Wake-up method and device

Abstract

The invention discloses an awakening method, which comprises the following steps: acquiring a current audio signal on a first processor side which is continuously powered, acquiring a current spectrum feature of the current audio signal based on the current audio signal, acquiring a target spectrum feature of a pre-acquired target audio signal, matching the current spectrum feature with the target spectrum feature, and outputting an awakening result. And under the condition of successful matching, awakening the second processor in the dormant state. According to the invention, through audio signal acquisition and feature comparison, detection omission of an intrusion detector is made up.

Description

Technical Field

[0001] The present invention relates to the field of electronic equipment, and in particular, to a wake-up method and device. Background Art

[0002] In order to reduce power consumption, more and more electronic devices have two states: sleep state and working state. Usually, they are in sleep state when no business processing is required. When an activation event is detected, the electronic device is activated, so that the electronic device switches from sleep state to working state. The above process of detecting an activation event to activate the device is usually called wake-up.

[0003] Take the image acquisition equipment used for monitoring, such as cameras, as an example. A battery camera is mainly powered by batteries and is very sensitive to power consumption. In order to extend its use time and reduce system power consumption, the business modules used to process business in the camera, such as image acquisition devices, are in a power-off or dormant state most of the time, and are activated and start working only after being triggered. In order to detect activation events, intrusion detectors for intrusion detection are usually used for detection, such as passive infrared (PIR) detectors and / or radar detectors.

[0004] Since the intrusion detector can only detect targets within its detection area and cannot detect situations outside the detection area, the service module cannot be awakened in time to effectively capture video and audio. Summary of the invention

[0005] The invention provides a wake-up method to make up for the failure of electronic equipment to be woken up in time due to missed detection by an intrusion detector.

[0006] A first aspect of the present invention provides a wake-up method, the method comprising:

[0007] On the first processor side which is continuously powered,

[0008] Get the current audio signal.

[0009] Based on the current audio signal, obtain the current spectrum characteristics of the current audio signal,

[0010] Obtain the target spectrum characteristics of the pre-collected target audio signal,

[0011] The current spectrum feature is matched with the target spectrum feature, and if the match is successful, the second processor in the dormant state is awakened.

[0012] Preferably, the first processor is externally connected to an audio acquisition circuit for collecting and amplifying audio signals, and the audio acquisition circuit is continuously powered.

[0013] The obtaining of the current audio signal comprises:

[0014] The first processor detects whether the input level from the audio acquisition circuit meets the trigger condition.

[0015] If yes, the first processor switches the pin connected to the audio acquisition circuit to an automatic data acquisition mode to collect the audio signal from the audio acquisition circuit;

[0016] The obtaining, based on the current audio signal, current frequency spectrum features of the current audio signal includes:

[0017] Perform Fourier transform on the collected audio signal to obtain the amplitude-frequency signal after Fourier transform.

[0018] Based on the amplitude-frequency signal after Fourier transformation, the frequency spectrum characteristics of the collected audio signal are obtained;

[0019] The step of waking up the second processor in a dormant state comprises:

[0020] The first processor sends a power-on instruction for supplying power to the second processor to the battery management module, so that the second processor is powered and converted from a sleep state to an operating state.

[0021] Preferably, the first processor is a microprocessor chip, and the second processor is a system-level chip.

[0022] The method of obtaining the frequency spectrum characteristics of the collected audio signal based on the amplitude-frequency signal after Fourier transformation includes:

[0023] Within the set frequency range, extract the signal amplitudes of m first frequency points from the amplitude-frequency signal after Fourier transformation.

[0024] Calculate the ratio of signal amplitudes between adjacent first frequency point signals to obtain m-1 first ratio results;

[0025] The step of obtaining a target frequency spectrum feature of a pre-collected target audio signal comprises:

[0026] Get the target audio signal that has been collected and stored in advance,

[0027] Perform Fourier transform on the target audio signal to obtain the target amplitude-frequency signal after Fourier transform.

[0028] Within the set frequency range, extract the signal amplitudes of m second frequency points from the target amplitude-frequency signal after Fourier transformation,

[0029] The ratio of the signal amplitudes between adjacent second frequency point signals is calculated to obtain m-1 second ratio results, wherein the extracted first frequency point is the same as the extracted second frequency point, and m is a natural number not less than 2.

[0030] Preferably, the target spectrum characteristics are pre-stored.

[0031] The target spectrum characteristics are determined as follows:

[0032] Perform Fourier transform on the pre-collected target audio signal to obtain the target amplitude-frequency signal after Fourier transform.

[0033] Within the set frequency range, extract the signal amplitudes of m second frequency points from the target amplitude-frequency signal after Fourier transformation.

[0034] Calculate the ratio of the signal amplitudes between adjacent second frequency point signals to obtain m-1 second ratio results;

[0035] The method of obtaining the frequency spectrum characteristics of the collected audio signal based on the amplitude-frequency signal after Fourier transformation includes:

[0036] Within the set frequency range, extract the signal amplitudes of m first frequency points from the amplitude-frequency signal after Fourier transformation.

