Self-adaptive multi-source level awakening method and system based on Hisilicon platform USB and medium

Abstract

The invention provides a self-adaptive multi-source level awakening method and system based on a Hisilicon platform USB and a medium, and the method comprises the steps: synchronously capturing first level information, first square wave information and first voltage information of a USB Hub, and generating a time-aligned trigger signal matrix through delay compensation; feature signals are extracted based on the trigger signal matrix, and a wake-up score is generated in combination with a weighted fusion algorithm; dynamically calculating an adaptive wake-up threshold based on a real-time noise baseline and a historical false wake-up record; when only the first slope information exceeds the limit, first-level awakening is judged, and the awakening score is up-regulated; when the first slope information and the first frequency information both exceed the limit, secondary awakening is judged, and an awakening action is triggered; when the wake-up score exceeds a wake-up threshold value, three-level wake-up is judged, a wake-up action is triggered, and wake-up diagnosis is started; noise interference is reduced through multi-source signal cross validation, a dynamic threshold algorithm is adopted to adapt to a complex industrial environment, and the success rate and reliability of awakening are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of level wake-up, more particularly, to a self-adaptive multi-source level wake-up method, system and medium based on HiSilicon platform USB. BACKGROUND

[0002] In the field of industrial control and embedded systems, HiSilicon platform is widely used due to its high performance and low power consumption. However, in the USB2.0 wake-up circuit of this platform, there is a significant technical bottleneck. The CPU does not support native USB 2.0 device wake-up function in S3 (sleep to memory) and S4 (suspend to hard disk) deep sleep states, including keyboard or mouse, which requires an external USB_Hub to indirectly achieve the purpose of wake-up. The traditional solution relies on single signal detection, which has serious defects in complex electromagnetic environments. Single PGANG pin level or DRV pin square wave detection is easily disturbed by environmental noise, leading to high false wake-up rate. In addition, in industrial scenarios, voltage fluctuations, power frequency interference, etc. cause signal distortion, and single threshold detection has failure phenomenon.

[0003] Therefore, there is an urgent need for a self-adaptive multi-source composite level wake-up technology based on HiSilicon platform USB.

[0004] Wake-up: refers to the process of waking up a computer system from a sleep state through specific devices or events.

[0005] ‌S3 (sleep to memory): In this mode, all devices except memory stop working, but the data in memory is saved to the hard disk. The system can quickly recover from S3 state because the data in memory is not cleared, and only needs to load the saved data.

[0006] ‌S4 (suspend to hard disk): In this mode, the data in memory is saved to the hard disk, and all devices stop working. When the system recovers from S4 state, it needs to read data from the hard disk to the memory, so the recovery time is longer.

[0007] EC: Full name is Embedded Controller, embedded controller. It is actually a single-chip microcomputer commonly used in laptops for keyboard control, touchpad, power management, fan control, notebook battery management, etc. SUMMARY

[0008] In view of the above problems, the purpose of the present application is to provide a HiSilicon platform USB-based adaptive multi-source level wake-up method, system and medium, which synchronously analyzes the PGANG level, DRV square wave and VSENSE voltage stroke multi-source composite level; extracts a characteristic signal and generates a wake-up score by combining a preset weighting algorithm; dynamically adjusts a threshold based on real-time noise and historical false positive rate; triggers a three-level wake-up mechanism according to the composite signal and the wake-up score; and reduces noise interference through multi-source signal cross-validation, adapts to complex industrial environments by using a dynamic threshold algorithm, and improves the success rate and reliability of wake-up.

[0009] The first aspect of the present application provides a HiSilicon platform USB-based adaptive multi-source level wake-up method, which comprises: Parallelly collecting a trigger signal, including first level information, first square wave information and first voltage information; Based on a preset delay compensation, a time-aligned trigger signal matrix is generated according to the trigger signal; According to the trigger signal matrix, a characteristic signal is extracted, including first slope information, first frequency information and first energy information; Based on a preset weighting fusion algorithm, a wake-up score is obtained according to the characteristic signal; According to the first voltage information, a noise baseline is determined, and a wake-up threshold is obtained by combining historical wake-up records; It is judged whether the first slope information exceeds a preset first slope threshold; If yes, it is judged whether the first frequency information is within a preset first frequency range; If no, it is determined as a first-level wake-up, and the wake-up score is adjusted upward; If yes, it is determined as a second-level wake-up, and a wake-up action is triggered; If the wake-up score exceeds the wake-up threshold, it is determined as a third-level wake-up, and a wake-up action and a wake-up diagnosis are triggered.

[0010] In the present scheme, based on the preset delay compensation, the time-aligned trigger signal matrix is generated according to the trigger signal, specifically: Based on a preset sampling circuit or sampling channel, first delay information and second delay information are determined; The first delay information is applied to the first level information for forward time shift compensation; The second delay information is applied to the first square wave information for forward time shift compensation; The compensated first level information and first square wave information are combined with the first voltage information to integrate into a trigger signal matrix according to the time axis, wherein the row vectors of the trigger signal matrix correspond to first level sequence, first square wave sequence and first voltage sequence respectively.

[0011] In the scheme, the feature signal is extracted according to the trigger signal matrix, specifically: The first voltage sequence is extracted based on a preset sliding window, and a slope value is extracted as first slope information according to a first-order differential operation; The first square wave sequence is extracted based on a preset time window, and first frequency information is obtained according to zero-crossing detection; The root mean square calculation is performed on the first voltage sequence, and the root mean square result is subjected to preset normalization processing to obtain first energy information.

[0012] In the scheme, the noise baseline is determined according to the first voltage information, and the wake-up threshold is obtained in combination with the historical wake-up record, specifically: The first voltage information is recorded when there is no wake-up event to obtain a second voltage sequence; The root mean square value of the difference value is calculated according to the difference value between the voltage value of the second voltage sequence and the standard voltage value, and the noise baseline is determined; Based on a preset statistical window, a historical false wake-up rate is generated according to the proportion of false wake-up events in total wake-up events in the historical wake-up record; Based on a pre-trained wake-up threshold association model, the wake-up threshold is obtained according to the noise baseline and the historical false wake-up rate.

