A human body inductive magnetic shaft keyboard sleep wake-up control system

Abstract

This invention discloses a sleep / wake-up control system for a magnetic axis keyboard integrating human body sensing, belonging to the field of peripheral control technology. It includes: a behavior timing analysis module, used to comprehensively analyze user operating habits and brief absences from the keyboard, accurately determining the user's state through timing behavior analysis algorithms and near-range radar micro-motion sensing algorithms; and a vital sign intention recognition module, used to continuously monitor biometrics and approach postures in front of the keyboard during sleep mode. By comprehensively analyzing long-term user operating habits and real-time near-range micro-motion sensing, this invention, compared to traditional single-time threshold sleep mechanisms, can intelligently distinguish between brief absences and long-term idle states. The system only enters deep sleep mode after confirming long-term idleness, thus avoiding unnecessary global wake-ups frequently triggered by user actions such as brief absences or getting water, significantly reducing unnecessary power consumption during standby, and extending the keyboard's overall battery life.

Description

Technical Field

[0001] This invention relates to the field of peripheral control technology, specifically to a magnetic axis keyboard sleep / wake-up control system integrating human body sensing. Background Technology

[0002] With the widespread use of mobile devices and portable computers, users have increasingly higher demands for device battery life. Magnetic axis keyboards use magnetic sensors to detect the status of keys and can automatically switch to sleep mode when not in use for a long time to reduce power consumption. Through a preset time delay, the keyboard status is dynamically monitored. When no user input is detected for a period of time, the system automatically enters a low-power state. Once the user triggers a key or other interaction, the system can be quickly woken up, thus achieving a seamless user experience.

[0003] For example, the row and column scanning control system based on a sleep and wake-up function chip, Chinese Patent Publication No. CN119758850A, supports three scanning modes and has low power consumption characteristics, making it particularly suitable for application scenarios such as three-mode keyboards.

[0004] In existing technologies, magnetic axis keyboards struggle to accurately distinguish between brief periods of user absence and prolonged periods of inactivity. This leads to the system frequently triggering unnecessary global wake-ups when the user briefly leaves their seat, increasing system power consumption and impacting battery life. Furthermore, when the system enters deep sleep due to prolonged inactivity, its wake-up mechanism relies solely on the direct trigger signal of the magnetic axis keys. If the user returns without immediately pressing a key, the system cannot detect the user's approach and perform a pre-wake-up, resulting in a significant delay in the first key press response and affecting the consistency and immediacy of the user experience. To address these issues, a magnetic axis keyboard sleep-wake control system integrating human body sensing is proposed. Summary of the Invention

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a magnetic axis keyboard sleep / wake-up control system integrating human body sensing, comprising: The behavior timing analysis module is used to integrate the user's operating habits and brief absence behavior of magnetic axis keyboards. Through timing behavior analysis algorithm and near-range radar micro-motion sensing algorithm, it accurately judges the user's state, avoids invalid wake-up triggered by brief absence, effectively distinguishes between brief absence and long-term idleness, and significantly reduces the number of invalid system wake-ups caused by misjudgment. The biometrics and intention recognition module is used to continuously monitor the biometrics and approach posture in front of the magnetic keyboard in sleep mode. It integrates infrared pyroelectric sequence matching and millimeter wave biometrics detection algorithms to identify the user's intention to return, provide a signal for pre-wake-up, and continuously sense the user's approach intention during sleep. The sleep strategy execution module, based on the behavior timing analysis results, triggers sleep mode determination analysis and controls the magnetic axis keyboard to enter the corresponding sleep mode, including shallow sleep, deep sleep and pre-wake state. Deep sleep is only triggered when it is confirmed to be idle for a long time. The sleep level is intelligently switched according to the user's status to ensure battery life while maintaining rapid recovery capability. The pre-wake-up trigger module is used to pre-wake up the magnetic axis scanning circuit and the main control unit in advance when the user's approach intention is detected, combined with the determined sleep mode, to shorten the first key response time and restore system functions in advance before the user actually presses the key, thus significantly shortening the first key response delay. The instant response coordination module is used to coordinate the rapid readiness of each module of the magnetic axis keyboard after pre-wake-up, ensuring that magnetic axis scanning, signal acquisition and main control processing are initialized before the user presses the key, optimizing the task processing flow, ensuring a seamless switch from sleep to response, and achieving zero-latency input response.

[0006] Preferably, the behavior timing analysis module includes a user habit modeling unit and a keyboard micro-motion sensing unit; The user habit modeling unit learns and models user keyboard usage habits based on pre-collected user operation habit data and combines time-series behavior analysis algorithms to construct a user behavior model, identify typical operation intervals and seat departure patterns, realize continuous learning and adaptive modeling of user operation habits, and improve the accuracy and personalization level of state judgment. The keyboard micro-motion sensing unit is used to detect the user's centimeter-level micro-motions in real time through the built-in near-range radar micro-motion sensing algorithm, and to determine the user's state by combining the user behavior model, distinguishing between short-term absence and long-term inactivity, reducing false wake-ups, monitoring the user's micro-motions in real time, and further reducing the probability of false wake-ups when the user is temporarily absent by combining the behavior model with the judgment.

[0007] Preferably, the user habit modeling unit performs the following steps: User operation data is collected by the built-in timing recording unit of the magnetic axis keyboard. The user operation data includes key timestamps, key duration, adjacent operation interval sequence and daily / weekly usage time distribution, to establish an objective and continuous user behavior data foundation and provide reliable input for intelligent judgment. Using time-series behavior analysis algorithms, pattern mining is performed on user operation data to identify user characteristics, including typical operation interval thresholds, high-frequency operation periods, and normal off-seat duration. Key features reflecting individual user habits are extracted to provide a quantitative basis for user status judgment. Based on the identified user characteristics, a dynamically updated user behavior model is constructed. The output of the user behavior model is used to determine whether the current user is in an operating state, a short-term away state, or a long-term idle state, enabling the system to accurately distinguish different intentions of leaving the seat and significantly reduce invalid wake-ups caused by short-term absence.