[0037] The ratio of the signal amplitudes between adjacent first frequency point signals is calculated to obtain m-1 first ratio results, wherein the extracted first frequency point is the same as the extracted second frequency point, and m is a natural number not less than 2.

[0038] Preferably, matching the current spectrum feature with the target spectrum feature includes:

[0039] Calculate the ratio between each first ratio result and its corresponding second ratio result to obtain m-1 third ratio results, wherein the first ratio result and its corresponding second ratio result correspond to the same frequency point,

[0040] Count the number of third ratio results within the set range,

[0041] When the counted number is greater than the set threshold, it is determined that the current spectrum feature matches the target spectrum feature successfully.

[0042] Preferably, the audio acquisition circuit includes a pickup for picking up an audio signal, and an amplifier circuit for amplifying the audio signal from the pickup, wherein the amplifier circuit is configured such that: when the voltage amplitude of the audio signal from the pickup is higher than a set voltage threshold, the level applied to the pin reaches the trigger condition; when the voltage amplitude of the audio signal from the pickup is not higher than the set voltage threshold, the level applied to the pin does not reach the trigger condition, and the trigger condition is a set trigger level threshold;

[0043] The target audio signal is input to the first processor through the audio acquisition circuit and stored, or is collected by an application program and then sent to the first processor.

[0044] A second aspect of the present invention provides a wake-up device, the device comprising:

[0045] The first processor is used to obtain a current audio signal, obtain a current frequency spectrum feature of the current audio signal based on the current audio signal, obtain a target frequency spectrum feature of a pre-collected target audio signal, match the current frequency spectrum feature with the target frequency spectrum feature, and wake up the second processor in a dormant state if the match is successful.

[0046] The first processor is continuously powered.

[0047] Preferably, the first processor is a microprocessor chip, and the second processor is a system-level chip.

[0048] The device also includes,

[0049] An audio acquisition circuit for collecting and amplifying audio signals. The audio signal amplified by the circuit is input to the first processor. The audio acquisition circuit is continuously powered.

[0050] The battery management module is used to continuously supply power to the first processor and receive a power-on instruction sent by the first processor to supply power to the second processor, so that the second processor is powered and converted from a sleep state to a working state.

[0051] A third aspect of the present invention provides an electronic device, comprising any of the wake-up devices.

[0052] Preferably, the electronic device is a camera.

[0053] The embodiment of the present application realizes detection of a desired audio signal by acquiring audio signals through a first processor that is continuously powered, and comparing the spectral characteristics of the acquired audio signals with the target spectral characteristics, so that the detected desired audio signal is used as an activation event to wake up the second processor that is in a dormant state. The present application makes up for the defect that the processor in a dormant state cannot be awakened in time due to missed detection by the intrusion detector, which is beneficial to reducing power consumption and improving the effectiveness of the business processed by the second processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The figure is a flowchart of a wake-up method according to an embodiment of the present application.

[0055] Figure 2 A schematic diagram of the functional modules of a camera.

[0056] Figure 3 A schematic diagram of an audio acquisition module.

[0057] Figures 4a to 4d To adopt Figure 3 A schematic diagram of the audio acquisition module simulation shown.

[0058] Figure 5 This is a schematic diagram of a process for MCU waking up the audio and video module in an embodiment of the present application.

[0059] Figure 6 A schematic diagram of a wake-up device according to an embodiment of the present application.

[0060] Figure 7 A schematic diagram of PIR, radar, and MIC wake-up tests.

[0061] Figure 8 A simplified diagram of the relationship between relative humidity, temperature and sound absorption / frequency. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical means and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings.

[0063] The embodiment of the present application realizes wake-up by collecting audio signals. Figure 1 As shown, Figure 1 The flowchart of a wake-up method according to an embodiment of the present application is shown in FIG. The method comprises: on a first processor side which is continuously powered and not in a sleep state,

[0064] Step 101, obtaining the current audio signal,

[0065] As an example, the first processor acquires the audio signal through an audio acquisition circuit for collecting and amplifying the audio signal, and the audio acquisition circuit is continuously powered to maintain a continuous working state without sleeping.

[0066] As an example, the audio acquisition circuit includes a pickup for picking up an audio signal, and an amplifier circuit for amplifying the audio signal from the pickup, the amplifier circuit being configured such that: when the voltage amplitude of the audio signal from the pickup is higher than a set voltage threshold, the level applied to the pin reaches the trigger condition, and when the voltage amplitude of the audio signal from the pickup is not higher than the set voltage threshold, the level applied to the pin does not reach the trigger condition, and the trigger condition may be a set trigger level threshold;

[0067] The first processor detects whether the input level from the audio acquisition circuit meets the trigger condition.

[0068] If yes, the first processor switches the pin connected to the audio acquisition circuit to an automatic data acquisition mode to collect the audio signal from the audio acquisition circuit;

[0069] Otherwise, the first processor processes the input level, for example, maintains the level trigger mode.

[0070] Step 102: based on the current audio signal, obtain the current frequency spectrum feature of the current audio signal.