[0013] In the scheme, the wake-up action specifically includes: Output a high-level pulse signal through a preset GPIO interface; Control the power management unit to switch the supply voltage from the S3 / S4 maintenance voltage to the rated voltage; According to the characteristic signal and the wake-up level when triggering the wake-up, an wake-up log is generated and stored.

[0014] In the scheme, it also includes: If the wake-up action is triggered, it is marked as a first device wake-up event by default; If the first slope information is greater than a preset second slope threshold, and the first frequency information is within a preset second frequency range, it is determined and recorded as a second device wake-up event; If the first slope information is greater than a preset second slope threshold, and the first energy information exceeds a preset first energy threshold, it is determined and recorded as a third device wake-up event.

[0015] The second aspect of the present application provides a self-adaptive multi-source level wake-up system based on HiSilicon platform USB, which includes a self-adaptive multi-source level wake-up method program based on HiSilicon platform USB, and the self-adaptive multi-source level wake-up method program based on HiSilicon platform USB is executed by the processor to realize the following steps: The trigger signal includes first level information, first square wave information and first voltage information; Based on the preset delay compensation, a time-aligned trigger signal matrix is generated according to the trigger signal; According to the trigger signal matrix, a feature signal is extracted, including first slope information, first frequency information and first energy information; Based on the preset weighted fusion algorithm, a wake-up score is obtained according to the feature signal; According to the first voltage information, a noise baseline is determined, and a wake-up threshold is obtained in combination with a historical wake-up record; Determine whether the first slope information exceeds a preset first slope threshold; If yes, determine whether the first frequency information is within a preset first frequency range; If no, determine that it is a first-level wake-up, and increase the wake-up score; If yes, determine that it is a second-level wake-up, and trigger a wake-up action; If the wake-up score exceeds the wake-up threshold, determine that it is a third-level wake-up, and trigger a wake-up action and start a wake-up diagnosis.

[0016] In the scheme, based on the preset delay compensation, a time-aligned trigger signal matrix is generated according to the trigger signal, specifically: Based on the preset sampling circuit or sampling channel, first delay information and second delay information are determined; The first delay information is applied to the first level information for forward time shift compensation; The second delay information is applied to the first square wave information for forward time shift compensation; The compensated first level information and first square wave information are combined with the first voltage information to integrate into a trigger signal matrix along the time axis, wherein the row vectors of the trigger signal matrix correspond to first level sequence, first square wave sequence and first voltage sequence respectively.

[0017] In the scheme, the feature signal is extracted according to the trigger signal matrix, specifically: The first level sequence is extracted based on the preset sliding window, and the slope value is extracted as the first slope information according to the first-order differential operation; The first square wave sequence is extracted based on the preset time window, and the first frequency information is obtained according to the zero-crossing point detection; The root mean square calculation is performed on the first voltage sequence, and the root mean square result is subjected to preset normalization processing to obtain the first energy information.

[0018] The third aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium comprises a HiSilicon platform USB based adaptive multi-source level wake-up method program, and when the HiSilicon platform USB based adaptive multi-source level wake-up method program is executed by a processor, the steps of the HiSilicon platform USB based adaptive multi-source level wake-up method according to any one of the preceding aspects are implemented.

[0019] The present application provides a HiSilicon platform USB based adaptive multi-source level wake-up method, system and medium, synchronously capturing first level information, first square wave information and first voltage information of a USB Hub, generating a time-aligned trigger signal matrix through delay compensation; extracting a characteristic signal based on the trigger signal matrix, and generating a wake-up score by combining a weighted fusion algorithm; dynamically calculating an adaptive wake-up threshold based on a real-time noise baseline and historical false wake-up records; when only the first slope information is out of limit, determining that it is a first-level wake-up, and increasing the wake-up score; when both the first slope information and the first frequency information are out of limit, determining that it is a second-level wake-up, and triggering a wake-up action; when the wake-up score exceeds the wake-up threshold, determining that it is a third-level wake-up, and triggering a wake-up action and starting a wake-up diagnosis; reducing noise interference through multi-source signal cross-validation, adapting to complex industrial environments by using a dynamic threshold algorithm, and improving the success rate and reliability of wake-up. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and should not be regarded as a limitation on the scope.

[0021] Figure 1 A flowchart of a HiSilicon platform USB based adaptive multi-source level wake-up method according to an embodiment of the present application is shown; Figure 2 A flowchart of generating a trigger signal matrix according to an embodiment of the present application is shown; Figure 3 A flowchart of extracting a characteristic signal according to an embodiment of the present application is shown; Figure 4 A block diagram of a HiSilicon platform USB based adaptive multi-source level wake-up system according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0022] The technical solutions of the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0023] Unless otherwise defined, all terms (including technical and scientific terms) used in the present application embodiments have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined in the present application embodiments.

[0024] The terms "first", "second", and similar terms used in the present application embodiments do not necessarily imply any order, quantity, or importance, but are used to distinguish different constituent parts. The terms "one", "a", or "the" and similar terms do not denote a limitation of quantity, but indicate the presence of at least one. Similarly, the terms "include", "comprise", and similar terms mean that the elements or objects before the term encompass the elements or objects listed after the term and equivalents thereof, without excluding other elements or objects. The terms "connected" or "coupled" and similar terms do not necessarily mean a physical or mechanical connection, but can include an electrical connection, whether direct or indirect. The steps of the method of the present application embodiments do not necessarily have to be performed in the order shown. On the contrary, the steps can be processed in reverse order or simultaneously. Other operations can also be added to these processes, or one or more steps can be removed from these processes.

[0025] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0026] Figure 1 A flowchart of a self-adaptive multi-source level wake-up method based on a HiSilicon platform USB is shown.

