A radar near-field saturation anti-false alarm control method and device based on touch physical interruption

Abstract

This application discloses a radar near-field saturation false alarm control method and device based on touch physical interruption. The method includes: acquiring touch status bits and mutual capacitance signals, and separating trend and residual components; when the impact judgment condition is met, executing a primary protection action by switching to a blocking level via a hardware pin to cut off the radar transmission power supply; dynamically selecting between maintaining primary protection and executing a signal degradation strategy based on the touch projection area ratio, the latter including restoring power supply and writing a saturation flag bit in the data frame; responding to the jump from the touch effective level to the release level, tracing back to obtain capacitance attenuation characteristics and fitting a comprehensive escape velocity; performing a recovery operation based on the presence or absence of a saturation flag bit, including clearing the flag bit to restore FFT processing, or restoring power supply after calculating the safe backoff time using the escape velocity. This application achieves adaptive control for radar near-field saturation false alarm prevention.

Description

Technical Field

[0001] This application relates to the fields of human-computer interaction and radar anti-interference technology for smart terminals, specifically to a radar near-field saturation anti-misjudgment control method and device based on touch physical interruption. Background Technology

[0002] In smart terminals equipped with large displays, millimeter-wave radar is often hidden behind the screen to achieve automatic energy-saving control by turning on the screen when someone approaches and turning it off when they leave. When a user touches or swipes the screen, their hand is extremely close to the radar's transceiver antenna, generating extremely strong electromagnetic wave reflections. This causes the analog-to-digital converter at the radar's receiving end to exceed its linear dynamic range, resulting in signal clipping distortion and a sharp increase in noise floor, i.e., near-field radio frequency saturation. This phenomenon can cause the radar's digital signal processing unit to misinterpret it as "target disappearance" or "severe interference," thus triggering a foolproof mechanism to forcibly turn off the screen, severely impacting the user experience.

[0003] Existing technologies attempt to address this issue through physical hardware interruption: when a touch is detected, the radar enable signal is directly lowered to forcibly stop transmission, and a fixed delay (e.g., 150 milliseconds) is set after the finger is removed before resuming. However, different users have vastly different hand-removal speeds: a fixed short delay when removing the hand slowly will cause the palm to remain in the near-field strong reflection zone, causing the radar to saturate again after recovery; if the delay is set too long, the response will be sluggish when removing the hand quickly.

[0004] Therefore, there is an urgent need for a solution that can adaptively and dynamically adjust the timing of radar recovery based on the speed of withdrawal. Summary of the Invention

[0005] This application aims to provide a radar near-field saturation anti-misjudgment control method and control device based on touch physical interruption, so as to solve the problem of improper radar recovery timing caused by the difference in user hand withdrawal speed in the prior art, and achieve the technical effect of avoiding secondary saturation when the hand is withdrawn slowly and responding quickly when the hand is withdrawn quickly.

[0006] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, a radar near-field saturation anti-false judgment control method based on touch physical interruption is provided, which is applied to an electronic device including a main control microprocessor, a touch acquisition unit and radar components, and includes the following steps: S1. Obtain the touch status bit signal and mutual capacitance signal output by the touch acquisition unit, and separate the trend component located in the first frequency range and the residual component located in the second frequency range from the mutual capacitance signal; when the trend component and the residual component meet the preset impact judgment condition, perform the primary protection action, that is, control the hardware pin to switch to the blocking level state to cut off the power supply to the transmitting circuit of the radar component. S2. In response to the primary protection action, based on the proportion of the projection area of ​​the two-dimensional touch coordinates in the radar field of view, dynamically select between maintaining the primary protection action and executing a signal degradation strategy; the signal degradation strategy is: switching to the enable level state to restore the power supply of the transmitting circuit, and writing a saturation flag bit in the data frame sent to the radar component to make the radar component enter the abnormal data discarding mode. S3. In response to the touch status bit signal changing from a touch valid level representing the presence of touch to a touch release level representing the disappearance of touch, the capacitance decay characteristics within a preset time period are obtained with the time of the change as the end point, and a weighted fusion calculation is performed using a preset recursive weight matrix to fit the comprehensive escape speed of the hand withdrawal. S4. Perform the corresponding recovery operation based on whether the saturation flag bit exists: If present, the saturation flag is cleared, thereby removing the abnormal data discarding mode and restoring the fast Fourier transform processing of the original data frame of the radar component. If it does not exist, the theoretical safe backoff time is calculated using the comprehensive escape velocity, and the hardware pin is switched to the enable level state after the theoretical safe backoff time expires.

[0007] In some embodiments, the first frequency range is 0 Hz to 2 Hz; the second frequency range is 10 Hz to 50 Hz. The impact determination condition is: the trend component exceeds the environmental baseline tolerance threshold, and the numerical change of the residual component in a single sampling period exceeds a preset threshold.

[0008] In some embodiments, the dynamic selection logic is as follows: if the projection area ratio reaches or exceeds the first occlusion threshold, the primary protection action is maintained; otherwise, the signal degradation strategy is executed. The abnormal data discarding mode specifically involves: when the radar component resolves the saturation flag bit, it executes abnormal frame discarding logic and stops performing Fast Fourier Transform processing on the original data of the current frame.