[0071] As an example, the collected audio signal is subjected to Fourier transformation to obtain an amplitude-frequency signal after Fourier transformation. Within a set frequency range, the signal amplitudes of m first frequency points are extracted from the amplitude-frequency signal after Fourier transformation, and the ratio of the signal amplitudes between adjacent first frequency point signals is calculated to obtain m-1 first ratio results, and the m-1 first ratio results form the current frequency spectrum features;

[0072] Step 103, obtaining the target frequency spectrum characteristics of the pre-collected target audio signal,

[0073] As an example, a target audio signal collected and stored in advance is obtained, and a Fourier transform is performed on the target audio signal to obtain a target amplitude-frequency signal after the Fourier transform. Within the set frequency range, the signal amplitudes of m second frequency points are extracted from the target amplitude-frequency signal after the Fourier transform, and the ratio of the signal amplitudes between adjacent second frequency point signals is calculated to obtain m-1 second ratio results, and the m-1 second ratio results form the target frequency spectrum features.

[0074] Among them, the extracted first frequency point is the same as the extracted second frequency point, m is a natural number not less than 2, the target audio signal can be input into the first processor and stored through the audio acquisition circuit, or, after being collected by the application, sent to the first processor, and the target audio signal includes but is not limited to a child's crying, knocking, alarm, falling, etc.

[0075] As another example, the target spectrum feature is pre-stored, so the current spectrum feature can be acquired in the same manner as the target spectrum feature.

[0076] Step 104: Match the current spectrum feature with the target spectrum feature, and if the match is successful, wake up the second processor in the dormant state.

[0077] As an example, the first processor calculates the ratio between each first ratio result and its corresponding second ratio result to obtain m-1 third ratio results, wherein the first ratio result and its corresponding second ratio result correspond to the same frequency point,

[0078] The number of third ratio results within a set range is counted. When the counted number is greater than a set threshold, it is determined that the current spectrum feature successfully matches the target spectrum feature. The first processor then sends a power-on instruction to the battery management module to supply power to the second processor, so that the second processor is powered and converted from a sleep state to a working state. Otherwise, it is determined that the current spectrum feature does not successfully match the target spectrum feature, and the first processor does not send a power-on instruction.

[0079] As an example, the second processor is a system on chip (SOC), and if the current spectrum feature successfully matches the target spectrum feature, the first processor enables the power supply circuit of the second processor, for example, enables the power supply pin of the second processor.

[0080] The wake-up method provided in the embodiment of the present application collects audio signals and matches the spectral characteristics of the collected audio signals with the target spectral characteristics, thereby detecting the expected audio signals in the environment as activation events for wake-up, effectively making up for the omissions of the intrusion detector.

[0081] To facilitate understanding of the embodiments of the present application, the wake-up method of the embodiments of the present application is described below using a camera as an example. It should be understood that the present application is not limited to cameras, but is also applicable to wake-up of other electronic devices.

[0082] See also Figure 2 As shown, Figure 2 Schematic diagram of the functional modules of a camera. As an example, it includes three parts: audio and video module, low power management module and battery management module.

[0083] The core device of the audio and video module is the system-on-chip (SOC), and the periphery may include image sensors, memory, microphones and speakers, etc. The SOC chip is mainly used for image and audio signal processing, audio and video data storage, wireless signal transmission, etc. The audio and video module is completely powered off when not enabled, and is powered on and processes related services only after being activated;

[0084] The low power management module is in a state of long-term continuous power supply. Its core components include a microprocessor (MCU), a network access module for connecting to the network, an intrusion detector for intrusion detection, and other buttons, instruction lights, and a power meter.

[0085] The battery management module is mainly composed of a battery and its charge and discharge management chip. The battery management module provides continuous power to the low power management module and intermittent power to the audio and video module.

[0086] As an example, the MCU issues instructions to control the power supply circuit of the audio and video module, such as controlling the power supply enable pin of the SOC chip, and then controlling the power on and off of the audio and video module.

[0087] After the intrusion detector detects an activation event, it sends a trigger signal to the MCU. In response to the trigger signal, the MCU outputs a power supply instruction to the power supply circuit of the audio and video module, so that the battery management module supplies power to the audio and video module, and the MCU wakes up the audio and video module.

[0088] Since intrusion detectors such as PIR detectors and / or radar detectors can only detect targets within their detection areas, they cannot detect situations outside the detection areas.

[0089] In view of this, an embodiment of the present application adds an audio acquisition module for collecting and amplifying audio signals in the low power management module, so as to obtain audio signals such as children crying, knocking on the door, elderly people falling, emergency calls, etc. in the family through the audio acquisition module.