[0008] Preferably, the keyboard micro-motion sensing unit performs the following steps: The built-in 60GHz millimeter-wave radar sensor collects micro-motion signals in a preset area in front of the magnetic axis keyboard in real time. The micro-motion signals include distance, speed and micro-motion amplitude information, realizing non-contact high-precision dynamic perception of the user's current state. The micro-motion signal is fused and analyzed with the current judgment state output by the user behavior model. If the current judgment is a brief departure state, the micro-motion signal is used to continuously determine whether the user is still in the vicinity of the keyboard operation area, which effectively prevents the system from being accidentally triggered to deep sleep due to the user's brief departure. When the micro-motion signal continuously exceeds the first preset duration and is below the activity threshold, and the user behavior model supports the determination of long-term idle state, it is confirmed as a long-term idle state. Combined with multi-source information, it is confirmed that the user has been away for a long time, providing a reliable basis for entering the ultra-low power mode. If the micro-motion signal is detected again within the second preset duration as a centimeter-level activity that conforms to human characteristics, it is determined that the user has returned to the nearby area. Before the user actually operates, the system can perceive the user's intention to approach in advance, creating conditions for the system to pre-wake up.

[0009] Preferably, the vital sign intention recognition module includes an infrared sequence matching unit and a millimeter-wave vital sign detection unit; The infrared sequence matching unit is used to detect the movement pattern of human heat source in front of the magnetic axis keyboard by using an infrared pyroelectric sensor sequence matching algorithm, analyze the user's return intention, determine whether the user is approaching the keyboard operation area, and identify the user's approach behavior through the heat source movement pattern to achieve low power consumption and high sensitivity preliminary intention judgment. The millimeter-wave vital sign detection unit is used to detect human biological signs through millimeter-wave radar, identify the approach of a living person and their posture intentions, enhance the accuracy of identification, and effectively distinguish between living users and non-living interference based on vital signs and posture analysis, thus greatly improving the accuracy of identification.

[0010] Preferably, the infrared sequence matching unit performs the following steps: By deploying a multi-channel infrared pyroelectric sensor array along the upper edge of the keyboard, a sequence of heat source signals in the space in front is collected at a fixed sampling period. Through synchronous acquisition by multiple sensors, continuous capture and preliminary screening of heat source signals in the space in front are achieved. Dynamic threshold filtering and motion pattern analysis are performed on the heat source signal sequence to extract heat source pattern features including movement direction, speed and spatial trajectory, effectively filtering out environmental noise interference, accurately extracting the motion features of the heat source, and improving the robustness of pattern recognition. The heat source pattern features are matched with a pre-stored return intent pattern library. If the matching degree exceeds the set matching threshold, it is determined that the user intends to approach the keyboard operation area. By judging the matching degree, the user's approach intent can be efficiently identified, reducing false positives and false negatives.

[0011] Preferably, the millimeter-wave vital sign detection unit performs the following steps: When the infrared sequence matching unit outputs an indication of approach intent, the millimeter-wave radar is activated to detect human vital signs, collecting vital sign signals within a micro-motion range to enhance detection accuracy and anti-interference capabilities, effectively eliminating false triggers from non-living entities. The system performs spectral analysis and extracts respiratory / heartbeat features from the vital signs signals to distinguish between the movement of a living human body and other heat sources, and identifies whether the user's posture is a ready-to-operate posture. This ensures that the wake-up is only for the actual user's operation intention, further reducing the false wake-up rate and improving the system's intelligence. By fusing infrared sequence matching results and vital sign recognition results, when both meet the preset return operation criteria, a high-confidence user return intent signal is output to the pre-wake-up trigger module, realizing dual-mode fusion decision-making, greatly improving the reliability of intent judgment, and providing accurate triggering basis for pre-wake-up.

[0012] Preferably, the hibernation strategy execution module performs the following steps: The magnetic axis keyboard is divided into sleep modes, including shallow sleep, deep sleep and pre-wake state, and the user state judgment result output by the behavior timing analysis module is received. If it is a brief absence state, the magnetic axis keyboard is controlled to switch to shallow sleep mode to keep the human body sensing circuit and part of the scanning circuit in low power standby mode, so as to maintain human body sensing and fast recovery capability with minimal power consumption and avoid invalid deep sleep caused by brief absence. If the user's status is determined to be long-term idle, the magnetic keyboard will be controlled to enter a deep sleep mode, retaining power only for the vital sign intention recognition module and interrupt wake-up function, minimizing the overall static power consumption of the system and significantly extending the keyboard's battery life in the absence of operation. During shallow or deep sleep, if the vital signs and intent recognition module outputs a user return intent signal, the magnetic axis keyboard is controlled to switch to the pre-wake state, and the power supply and initialization of the main control unit and scanning circuit are partially restored in advance. The system is silently ready before the user actually touches the key, achieving zero-perceptible delay from sleep to response.

[0013] Preferably, the pre-wake-up trigger module performs the following steps: When the sleep strategy execution module controls the magnetic axis keyboard to enter the pre-wake state, it immediately sends a wake-up interrupt signal to the main control unit of the magnetic axis scanning circuit and restores its basic clock and power domain, ensuring that the main control unit can be activated in real time at the hardware level, providing the core operating foundation for subsequent software initialization and task scheduling. The row and column scanning control chip in the magnetic axis scanning circuit exits the sleep mode and is configured to high-speed progressive scanning ready state, so that the magnetic axis scanning circuit enters the low-latency standby mode to prepare for a fast response to the upcoming key signal. The task scheduler of the main control unit is preheated and initialized, and the keyboard mapping table and response processing program are loaded to ensure that the system has completed the hardware and software readiness from sleep to working state before the user actually touches the key. The software system is pre-set to standby state to eliminate the software startup delay after the system wakes up and ensures immediate processing and response to the first key press.

[0014] Preferably, the instantaneous response coordination module performs the following steps: After the pre-wake-up trigger module completes hardware readiness, it coordinates the magnetic axis scanning circuit to perform a full matrix fast self-test scan to initialize the signal output state of all Hall chips, ensure the stability of the Hall sensor array output, and eliminate the interference of hardware abnormalities on subsequent signal acquisition. The analog-to-digital converter sampling circuit of the signal acquisition module is started synchronously, and benchmark calibration and noise filtering initialization are performed on it to ensure the stability of the acquisition link, improve the signal acquisition accuracy and anti-interference capability, and ensure the true reproduction of the button signal; Through the task coordination mechanism of the main control unit, the pipeline scheduling and synchronization between scanning, acquisition and processing tasks are completed before the user's first key press signal arrives, realizing zero-wait link readiness from key signal generation to system response.