[0009] In some embodiments, obtaining the overall escape velocity includes: acquiring the attenuation slope of the mutual capacitance signal and the area shrinkage rate of the contact area within a preset time period before the touch effective level jumps to the touch release level; using a preset recursive weight matrix to weight and fuse the attenuation slope of the mutual capacitance signal and the area shrinkage rate to calculate a spatial displacement vector characterizing the hand withdrawal speed, which is used as the overall escape velocity.

[0010] In some embodiments, the calculation of the theoretical safe backoff time using the comprehensive escape velocity includes: using the full-scale input voltage value of the analog-to-digital converter of the radar component receiver as a reference upper limit, cyclically performing numerical simulations for the expected value of the echo voltage at the delay time; when the expected value of the echo voltage is less than the full-scale input voltage value, and the absolute value of the numerical difference between two adjacent simulation cycles is less than a preset convergence threshold, the sum of the step times accumulated in the current cycle is determined as the theoretical safe backoff time.

[0011] Secondly, this application provides a radar near-field saturation anti-false judgment control device based on touch physical interruption, including a touch acquisition unit, a radar component, and a main control microprocessor. The control device implements the method described above through logical cooperation between the components, wherein: The touch acquisition unit is configured as the data source of the control device, used to provide the mutual capacitance signal and the touch status bit signal to the main control microprocessor, and includes a storage unit for temporarily storing the saturation flag bit; The radar component is configured as a controlled terminal of the control device, used to perform data frame discarding according to the state of the saturation flag bit, or to cut off beam transmission by responding to the physical pin level of the main control microprocessor through its hardware enable terminal. The main control microprocessor is configured as the control center of the control device and is electrically connected to the touch acquisition unit and the radar component, respectively.

[0012] Furthermore, the main control microprocessor is equipped with a coordinate mapping logic unit, which is used to store the relative position offset parameters between the touch acquisition unit and the radar component, and to map the two-dimensional coordinates in the original touch data to the mask area in the field of view of the radar component, so that the main control microprocessor can calculate the projection area ratio.

[0013] Furthermore, a hardware interrupt link is formed between the hardware pins of the main control microprocessor and the enable pin of the radar component; when executing the primary protection action, the main control microprocessor outputs the blocking level state through the hardware pin to cut off the power supply to the transmitting circuit of the radar component within a microsecond time; when executing the signal degradation strategy or meeting the power supply restoration conditions, the main control microprocessor outputs the enable level state through the hardware pin to restore the power supply to the transmitting circuit.

[0014] Compared to existing technologies, this application discloses a radar near-field saturation anti-false alarm control method and device based on touch physical interruption. The method includes: acquiring touch status bit signals and mutual capacitance signals, separating trend components and residual components; when both meet the impact judgment conditions, executing a primary protection action, controlling the hardware pin to switch to the blocking level state to cut off radar transmission; based on the proportion of the projection area of ​​the touch two-dimensional coordinates within the radar field of view, selecting to maintain the primary protection action or execute a signal degradation strategy, i.e., resuming transmission and writing a saturation flag bit in the data frame; responding to the touch status bit switching from an active level to a release level, acquiring capacitance attenuation characteristics, and fitting the comprehensive escape velocity using a recursive weight matrix; performing recovery based on the presence or absence of a saturation flag bit: if present, clearing the flag bit and resuming Fast Fourier Transform processing; if absent, calculating the theoretical safe backoff time using the comprehensive escape velocity, and switching the hardware pin to the enable level state after the timeout. This achieves adaptive control for radar near-field saturation anti-false alarm. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort. Figure 1 This is a schematic diagram of the hardware architecture of a radar near-field saturation anti-false judgment control device based on touch physical interruption, provided as an embodiment of this application.

[0016] Figure 2 This is a flowchart illustrating a radar near-field saturation anti-false alarm control method based on touch physical interruption, provided as an embodiment of this application.

[0017] Figure 3 This is a hardware timing diagram illustrating the triggering of a primary protection action according to an embodiment of this application.

[0018] Figure 4 This is a schematic diagram illustrating the principle of capacitor attenuation feature extraction based on tracing back a preset time from the jump moment, as provided in one embodiment of this application.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0021] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0022] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0023] To facilitate understanding of the execution flow and hardware foundation of the control method of this application, the control device implementing the method is first described. This device is the physical carrier for implementing steps S1 to S4, and the signal interaction and control logic between its constituent units constitute the underlying support for the operation of the method of this application. The following section combines... Figure 1 The specific structure of the device and the function of each unit are described in detail.

[0024] like Figure 1 As shown, the radar near-field saturation anti-false judgment control device based on touch physical interruption provided in this embodiment includes: touch acquisition unit 100, radar component 200 and main control microprocessor 300.

[0025] The touch acquisition unit 100, configured as the data source of the device, provides mutual capacitance signals and touch status bit signals to the main control microprocessor 300. Specifically, the touch acquisition unit 100 is typically integrated into the touchscreen driver chip, periodically scanning the mutual capacitance matrix of the entire screen to generate the original mutual capacitance signals and producing a touch status bit signal reflecting whether a finger is currently touching the screen (e.g., a high level indicates a touch, and a low level indicates no touch). The touch acquisition unit 100 also includes a storage unit 110 (e.g., a small SRAM or register), which is configured to temporarily store a saturation flag bit written by the main control microprocessor 300. This saturation flag bit is a single-bit data used to indicate that the data frame currently received by the radar component 200 may be contaminated by near-field reflections.