[0090] See also Figure 3 As shown, Figure 3 A schematic diagram of an audio acquisition module. The module is an audio acquisition circuit, including a pickup for picking up an audio signal, such as a microphone (MIC), and an amplifier circuit for amplifying an audio signal from the pickup, wherein the amplifier circuit includes a triode, a branch for isolating direct current, and a bias circuit for providing a static working voltage for the triode. As an example, in the bias circuit, one end of the bias resistor R2 is connected to the working power supply of the amplifier circuit, and the other end is connected to the collector of the triode. One end of the bias resistor R3 is connected to the working power supply, and the other end is connected to the base of the triode. The collector of the triode is also connected to the MCU. The branch for isolating direct current is connected in parallel with the bias resistor R3. The branch is formed by a resistor R1 and a capacitor in series. The audio signal output by the microphone is input to the base of the triode through the capacitor.

[0091] The transistor amplification factor is β, and the resistors R2 and R3 are selected with appropriate resistance values. In static state, that is, when the MIC has no output, the transistor is in the amplification state. At this time, the transistor output voltage is the voltage U between the collector and the emitter. CEQ It can be calculated as follows:

[0092] Assume the transistor gain is β, and the voltage between the transistor base and emitter is U BE , then:

[0093] β×(V CC -U BE ) / R3=(V CC -U CEQ ) / R2

[0094] Thus, we can get: U CEQ =V CC -R2×β×(V CC -U BE ) / R3

[0095] Among them, V CC It is the working power supply voltage of the amplifier circuit.

[0096] For the selected transistor, U BE and β are fixed, select appropriate R2 and R3 values ​​to make the trigger level U CEQ When the set trigger level threshold is reached, the trigger level is used to trigger the MCU to collect audio signals through the audio acquisition module.

[0097] As an example, the initial value of the pin connected to the MCU and the amplifier circuit is set to: low level trigger, for example, it is triggered after detecting a level lower than 0.7V. It should be understood that the initial value of the pin connected to the MCU and the amplifier circuit can also be set to high level trigger. In this case, as an example, it is only necessary to add an inverting circuit to the output end of the amplifier circuit.

[0098] See also Figures 4a to 4d As shown, Figures 4a to 4d To adopt Figure 3 A schematic diagram of the audio acquisition module simulation is shown. In the figure, XFG1 is the simulation signal source, and XSC1 is the simulation oscilloscope. The simulation signal source simulates the AC peak-to-peak voltage output by the MIC. Taking the electret MIC as an example, if the sensitivity of the MIC is -38dB and the maximum input sound pressure level is 110dB SPL, the maximum output peak-to-peak voltage is 224mV, and the MIC output AC peak-to-peak voltage range is 0 to 224mV. The sound frequency range that can be heard by the human ear is 20HZ to 20KHZ. During the simulation, signals within the four ranges of 20KHZ 10mv, 20HZ 10mV, 20KHZ 1mv, and 20HZ 1mV were selected.

[0099] like Figure 4a As shown in the figure, when the audio acquisition module inputs an audio signal of 20KHZ and a voltage amplitude of 10mV, U CE The maximum value is 2.47V, U CE The minimum value is 0.103V. If there is a low-level signal less than the MCU low-level threshold of 0.7V, the MCU will be triggered;

[0100] like Figure 4b As shown in the figure, when the audio acquisition module inputs an audio signal of 20HZ and a voltage amplitude of 10mV, U CE The maximum value is 1.97V, U CEThe minimum value is 0.209V. If there is a low-level signal less than the MCU low-level threshold of 0.7V, the MCU will be triggered;

[0101] like Figure 4c As shown in the figure, when the audio acquisition module inputs an audio signal of 20KHZ and a voltage amplitude of 1mV, U CE The maximum value is 1.21V, U CE The minimum value is 0.925V. If there is no low-level signal less than the MCU low-level threshold of 0.7V, the MCU will not be triggered;

[0102] like Figure 4d As shown in the figure, when the audio acquisition module inputs an audio signal of 20HZ and a voltage amplitude of 1mV, U CE The maximum value is 1.16V, U CE The minimum value is 0.976V. If there is no low-level signal less than the MCU low-level threshold of 0.7V, the MCU will not be triggered.

[0103] It can be seen from the simulation results that the larger the voltage amplitude of the audio signal input to the audio acquisition module, the more likely the MCU will be triggered. In other words, the louder the sound, the easier it is for the MCU to be triggered. In this way, large audio signals are collected and small audio signals are filtered, thereby taking into account both the effectiveness of wake-up and the reduction of power consumption. It can be seen that the amplifier circuit makes: when the voltage amplitude of the audio signal from the pickup is higher than the set voltage threshold, the level applied to the pin reaches the trigger condition, and when the voltage amplitude of the audio signal from the pickup is not higher than the set voltage threshold, the level applied to the pin does not reach the trigger condition, and the trigger condition can be the set trigger level threshold.

[0104] See also Figure 5 As shown, Figure 5 This is a flow chart of MCU waking up the audio and video module in the embodiment of the present application. It includes:

[0105] Step 501: The MCU determines whether the level on the pin connected to the audio acquisition module meets the trigger condition.

[0106] When the level input from the audio acquisition module to the MCU meets the trigger condition, the MCU is triggered. After the trigger, the MCU changes to the automatic data collection (ADC) sampling mode and samples the current audio signal through the audio acquisition module.