[0027] As shown in Figure 1 The first aspect of the present application discloses a self-adaptive multi-source level wake-up method based on a HiSilicon platform USB, which comprises: Parallelly collecting a trigger signal, including first level information, first square wave information, and first voltage information; Based on a preset delay compensation, generating a time-aligned trigger signal matrix according to the trigger signal; According to the trigger signal matrix, extracting a feature signal, including first slope information, first frequency information, and first energy information; Based on a preset weighted fusion algorithm, obtaining a wake-up score according to the feature signal; According to the first voltage information, determining a noise baseline, and combining a historical wake-up record to obtain a wake-up threshold; determining whether the first slope information exceeds a preset first slope threshold value; If yes, determining whether the first frequency information is within a preset first frequency range; If no, determining as a first level wake-up, and increasing the wake-up score; If yes, determining as a second level wake-up, and triggering a wake-up action; If the wake-up score exceeds the wake-up threshold value, determining as a third level wake-up, triggering a wake-up action and starting a wake-up diagnosis.

[0028] It should be noted that the USB Hub includes a PGANG pin, a DRV pin and a VSENSE pin; the first level information is a level value of the PGANG pin; the first square wave information is a square wave value of the DRV pin; the first voltage information is a voltage value of the VSENSE pin. The first slope information is a change slope of the PGANG level; the first frequency information is a frequency of the DRV pin square wave; and the first energy information is a fluctuation energy reflected by the VSENSE pin voltage.

[0029] In the embodiment, when the system enters the S3 / S4 hibernation state, the processor starts three-channel parallel sampling, the PGANG pin level signal is input to the ADC channel after being stabilized by the 100kΩ pull-down resistor, the DRV pin square wave signal is captured by the timer, and the VSENSE pin voltage is sampled after being amplified by the double-stage operational amplifier. Since the three signals pass through different hardware circuits and detection structures, delay compensation is performed based on the corresponding delay conditions. As an implementation, the PGANG level signal is delayed by 0.2ms, and the DRV square wave signal is delayed by 1.5ms. Subsequently, based on the sampling data after delay compensation, a two-dimensional matrix with strictly aligned time axes is constructed in the memory. In the two-dimensional matrix, the PGANG level sequence, the DRV square wave sequence and the VSENSE voltage sequence are included. Secondly, based on the preset feature extraction logic, the two-dimensional matrix is analyzed in real time. The differential derivative or differential value of the PGANG level sequence is calculated as the slope feature to obtain the PGANG slope; the number of zero-crossing points of the DRV square wave sequence in the set time window is counted to derive the frequency to obtain the DRV frequency; and after the root mean square operation is performed on the VSENSE sequence, normalization processing is performed to obtain the VSENSE energy value. Then, based on the preset weighted fusion algorithm, the PGANG slope, the DRV frequency and the VSENSE energy value corresponding to the weighted weight are calculated to obtain the wake-up score. According to the dynamic decision mechanism, the adaptive threshold is generated by fusing the environmental noise baseline and the historical false wake-up rate. Based on the preset hierarchical wake-up triggering mechanism, the wake-up level is determined. Only when the PGANG slope exceeds the preset first slope threshold, the first-level wake-up is started, and the wake-up score is increased. When the PGANG slope exceeds the preset first slope threshold and the DRV frequency is in the set first frequency range, the second-level wake-up is started, the EC controller is woken up, and the supply voltage is switched. When the above two conditions are not met, but the wake-up score exceeds the wake-up threshold, the third-level wake-up is activated, the wake-up action is performed, and the wake-up parameters are diagnosed. The embodiment reduces the false wake-up problem by using multiple source composite signals, reduces the false wake-up rate based on the three-level wake-up decision, and improves the reliability of the wake-up operation.

[0030] Figure 2 A flowchart for generating a trigger signal matrix is shown.

[0031] According to the embodiment of the present application, as Figure 2 shown, the time-aligned trigger signal matrix is generated based on the preset delay compensation according to the trigger signal, specifically: The first delay information and the second delay information are determined based on the preset sampling circuit or sampling channel; The first level information is subjected to forward time shift compensation of the first delay information; The first square wave information is subjected to forward time shift compensation of the second delay information; The compensated first level information and the first square wave information are combined with the first voltage information and integrated into a trigger signal matrix according to a time axis, wherein row vectors of the trigger signal matrix correspond to the first level sequence, the first square wave sequence and the first voltage sequence respectively.

[0032] It should be noted that the first delay information is the delay caused by the PGANG level sampling circuit and logic; and the second delay information is the delay caused by the DRV square wave sampling circuit and logic. The embodiment provides a generation process of a trigger signal matrix. In order to eliminate the difference in signal transmission delay, the system allocates independent ADC channels for the PGANG level and the VSENSE voltage, and allocates a timer for the DRV square wave. As an implementation, after the PGANG signal is input, a 0.2 ms time shift compensation is applied to offset the internal sampling circuit delay; the DRV signal is applied with 1.5 ms compensation to solve the square wave generation circuit response lag; and the VSENSE signal is processed directly. The time stamps of the three compensated data are aligned to construct a trigger signal matrix with fixed row vector dimension, wherein the first row is the PGANG level sequence, the second row is the DRV square wave sequence, and the third row is the VSENSE voltage sequence. Based on the delay compensation, the embodiment ensures the timing consistency of the wake-up event analysis and avoids the feature extraction distortion caused by the asynchronous signals.

[0033] Figure 3 A flowchart of extracting a feature signal is shown.

[0034] According to the embodiment of the present application, as shown in Figure 3 According to the trigger signal matrix, the feature signal is extracted, specifically: The first level sequence is extracted based on a preset sliding window, and a slope value is extracted as first slope information according to a first-order differential operation; The first square wave sequence is extracted based on a preset time window, and first frequency information is obtained according to zero-crossing detection; The first voltage sequence is subjected to a root mean square calculation, and a root mean square result is subjected to a preset normalization processing to obtain first energy information.

[0035] It should be noted that the embodiment provides a feature extraction process of a wake-up signal. As an implementation manner, each sequence is separated from a trigger signal matrix. A PGANG level sequence is based on preset first-order differentiation, derivative or difference operation, and an instantaneous slope is calculated through an adjustable sliding window, which is used to lock the maximum positive slope value as a wake-up sensitive index. A DRV square wave sequence is counted through a zero-crossing detection mechanism, and the number of rising and falling edge transitions in a preset time window is counted. The DRV square wave frequency feature is calculated according to the transition number and the time window length. The VSENSE voltage sequence first performs a square operation to eliminate negative values, then calculates the average value in the window, and finally obtains the root mean square energy value by taking the square root. The root mean square energy value is normalized to map the root mean square energy value to the range [0, 1]. The processing result is packaged as a feature vector for the decision layer to call. The embodiment accurately captures the PGANG level transition through differential operation, enhances the anti-amplitude interference of the DRV square wave through zero-crossing detection, and eliminates individual differences of devices by using energy standardization, thereby improving the accuracy of subsequent analysis and decision.

