human body is in a walking state, calculating a walking step number of the human body, and calculating a walking step frequency according to the walking step number; calculating a corresponding physical sign frequency during the walking process according to a physical sign signal collected by the human physical sign sensor; and comparing the walking step frequency and the physical sign frequency that are obtained by calculating respectively with a step frequency threshold and a physical sign frequency threshold, and if the walking step frequency is greater than the step frequency threshold, and the physical sign frequency is greater than the physical sign frequency threshold, determining that the human movement state is a running state, and recording the calculated walking step number to be a running step number; and if not, determining that the human movement state is a pacing state, and recording the calculated walking step number to be a pacing step number. ### Claim 2 (depends on 1) The method according to claim 1, wherein the determining, according to acceleration signals that are collected by the three-axis acceleration sensor, that a human body is in a walking state, calculating a walking step number of the human body, and calculating a walking step frequency according to the walking step number comprises: performing a filtering processing on the acceleration signals, and removing an interfering extremum point in the acceleration signals that have been processed by filtering; calculating the number of valid extremum points in the acceleration signals with the interfering extremum point removed, the number of the valid extremum points being the walking step number; counting the number of signal sampling points between two neighboring steps; multiplying the number of the signal sampling points by a signal sampling duration, to obtain a walking cycle, a reciprocal of the walking cycle being the walking step frequency. ### Claim 3 (depends on 1) The method according to claim 1, wherein the human physical sign sensor is a heart rate sensor, and the calculating a corresponding physical sign frequency during the walking process according to a physical sign signal that is collected by the human physical sign sensor is specifically: calculating a heart rate during the walking process according to a periodic fluctuation of a heart rate signal that is collected by the heart rate sensor. ### Claim 4 A method for identifying a human movement state, wherein the method comprises: providing a three-axis acceleration sensor and a human physical sign sensor in a wearable device; detecting an instantaneous abnormal motion of a human body according to acceleration signals that are collected by the three-axis acceleration sensor; judging, according to a physical sign signal that is collected by the human physical sign sensor, whether the wearable device is being worn; if a number of the instantaneous abnormal motions that are detected within a sleeping state statistic duration is less than or equal to a preset first number threshold, and it is judged that the human body is wearing the wearable device, determining that the human movement state is a sleeping state; and if not, determining that the human movement state is a waking state. ### Claim 5 (depends on 4) The method according to claim 4, wherein the detecting an instantaneous abnormal motion of a human body according to acceleration signals that are collected by the three-axis acceleration sensor comprises: performing a filtering processing on the acceleration signals that are collected by the three-axis acceleration sensor, to filter out a direct current signal; calculating a plurality of instantaneous energies within a unit duration of the acceleration signals that have been processed by filtering, the instantaneous energies being energies of the acceleration signals in each of sub-durations that are obtained by evenly dividing the unit duration; comparing the plurality of instantaneous energies individually with a first energy threshold, and if the plurality of instantaneous energies are all less than the first energy threshold, determining that the instantaneous abnormal motions do not happen within the unit duration; and if not, determining that the instantaneous abnormal motions happen within the unit duration. ### Claim 6 (depends on 5) The method according to claim 5, wherein the if a number of the instantaneous abnormal motions that are detected within a sleeping state statistic duration is less than or equal to a preset first number threshold, and it is judged that the human body is wearing the wearable device, determining that the human movement state is a sleeping state comprises: evenly dividing the sleeping state statistic duration into a plurality of movement amount statistic durations, and counting the number of the instantaneous abnormal motions that happen within each of the movement amount statistic durations; comparing the number of the instantaneous abnormal motions within each of the movement amount statistic durations individually with the preset first number threshold, and if the number of the instantaneous abnormal motions within each of the movement amount statistic durations is less than or equal to the first number threshold, and it is judged that the human body is wearing the wearable device, determining that the human movement state is a sleeping state. ### Claim 7 (depends on 6) The method according to claim 6, wherein the determining that the human movement state is a sleeping state further comprises: evenly dividing the sleeping state statistic duration into a plurality of deep/light sleeping statistic duration sections, and counting the number of the instantaneous abnormal motions that happen within each of the duration sections; the duration lengths of the deep/light sleeping statistic duration sections being greater than those of the movement amount statistic durations; comparing sequentially the number of the instantaneous abnormal motions within each of the deep/light