[0107] Otherwise, the MCU is not triggered, the level trigger mode is maintained, and the process returns to step 501 .

[0108] Step 502: The MCU performs Fourier transform on the sampled audio signal to obtain an amplitude-frequency signal of the sampled audio signal.

[0109] Step 503, based on the amplitude-frequency signal after Fourier transformation, obtain the frequency spectrum characteristics of the collected audio signal,

[0110] As an example, the signal amplitude ratios between m set adjacent first frequency point signals are calculated to obtain m-1 first ratios, where m is a natural number not less than 2 and can be specifically determined according to the amount of calculation.

[0111] For example, extract the signal amplitudes of m first frequency points within the set frequency range, for example, extract the signal amplitude u1 of the first frequency of 20HZ, the signal amplitude u2 of the first frequency of 100HZ, the signal amplitude u3 of the first frequency of 500HZ, the signal amplitude u4 of the first frequency of 1KHZ, ... the signal amplitude u23 of the first frequency of 20KHz within the range of 20HZ to 20KHZ, calculate the signal amplitude ratios between adjacent first frequency points, and obtain m-1 first ratios, for example, a1 = u2 / u1, a2 = u3 / u2 ... a22 = u23 / u22. In this way, the main part of the collected audio signal can be restored through the signal amplitude ratios between multiple adjacent frequency signals.

[0112] Step 504: obtaining a target frequency spectrum feature for comparison with the frequency spectrum feature of the collected audio signal.

[0113] As an example, the MCU obtains a target audio signal that has been pre-collected and stored, such as a child's crying, a knock on the door, an alarm, etc., performs Fourier operation on the target audio signal to obtain a target amplitude-frequency signal after Fourier transformation, and calculates the signal amplitude ratio between m set adjacent second frequency points in the same manner as step 503. For example, extract the signal amplitude u1' of the second frequency of 20 Hz, the signal amplitude u2' of the second frequency of 100 Hz, the signal amplitude u3' of the second frequency of 500 Hz, the signal amplitude u4' of the second frequency of 1 kHz, ... the signal amplitude u23' of the second frequency of 20 kHz, calculate the signal amplitude ratio between adjacent frequency signals, and obtain m-1 second ratios as target frequency spectrum features, for example, a1'=u2' / u1', a2'=u3' / u2'...a22'=u23' / u22'.

[0114] In this step, the second frequency point extracted is the same as the first frequency point extracted in step 503 .

[0115] Step 505, compare the frequency spectrum characteristics obtained in step 503 with the target frequency spectrum characteristics obtained in step 504, and identify whether the collected audio signal matches the target audio signal based on the comparison result. If so, as an example, the MCU sends a power-on instruction to the battery management module, so that the battery management module supplies power to the audio and video module, thereby waking up the audio and video module. As another example, the MCU enables the power supply circuit of the audio and video module, for example, controlling the power enable pin of the SOC chip.

[0116] As an example, the ratio between each first ratio in the spectral feature and the second ratio in the target spectral feature corresponding to the ratio is calculated to obtain m-1 third ratios. If the calculated third ratios are within a set range, it is judged to be true; otherwise, it is judged to be false. The number of true judgments is counted. If the number is greater than a set threshold, it is judged that the collected audio signal matches the target audio signal; otherwise, it is judged that the collected audio signal does not match the target audio signal.

[0117] It should be understood that the MCU obtaining the pre-collected and stored target audio signal can also be transformed into the MCU obtaining the spectral characteristics of the pre-stored target audio signal, that is, after the MCU collects the target audio signal in advance, it performs Fourier transform on the collected target audio signal, and then obtains the target spectral characteristics based on the target amplitude-frequency signal after Fourier transform, and stores the obtained target spectral characteristics. Accordingly, step 503 obtains the spectral characteristics of the collected audio signal in the same manner as the target spectral characteristics. In this way, since the target spectral characteristics do not need to be calculated in real time, it is beneficial to improve the timeliness of wake-up.

[0118] As an example, the target audio signal can be recorded by controlling the camera through the camera client, and the recorded target audio signal is uploaded to the server side. For example, the camera is controlled to record through the feature value acquisition function control of the camera client, and the recorded target audio signal is uploaded to the cloud; the audio and video module in the camera obtains the recorded target audio signal from the server through the network interface, and sends the obtained target audio signal to the MCU to calculate the target spectrum characteristics of the target audio signal.

[0119] As an example, in order to facilitate the combination with the first activation event detected by the existing intrusion detector, further, when the collected audio signal is identified to match the target audio signal, a second activation event is generated, and the MCU performs logical calculations based on the first activation event and the second activation event. For example, the MCU sets the first activation event to 1 and the second activation event to 1, performs an OR operation on the two logical values, and controls the sending of the power-on command according to the result of the OR operation.

[0120] The embodiment of the present application realizes double filtering of the audio signal through trigger level and spectrum feature comparison, which can effectively reduce false triggering of audio, effectively reduce system power consumption, and extend the single use time of the battery camera.