[0036] According to the embodiment of the present application, the noise baseline is determined according to the first voltage information, and the wake-up threshold is obtained in combination with the historical wake-up record, specifically: The first voltage information is recorded when there is no wake-up event to obtain a second voltage sequence; According to the difference between the voltage value of the second voltage sequence and the standard voltage value, the root mean square value of the difference is calculated to determine the noise baseline; Based on a preset statistical window, the historical false wake-up rate is generated according to the proportion of false wake-up events in total wake-up events in the historical wake-up record; The wake-up threshold is obtained based on a pre-trained wake-up threshold association model and the noise baseline and the historical false wake-up rate.

[0037] It should be noted that the embodiment provides a dynamic threshold self-adaptive setting mechanism. The noise baseline is calculated automatically during the system idle period, that is, the VSENSE voltage is recorded when there is no wake-up event, and the record is recorded as a second voltage sequence. The deviation value of the second voltage sequence and the standard voltage value is calculated, and the root mean square value of the deviation value is calculated as the environmental noise baseline. As an implementation manner, 24 hours are taken as a statistical window, and all wake-up records are retrieved. The percentage of records marked as false wake-up events in the total number of wake-up times is calculated to obtain a historical false wake-up rate. The environmental noise baseline and the historical false wake-up rate are input into a pre-trained model, and the superposition effect of the two is adjusted through a negative feedback coefficient to dynamically generate a wake-up threshold. As an implementation manner, the wake-up threshold association model is a neural network model; as another implementation manner, the wake-up threshold association model is a mapping relationship. The embodiment adopts a dynamic adaptive working voltage fluctuation environment to cooperate with the threshold association model, thereby improving the accuracy of wake-up judgment.

[0038] According to the embodiment of the present application, the wake-up action specifically includes: outputting a high-level pulse signal through a preset GPIO interface; controlling the power management unit to switch the supply voltage from the S3 / S4 maintenance voltage to the rated voltage; generating and storing a wake-up log according to the characteristic signal and the wake-up level when the wake-up is triggered.

[0039] It should be noted that, as an implementation, the wake-up action specifically includes pulse output, voltage switching, and log recording. When determining the effective wake-up, the GPIO_CTRL pin outputs a high-level pulse with a set duration. As an implementation, the pulse is transmitted to the power management unit after photoelectric isolation. Based on the pulse signal, the S3 / S4 energy-saving power supply circuit is first cut off, and then the 3.3V rated voltage converter is activated, and the CPU core power supply switching is completed within the safe switching time. At the same time, the log engine records the wake-up event characteristics, including the timestamp, stores the characteristic information such as the PGANG slope peak value, the DRV dominant frequency, and the VSENSE energy surge value, and the wake-up level code. The present embodiment uses log recording for traceability for later analysis.

[0040] According to the embodiment of the present application, it further includes: if the wake-up action is triggered, it is marked as a first device wake-up event by default; if the first slope information is greater than a preset second slope threshold value, and the first frequency information is within a preset second frequency range, it is determined and recorded as a second device wake-up event; if the first slope information is greater than a preset second slope threshold value, and the first energy information exceeds a preset first energy threshold value, it is determined and recorded as a third device wake-up event.

[0041] It should be noted that, the first device wake-up event represents an event triggered by an unknown device; the second device wake-up event represents an event triggered by a mouse; and the third device wake-up event represents an event triggered by a keyboard. The present embodiment provides a wake-up device classification mechanism. When the wake-up is triggered, it is by default an unknown device event. As an implementation, if the PGANG slope is greater than 0.8V / ms and the DRV frequency is within the range of 950-1050Hz, it is marked as a mouse device; and if the slope is greater than 0.8V / ms and the VSENSE energy is greater than 50mV, it is marked as a keyboard device. After being marked as a mouse device or a keyboard device, it can be used to formulate a communication protocol selection and shorten the device activation time after the wake-up operation.

[0042] It is worth mentioning that, after determining the three-level wake-up, it further includes: obtaining a slope change rate according to the first level sequence; If the slope change rate is lower than a preset change rate threshold, the wake-up action is denied, and it is determined as an interference event; Performing fast Fourier transform on the first square wave sequence, and calculating the energy proportion of a preset frequency band according to frequency domain data statistics; If the energy proportion exceeds a preset proportion threshold, the wake-up action is denied, and it is determined as an interference event; If the first energy information continuously is lower than a preset second energy threshold, the wake-up action is denied, and it is determined as an interference event.

[0043] It should be noted that the embodiment provides a triple interference detection mechanism. In the embodiment, after three levels of wake-up triggering, the triple anti-interference detection is started. First, the PGANG slope change rate analysis is performed, and if the slope change rate is lower than the change rate threshold, it is determined as a slow-changing interference, for example, a level shift caused by temperature drift, and therefore, the current wake-up is denied. Then, the DRV square wave is subjected to 512-point fast Fourier transform (FFT), and the energy proportion in the 50Hz-60Hz frequency band is calculated. When the energy proportion exceeds the set 30% threshold, it is determined as a power frequency interference, and the current wake-up is denied. Finally, the VSENSE energy is subjected to continuous detection, and if the normalized energy value identified in the VSENSE sequence continuously is lower than 0.3, it is determined as a transient interference, and the current wake-up is denied. The wake-up process is terminated if any of the above steps is denied, and the event is recorded as a "interference denial" type. The embodiment improves the anti-interference ability through triple verification; among them, the frequency domain analysis solves the influence of power frequency interference, and the energy continuous monitoring prevents the contact from shaking and false touch.