[0026] The radar component 200, configured as the controlled terminal of the device, is used to discard data frames based on the state of the saturation flag bit, or to cut off beam transmission in response to the physical pin level of the main microprocessor 300 via its hardware enable pin. Specifically, the radar component 200 includes a millimeter-wave radar transceiver front-end (containing a voltage-controlled oscillator (VCO), power amplifier (PA), low-noise amplifier (LNA), and analog-to-digital converter (ADC)) and an internal digital signal processing unit (DSP). The radar component 200 receives physical level control signals from the main microprocessor 300 via its hardware enable pin (EN pin): when the enable pin is at a blocking level, the power supply to the radar's transmitting circuit is cut off, and electromagnetic wave transmission stops; when the enable pin is at an enabling level, the power supply to the transmitting circuit is restored, and the radar operates normally. Furthermore, the radar component 200 interacts with the main microprocessor 300 via a serial peripheral interface (SPI) or an internal integrated circuit bus (I2C), receiving configuration commands and uploading raw radar data frames. The radar component 200 is configured to automatically execute the abnormal frame discarding logic when a saturation flag bit is detected in the header of the received data frame. That is, it does not perform Fast Fourier Transform (FFT) and subsequent target detection algorithms on the original data of the current frame, but directly discards the data of the frame and maintains the target tracking result of the previous frame.

[0027] The main control microprocessor 300, configured as the control center of the device, is electrically connected to both the touch acquisition unit 100 and the radar component 200. Specifically, the main control microprocessor 300 can be a single-chip microcomputer (MCU) or an embedded processor. It is connected to the touch acquisition unit 100 via a serial bus (such as I2C or SPI) to read mutual capacitance signals and touch status bit signals at fixed time intervals (e.g., every 5 milliseconds). The main control microprocessor 300 is connected to the radar component 200 via two interfaces: one is a general-purpose input / output (GPIO) pin, directly connected to the hardware enable pin of the radar component 200, used to output a blocking or enabling level; the other is a serial data interface (such as SPI), used to send configuration commands and data frame header tags to the radar component 200.

[0028] Furthermore, the main control microprocessor 300 is equipped with a coordinate mapping logic unit 310. This coordinate mapping logic unit 310 is used to store the relative position offset parameters (i.e., a spatial rigid body transformation matrix containing translation and scaling parameters) between the touch acquisition unit 100 and the radar component 200, and to map the two-dimensional coordinates in the original touch data to a mask area within the field of view of the radar component 200, so that the main control microprocessor 300 can calculate the projection area ratio.

[0029] Furthermore, a hardware interrupt link is formed between the hardware pins (i.e., GPIO pins) of the main control microprocessor 300 and the enable pin of the radar component 200. When performing primary protection actions, the main control microprocessor 300 outputs a blocking level state through the hardware pins to cut off the power supply to the transmitting circuit of the radar component 200 within a microsecond time. When performing signal degradation strategies or meeting the power restoration conditions, the main control microprocessor 300 outputs an enable level state through the hardware pins to restore the power supply to the transmitting circuit.

[0030] In some embodiments, the main control microprocessor 300 also integrates a hardware timer for timing when precise delay is required.

[0031] Based on the aforementioned control device, this application provides a radar near-field saturation anti-false alarm control method based on touch physical interruption. This method sequentially executes the following steps S1 to S4. Figure 2 The overall flow of the method is shown. The method includes the following steps: S1. Obtain the touch status bit signal and mutual capacitance signal output by the touch acquisition unit, and separate the trend component located in the first frequency range and the residual component located in the second frequency range from the mutual capacitance signal; when the trend component and the residual component meet the preset impact judgment conditions, perform the primary protection action, that is, control the hardware pin to switch to the blocking level state to cut off the power supply to the transmitting circuit of the radar component.

[0032] Please see Figure 3 The diagram illustrates the hardware timing of the primary protection action triggering in this embodiment. In this embodiment, the hardware pins (such as GPIO ports) of the main control microprocessor are directly coupled to the enable pin (EN pin) of the radar component via a hardware interrupt link.

[0033] like Figure 3 As shown, the physical jump response time T from the moment the strike determination condition is met to the moment the hardware pin outputs the blocking level signal is strictly controlled to the microsecond level (e.g., less than 10μs). This high-speed physical interruption mechanism ensures that beam transmission is shut down at the source before the radar component generates saturated echo data packets and sends them to the resolution link, thereby achieving real-time physical isolation from near-field saturation interference.

[0034] Specifically, the main control microprocessor 300 continuously reads the mutual capacitance matrix data output by the touch acquisition unit 100 via a serial bus (such as I2C or SPI) at a fixed sampling period (e.g., once every 5 milliseconds), and simultaneously reads the touch status bit signal. To eliminate the interference of slow changes in the environment such as temperature drift and water stains on touch detection, the main control microprocessor 300 first performs dynamic baseline calibration on the mutual capacitance signal: taking the average value of the capacitance matrix within a relatively long time window (e.g., 1000 milliseconds) in the past as the real-time static background baseline of the current environment, and subtracting this baseline from the currently sampled mutual capacitance matrix to obtain the net capacitance change matrix.