[0121] See also Figure 6 As shown, Figure 6 This is a schematic diagram of a wake-up device according to an embodiment of the present application. The device includes:

[0122] The first processor is used to obtain a current audio signal, obtain a current frequency spectrum feature of the current audio signal based on the current audio signal, obtain a target frequency spectrum feature of a pre-collected target audio signal, match the current frequency spectrum feature with the target frequency spectrum feature, and wake up the second processor in a dormant state if the match is successful.

[0123] The first processor is continuously powered.

[0124] The device also includes,

[0125] An audio acquisition circuit for collecting and amplifying audio signals. The audio signal amplified by the circuit is input to the first processor. The audio acquisition circuit is continuously powered.

[0126] The battery management module is used to continuously supply power to the first processor and receive a power-on instruction sent by the first processor to supply power to the second processor, so that the second processor is powered and converted from a dormant state to a working state.

[0127] The memory is used to store the pre-collected target audio signal and / or target spectrum characteristics.

[0128] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0129] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0130] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the wake-up method of the embodiment of the present application are implemented.

[0131] As for the apparatus / network-side device / storage medium embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0132] The following is a test comparison of the wake-up method of the present invention and the PIR and radar wake-up methods.

[0133] The PIR, radar, and MIC of the electronic device are facing forward at 0°, and the movement of people and the sound source are tested at various angles at a distance of 4 meters and 5 meters from the electronic device. Figure 7 .

[0134] The wake-up results of PIR at 4 meters and 5 meters are shown in Table 1. × represents no triggering, and √ represents triggering. It can be seen that the wake-up angle range of the PIR used in the test at 4 meters is 120°~150°, and the wake-up angle range of 5 meters is 0°~40°.

[0135] Table 1: PIR wake-up range

[0136]

[0137]

[0138] The wake-up results of the radar at 4 meters and 5 meters are shown in Table 2. × represents no triggering, and √ represents triggering. It can be seen that the wake-up angle range of the radar used in the test is 40°~80° at 4 meters, and the wake-up angle range of 5 meters is 40°~80°.

[0139] Table 2: Radar wake-up angle range

[0140] Distance\Angle -75° -60° -40° -20° 0° 20° 40° 60° 75° 4 meters X X X √ √ √ X X X 5 meters X X X √ √ √ X X X

[0141] The wake-up results of a small audio signal (as an example, the sound intensity is 10db SPL at 1 meter away from the sound source) at 4 meters and 5 meters are shown in Table 3. × represents no triggering, and √ represents triggering. It can be seen that the wake-up angle of the audio discharge circuit used in the test is 0° at 4 meters and 0° at 5 meters.

[0142] Table 3: Small audio signal wake-up angle range

[0143] Distance\Angle -180° -75° -60° -40° -20° 0° 20° 40° 60° 75° 180° 4 meters X X X X X X X X X X X 5 meters X X X X X X X X X X X

[0144] A large audio signal (as an example, the sound intensity is 104db SPL at 1 meter away from the sound source) but the spectrum characteristics are unqualified. The wake-up results at 4 meters and 5 meters are shown in Table 4. × represents no triggering and √ represents triggering. It can be seen that the audio discharge circuit used in the test has a wake-up angle of 0° at 4 meters and a wake-up angle of 0° at 5 meters.

[0145] Table 4: Unqualified wake-up angle range for large audio signal spectrum characteristics

[0146] Distance\Angle -180° -75° -60° -40° -20° 0° 20° 40° 60° 75° 180° 4 meters X X X X X X X X X X X 5 meters X X X X X X X X X X X

[0147] For a large audio signal (as an example, the sound intensity is 104db SPL at 1 meter away from the sound source) with qualified spectrum characteristics, the triggering results at 4 meters and 5 meters are shown in Table 5, where × represents no triggering and √ represents triggering. It can be seen that the audio discharge circuit used in the test has a wake-up angle of 360° at 4 meters and a wake-up angle of 360° at 5 meters.

[0148] Table 5: Qualified trigger angle range for large audio signal spectrum characteristics

[0149] Distance\Angle -180° -75° -60° -40° -20° 0° 20° 40° 60° 75° 180° 4 meters √ √ √ √ √ √ √ √ √ √ √ 5 meters √ √ √ √ √ √ √ √ √ √ √

[0150] The voltage trigger of the audio signal is derived from the following formula.

[0151] The attenuation of sound intensity in the open air of an ideal room comes from the following two factors:

[0152] 1. Spherical waves

[0153] Under the condition of free sound field, the sound wave of point sound source follows the law of spherical divergence. Taking the sound power level as the evaluation quantity of point sound source, the expression of spherical sound wave attenuation with distance is:

[0154] Lp = Lw-20lgr-k;

[0155] Where k is the correction factor, k = 11 in free field, k = 8 in semi-free field. Lp represents the sound pressure level, Lw represents the sound power, and r represents the distance.

[0156] The difference in sound pressure level between distances r1 and r2 is:

[0157] Lp1-Lp2=20Lg(r2 / r1),

[0158] Therefore, the sound attenuation of a sound source at a distance of 5 meters reaching the MIC is 20lg5, that is, 13.979db; the sound attenuation of a sound source at a distance of 4 meters reaching the MIC is 20lg4, that is, 12.04db.