[0015] This invention provides a sleep / wake control system for magnetic axis keyboards that integrates human body sensing. It has the following beneficial effects: (i) The integrated human body sensing magnetic axis keyboard sleep and wake-up control system, by comprehensively analyzing the user's long-term operating habits and close-range real-time micro-motion sensing, compared with the traditional single time threshold sleep mechanism, can intelligently distinguish between the two states of the user's short-term absence and long-term idleness. The system only enters deep sleep mode after confirming that the user has been idle for a long time, thereby avoiding unnecessary global wake-ups frequently triggered by the user's short-term departure, pouring water, etc., significantly reducing unnecessary power consumption of the system during standby, and extending the overall battery life of the keyboard.

[0016] (II) The integrated human body sensing magnetic axis keyboard sleep-wake control system continuously monitors the bio-signals and approach posture in front of the keyboard in sleep mode. When the system recognizes the user's high confidence intention to return and prepare for operation by integrating infrared sequence matching and millimeter wave bio-signal detection algorithms, it triggers the pre-wake process. This allows the magnetic axis scanning circuit, main control unit and signal acquisition module to be initialized and ready in advance before the user actually presses the first key. This reduces the significant response delay of the first key press after traditional deep sleep to near zero, ensuring the immediacy and continuity of the user experience.