[0035] Based on this, the main control microprocessor 300 extracts the most significant regions of change (i.e., touch extreme points) from the net capacitance change matrix, forming a one-dimensional capacitance intensity time series signal that varies with time. Then, a preset first-order or second-order infinite impulse response (IIR) digital filter is used to perform frequency decomposition on this sequence, separating it into two components: Low-frequency trend component: This corresponds to signal components with frequencies below 2 Hz (i.e., the first frequency range of 0-2 Hz). Physically, this component reflects slow-changing events such as slow drift in ambient temperature and humidity, water stains spreading on a screen, or a large area of ​​a human torso slowly approaching.

[0036] High-frequency residual components: These correspond to signal components with frequencies higher than 10 Hz (i.e., the second frequency range of 10-50 Hz). This component only produces significant abrupt changes when the hand strikes, clicks, or swipes the screen at a relatively fast speed.

[0037] After the decomposition is completed, the main control microprocessor 300 performs dual-channel threshold comparison in parallel: First threshold test: Determine whether the current low-frequency trend component exceeds the preset environmental baseline tolerance threshold (e.g., 50 femtoseconds). This condition confirms that there is indeed a macroscopic conductive object (such as a human body) in front of the screen, excluding pure radio frequency noise.

[0038] Second threshold test: Within an extremely short microsecond time window (e.g., 10 microseconds to 50 microseconds), calculate the instantaneous rate of change (first-order difference value) of the high-frequency residual components. If this rate of change exceeds the preset transient impact slope threshold (e.g., 30 flyweights / sampling period), it indicates that a conductive object is pressing against the screen at a relatively fast speed and at a very close distance.

[0039] Only when both conditions are met simultaneously does the main control microprocessor 300 determine that the current event is a genuine, near-field saturation-damaging finger touch. At this point, to prevent the radar receiver from saturating due to strong reflected signals after tens of microseconds, the main control microprocessor 300 bypasses all operating system-level task scheduling and directly outputs a blocking level (low level) through its GPIO pin. This GPIO pin is directly connected to the hardware enable pin (EN pin) of the radar component 200. The low level instantly cuts off the power supply to the voltage-controlled oscillator (VCO) and power amplifier (PA) inside the radar component 200, causing the radar to stop emitting electromagnetic waves. The entire physical blocking process is completed within microseconds, much faster than the time it takes for the radar signal to travel from transmission to reception of the reflected echo.

[0040] S2. In response to the primary protection action, based on the proportion of the projection area of ​​the touch two-dimensional coordinates in the radar field of view, dynamically select between maintaining the primary protection action and executing a signal degradation strategy; the signal degradation strategy is: switching to the enable level state to restore the power supply of the transmitting circuit, and writing a saturation flag bit in the data frame sent to the radar component to make the radar component enter the abnormal data discard mode.

[0041] Specifically, while the user's finger is continuously pressing or swiping the screen, the main control microprocessor 300 cannot simply maintain the radar in a power-off state; otherwise, if the user only operates at the edge of the screen, excessive radar shutdown will lead to loss of target tracking upon subsequent recovery. Therefore, this step introduces a graded protection mechanism based on spatial occlusion assessment.

[0042] The main control microprocessor 300 periodically reads the absolute two-dimensional pixel coordinates (X, Y) of all current touch points and the capacitive coverage area parameters representing the pressing area from the registers of the touch acquisition unit 100. Then, it calls the spatial rigid body transformation matrix (containing translation and scaling parameters) pre-installed in the flash memory. This matrix describes the geometric correspondence between the physical two-dimensional plane of the touch screen and the three-dimensional beam field of view of the radar transceiver antenna array installed behind the screen. Through linear projection calculation, the main control microprocessor 300 maps the two-dimensional touch coordinates and contact area to the front of the radar antenna array, and calculates the physical obstruction projection contour of the current palm on the radar antenna radiation surface.

[0043] Next, the main control microprocessor 300 calculates the area percentage of the radar antenna radiating surface covered by the projected outline, and makes dynamic decisions based on this percentage: If the projected area ratio reaches or exceeds the first occlusion threshold (e.g., 80%), it is determined to be a high-confidence global occlusion. At this time, the low-noise amplifier and analog-to-digital converter of the radar receiver front-end have most likely experienced severe clipping distortion. Therefore, the main control microprocessor 300 maintains the primary protection action in step S1, that is, the GPIO pin continues to maintain the blocking level, and the radar transmitting circuit is continuously disconnected. During this period, the radar's digital signal processing unit is in an idle waiting state with no data input, completely shielding the erroneous point cloud caused by ADC overflow.