sleeping statistic duration sections with a preset second number threshold, the second number threshold being less than the first number threshold; if the number of the instantaneous abnormal motions within a deep/light sleeping statistic duration section is less than or equal to the second number threshold, determining that the sleeping state in the deep/light sleeping statistic duration section is a deep sleeping state; and if not, determining that the sleeping state in the deep/light sleeping statistic duration section is a light sleeping state; Or, evenly dividing the sleeping state statistic duration into a plurality of deep/light sleeping statistic duration sections, and calculating energies of acceleration signals that have been processed by filtering within each of the deep/light sleeping statistic duration sections; the duration lengths of the deep/light sleeping statistic duration sections being greater than those of the movement amount statistic durations; comparing sequentially the energies of the acceleration signals within each of the deep/light sleeping statistic duration sections with a preset second energy threshold, the second energy threshold being less than the first energy threshold; if the energies of the acceleration signals within a deep/light sleeping statistic duration section are less than or equal to the second energy threshold, determining that the sleeping state in the deep/light sleeping statistic duration section is a deep sleeping state; and if not, determining that the sleeping state in the deep/light sleeping statistic duration section is a light sleeping state. ### Claim 8 (depends on 4) The method according to claim 4, wherein the human physical sign sensor is a heart rate sensor, and the judging according to a physical sign signal that is collected by the human physical sign sensor whether the wearable device is being worn is specifically: judging, according to an energy or amplitude of a heart rate signal that is collected by the heart rate sensor, whether the wearable device is being worn; or, the human physical sign sensor is a capacitance sensor, and the judging according to a physical sign signal that is collected by the human physical sign sensor whether the wearable device is being worn is specifically: judging, according to an energy or amplitude of a capacitance signal that is collected by the capacitance sensor, whether the wearable device is being worn. ### Claim 9 A device for identifying a human movement state, adapted for a wearable device having multiple sensors, wherein the device for identifying comprises a processor and at least one computer-readable storage medium, the at least one computer-readable storage medium stores computer-executable instructions which, when executed by the processor, cause the processor to perform operations comprising: when it is determined according to acceleration signals that are collected by a three-axis acceleration sensor in the wearable device that the human body is in a walking state, calculating a walking step number of the human body according to the acceleration signals that are collected by the three-axis acceleration sensor, and calculating a walking step frequency according to the walking step number; calculating a corresponding physical sign frequency during the walking process according to a physical sign signal that is collected by a human physical sign sensor in the wearable device; comparing the calculated walking step frequency and the physical sign frequency respectively with a step frequency threshold and a physical sign frequency threshold; and if the walking step frequency is greater than the step frequency threshold, and the physical sign frequency is greater than the physical sign frequency threshold, determining that the human body movement state is a running state, and recording the calculated walking step number to be a running step number; and if not, determining that the human body movement state is a pacing state, and recording the calculated walking step number to be a pacing step number. ### Claim 10 (depends on 9) The device for identifying according to claim 9, wherein the when it is determined according to acceleration signals that are collected by a three-axis acceleration sensor in the wearable device that the human body is in a walking state, calculating a walking step number of the human body according to the acceleration signals that are collected by the three-axis acceleration sensor, and calculating a walking step frequency according to the walking step number comprises performing a filtering processing on the acceleration signals; removing an interfering extremum point in the acceleration signals that have been processed by filtering; calculating the number of valid extremum points in the acceleration signals with the interfering extremum point removed, wherein the number of the valid extremum points is the walking step number; counting the number of signal sampling points between two neighboring steps, and multiplying the number of the signal sampling points by a signal sampling duration, to obtain a walking cycle; and for obtaining the walking step frequency by taking a reciprocal of the walking cycle; wherein the human body physical sign sensor is a heart rate sensor, and the calculating a corresponding physical sign frequency during the walking process according to a physical sign signal that is collected by the human physical sign sensor comprise: calculating a heart rate during the walking process according to a periodic fluctuation of a heart rate signal that is collected by the heart rate sensor. ### Claim 11 A device for identifying a human movement state, adapted for a wearable device having multiple sensors, wherein the device for identifying comprises a processor and at least one computer-readable storage medium, the at least one computer-readable storage medium stores computer-executable instructions which, when executed by the processor, cause the processor to perform operations comprising: detecting an instantaneous