[0017] (III) The integrated human body sensing magnetic axis keyboard sleep and wake-up control system defines three core sleep states: shallow sleep, deep sleep, and pre-wake-up. The sleep strategy execution module performs dynamic and fine-grained control based on upstream sensing results. In shallow sleep, it maintains the fast response capability of some sensing circuits. In deep sleep, it maintains only the minimum biological sensing and interrupt wake-up path. In pre-wake-up, it efficiently and orderly restores each functional module. Through the strategy of allocating power resources on demand, it achieves the optimal balance between extremely low static power consumption and instantaneous high-performance response. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the workflow of a magnetic axis keyboard sleep / wake-up control system integrating human body sensing according to the present invention. Figure 2 This is a data flow diagram of a magnetic axis keyboard sleep / wake-up control system integrating human body sensing according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, please refer to Figure 1 , Figure 2 This invention provides a technical solution: a magnetic axis keyboard sleep / wake control system integrating human body sensing, comprising: The behavior timing analysis module is used to integrate the user's operating habits and brief absence behavior of magnetic axis keyboards. Through timing behavior analysis algorithm and close-range radar micro-motion sensing algorithm, it accurately judges the user's state, avoids invalid wake-up triggered by brief absence, effectively distinguishes between brief absence and long-term idleness, and significantly reduces the number of invalid system wake-ups caused by misjudgment. The behavior timing analysis module includes a user habit modeling unit and a keyboard micro-motion sensing unit. The user habit modeling unit, based on pre-collected user operation habit data, learns and models user keyboard usage habits using a time-series behavior analysis algorithm, constructs a user behavior model, identifies typical operation intervals and departure patterns, and achieves continuous learning and adaptive modeling of user operation habits, improving the accuracy and personalization of state judgment. It collects user operation data through the time-series recording unit built into the magnetic axis keyboard, including key timestamps, key duration, adjacent operation interval sequences, and daily / weekly usage time distribution, establishing an objective and continuous user behavior data foundation to provide reliable input for intelligent judgment. The time-series behavior analysis algorithm performs pattern mining on the user operation data, identifying user characteristics including typical operation interval thresholds, high-frequency operation periods, and normal departure duration, extracting key features reflecting individual user habits, and providing a quantitative basis for user state judgment. Based on the identified user characteristics, a dynamically updated user behavior model is constructed. The output of the user behavior model is used to determine whether the current user is in an operation state, a brief departure state, or a long-term idle state, enabling the system to accurately distinguish different departure intentions and significantly reduce invalid wake-ups caused by brief departures. Furthermore, the system continuously collects and stores user operation data through the high-precision real-time clock and timing recording firmware built into the keyboard main control unit. Specific parameters collected include: the precise timestamp of each key press, the duration of the key press from press to release, the time sequence of intervals between two consecutive valid key presses, and a histogram of usage time distribution on a daily and weekly basis. The user's raw operation data is stored in the keyboard's non-volatile memory in the form of a circular buffer, with a storage depth covering the operation records of the most recent 30 days to ensure the model has sufficient recent and long-term learning samples. The collection process is entirely local, and all data processing adheres to privacy protection principles, without involving the extraction or uploading of personally identifiable information. After the system is powered on, the timing recording unit automatically runs in the background. When sufficient new data accumulation is detected, the timing behavior analysis algorithm is activated to perform offline mode mining on the collected operation data. First, statistical analysis is performed on the operation interval sequence, using percentile methods to determine typical operation interval thresholds. Simultaneously, the daily / weekly usage time distribution is analyzed to identify the daily average cumulative... The system calculates high-frequency operation periods exceeding 30 minutes. Furthermore, it extracts typical ranges of normal absence duration from long-interval events through cluster analysis, constructing and dynamically updating a lightweight user behavior model. Using the current time and the most recent operation interval as input, it calculates the confidence probability of a user being in an active operation, a brief absence, or a long-term idle state in real time. The user behavior model uses a sliding time window mechanism for dynamic updates, with a time window length set to 7 days to ensure the model can adapt to slow changes in user habits. During model updates, recent data is given higher weight. In actual operation, the model receives real-time operation streams as input. When an operation interval exceeds the typical operation interval threshold but is below the lower limit of the normal absence duration cluster, it outputs a high-confidence judgment of a brief absence state, considering whether the current period is a high-frequency operation period. If the operation interval continuously exceeds the upper limit of the normal absence duration cluster, and the current period is a non-high-frequency period, the model outputs a judgment of a long-term idle state. All judgment logic has a hysteresis buffer to prevent frequent state oscillations caused by accidental operation pauses. The keyboard micro-motion sensing unit is used to detect the user's centimeter-level micro-motions in real time through the built-in near-range radar micro-motion sensing algorithm. Combined with the user behavior model, it judges the user's state, distinguishes between short-term absence and long-term inactivity, reduces false wake-ups, monitors the user's micro-motions in real time, and further reduces the probability of false wake-ups when the user is away for a short time by combining the behavior model with the judgment. The built-in 60GHz millimeter-wave radar sensor collects the micro-motion signals in the preset area in front of the magnetic axis keyboard in real time. The micro-motion signals include distance, speed and micro-motion amplitude information, realizing non-contact high-precision dynamic perception of the user's current state. By fusing and analyzing the micro-motion signal with the current judgment state output by the user behavior model, if the current judgment is a brief departure state, the micro-motion signal is used to continuously determine whether the user is still in the vicinity of the keyboard operation area, effectively preventing the system from being accidentally triggered to deep sleep due to the user's brief departure. When the micro-motion signal continuously exceeds the first preset duration and is below the activity threshold, and the user behavior model supports the determination of long-term idle state, it is confirmed as a long-term idle state. Combined with multi-source information, it is confirmed that the user has been away for a long time, providing a reliable basis for entering the ultra-low power mode. If the micro-motion signal is detected again within the second preset duration as a centimeter-level activity that conforms to human characteristics, it is determined that the user has returned to the nearby area. Before the user actually operates, the system can perceive the user's intention to approach in advance, creating conditions for the system to pre-wake up. Furthermore, a 60GHz millimeter-wave radar sensor is integrated along the inner edge of the magnetic keyboard. Its detection range is set as a fan-shaped area with a radius of 0.8 meters and an angle of 60°, centered on the keyboard center. This sensor continuously acquires information on the distance, radial velocity, and micro-motion amplitude of targets within the area at a rate of 20 frames per second. The distance resolution reaches 5 millimeters, and the velocity detection lower limit is 0.01 meters per second. It processes the radar point cloud data in real time, filters out static background noise, and extracts signal components that match the characteristics of human micro-motions (chest rise and fall, slight limb swaying, etc.). When the user habit modeling unit determines that the current movement is short... When temporarily away from the user, the micro-motion sensing algorithm continuously monitors the area for periodic activities exceeding a speed threshold of 0.05 m / s and an amplitude greater than 3 mm. This confirms whether the user remains in the vicinity of the keyboard's operable area. All signal processing is completed in a dedicated DSP core built into the keyboard's main control unit, ensuring real-time performance and low power consumption. If the radar continuously detects micro-motion signals below the activity threshold (speed continuously below 0.01 m / s and amplitude less than 2 mm) for a first preset duration of 120 seconds, the user behavior model calculates the duration of long-term idle state based on the operation interval and time period characteristics. If the confidence level exceeds 85%, the system confirms that it has entered a long-term idle state. During this period, the radar sensor switches to an ultra-low power monitoring mode, reducing the sampling rate to 2 frames per second, retaining only basic motion detection functions. When the radar re-captures a motion pattern consistent with human biometrics within a subsequent second preset duration of 30 seconds, including a distance between 0.2 meters and 0.8 meters, a speed between 0.05 and 0.5 meters per second, and a typical human micro-motion frequency (corresponding to breathing and minor movements) of 10-60 times per minute, it is determined that the user has returned to the nearby area. This determination requires five consecutive sampling periods. Consistency of signal characteristics within a sampling period (2.5 seconds); The radar sensing module and the user behavior model are fused and analyzed through a weighted decision mechanism. In the short-term absence state, the radar data weight is set to 70%, and the behavior model confidence weight is 30%; When entering the long-term idle judgment stage, the weights of the two are adjusted to 50% each. All state transitions are equipped with a 3-second hysteresis buffer to prevent misjudgment caused by instantaneous signal fluctuations. The working parameters of the millimeter-wave radar can be adaptively adjusted according to the environment: a high-sensitivity mode is adopted in a low-interference environment, and an anti-interference mode is automatically switched in a complex electromagnetic environment. The vital signs and intentions recognition module is used to continuously monitor the bio-signals and approach postures in front of the magnetic axis keyboard in sleep mode. It integrates infrared pyroelectric sequence matching and millimeter-wave bio-signal detection algorithms to identify the user's return intention and provide a signal for pre-wake-up. During sleep, it continuously senses the user's approach intention and provides a highly reliable trigger signal for system pre-wake-up. The vital signs and intentions recognition module includes an infrared sequence matching unit and a millimeter-wave vital signs detection unit. The infrared sequence matching unit utilizes an infrared pyroelectric sensor sequence matching algorithm to detect the movement pattern of human heat sources in front of the magnetic keyboard, analyzes