[0044] If the projected area percentage is lower than the first occlusion threshold (e.g., only 5%-15%), it is determined to be a slight edge occlusion with low confidence. At this time, the radar is not completely disabled, and some antenna channels can still provide valid signals. Therefore, the main control microprocessor 300 executes a signal degradation strategy: first, it switches the GPIO pin to the enable level (high level) to restore power to the radar transmitting circuit; then, it forcibly writes a single-bit saturation flag bit (e.g., set to 1) into the reserved bits of the protocol header of the current data frame sent to the radar component 200 via a serial interface (e.g., SPI). When the digital signal processing unit inside the radar component 200 parses the data frame, once it detects this flag bit, it executes the abnormal frame discarding logic: it does not perform Fast Fourier Transform (FFT) and constant false alarm rate target detection on the original data of the current frame, directly discards the frame data, and maintains the target tracking result of the previous frame. This soft-marking strategy avoids contaminated echoes being misidentified as false targets and does not require completely shutting down the radar, thus ensuring the ability to recover quickly afterwards.

[0045] S3. In response to the touch status bit signal changing from a touch valid level representing the presence of touch to a touch release level representing the disappearance of touch, the capacitance decay characteristics within a preset time period are obtained with the time of the change as the end point, and a weighted fusion calculation is performed using a preset recursive weight matrix to fit the comprehensive escape speed of the hand withdrawal.

[0046] Please see Figure 4 This diagram illustrates the principle of capacitance decay feature extraction based on tracing back a preset time period from the jump moment in this embodiment of the application. The main control microprocessor uses the jump moment T_jump as the time endpoint to obtain the capacitance decay features within a preset tracing period (e.g., 50ms-100ms). Figure 4 As shown in the observation window indicated by the dashed box, this feature specifically includes the attenuation slope of the mutual capacitance signal and the area shrinkage rate of the contact region. The main control microprocessor uses a preset recursive weight matrix to perform weighted fusion calculations to fit a comprehensive escape velocity reflecting the physical characteristics of the touch release. Through this 'backtracking sampling' method, this application solves the sampling distortion problem caused by electrical signal fluctuations at the moment of touch release in traditional methods, significantly improving the accuracy of velocity prediction.

[0047] Specifically, when a user lifts their finger from the screen surface, the touch status bit signal of the touch acquisition unit 100 jumps from an active level (e.g., high level) representing the presence of a touch to a release level (e.g., low level) representing the disappearance of the touch. In response to this jump (usually triggered by a hardware interrupt), the main control microprocessor 300 retrieves historical capacitance data from its internal static random access memory (SRAM) ring buffer within a preset very short time window (e.g., the past 50 milliseconds) before the jump, with the jump time T_jump as the time endpoint.

[0048] From this continuous time series, the main control microprocessor 300 extracted two features that directly characterize the kinematic velocity of hand withdrawal: The attenuation slope of the mutual capacitance signal is calculated as the first-order difference of the net mutual capacitance value between each sampling period. The absolute value of this attenuation slope physically corresponds to the instantaneous velocity at which the finger is withdrawn along the screen normal (Z-axis). The faster the finger is withdrawn, the steeper the attenuation slope.

[0049] Contact area shrinkage rate: The rate of change in the area of ​​the bounding rectangle of the two-dimensional touch projection contour within the calculation time window. This shrinkage rate maps the lateral escape velocity of the entire three-dimensional volume of the palm as it leaves the radar beam coverage area (XY plane).

[0050] The two eigenvalues ​​are combined to form a transient feature input vector for the current moment. To achieve millisecond-level trajectory extrapolation on microprocessors with limited computing power (typically with clock speeds of only tens to hundreds of megahertz), this step employs a discrete-time recursive state mapping unit with extremely low computational overhead. The mathematical essence of this unit is equivalent to a lightweight echo state network mechanism: a fixed, high-dimensional sparsely coupled weight matrix is ​​pre-programmed into read-only memory (Flash memory) (internal node connections are sparse and weights are fixed, without online learning). The microprocessor multiplies the input vector with the input weight matrix and injects it into a dynamic state pool composed of this high-dimensional sparse matrix. The state pool iterates in discrete time steps to generate transient nonlinear responses to the timing fluctuations of the input capacitance. Subsequently, the internal state vectors in the current state pool are extracted, and only one linear weighted summation operation is performed (i.e., the state vector is multiplied by the one-dimensional output weight vector). This output weight vector has been pre-trained and embedded in the microprocessor during the product calibration phase.

[0051] Through this hardware-level minimalist matrix operation with "fixed internal weights and only linear mapping of the output layer," the main control microprocessor can map the nonlinear temporal decay characteristics of the capacitor into a specific physical prediction value within a few microseconds: the three-dimensional spatial comprehensive escape velocity (denoted as V_escape) of the hand leaving the screen. This velocity comprehensively reflects the speed of finger withdrawal in both the normal and tangential directions.

[0052] S4. Perform the corresponding recovery operation based on whether the saturation flag bit exists: If present, the saturation flag is cleared, thereby removing the abnormal data discarding mode and restoring the fast Fourier transform processing of the original data frame of the radar component. If it does not exist, the theoretical safe backoff time is calculated using the comprehensive escape velocity, and the hardware pin is switched to the enable level state after the theoretical safe backoff time expires.

[0053] Specifically, after obtaining the overall escape velocity V_escape, the main control microprocessor 300 first checks whether there is a saturation flag bit (i.e. whether the signal degradation strategy was executed in step S2).