abnormal motion of a human body according to acceleration signals that are collected by a three-axis acceleration sensor in the wearable device; judging, according to a physical sign signal that is collected by a human physical sign sensor in the wearable device, whether the wearable device is being worn; if a number of the instantaneous abnormal motions that are detected within a sleeping state statistic duration is less than or equal to a preset first number threshold, and it is judged that the human body is wearing the wearable device, determining that the human movement state is a sleeping state; and if not, determining that the human movement state is a waking state. ### Claim 12 (depends on 11) The device for identifying according to claim 11, wherein the detecting an instantaneous abnormal motion of a human body according to acceleration signals that are collected by a three-axis acceleration sensor in the wearable device instantaneous comprises: performing a filtering processing on the acceleration signals that are collected by the three-axis acceleration sensor, to filter out a direct current signal; calculating a plurality of instantaneous energies within a unit duration of the acceleration signals that have been processed by filtering, the instantaneous energies being energies of the acceleration signals in each of sub-durations that are obtained by evenly dividing the unit duration; comparing the plurality of instantaneous energies individually with a first energy threshold, and if the plurality of instantaneous energies are all less than the first energy threshold, determining that the instantaneous abnormal motions do not happen within the unit duration; and if not, determining that the instantaneous abnormal motions happen within the unit duration. ### Claim 13 (depends on 12) The device for identifying according to claim 12, wherein the if a number of the instantaneous abnormal motions that are detected within a sleeping state statistic duration is less than or equal to a preset first number threshold, and it is judged that the human body is wearing the wearable device, determining that the human movement state is a sleeping state comprises: evenly dividing the sleeping state statistic duration into a plurality of movement amount statistic durations, and counting the number of the instantaneous abnormal motions that happen within each of the movement amount statistic durations; comparing the number of the instantaneous abnormal motions within each of the movement amount statistic durations individually with the preset first number threshold, and if the number of the instantaneous abnormal motions within each of the movement amount statistic durations is less than or equal to the first number threshold, and it is determined that the human body is wearing the wearable device, determining that the human movement state is a sleeping state. ### Claim 14 (depends on 13) The device for identifying according to claim 13, wherein the determining that the human movement state is a sleeping state further comprises: evenly dividing the sleeping state statistic duration into a plurality of deep/light sleeping statistic duration sections, and counting the number of the instantaneous abnormal motions that happen within each of the deep/light sleeping statistic duration sections; and comparing sequentially the number of the instantaneous abnormal motions within each of the deep/light sleeping statistic duration sections with a preset second number threshold, and if the number of the instantaneous abnormal motions within a deep/light sleeping statistic duration section is less than or equal to the second number threshold, determining that the sleeping state in the deep/light sleeping statistic duration section is a deep sleeping state; and if not, determining that the sleeping state in the deep/light sleeping statistic duration section is a light sleeping state; the duration lengths of the deep/light sleeping statistic duration sections being greater than those of the movement amount statistic durations, and the second number threshold being less than the first number threshold; or, evenly dividing the sleeping state statistic duration into a plurality of deep/light sleeping statistic duration sections, and calculating energies of acceleration signals that have been processed by filtering within each of the deep/light sleeping statistic duration sections; and comparing sequentially the energies of the acceleration signals within each of the deep/light sleeping statistic duration sections with a preset second energy threshold, and if the energies of the acceleration signals within a deep/light sleeping statistic duration section are less than or equal to the second energy threshold, determining that the sleeping state in the deep/light sleeping statistic duration section is a deep sleeping state; and if not, determining that the sleeping state in the deep/light sleeping statistic duration section is a light sleeping state; the duration lengths of the deep/light sleeping statistic duration sections being greater than those of the movement amount statistic durations, the second energy threshold being less than the first energy threshold. ### Claim 15 (depends on 11) The device for identifying according to claim 11, wherein the human physical sign sensor is a heart rate sensor, and the judging according to a physical sign signal that is collected by the human physical skin sensor whether the wearable device is being worn comprise: judging, according to an energy or amplitude of a heart rate signal that is collected by the heart rate sensor, whether the wearable device is being worn; or, the human physical sign sensor is a capacitance sensor, and the judging according to a physical sign signal that is collected by the human physical sign sensor whether the wearable device is being worn comprise: judging, according to an energy or amplitude of a capacitance signal that is collected by the capacitance sensor, whether the wearable device is being worn.