[0044] It is worth mentioning that it also includes: Establishing a wake-up event database for recording the characteristic signal, wake-up level and wake-up device type; When it is determined as a false wake-up event, the database record of the current event is recorded to the interference mode characteristic database; According to the characteristic signal and wake-up level of the current event, the weight coefficient in the weighted fusion algorithm is adjusted; The updated weight coefficient and the interference mode characteristic library are written to the non-volatile memory.

[0045] It should be noted that the embodiment provides a self-learning mechanism for adaptively adjusting the weight coefficient in the weighted fusion algorithm. In the embodiment, the adaptive learning mechanism includes two stages of online learning and offline storage. After each wake-up event, the slope, frequency and energy value are packaged as a feature vector, a decision result and a device type into a data packet and stored in a RAM buffer area. If the event is denied by triple verification, the data packet is appended with a false wake-up label and migrated to the interference feature library; if the wake-up is successful, it is marked as a positive sample and added to the positive sample library. Based on a preset self-learning period, the occurrence frequency of similar feature patterns in the false wake-up library is counted, and the corresponding signal weight is reduced. As an implementation, if the PGANG slope is determined to be frequent false touch, the PGANG slope weight coefficient is reduced; and the positive sample library is analyzed to optimize the recognition threshold. Finally, the updated weight coefficient and the interference mode feature library are written into the local non-volatile memory, realizing offline storage.

[0046] It is worth mentioning that it also includes: When the ambient temperature exceeds the preset temperature threshold, the interval of the first frequency range and the second frequency range is expanded and a safety event log is recorded; When it is determined as an environmental surge according to the first energy information, the wake-up function is turned off based on a preset shutdown delay and a safety event log is recorded.

[0047] It should be noted that the embodiment also provides an anti-environmental interference mechanism. For high-temperature environmental conditions, when the temperature sensor detects that the ambient temperature exceeds 85℃, the frequency tolerance is automatically expanded to ±10% and a high-temperature compensation log is recorded, wherein the log includes temperature value, compensation time and original frequency data. As an implementation, when the temperature exceeds 85℃, the first frequency range is expanded from 1KHz±5% to 1KHz±10% for compensating for the temperature drift of the crystal oscillator. For power surge conditions, as an implementation, when the normalized energy value of the first energy information exceeds 0.8 for 100ms, it is determined as an environmental surge, and the wake-up function is automatically turned off for 200ms, and a surge event log is generated, wherein the log includes peak voltage, duration and spectral characteristics.

[0048] Figure 4 A block diagram of an adaptive multi-source level wake-up system based on a HiSilicon platform USB is shown.

[0049] As Figure 4 shown, the second aspect of the present application discloses an adaptive multi-source level wake-up system 4 based on a HiSilicon platform USB, comprising a memory 41 and a processor 42, the memory comprising an adaptive multi-source level wake-up method based on a HiSilicon platform USB program, the adaptive multi-source level wake-up method based on a HiSilicon platform USB program is executed by the processor to realize the following steps: The parallel acquisition trigger signal includes first level information, first square wave information and first voltage information; Based on the preset delay compensation, a time-aligned trigger signal matrix is generated according to the trigger signal; According to the trigger signal matrix, a feature signal is extracted, including first slope information, first frequency information and first energy information; Based on the preset weighted fusion algorithm, a wake-up score is obtained according to the feature signal; According to the first voltage information, a noise baseline is determined, and a wake-up threshold is obtained in combination with a historical wake-up record; It is judged whether the first slope information exceeds a preset first slope threshold; If yes, it is judged whether the first frequency information is within a preset first frequency range; If no, it is determined as a first-level wake-up, and the wake-up score is adjusted upward; If yes, it is determined as a second-level wake-up, and a wake-up action is triggered; If the wake-up score exceeds the wake-up threshold, it is determined as a third-level wake-up, and a wake-up action and a wake-up diagnosis are triggered.

[0050] It should be noted that the USB Hub includes a PGANG pin, a DRV pin and a VSENSE pin; the first level information is the level value of the PGANG pin; the first square wave information is the square wave value of the DRV pin; and the first voltage information is the voltage value of the VSENSE pin. The first slope information is the change slope of the PGANG level; the first frequency information is the frequency of the square wave of the DRV pin; and the first energy information is the fluctuation energy reflected by the voltage of the VSENSE pin.

[0051] In the embodiment, when the system enters the S3 / S4 hibernation state, the processor starts three-channel parallel sampling, the PGANG pin level signal is input to the ADC channel after being stabilized by the 100kΩ pull-down resistor, the DRV pin square wave signal is captured by the timer, and the VSENSE pin voltage is sampled after being amplified by the double-stage operational amplifier. Since the three signals pass through different hardware circuits and detection structures, delay compensation is performed based on the corresponding delay conditions. As an implementation, the PGANG level signal is delayed by 0.2ms, and the DRV square wave signal is delayed by 1.5ms. Subsequently, based on the sampling data after delay compensation, a two-dimensional matrix with strictly aligned time axes is constructed in the memory. In the two-dimensional matrix, the PGANG level sequence, the DRV square wave sequence and the VSENSE voltage sequence are included. Secondly, based on the preset feature extraction logic, the two-dimensional matrix is analyzed in real time. The differential derivative or differential value of the PGANG level sequence is calculated as the slope feature to obtain the PGANG slope; the number of zero-crossing points of the DRV square wave sequence in the set time window is counted to derive the frequency to obtain the DRV frequency; after the root mean square operation is performed on the VSENSE sequence, normalization processing is performed to obtain the VSENSE energy value. Then, based on the corresponding weighted weights of the PGANG slope, the DRV frequency and the VSENSE energy value in the preset weighted fusion algorithm, the wake-up score is calculated. According to the dynamic decision mechanism, the adaptive threshold is generated by fusing the environmental noise baseline and the historical false wake-up rate. Based on the preset hierarchical wake-up triggering mechanism, the wake-up level is determined. Only when the PGANG slope exceeds the preset first slope threshold, the first-level wake-up is started, and the wake-up score is increased. When the PGANG slope exceeds the preset first slope threshold and the DRV frequency is in the set first frequency range, the second-level wake-up is started, the EC controller is woken up, and the supply voltage is switched. When the above two conditions are not met, but the wake-up score exceeds the wake-up threshold, the third-level wake-up is activated, and the wake-up action and the diagnosis of the wake-up parameters are performed. The embodiment reduces the false wake-up problem by using multiple source composite signals; and reduces the false wake-up rate and improves the reliability of the wake-up operation based on the three-level wake-up decision.