[0054] Branch 1: Saturation flag exists This indicates that during the finger touch, the system is in a state of slight edge occlusion, and the radar transmitting circuit is never physically disconnected (or has been restored after a brief disconnection). Therefore, no additional delay is required after the user removes their hand. The main control microprocessor 300 directly clears the saturation flag bit stored in the touch acquisition unit 100's storage unit 110 (essentially, this flag bit is no longer written in subsequent data frames sent to the radar component 200). When the digital signal processing unit of the radar component 200 parses the next frame and finds no flag bit, it automatically deactivates the abnormal data discarding mode and resumes normal Fast Fourier Transform processing for all original data frames. The entire process is completed within one radar frame period (typically tens of milliseconds), achieving seamless recovery.

[0055] Branch 2: No saturation flag bit exists This indicates that during the finger touch, the system is in a state of high-confidence global blockage, and the radar transmission circuit is physically disconnected. If radar transmission were immediately resumed at this time, the user's palm might still be within the near-field high-reflection zone, only millimeters to centimeters from the antenna, causing the radar to saturate again. Therefore, a dynamic safe backoff time must be calculated.

[0056] The main control microprocessor 300 first reads a core physical parameter from the radar chip's hardware specifications: the maximum undistorted peak input voltage (denoted as V_adc_max) of the radar receiver front-end analog-to-digital converter (ADC). This threshold is an inherent property of the radar analog circuitry and represents the maximum echo signal strength that the radar can linearly process.

[0057] Then, the main control microprocessor 300 takes the comprehensive escape velocity V_escape and the preset radar near-field saturation critical distance d_safe (e.g., 10 cm, which is physically deduced from the maximum input power of the radar front-end low-noise amplifier and the antenna gain) as the initial values, and uses the method of cyclic numerical simulation to calculate the theoretical safe avoidance time. The specific process is as follows: Set an initial delay time T = 0 and a step size ΔT = 1 ms.

[0058] In each loop, calculate the predicted distance of the hand withdrawal after the current delay time T. d = V_escape × T.

[0059] According to the radar equation, inversely calculate the theoretical peak voltage V_echo(d) of the echo signal received by the radar when the hand is at a distance d.

[0060] Compare V_echo(d) with V_adc_max: If V_echo(d) ≥ V_adc_max, increase the step size by T = T + ΔT and continue the loop.

[0061] If V_echo(d) < V_adc_max, record the current T value, and further check whether the absolute value of the difference in echo voltages between two adjacent simulation periods (T and T - ΔT) is lower than the preset convergence threshold (e.g., 1 mV). If the convergence condition is met, the current T value is determined as the theoretical safe avoidance time T_safe.

[0062] The above cyclic simulation usually only needs dozens of iterations to complete (millisecond-level time-consuming), and the computational amount is extremely small. In order to further filter out the single prediction fluctuations caused by physiological jitter or environmental noise, the main control microprocessor 300 can also use the numerical iterative weighted average algorithm (equivalent to lightweight particle filtering) to smooth the T_safe predicted in multiple consecutive sampling periods and output the final adaptive avoidance delay time T_dynamic_delay.

[0063] After obtaining T_dynamic_delay, the main control microprocessor 300 starts its internal hardware timer (or uses the system tick for timing), and sets the delay to T_dynamic_delay. During the delay period, the GPIO pin maintains the blocking level and the radar remains silent. When the timer expires, the main control microprocessor 300 automatically raises the GPIO pin from the low level to the enabling level, and the radar transmitting circuit resumes power supply and re-transmits electromagnetic waves. Since the user's palm has completely withdrawn from the near-field strong reflection physical dead zone according to the optimal timing at this time, the intensity of the first frame of echo signal received by the radar is perfectly controlled within the linear dynamic range of the ADC, thus achieving safe and seamless recovery at any hand withdrawal speed.

[0064] To more clearly demonstrate the working process of this method in actual equipment, the following five specific application scenario examples are provided for detailed explanation. Scenario 1: A child quickly taps the screen of a smart refrigerator. Scene description: In front of a smart refrigerator equipped with a 21.5-inch touchscreen, a child is rapidly tapping cartoon icons on the screen. The child's finger touches and quickly leaves the screen at a high speed (approximately 10 milliseconds to complete the tap).

[0065] Phase S1: The main control microprocessor reads the mutual capacitance signal output by the touch acquisition unit via the serial bus. After dual-frequency decomposition, the low-frequency trend component slowly rises due to the child's proximity and exceeds the environmental baseline tolerance threshold (e.g., 50 fF); simultaneously, the high-frequency residual component's value change increases dramatically within a single sampling period (10 microseconds), far exceeding the transient impact slope threshold (e.g., 30 fF / sampling period). With both threshold conditions met, the main control microprocessor determines it as a genuine finger strike event. To prevent the impending strong reflection from causing radar saturation, the main control microprocessor immediately switches the GPIO pin to a blocking level (low level), directly cutting off the power supply to the radar component's transmitting circuit. The entire blocking process is completed within microseconds.

[0066] Phase S2: During the brief period of continuous finger pressure on the screen (approximately 30 milliseconds), the main control microprocessor acquires the two-dimensional coordinates and contact area of ​​the touch point, and calculates the projected area percentage through the coordinate mapping logic unit. Since the click is located in the center of the screen and has a small area, the percentage exceeds 85%, which is higher than the preset first occlusion threshold (80%). Therefore, the system chooses to maintain the primary protection action, the GPIO pins continue to maintain the blocking level, and the radar remains silent.