the user's return intention, and determines whether the user is approaching the keyboard operation area. By identifying the user's approach behavior through heat source movement patterns, it achieves low-power, high-sensitivity preliminary intention judgment. Through a multi-channel infrared pyroelectric sensor array deployed along the upper edge of the keyboard, it collects heat source signal sequences in the space in front at a fixed sampling period. Through synchronous acquisition by multiple sensors, it achieves continuous capture and preliminary screening of heat source signals in the space in front. It performs dynamic threshold filtering and movement pattern analysis on the heat source signal sequences, extracting heat source pattern features including movement direction, speed, and spatial trajectory, effectively filtering out environmental noise interference, accurately extracting the motion features of the heat source, and improving the robustness of pattern recognition. It performs sequence matching of the heat source pattern features with a pre-stored return intention pattern library. If the matching degree exceeds the set matching threshold, it is determined that there is an intention of the user approaching the keyboard operation area. Through matching degree judgment, it achieves efficient recognition of the user's approach intention and reduces false and false judgments. Furthermore, within the casing along the upper edge of the magnetic keyboard, three infrared pyroelectric sensors are embedded at even intervals, forming a linear array. The array's horizontal orientation is parallel to the keyboard's main plane, covering a horizontal sector of ±60° directly in front and a vertical detection angle of ±30°. The effective detection distance is 0.1 meters to 2.5 meters. Each sensor employs a dual-differential structure to suppress common-mode interference caused by uniform changes in ambient temperature. The sensor array uses a fixed sampling period of 50 milliseconds to synchronously acquire dynamic infrared radiation signals generated by moving heat sources within the front sector detection area, generating three independent and... The original analog voltage signal sequence, which has spatial location correlation, is acquired and processed by a built-in low-noise preamplifier and bandpass filter. The bandpass filter has a passband of 0.1 Hz to 10 Hz to retain signal components corresponding to typical human movement frequencies while suppressing high-frequency circuit noise and extremely low-frequency temperature drift interference. The processed digital signal sequence is transmitted in real time to the dedicated signal processing module of the main control unit for subsequent analysis. After receiving the digital signal sequence from the infrared sensor array, the dedicated signal processing module of the main control unit performs dynamic thresholding on each signal. The filtering dynamic threshold is calculated in real time based on the short-time statistical characteristics of the signal itself. The specific formula is: Current dynamic threshold = current window signal mean + 3 × current window signal standard deviation. It adapts to the changes in the ambient background infrared noise level and effectively filters out random noise pulses. Subsequently, the three filtered signals are jointly analyzed. By comparing the peak arrival time difference of adjacent sensor signals in time sequence and combining the fixed spatial spacing of the sensors (design value is 75 mm), the moving direction and tangential moving speed of the heat source are calculated in real time. The speed calculation accuracy is better than 0.05 m / s. At the same time, by tracking the intensity change and temporal relationship of a specific heat source signal in the array, its moving trajectory segment in the approximate spatial plane is reconstructed. A set of heat source pattern feature vectors containing the moving direction angle, average moving speed and trajectory segment coordinate sequence is output for each tracked heat source target. The system has a built-in pre-stored return intention pattern feature library. This feature library is calibrated through previous experiments and stores various typical user behavior pattern feature templates for approaching the keyboard operation area. The behavior pattern feature templates are stored in the form of feature vector sequences, describing the forward approach from a distance (1.5 meters away) to the operation distance (0.During the process (within 3 meters), the typical changes in the direction, speed, and spatial trajectory of the heat source are analyzed. In real-time operation, the extracted heat source pattern feature vector sequence is dynamically time-warped and matched with each template in the pattern library. The matching process calculates the similarity score between the real-time sequence and each template sequence. The score calculation comprehensively considers directional consistency, velocity curve matching degree, and trajectory spatial correlation, and uses weighted Euclidean distance for quantization. The weights are pre-calibrated according to the distinguishability of the features, and the matching degree threshold is set to 85%. If the matching degree calculation result of the current heat source feature sequence and any approach intention template exceeds this matching degree threshold, and the matching state lasts for more than 3 consecutive sampling periods (i.e., 150 milliseconds), it is determined that there is a clear intention of the user to approach the keyboard operation area. This determination result serves as a trigger signal and is output to the subsequent process to initiate a millimeter-wave vital sign verification process with higher confidence. The millimeter-wave vital sign detection unit is used to detect human biological signs using millimeter-wave radar, identify the approach and posture intentions of living beings, and enhance the accuracy of recognition. Based on vital signs and posture analysis, it effectively distinguishes between living users and non-living interference, significantly improving the accuracy of recognition. When the infrared sequence matching unit outputs that there is an approach intention, the millimeter-wave radar is activated to detect human vital signs, collect vital sign signals within the micro-motion range, enhance the accuracy of detection and anti-interference ability, effectively eliminate false triggers from non-living beings, perform spectrum analysis and respiratory / heartbeat feature extraction on vital sign signals, distinguish the movement of living human beings from other heat sources, and identify whether the user's posture is a ready-to-operate posture, ensuring that wake-up is only for the operation intention of real users, further reducing the false wake-up rate and improving the intelligence of the system. It integrates the infrared sequence matching results and vital sign recognition results. When both meet the preset return operation criteria, it outputs a high-confidence user return intention signal to the pre-wake-up trigger module, realizing dual-mode fusion decision-making, greatly improving the reliability of intention judgment, and providing accurate triggering basis for pre-wake-up. Furthermore, once the infrared sequence matching unit determines that a user is approaching and outputs a trigger signal, the millimeter-wave radar sensor is immediately activated, switching it from basic monitoring mode to high-precision vital sign detection mode. This mode is specifically optimized for analyzing biometric features within a close-range (0.2 m to 1.0 m) micro-motion range. The radar transmits a 60 GHz frequency-modulated continuous wave at a sampling rate of 10 frames per second and receives reflected signals from targets ahead. The received raw intermediate frequency signal undergoes fast Fourier transform processing in the range and velocity dimensions via a built-in hardware accelerator, generating a point cloud spectrum containing target range, radial velocity, and micro-motion phase information. The system focuses on a range threshold set at 0.5 m ± 0. The target point cloud within a 0.3-meter range is continuously acquired, and its ultra-low velocity (±0.01 m / s to ±0.1 m / s) spectral components are continuously collected in the velocity dimension, with a time window length of no less than 8 seconds to ensure coverage of at least two complete human respiratory cycles, forming a stable, low-noise original vital sign signal sequence. The acquired vital sign signal sequence is sent to a dedicated digital signal processor in the main control unit for in-depth analysis. Windowed Fourier transforms are performed on the signal sequence, and its power spectral density in the 0.1 Hz to 3 Hz frequency band is calculated. By identifying significant peaks in the power spectrum, the frequencies corresponding to the respiratory rate (typically 0.1 Hz to 0.5 Hz) and heart rate harmonic frequencies (typically 0.5 Hz) are extracted. The system uses characteristic spectral lines (from 0.8 Hz to 2.5 Hz) to distinguish living human bodies from non-living heat sources with similar movement characteristics (swaying plants or pets). It also assesses the physiological coupling between respiratory and heart rate harmonic frequencies and examines their rhythmic stability. Furthermore, by analyzing the distribution changes of radar point clouds along the pitch angle dimension, it infers the user's rough posture information, determining whether the target is standing, bent over, or sitting. Finally, it outputs a vital sign recognition conclusion that includes a liveness verification flag (yes / no) and posture classification results. Using a pre-defined decision logic unit, it integrates the initial proximity intent judgment from the infrared sequence matching unit with the liveness and posture verification conclusions from the millimeter-wave vital sign detection unit. The decision logic unit is pre-set with dual-path joint criteria: First, the infrared matching score must be consistently higher than 85%; second, the millimeter-wave vital sign detection must confirm that the target is a living human body and that its posture is identified as a ready-to-operate posture. Only when both criteria are met simultaneously within the same decision cycle (set to 500 milliseconds) will the decision logic unit generate a high-confidence user return intent signal. This signal, as a clear digital trigger instruction, is immediately sent to the pre-wake trigger module to ensure that the system only initiates the pre-wake process when it is highly certain that the user is returning to the nearby area in a posture ready to operate the keyboard. This maximizes the reduction of false trigger rate while ensuring the accuracy and immediacy of the response trigger. The sleep strategy execution module, based on the behavior timing analysis results, triggers sleep mode determination analysis and controls the magnetic axis keyboard to enter the corresponding sleep mode, including shallow sleep, deep sleep and pre-wake state. Deep sleep is only triggered when it is confirmed to be idle for a long time. The sleep level is intelligently switched according to the user's status to ensure battery life while maintaining rapid recovery capability. The pre-wake-up trigger module is used to pre-wake up the magnetic axis scanning circuit and the main control unit in advance when the user's approach intention is detected, combined with the determined sleep mode, to shorten the first key response time and restore system functions in advance before the user actually presses the key, thus significantly shortening the first key response delay. The instant response coordination module is used to coordinate the rapid readiness of each module of the magnetic axis keyboard after pre-wake-up, ensuring that magnetic axis scanning, signal acquisition and main control processing are initialized before the user presses the key, optimizing the task processing flow, ensuring seamless switching of the system from sleep to response, achieving zero-delay input response, and coordinating each module to complete the full-link initialization before the key is pressed, achieving seamless instant switching from sleep to response.