[0052] According to the embodiment of the present application, based on the preset delay compensation, the time-aligned trigger signal matrix is generated according to the trigger signal, specifically: Based on the preset sampling circuit or sampling channel, the first delay information and the second delay information are determined; The first level information is subjected to forward time shift compensation of the first delay information; The first square wave information is subjected to forward time shift compensation of the second delay information; The compensated first level information and the first square wave information are combined with the first voltage information to integrate into a trigger signal matrix according to the time axis, wherein the row vectors of the trigger signal matrix correspond to the first level sequence, the first square wave sequence and the first voltage sequence respectively.

[0053] It should be noted that the first delay information is the delay caused by the PGANG level sampling circuit and logic; and the second delay information is the delay caused by the DRV square wave sampling circuit and logic. The embodiment provides a generation process of a trigger signal matrix. In order to eliminate the difference in signal transmission delay, the system allocates independent ADC channels for the PGANG level and the VSENSE voltage, and allocates a timer for the DRV square wave. As an implementation, after the PGANG signal is input, a 0.2 ms time shift compensation is applied to offset the internal sampling circuit delay; the DRV signal is applied with a 1.5 ms compensation to solve the square wave generation circuit response lag; and the VSENSE signal is processed directly. The timestamps of the three compensated data are aligned, and a trigger signal matrix with fixed row vector dimension is constructed, wherein the first row is a PGANG level sequence, the second row is a DRV square wave sequence, and the third row is a VSENSE voltage sequence. Based on the delay compensation, the embodiment ensures the timing consistency of the wake-up event analysis, and avoids feature extraction distortion caused by asynchronous signals.

[0054] According to the embodiment of the present application, the feature signal is extracted according to the trigger signal matrix, specifically: The first level sequence is extracted based on a preset sliding window, and a slope value is extracted as first slope information according to a first-order differential operation; The first square wave sequence is extracted based on a preset time window, and first frequency information is obtained according to zero-crossing detection; The first voltage sequence is subjected to a root mean square calculation, and a root mean square result is subjected to a preset normalization processing to obtain first energy information.

[0055] It should be noted that the embodiment provides a feature extraction process of a wake-up signal. As an implementation, each sequence is separated from the trigger signal matrix. The PGANG level sequence is based on a preset first-order differential, derivative or difference operation, and the instantaneous slope is calculated through an adjustable sliding window to lock the maximum positive slope value as a wake-up sensitive index. The DRV square wave sequence is subjected to a zero-crossing detection mechanism to count the rising edge and falling edge transition times in a preset time window, and the DRV square wave frequency feature is calculated according to the transition times and the time window length. The VSENSE voltage sequence is first subjected to a square operation to eliminate negative values, then the average value in the window is calculated, and finally the root mean square energy value is obtained; the root mean square energy value is subjected to a normalization processing to map the root mean square energy value to the range of [0, 1]. The processing result is packaged as a feature vector for calling by a decision layer. The embodiment accurately captures the PGANG level transition through the differential operation, enhances the anti-amplitude interference of the DRV square wave through the zero-crossing detection, and eliminates the individual differences of devices by using the energy standardization; and then the accuracy of subsequent analysis and decision is improved.

[0056] According to the embodiment of the present application, the noise baseline is determined according to the first voltage information, and the wake-up threshold is obtained in combination with the historical wake-up record, specifically as follows: The first voltage information is recorded in the absence of a wake-up event to obtain a second voltage sequence; The noise baseline is determined according to the root mean square value of the difference between the voltage value of the second voltage sequence and the standard voltage value; The historical false wake-up rate is generated according to the proportion of false wake-up events in total wake-up events in the historical wake-up record based on a preset statistical window; The wake-up threshold is obtained according to the noise baseline and the historical false wake-up rate based on a pre-trained wake-up threshold association model.

[0057] It should be noted that the embodiment provides a dynamic threshold self-adaptive setting mechanism. The noise baseline is calculated automatically during the system idle period, that is, the VSENSE voltage is recorded in the absence of a wake-up event, and the recording is a second voltage sequence. The deviation value of the second voltage sequence from the standard voltage value is calculated, and the root mean square value of the deviation value is taken as the environmental noise baseline. As an implementation manner, all wake-up records are retrieved with a 24-hour statistical window, and the percentage of records marked as false wake-up events in the total number of wake-up times is calculated to obtain the historical false wake-up rate. The environmental noise baseline and the historical false wake-up rate are input into the pre-trained model, and the superimposed effect of the two is adjusted by a negative feedback coefficient to dynamically generate a wake-up threshold. As an implementation manner, the wake-up threshold association model is a neural network model; as another implementation manner, the wake-up threshold association model is a mapping relationship formula. The embodiment adopts a dynamic adaptive working voltage fluctuation environment to cooperate with the threshold association model to improve the accuracy of wake-up judgment.

[0058] According to the embodiment of the present application, the wake-up action specifically includes: Outputting a high-level pulse signal through a preset GPIO interface; Controlling the power management unit to switch the supply voltage from the S3 / S4 maintenance voltage to the rated voltage; Generating and storing a wake-up log according to the characteristic signal and the wake-up level when the wake-up is triggered.

[0059] It should be noted that, as an implementation, the wake-up action specifically includes pulse output, voltage switching and log recording. When determining the effective wake-up, the GPIO_CTRL pin outputs a high-level pulse with a set duration. As an implementation, the pulse is transmitted to the power management unit after photoelectric isolation. Based on the pulse signal, first, the S3 / S4 energy-saving power supply circuit is cut off, and then the 3.3V rated voltage converter is activated, and the CPU core power supply switching is completed within the safe switching time. At the same time, the log engine records the wake-up event characteristics, including the time stamp, stores the characteristic information such as the PGANG slope peak value, the DRV dominant frequency, the VSENSE energy surge value, and the wake-up level code. The embodiment uses log recording for later analysis of the trace.