[0067] In phase S3: the child quickly lifts their finger off the screen, and the touch status signal transitions from an active level (high) to a release level (low). The main microprocessor uses this transition moment T_jump as the endpoint and retrieves historical capacitance data from the previous 50 milliseconds. The extracted features show that the mutual capacitance signal has an extremely steep attenuation slope (approximately -200 fF / ms) and an extremely high contact area shrinkage rate (approximately -150 mm² / ms). These two feature values ​​are input into the discrete-time recursive state mapping unit, and weighted fusion is performed using a preset recursive weight matrix, outputting a comprehensive escape velocity V_escape = 1.5 m / s within a few microseconds.

[0068] Phase S4: Since Phase S2 maintains the primary protection action (global blocking) and never writes a saturation flag, there is no saturation flag. The main microprocessor executes the recovery branch: using the radar ADC full-scale input voltage (e.g., 1V) as the upper limit, it iteratively simulates the expected echo voltage value. Each step increases by 1 millisecond, calculating the echo voltage at the corresponding distance. When the delay time reaches 67 milliseconds, the echo voltage drops below full scale and the voltage difference between adjacent steps is less than the convergence threshold (1mV). 67 milliseconds is determined as the theoretical safe backoff time. The main microprocessor starts an internal timer, automatically pulling the GPIO pin high to the enable level after 67 milliseconds, and the radar resumes transmission. At this time, the child's finger is far from the near field, the radar receives the signal within the linear range, and the system responds quickly.

[0069] Scenario 2: Slow swiping and releasing for users with limited mobility Scenario description: In front of the smart control screen on the wall, a user with limited mobility slowly slides the curtain control bar on the screen and withdraws his hand at an extremely slow speed. The whole process takes about 600 milliseconds.

[0070] Phase S1: The user's finger slowly touches the screen and begins to slide. The low-frequency trend component gradually increases, and the high-frequency residual component also triggers the threshold (because there is still a relatively fast speed component when the finger finally presses firmly on the screen). When the dual threshold conditions are met, the main control microprocessor immediately executes the primary protection action, the GPIO outputs a blocking level, and the radar stops transmitting.

[0071] Phase S2: When the user slides their finger over a large area, the coordinate mapping calculation shows that the projected area accounts for more than 80%. The system maintains basic protective actions, and the radar remains physically silent throughout the entire sliding process.

[0072] Phase S3: The user withdraws their finger at an extremely slow speed (approximately 600 milliseconds). After detecting the touch release level transition, the main control microprocessor extracts the capacitance history data from the previous 50 milliseconds and finds that the mutual capacitance signal attenuation slope is extremely gentle (approximately -5fF / ms) and the area shrinkage rate is extremely low (approximately -2mm² / ms). The recursive weight matrix outputs a comprehensive escape velocity V_escape = 0.15m / s.

[0073] Phase S4: No saturation flag is present; the recovery branch is executed. Numerical simulation loop: With d_safe=10cm as the target, the theoretical backoff time T=0.1m / 0.15m / s≈667ms is calculated. Step-by-step simulation confirms that the echo voltage is below full scale after 667 milliseconds, which is determined to be the theoretical safe backoff time. The main control microprocessor delays for 667 milliseconds before pulling high the GPIO pin, and the radar resumes transmission. At this time, the user's hand has completely withdrawn from the near-field strong reflection zone, avoiding the secondary saturation crash caused by a fixed short delay (e.g., 150 milliseconds) for recovery. If a fixed long delay (e.g., 1000 milliseconds) is used, the system response is sluggish; this method achieves adaptive adjustment.

[0074] Scenario 3: Slight edge obstruction; wipe the screen edge with a damp cloth. Scenario description: A cleaning staff member wipes the edge of the refrigerator screen with a damp cloth. The cloth occasionally sweeps across the touch-sensitive area, but the area it covers is very small, and the radar is not fully saturated.

[0075] S1 phase: Touching with a wet cloth triggers primary protection, the GPIO pin switches to blocking level, and the radar briefly stops transmitting.

[0076] In Phase S2: The main control microprocessor calculates that the projected area ratio is only 5%, which is lower than the first occlusion threshold (80%). Therefore, a signal degradation strategy is implemented: First, the GPIO pin is switched to the enable level (high) to restore power to the radar transmitting circuit; then, a single-bit saturation flag bit (set to 1) is written into the reserved bits of the protocol header of the current data frame sent to the radar component via the SPI interface. After the digital signal processing unit inside the radar component parses the flag bit, it automatically executes the abnormal frame discarding logic: it does not perform FFT and constant false alarm rate detection on the original data of the current frame, directly discards the frame, and maintains the target tracking result of the previous frame. This avoids false targets (such as ghost obstacles) caused by edge reflections, and does not require completely shutting down the radar, thus preserving the radar's continuous perception capability of the background environment.

[0077] S3 stage: After the cloth leaves the screen, since the radar is never physically disconnected (only briefly powered off and then immediately restored), there is no need to perform hand release trajectory prediction after the touch release event occurs.