[0021] Example 2, as Figure 1 , Figure 2 As shown, based on Embodiment 1, the present invention provides a technical solution: the sleep strategy execution module performs the following steps: dividing the sleep mode of the magnetic axis keyboard, including shallow sleep, deep sleep and pre-wake state, and receiving the user state judgment result output by the behavior timing analysis module. If it is a brief absence state, the magnetic axis keyboard is controlled to switch to shallow sleep mode to keep the human body sensing circuit and part of the scanning circuit in low-power standby mode, so as to maintain the human body sensing and rapid recovery capability with minimal power consumption and avoid invalid deep sleep caused by brief absence. If the user state is judged to be long-term idle, the magnetic axis keyboard is controlled to enter deep sleep mode, only retaining the power supply of the vital sign intention recognition module and the interrupt wake-up function, minimizing the overall static power consumption of the system, and significantly extending the battery life of the keyboard in the no-operation state. During shallow sleep or deep sleep, if the vital sign intention recognition module outputs the user return intention signal, the magnetic axis keyboard is controlled to switch to pre-wake state to partially restore the power supply and initialization of the main control unit and scanning circuit in advance, and silently complete the system readiness before the user actually touches the key, so as to achieve zero-perceptible delay from sleep to response. Furthermore, the magnetic axis keyboard system defines three core sleep modes: shallow sleep, deep sleep, and pre-wake. The state transitions are entirely controlled by the sleep strategy execution module based on the output of the upstream behavior timing analysis module. When the sleep strategy execution module receives a judgment result indicating that the user is in a brief absence state, it immediately enters shallow sleep mode. In this mode, the core processing functions of the main control unit are suspended, and most of its clock domain and power supply are shut down. To ensure rapid recovery and proximity state perception, and to maintain the power supply and basic operation of key sensors in the human body sensing circuit: the 60GHz millimeter-wave radar sensor switches to its basic monitoring mode, maintaining a sampling rate of 2 frames per second for continuous micro-motion sensing; the infrared pyroelectric sensor array maintains a sampling rate of 20 frames per second, executing initial... During the proximity detection, the row and column scanning control chip in the magnetic axis scanning circuit is placed in a minimum power standby state. Its internal oscillator stops oscillating but maintains the register and I / O levels, waiting for a quick wake-up command. Throughout the shallow sleep state, the static operating current is controlled below 2 mA. When the sleep strategy execution module receives a high-confidence judgment result that the user is in a long-term idle state, it executes the deep sleep process, retaining only the circuit power supply necessary to maintain the most basic bio-sensing and hardware interrupt wake-up capabilities. Specifically, except for the real-time clock and a few general-purpose input / output pins used for interrupt detection, the main control unit is completely powered off. In the human body sensing circuit, the infrared pyroelectric sensor array is completely turned off to save power, and the millimeter-wave radar sensor is further equipped with... Set to ultra-low power monitoring mode, its transmission power is reduced and the sampling interval is extended, used only for detecting large-scale approach movements. The entire system's wake-up link is simplified to rely on only two physical paths: one is the hardware interrupt line generated when the millimeter-wave radar detects a strong return signal consistent with biometrics; the other is the interrupt line connected to the physical wake-up button on the keyboard. In this state, the overall static current of the system is targeted to be less than 500 microamps. During shallow or deep sleep, once the decision logic unit in the biometric intent recognition module fuses infrared and millimeter-wave signals to generate and output a high-confidence user return intent signal, the sleep strategy execution module immediately triggers the system to switch to the pre-wake state. The core objective is to ensure that the user receives the system from the moment the first button is pressed. The time difference between responses (first key delay) approaches zero. In this state, the power supply to the main control unit core, magnetic axis scanning circuit and its related peripheral circuits is restored sequentially through the power management unit. This process is completed within 50 milliseconds. Subsequently, the main control unit starts from a low-power state, initializes its system clock and loads a simplified operating system kernel. At the same time, it sends a specific instruction sequence to the row and column scanning control chip through the serial communication interface, commands it to exit the sleep mode and pre-configures its scanning mode to a high-speed progressive scan ready state. The main control unit's own task scheduler completes the warm-up. The critical interrupt service routine and key scanning task are loaded into memory first and are in a standby state. The entire pre-wake-up process ensures that it is fully ready before the user actually sits down and touches the key. The pre-wake-up trigger module performs the following steps: When the sleep strategy execution module controls the magnetic axis keyboard to enter the pre-wake-up state, it immediately sends a wake-up interrupt signal to the main control unit of the magnetic axis scanning circuit and restores its basic clock and power domain to ensure that the main control unit can be activated in real time at the hardware level, providing the core operating foundation for subsequent software initialization and task scheduling. It controls the row and column scanning control chip in the magnetic axis scanning circuit to exit the sleep mode and configures its scanning mode to high-speed progressive scan ready state, so that the magnetic axis scanning circuit enters the low-latency pre-working mode from standby, preparing for a rapid response to the upcoming key signals. It performs preheating initialization of the task scheduler of the main control unit, loads the keyboard mapping table and response processing program, and ensures that the system has completed the hardware and software readiness from sleep to working state before the user actually touches the key. It pre-places the software system in the standby state, eliminates the software startup delay after the system wakes up, and ensures the immediate processing and response of the first key press. Furthermore, when the hibernation strategy execution module determines that the system needs to enter a pre-wake-up state, it first restores the power supply to the main control unit core and various peripheral modules sequentially through the power management unit. The wake-up process of the main control unit begins when a dedicated low-power wake-up interrupt pin receives a high-level valid signal. This signal is set high by the power management integrated circuit inside the hibernation strategy execution module after the output voltage stabilizes (typically with a delay of less than 5 milliseconds). Subsequently, the main control unit activates the main phase-locked loop circuit through an independent low-speed 32.768kHz real-time clock source. The phase-locked loop locks and outputs the system master clock within 1 millisecond, completing the power-on reset sequence of the core power domain. During this process, the main control unit reads the preset boot vector from the static random access memory that maintains power supply during deep hibernation. A streamlined startup loading procedure is executed, necessary system control registers are initialized, and the core voltage is adjusted to the normal operating level. While the main control unit's basic functions are restored, the pre-wake-up process pre-configures the row and column scanning control chip in the magnetic axis scanning circuit via the serial peripheral interface. The main control unit first sends a low-level strobe signal to the chip select pin of this chip, and then sends a specific 4-byte instruction sequence via the serial clock line (frequency preset to 2MHz) and data output line. The first byte of this instruction sequence is the exit sleep mode command code, and subsequent bytes contain mode configuration parameters: setting the scan mode register to high-speed progressive scan mode, configuring the row drive current to 8mA, and setting the blanking time to 10 microseconds. Upon receiving the complete instruction... After execution, its internal 12MHz oscillator starts oscillating, and the row drive and column sampling modules enter standby mode. However, the scan timing generator is not started yet. This configuration process is completed through a hardware state machine. After configuration, the output pin of the scan control chip enters a known high-impedance state, waiting for the main control unit to send the final trigger command to start scanning, ensuring that it is in a low-power ready-to-trigger state before receiving user key presses. After the hardware is ready, the main control unit immediately performs software warm-up initialization, starts the priority-based preemptive real-time kernel task scheduler, and creates and initializes three core tasks: high-speed scan trigger task (highest priority), key signal acquisition and debouncing task, and key value encoding and reporting task. The keyboard is loaded through direct memory access. The mapping table is mapped to a designated area of ​​static random access memory. At the same time, the key response handler (containing the firmware debouncing algorithm, with the debouncing time set to 5 milliseconds) and the interrupt service routine (used to handle timer interrupts and external interrupts) are linked and reside in the kernel executable area. The system timer is configured to generate a periodic tick interrupt of 1ms to drive task scheduling. Finally, the main control unit sends a high-level pulse to the scan enable pin of the scan control chip through the general-purpose input / output pin to officially start the high-speed progressive scan timing. The entire signal link, from the generation of magnetic flux change signal, through the Hall sensor array, the scan control chip, the analog-to-digital converter, to the main control unit's task processing and report output, is initialized and ready before the user actually applies the pressing pressure. The instantaneous response coordination module performs the following steps: After the pre-wake-up trigger module completes hardware readiness, it coordinates the magnetic axis scanning circuit to perform a full-matrix fast self-test scan to initialize the signal output state of all Hall chips, ensure the stability of the Hall sensor array output, eliminate interference from hardware abnormalities to subsequent signal acquisition, and synchronously start the analog-to-digital converter sampling circuit of the signal acquisition module, and perform benchmark calibration and noise filtering initialization to ensure the stability of the acquisition link, improve signal acquisition accuracy and anti-interference capability, and ensure the true reproduction of the key signal. Through the task coordination mechanism of the main control unit, before the user's first key signal arrives, it completes the pipelined scheduling and synchronization between scanning tasks, acquisition tasks and processing tasks, and realizes zero-wait link readiness from key signal generation to system response. Furthermore, after the pre-wake-up trigger module completes hardware power-on and configuration, a full-matrix fast self-test scan is performed. This scan is driven by the row and column scanning control chip, operating in high-speed interval row mode with a row switching cycle set to 5 microseconds, covering all N rows × M columns of Hall sensor units. During the self-test, the main control unit applies a uniform wake-up level to the sleep pins of each Hall chip to put them into working state, and simultaneously configures the wake-up pins to activate the output row by row. The activation duration of each row is 20 microseconds. During this period, the basic output voltage of each Hall chip is read through the column acquisition port. This voltage value should be within the reference range of 1.65V ± 0.05V when there is no magnetic field interference. If a certain channel reads... If the value deviates from the reference by more than ±100mV, it is marked as abnormal and recorded in non-volatile memory. The entire self-test takes no more than (N×20+M×5) microseconds. After completion, all Hall chips return to high-impedance standby state. Simultaneously, when the self-test scan starts, the multi-channel analog-to-digital converter in the signal acquisition module is activated. The analog-to-digital converter adopts a successive approximation architecture, with the reference voltage set to 3.3V, the resolution to 12 bits, and the sampling rate configured to 500kSPS. During the initialization phase, internal reference calibration is first performed: by shorting the input pin to ground, 32 samples are acquired and the zero offset value is calculated and stored in the calibration register. Then, a 1.65V medium-voltage reference source is connected for full-scale calibration. Gain calibration is performed, and each acquisition channel uses a built-in digital filter to suppress power supply and high-frequency interference. The filter is configured as a second-order Butterworth low-pass structure with a cutoff frequency of 2kHz and a stopband attenuation of no less than 40dB. After calibration and filtering parameters are loaded, the analog-to-digital converter enters continuous sampling standby mode. Its timing is triggered by the horizontal synchronization signal of the scan control chip, ensuring synchronization between acquisition and scanning, and establishing a low-noise, high-consistency signal acquisition link. Based on the hardware ready state, the main control unit constructs a pipelined collaboration mechanism between scanning, acquisition, and processing through a real-time task scheduler, creating three tasks with distinct priorities: the scan trigger task (priority 3) is responsible for scheduling tasks in 1-millisecond cycles. The scanning sequence is started in real time; the data acquisition and transfer task (priority 2) transfers the ADC results to the double-buffered memory area in real time through the DMA controller; the key processing task (priority 1) performs real-time debouncing, threshold comparison and key mapping on the buffered data. The tasks are synchronized through event flag groups and semaphores. At the same time, the system timer is configured to generate a 100-microsecond interrupt to monitor task execution timeout and link delay. This pipeline has completed no-load operation verification before the user presses the key for the first time. All tasks are in a suspended waiting event state. Once the Hall chip outputs a valid signal, the entire processing link completes the entire process from signal acquisition to USB report output within a fixed delay of less than 50 microseconds.