[0159] 2. Sound absorption in the atmosphere

[0160] Sound absorption in the atmosphere is related to temperature, humidity, pressure and frequency. There is an approximate simplified relationship as follows:

[0161] When H【1.8t+32】≥4000, the sound absorption equation is:

[0162] ai0=fi / 500

[0163] Unit: db / 305m

[0164] When H[1.8t+32]≤4000, the sound absorption equation is: ai=(fi / 750)*[5.50-H(1.8t+3.2) / 1000]

[0165] Where: ai is the sound attenuation in the ith 1 / 3 octave band when the product of temperature and humidity is equal to 4000 (dB / 305m); fi is the center frequency of the ith 1 / 3 octave band (Hz); H is the relative humidity (%); t is the temperature (℃). The simplified relationship is shown in Figure 8 As shown, it can be seen that the sound absorption in the atmosphere is less than 3db / 305m. The distances used in the test are 4 meters and 5 meters, and the attenuation is small and can be ignored.

[0166] Taking the above two factors into consideration,

[0167] The sound attenuation of a sound source at a distance of 5 meters reaching the MIC is 20lg5, or 13.979db;

[0168] The sound attenuation of a sound source at a distance of 4 meters reaching the MIC is 20lg4, or 12.04db.

[0169] The sound intensity of a small audio signal of 10db SPL and a sound source at a distance of 4 meters reaching the MIC is: 10db-12.4db=-2.04dB SPL

[0170] The sound intensity of a small audio signal of 10db SPL and a sound source at a distance of 5 meters reaching the MIC is: 10db-13.979db=-3.979dB SPL

[0171] The maximum audio signal is 104db SPL, and the sound intensity of the sound source reaching the MIC at a distance of 4 meters is: 104db-12.04db=91.96dB SPL

[0172] The maximum audio signal is 104db SPL, and the sound intensity of the sound source reaching the MIC at a distance of 5 meters is: 104db-13.979db=90.02dB SPL

[0173] As an example, the MIC specifications used in the audio circuit are: sensitivity -38dB, reference sound pressure Pr = 2*10 -5 Pa; the sensitivity of MIC is defined as: the ratio of output voltage to input sound pressure under unit sound pressure excitation, that is, when a sound pressure of 1Pa (94dB SPL sound pressure level) is applied to the MIC, the voltage (dBV) output by the microphone is:

[0174]

[0175] Thus, the sensitivity of the MIC can be obtained:

[0176] x=10 -38 / 20 =12.59mV(rms) / Pa=12.59*2*1.414mV=35.60mV(peak-to-peak) / Pa.

[0177] The sound pressure level is represented by the symbol SPL, which is defined as the ratio of the effective value of the sound pressure to be measured P(e) to the reference sound pressure P(ref). Then the sound pressure P = 10 SPL / 20 *Pr.

[0178] Small audio signal 10db SPL, sound source sound pressure P=10 collected by the sound source MIC at a distance of 4 meters -2.04 / 20 *2*10 -5 =1.58*10 -5 Pa, MIC output peak-to-peak voltage = 1.58*10 -5 *35.6mV<1mV, MCU will not be triggered.

[0179] Small audio signal 10db SPL, sound source sound pressure P=10 collected by the sound source MIC at a distance of 5 meters -3.979 / 20 *2*10 -5 =1.26*10 -5 Pa, MIC output peak-to-peak voltage = 1.26*10 -5 *35.6mV<1mV, MCU will not be triggered.

[0180] The maximum audio signal is 104db SPL, and the sound pressure P collected by the sound source MIC at a distance of 4 meters is 10 91.96 / 20 *2*10-5 =0.79Pa, the peak-to-peak voltage output by MIC = 0.79*35.6mV = 28.124mV>10mV, and the MCU will be triggered.

[0181] The maximum audio signal is 104db SPL, and the sound pressure P collected by the sound source MIC at a distance of 5 meters is 10 90.02 / 20 *2*10 -5 =0.63Pa, the peak-to-peak voltage output by MIC = 0.63*35.6mV = 22.482mV>10mV, and the MCU will be triggered.

[0182] From the comparison of the above results, we can see that:

[0183] A. The triggering angles of PIR and radar cannot reach 360°, that is, there are certain blind spots. The present invention can effectively solve the defects of the blind spots.

[0184] B. In the present invention, when the audio signal is large enough and the spectrum characteristics are qualified, the MCU can be effectively triggered, realizing double filtering of the audio signal, which can effectively reduce false triggering of the audio.

[0185] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0186] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A wake-up method, characterized in that: The method includes: On the first processor side which is continuously powered, Get the current audio signal. Based on the current audio signal, obtain the current spectrum characteristics of the current audio signal, Obtain the target spectrum characteristics of the pre-collected target audio signal, The current spectrum feature is matched with the target spectrum feature, and if the match is successful, the second processor in the dormant state is awakened.