[0060] According to the embodiment of the present application, further comprising: If the wake-up action is triggered, the default is marked as a first device wake-up event; If the first slope information is greater than a preset second slope threshold, and the first frequency information is within a preset second frequency range, it is determined and recorded as a second device wake-up event; If the first slope information is greater than a preset second slope threshold, and the first energy information exceeds a preset first energy threshold, it is determined and recorded as a third device wake-up event.

[0061] It should be noted that the first device wake-up event represents an event triggered by an unknown device; the second device wake-up event represents an event triggered by a mouse; and the third device wake-up event represents an event triggered by a keyboard. The embodiment provides a wake-up device classification mechanism. When the wake-up is triggered, the default is an unknown device event. As an implementation, if the PGANG slope is greater than 0.8V / ms and the DRV frequency is within the range of 950-1050Hz, it is marked as a mouse device; if the slope is greater than 0.8V / ms and the VSENSE energy is greater than 50mV, it is marked as a keyboard device. After being marked as a mouse device or a keyboard device, it can be used to formulate a communication protocol selection and shorten the device activation time after the wake-up operation.

[0062] It is worth mentioning that after determining the three-level wake-up, further comprising: According to the first level sequence, a slope change rate is obtained; If the slope change rate is lower than a preset change rate threshold, the wake-up action is denied, and it is determined as an interference event; Performing fast Fourier transform on the first square wave sequence, and according to the frequency domain data, the energy proportion of a preset frequency band is counted; If the energy proportion exceeds a preset proportion threshold, the wake-up action is denied, and it is determined as an interference event; If the first energy information continues to be lower than the preset second energy threshold, the wake-up action is denied, and it is determined that it is an interference event.

[0063] It should be noted that the embodiment provides a triple interference detection mechanism. In the embodiment, after the three-level wake-up trigger, the triple anti-interference detection is started. First, the PGANG slope change rate analysis is performed. If the slope change rate is lower than the change rate threshold, it is determined that it is a slow-changing interference, for example, a level shift caused by temperature drift, and therefore, the current wake-up is denied. Then, the DRV square wave is subjected to 512-point fast Fourier transform (FFT), and the energy proportion in the 50Hz-60Hz frequency band is calculated. When the energy proportion exceeds the set 30% threshold, it is determined that it is a power frequency interference, and the current wake-up is denied. Finally, the VSENSE energy is subjected to continuous detection. If the normalized energy value identified in the VSENSE sequence continuously is lower than 0.3, it is determined that it is a transient interference, and the current wake-up is denied. The wake-up process is terminated if any of the above steps is denied. The event is recorded as a "interference denial" type. The embodiment improves the anti-interference capability through triple verification; among them, the frequency domain analysis solves the influence of power frequency interference, and the energy continuous monitoring prevents the contact from shaking and being mistaken.

[0064] It is worth mentioning that it also includes: An wake-up event database is established to record the feature signal, wake-up level and wake-up device type; When it is determined to be a false wake-up event, the database record of the current event is recorded to the interference mode feature database; According to the feature signal and the wake-up level of the current event, the weight coefficient in the weighted fusion algorithm is adjusted; The updated weight coefficient and the interference mode feature library are written to the non-volatile memory.

[0065] It should be noted that the embodiment provides a self-learning mechanism to adaptively adjust the weight coefficient in the weighted fusion algorithm. In the embodiment, the adaptive learning mechanism includes two stages of online learning and offline storage. After each wake-up event, the slope, frequency and energy value are taken as a feature vector, a decision result and a device type, and are packaged as a data packet and stored in a RAM cache area. If the event is denied by the triple verification, the data packet is appended with a false wake-up label and is migrated to the interference feature library; if the wake-up is successful, it is marked as a positive sample and is entered into a positive sample library. Based on a preset self-learning period, the occurrence frequency of similar feature modes in the false wake-up library is counted, and the weight of the corresponding signal is reduced. As an implementation manner, if the PGANG slope determines that the false touch is frequent, the PGANG slope weight coefficient is lowered. Moreover, the positive sample library is analyzed to optimize the recognition threshold. Finally, the updated weight coefficient and the interference mode feature library are written to the local non-volatile memory to realize offline storage.

[0066] It is worth mentioning that it also includes: When the ambient temperature exceeds the preset temperature threshold, then the interval of the first frequency range and the second frequency range is expanded and a safety event log is recorded; When it is determined according to the first energy information that the ambient surge, then the wake-up function is turned off based on a preset shutdown delay and a safety event log is recorded.

[0067] It should be noted that the embodiment also provides an anti-environmental interference mechanism. For high-temperature environmental conditions, when the temperature sensor detects that the ambient temperature exceeds 85℃, the frequency tolerance is automatically expanded to ±10% and a high-temperature compensation log is recorded, wherein the log includes temperature value, compensation time and original frequency data. As an embodiment, when the temperature exceeds 85℃, the first frequency range is expanded from 1KHz±5% to 1KHz±10% for compensating the crystal oscillator temperature drift. For power surge conditions, as an embodiment, when the normalized energy value of the first energy information exceeds 0.8 for 100ms, it is determined that the ambient surge, then the wake-up function is automatically turned off for 200ms, and a surge event log is generated, wherein the log includes peak voltage, duration, and spectral characteristics.

[0068] The third aspect of the present application provides a computer readable storage medium, the computer readable storage medium comprises a HiSilicon platform USB based adaptive multi-source level wake-up method program, when the HiSilicon platform USB based adaptive multi-source level wake-up method program is executed by a processor, the steps of the HiSilicon platform USB based adaptive multi-source level wake-up method are realized.

[0069] In summary, the present application provides a HiSilicon platform USB based adaptive multi-source level wake-up method, system and medium, synchronously capturing the first level information, the first square wave information and the first voltage information of the USB Hub, generating a time-aligned trigger signal matrix through delay compensation; extracting feature signals based on the trigger signal matrix, generating a wake-up score by combining a weighted fusion algorithm; dynamically calculating an adaptive wake-up threshold based on a real-time noise baseline and historical false wake-up records; when only the first slope information is out of limit, determining that it is a first-level wake-up and increasing the wake-up score; when both the first slope information and the first frequency information are out of limit, determining that it is a second-level wake-up and triggering a wake-up action; when the wake-up score exceeds the wake-up threshold, determining that it is a third-level wake-up and triggering a wake-up action and starting a wake-up diagnosis; reducing noise interference through multi-source signal cross-validation, adapting to complex industrial environments by using a dynamic threshold algorithm, and improving the success rate and reliability of wake-up.