[0022] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0023] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A sleep / wake control system for a magnetic axis keyboard integrating human body sensing, characterized in that, include: The behavior timing analysis module is used to comprehensively analyze the user's operating habits and brief absence from the seat of the magnetic axis keyboard. Through timing behavior analysis algorithm and near-range radar micro-motion sensing algorithm, it accurately determines the user's state. The vital signs and intention recognition module is used to continuously monitor the bio-signs and approach posture in front of the magnetic axis keyboard in sleep mode, and integrates infrared pyroelectric sequence matching and millimeter wave bio-sign detection algorithms to identify the user's return intention; The sleep strategy execution module, based on the behavior timing analysis results, triggers sleep mode determination analysis and controls the magnetic switch keyboard to enter the corresponding sleep mode, and only triggers deep sleep when it is confirmed to be idle for a long time; The pre-wake-up trigger module is used to pre-wake up the magnetic axis scanning circuit and the main control unit in advance when the user's approach intention is detected, in combination with the determined sleep mode, so as to shorten the first key response time; The instant response coordination module is used to coordinate the rapid readiness of each module of the magnetic axis keyboard after pre-wake-up, ensuring that magnetic axis scanning, signal acquisition and main control processing are initialized before the user presses the key, thus optimizing the task processing flow.

2. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 1, characterized in that: The behavior timing analysis module includes a user habit modeling unit and a keyboard micro-motion sensing unit; The user habit modeling unit learns and models user keyboard usage habits based on pre-collected user operation habit data and combines time-series behavior analysis algorithms to construct a user behavior model and identify typical operation intervals and seat-leaving patterns. The keyboard micro-motion sensing unit is used to detect centimeter-level micro-motions of the user in real time through a built-in near-range radar micro-motion sensing algorithm, and to determine the user's state by combining the user behavior model, distinguishing between short-term absence and long-term inactivity, thereby reducing false wake-ups.

3. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 2, characterized in that: The user habit modeling unit performs the following steps: User operation data is collected through the timing recording unit built into the magnetic axis keyboard. The user operation data includes key timestamps, key duration, adjacent operation interval sequence, and daily / weekly usage time distribution. Using time-series behavior analysis algorithms, we can perform pattern mining on user operation data to identify user characteristics, including typical operation interval thresholds, high-frequency operation periods, and normal sitting time. Based on the identified user characteristics, a dynamically updated user behavior model is constructed. The output of the user behavior model is used to determine whether the current user is in an active state, a briefly away state, or a long-term idle state.

4. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 2, characterized in that: The keyboard micro-motion sensing unit performs the following steps: The built-in 60GHz millimeter-wave radar sensor collects micro-motion signals in a preset area in front of the magnetic axis keyboard in real time. The micro-motion signals include distance, speed and micro-motion amplitude information. The micro-motion signal is fused and analyzed with the current judgment state output by the user behavior model. If the current judgment is a brief departure state, the micro-motion signal is used to continuously determine whether the user is still in the vicinity of the keyboard operation area. When the micro-motion signal continuously exceeds the first preset duration and is below the activity threshold, and the user behavior model supports the determination of long-term idleness, it is confirmed as a long-term idle state; if the micro-motion signal is detected again within the second preset duration as centimeter-level activity that conforms to human characteristics, it is determined that the user has returned to a nearby area.

5. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 2, characterized in that: The vital sign intention recognition module includes an infrared sequence matching unit and a millimeter-wave vital sign detection unit; The infrared sequence matching unit is used to detect the movement pattern of human heat source in front of the magnetic axis keyboard by using an infrared pyroelectric sensor sequence matching algorithm, analyze the user's return intention, and determine whether the user is close to the keyboard operation area. The millimeter-wave biometrics detection unit is used to detect human biometrics using millimeter-wave radar and to identify the approach and posture intentions of a living person.

6. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 5, characterized in that: The infrared sequence matching unit performs the following steps: A multi-channel infrared pyroelectric sensor array deployed along the upper edge of the keyboard is used to collect heat source signal sequences from the space in front at a fixed sampling period. Dynamic threshold filtering and movement pattern analysis are performed on the heat source signal sequence to extract heat source pattern features including movement direction, speed and spatial trajectory; The heat source pattern features are matched with a pre-stored return intent pattern library. If the matching degree exceeds the set matching threshold, it is determined that there is an intent for the user to approach the keyboard operation area.

7. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 6, characterized in that: The millimeter-wave vital sign detection unit performs the following steps: When the infrared sequence matching unit outputs an indication of approach intent, the millimeter-wave radar is activated to detect human vital signs and collect vital sign signals within the micro-motion range. The vital signs signals are subjected to spectral analysis and respiratory / heartbeat feature extraction to distinguish the movement of a living human body from other heat sources and to identify whether the user's posture is a ready-to-operate posture. By integrating infrared sequence matching results and vital sign recognition results, when both meet the preset return operation criteria, a high-confidence user return intent signal is output to the pre-wake-up trigger module.

8. The integrated human body sensing magnetic axis keyboard sleep / wake control system according to claim 5, characterized in that: The hibernation strategy execution module performs the following steps: The magnetic axis keyboard is divided into sleep modes, including shallow sleep, deep sleep and pre-wake state, and the user state judgment result output by the behavior timing analysis module is received. If it is a brief absence state, the magnetic axis keyboard is controlled to switch to shallow sleep mode to keep the human body sensing circuit and some scanning circuits in low power standby mode. If the user's status is determined to be long-term idle, the magnetic keypad will be controlled to enter deep sleep mode, retaining power only for the vital sign intention recognition module and the interrupt wake-up function; During shallow or deep sleep, if the vital signs intention recognition module outputs a user return intention signal, the magnetic axis keyboard is controlled to switch to the pre-wake state, and the power supply and initialization of the main control unit and scanning circuit are restored in advance.

9. A magnetic axis keyboard sleep / wake control system integrating human body sensing according to claim 8, characterized in that: The pre-wake-up trigger module performs the following steps: When the sleep strategy execution module controls the magnetic axis keyboard to enter the pre-wake state, it immediately sends a wake-up interrupt signal to the main control unit of the magnetic axis scanning circuit and restores its basic clock and power domain. The row and column scanning control chip in the control magnetic axis scanning circuit exits the sleep mode and is configured to high-speed progressive scan ready state. The task scheduler of the main control unit is preheated and initialized, and the keyboard mapping table and response processing program are loaded to ensure that the system is ready in terms of hardware and software from sleep to working state before the user actually touches the key.

10. A magnetic axis keyboard sleep / wake control system integrating human body sensing according to claim 9, characterized in that: The instantaneous response coordination module performs the following steps: After the pre-wake-up trigger module completes hardware readiness, it coordinates the magnetic axis scanning circuit to perform a full matrix fast self-test scan to initialize the signal output state of all Hall chips. The analog-to-digital converter sampling circuit of the signal acquisition module is started synchronously, and reference calibration and noise filtering initialization are performed on it to ensure the stability of the acquisition link; Through the task coordination mechanism of the main control unit, the pipeline scheduling and synchronization between scanning tasks, acquisition tasks and processing tasks are completed before the user's first key press signal arrives.

Images