2. The wake-up method according to claim 1, characterized in that: The first processor is externally connected to an audio acquisition circuit for collecting and amplifying audio signals, and the audio acquisition circuit is continuously powered. The obtaining of the current audio signal comprises: The first processor detects whether the input level from the audio acquisition circuit meets the trigger condition. If yes, the first processor switches the pin connected to the audio acquisition circuit to an automatic data acquisition mode to collect the audio signal from the audio acquisition circuit; The obtaining, based on the current audio signal, current frequency spectrum features of the current audio signal includes: Perform Fourier transform on the collected audio signal to obtain the amplitude-frequency signal after Fourier transform. Based on the amplitude-frequency signal after Fourier transformation, the frequency spectrum characteristics of the collected audio signal are obtained; The step of waking up the second processor in a dormant state comprises: The first processor sends a power-on instruction for supplying power to the second processor to the battery management module, so that the second processor is powered and converted from a sleep state to an operating state.

3. The wake-up method according to claim 2, characterized in that: The first processor is a microprocessor chip, and the second processor is a system-on-chip. The method of obtaining the frequency spectrum characteristics of the collected audio signal based on the amplitude-frequency signal after Fourier transformation includes: Within the set frequency range, extract the signal amplitudes of m first frequency points from the amplitude-frequency signal after Fourier transformation. Calculate the ratio of signal amplitudes between adjacent first frequency point signals to obtain m-1 first ratio results; The step of obtaining a target frequency spectrum feature of a pre-collected target audio signal comprises: Get the target audio signal that has been collected and stored in advance, Perform Fourier transform on the target audio signal to obtain the target amplitude-frequency signal after Fourier transform. Within the set frequency range, extract the signal amplitudes of m second frequency points from the target amplitude-frequency signal after Fourier transformation, Calculate the ratio of the signal amplitudes between adjacent second frequency point signals to obtain m-1 second ratio results. The extracted first frequency point is the same as the extracted second frequency point, and m is a natural number not less than 2.

4. The wake-up method according to claim 2, characterized in that: The target spectrum characteristics are pre-stored, The target spectrum characteristics are determined as follows: Perform Fourier transform on the pre-collected target audio signal to obtain the target amplitude-frequency signal after Fourier transform. Within the set frequency range, extract the signal amplitudes of m second frequency points from the target amplitude-frequency signal after Fourier transformation. Calculate the ratio of the signal amplitudes between adjacent second frequency point signals to obtain m-1 second ratio results; The method of obtaining the frequency spectrum characteristics of the collected audio signal based on the amplitude-frequency signal after Fourier transformation includes: Within the set frequency range, extract the signal amplitudes of m first frequency points from the amplitude-frequency signal after Fourier transformation. Calculate the ratio of the signal amplitudes between adjacent first frequency point signals to obtain m-1 first ratio results. The extracted first frequency point is the same as the extracted second frequency point, and m is a natural number not less than 2.

5. The wake-up method according to claim 3 or 4, characterized in that: The matching of the current spectrum feature with the target spectrum feature includes: Calculate the ratio between each first ratio result and its corresponding second ratio result to obtain m-1 third ratio results, wherein the first ratio result and its corresponding second ratio result correspond to the same frequency point, Count the number of third ratio results within the set range, When the counted number is greater than the set threshold, it is determined that the current spectrum feature matches the target spectrum feature successfully.

6. The wake-up method according to claim 2, characterized in that: The audio acquisition circuit includes a pickup for picking up an audio signal, and an amplifier circuit for amplifying the audio signal from the pickup, wherein the amplifier circuit is configured such that: when the voltage amplitude of the audio signal from the pickup is higher than a set voltage threshold, the level applied to the pin reaches the trigger condition; when the voltage amplitude of the audio signal from the pickup is not higher than the set voltage threshold, the level applied to the pin does not reach the trigger condition, and the trigger condition is a set trigger level threshold; The target audio signal is input to the first processor through the audio acquisition circuit and stored, or is collected by an application program and then sent to the first processor.

7. A wake-up device, characterized in that: The device includes: The first processor is used to obtain a current audio signal, obtain a current frequency spectrum feature of the current audio signal based on the current audio signal, obtain a target frequency spectrum feature of a pre-collected target audio signal, match the current frequency spectrum feature with the target frequency spectrum feature, and wake up the second processor in a dormant state if the match is successful. The first processor is continuously powered.

8. The wake-up device according to claim 7, characterized in that: The first processor is a microprocessor chip, and the second processor is a system-on-chip. The device also includes, An audio acquisition circuit for collecting and amplifying audio signals. The audio signal amplified by the circuit is input to the first processor. The audio acquisition circuit is continuously powered. The battery management module is used to continuously supply power to the first processor and receive a power-on instruction sent by the first processor to supply power to the second processor, so that the second processor is powered and converted from a sleep state to a working state.

9. An electronic device, characterized in that: The device comprises the wake-up device as claimed in any one of claims 7 to 8.

10. The electronic device according to claim 9, characterized in that: The electronic device is a camera.

Images