[0070] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0071] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for adaptive multi-source level wake-up based on HiSilicon platform USB, characterized in that, The method comprises: a trigger signal is collected in parallel, including first level information, first square wave information and first voltage information; based on preset delay compensation, a time-aligned trigger signal matrix is generated according to the trigger signal; based on the trigger signal matrix, feature signals are extracted, including first slope information, first frequency information and first energy information; based on a preset weighted fusion algorithm, a wake-up score is obtained according to the feature signals; based on the first voltage information, a noise baseline is determined, and a wake-up threshold is obtained in combination with historical wake-up records; it is judged whether the first slope information exceeds a preset first slope threshold; if yes, it is judged whether the first frequency information is within a preset first frequency range; if no, it is determined as a first-level wake-up, and the wake-up score is increased; if yes, it is determined as a second-level wake-up, and a wake-up action is triggered; if the wake-up score exceeds the wake-up threshold, it is determined as a third-level wake-up, and a wake-up action and a wake-up diagnosis are triggered.

2. The USB-based adaptive multi-source level wake-up method based on HiSilicon platform according to claim 1, characterized in that, The method comprises: based on a preset sampling circuit or sampling channel, first delay information and second delay information are determined; a forward time shift compensation of the first delay information is applied to the first level information; a forward time shift compensation of the second delay information is applied to the first square wave information; the compensated first level information and the first square wave information are combined with the first voltage information, and are integrated into a trigger signal matrix according to a time axis, wherein row vectors of the trigger signal matrix correspond to first level sequences, first square wave sequences and first voltage sequences respectively.

3. The method of claim 2, wherein the method is based on HiSilicon platform USB adaptive multi-source level wake-up. The method comprises: based on a preset sliding window, the first level sequences are extracted, and slope values are extracted as first slope information according to a first-order differential operation; based on a preset time window, the first square wave sequences are extracted, and first frequency information is obtained according to zero-crossing point detection; root mean square calculation is performed on the first voltage sequences, and a preset normalization processing is performed on the root mean square result to obtain first energy information.

4. The USB-based adaptive multi-source level wake-up method based on HiSilicon platform according to claim 1, characterized in that, The method comprises: when there is no wake-up event, the first voltage information is recorded to obtain a second voltage sequence; a difference between voltage values of the second voltage sequence and a standard voltage value is calculated, and a root mean square value of the difference is calculated to determine a noise baseline; based on a preset statistical window, a historical false wake-up rate is generated according to a proportion of false wake-up events to total wake-up events in the historical wake-up records; based on a pre-trained wake-up threshold association model, a wake-up threshold is obtained according to the noise baseline and the historical false wake-up rate.

5. The USB-based adaptive multi-source level wake-up method based on HiSilicon platform according to claim 1, characterized in that, The wake-up action specifically comprises: a high-level pulse signal is output through a preset GPIO interface; a power management unit is controlled to switch a power supply voltage from an S3 / S4 maintenance voltage to a rated voltage; a feature signal and a wake-up level when a wake-up is triggered are generated and stored as a wake-up log.

6. The USB-based adaptive multi-source level wake-up method based on HiSilicon platform according to claim 1, wherein, The method further comprises: if a wake-up action is triggered, a first device wake-up event is marked by default. If the first slope information is greater than a preset second slope threshold value and the first frequency information is within a preset second frequency range, a second device wake-up event is determined and recorded; If the first slope information is greater than a preset second slope threshold value and the first energy information exceeds a preset first energy threshold value, a third device wake-up event is determined and recorded.

7. A USB-based adaptive multi-source level wake-up system based on HiSilicon platform, characterized in that, The system comprises a memory and a processor, the memory comprising a HiSilicon platform USB-based adaptive multi-source level wake-up method program, which, when executed by the processor, implements the following steps: Parallelly collecting trigger signals, including first level information, first square wave information, and first voltage information; Based on preset delay compensation, generating a time-aligned trigger signal matrix according to the trigger signals; Based on a preset weighted fusion algorithm, obtaining a wake-up score according to the feature signals; Determining a noise baseline according to the first voltage information, and obtaining a wake-up threshold value in combination with historical wake-up records; Determining whether the first slope information exceeds a preset first slope threshold value; If yes, determining whether the first frequency information is within a preset first frequency range; If no, determining a first-level wake-up and increasing the wake-up score; If yes, determining a second-level wake-up and triggering a wake-up action; If the wake-up score exceeds the wake-up threshold value, determining a third-level wake-up, triggering a wake-up action, and starting a wake-up diagnosis. The delay compensation is based on a preset sampling circuit or sampling channel to determine first delay information and second delay information.

8. The USB-based adaptive multi-source level wake-up system based on HiSilicon platform according to claim 7, characterized in that, The first level information is subjected to forward time shift compensation of the first delay information. The first square wave information is subjected to forward time shift compensation of the second delay information. The compensated first level information and first square wave information are combined with the first voltage information to form a trigger signal matrix along a time axis, wherein the row vectors of the trigger signal matrix correspond to first level sequences, first square wave sequences, and first voltage sequences. The feature signals are extracted based on a preset sliding window to extract slope values as first slope information through first-order differential operation. The first square wave sequences are extracted based on a preset time window to obtain first frequency information through zero-crossing point detection.

9. The USB-based adaptive multi-source level wake-up system based on HiSilicon platform of claim 7, wherein, The first voltage sequences are subjected to root mean square calculation, and the root mean square results are subjected to preset normalization processing to obtain first energy information. The computer readable storage medium comprises a HiSilicon platform USB-based adaptive multi-source level wake-up method program, which, when executed by a processor, implements the steps of the HiSilicon platform USB-based adaptive multi-source level wake-up method according to any one of claims 1 to 6. ​ ​ 10. A computer-readable storage medium having stored thereon a computer program, characterized in that, ​

Images