[0078] Phase S4: The main control microprocessor clears the saturation flag bit (it will not be written to in subsequent data frames). The radar component detects the disappearance of the flag bit in the next frame cycle, automatically deactivates the abnormal data discarding mode, and resumes normal FFT processing for all original data frames. The entire process is seamless, and the radar's detection of the presence of people in the surrounding area is not interfered with.

[0079] Scenario 4: Capacitive stylus writing scenario Scenario Description: On a large conference tablet, a user uses a thin capacitive stylus to write or annotate. The contact area of ​​the stylus tip is much smaller than that of a finger, resulting in extremely weak electromagnetic wave reflection, insufficient to saturate the radar ADC.

[0080] S1 stage: Capacitive pen tip touch triggers primary protection (high frequency sudden change will still be detected by dual frequency decomposition), GPIO pin switches to blocking level, radar briefly stops transmitting.

[0081] In Phase S2: The main control microprocessor calculates that the projected area is almost zero, and the contact area is much smaller than the preset minimum finger area threshold (e.g., 20mm²). Under these special conditions, the system is configured not to maintain primary protection actions or execute signal degradation strategies, but instead immediately switches the GPIO pins to the enable level and resumes radar transmission. This means skipping the S2 intervention.

[0082] Phases S3 and S4: No triggering. The radar component remains operational throughout the capacitive pen writing process, continuously monitoring the presence of other personnel in the meeting room. When the user leaves the meeting room, the radar immediately detects this and triggers a screen-off event. This embodiment demonstrates the application's intelligent recognition and differentiated processing capabilities for different interaction tools (finger vs. capacitive pen).

[0083] Scenario 5: Realistic Touch Control in a High-Temperature and High-Humidity Kitchen Environment Scenario Description: In a hot and humid kitchen environment, a user operates a range hood with a screen. Moisture or grease may adhere to the screen surface, causing the overall mutual capacitance signal baseline to slowly rise.

[0084] Phase S1: Ambient temperature and humidity cause a slow drift in the mutual capacitance signal baseline. After dual-frequency decomposition, this slow change is completely attributed to the low-frequency trend component, but the high-frequency residual component does not fluctuate drastically (because water vapor diffusion is a slow process). Therefore, the impact judgment condition is not met—although the low-frequency trend component may exceed the first threshold, the high-frequency rate of change does not exceed the second threshold. The main control microprocessor does not execute primary protection actions, the radar continues to operate normally, and the screen will not erroneously cut off radar transmission due to environmental drift.

[0085] When a user touches the screen with their actual finger, the high-frequency residual component increases dramatically within a very short time, simultaneously meeting both thresholds, and the system triggers the initial protection normally. Subsequent steps S2, S3, and S4 are executed automatically based on the touch area ratio and the speed of finger withdrawal, similar to scenario one or two.

[0086] This embodiment verifies the application's ability to accurately eliminate environmental interference (temperature drift, moisture), ensuring that the radar operation will only be intervened when a truly destructive touch event occurs.

[0087] The above five embodiments cover typical scenarios such as rapid clicking, slow hand withdrawal, slight edge occlusion, capacitive pen writing, and high temperature and humidity environments, fully demonstrating the adaptability and robustness of the control method of this application under different user behaviors, different contact tools, and different environmental conditions.

[0088] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0089] The above are merely optional embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made based on the inventive concept of this application and the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the patent protection scope of this application.

Claims

1. A radar near-field saturation anti-false judgment control method based on touch physical interruption, characterized in that, The method, applied to an electronic device including a main control microprocessor, a touch-sensitive data acquisition unit, and a radar component, includes the following steps: S1. Obtain the touch status bit signal and mutual capacitance signal output by the touch acquisition unit, and separate the trend component located in the first frequency range and the residual component located in the second frequency range from the mutual capacitance signal; when the trend component and the residual component meet the preset impact judgment condition, perform the primary protection action, that is, control the hardware pin to switch to the blocking level state to cut off the power supply to the transmitting circuit of the radar component. S2. In response to the primary protection action, based on the proportion of the projection area of ​​the two-dimensional touch coordinates in the radar field of view, dynamically select between maintaining the primary protection action and executing a signal degradation strategy; the signal degradation strategy is: switching to the enable level state to restore the power supply of the transmitting circuit, and writing a saturation flag bit in the data frame sent to the radar component to make the radar component enter the abnormal data discarding mode. S3. In response to the touch status bit signal changing from a touch valid level representing the presence of touch to a touch release level representing the disappearance of touch, the capacitance decay characteristics within a preset time period are obtained with the time of the change as the end point, and a weighted fusion calculation is performed using a preset recursive weight matrix to fit the comprehensive escape speed of the hand withdrawal. S4. Perform the corresponding recovery operation based on whether the saturation flag bit exists: If present, the saturation flag is cleared, thereby removing the abnormal data discarding mode and restoring the fast Fourier transform processing of the original data frame of the radar component. If it does not exist, the theoretical safe backoff time is calculated using the comprehensive escape velocity, and the hardware pin is switched to the enable level state after the theoretical safe backoff time expires.

2. The control method according to claim 1, characterized in that: The first frequency range is 0 Hz to 2 Hz; the second frequency range is 10 Hz to 50 Hz.

3. The control method according to claim 1, characterized in that: The impact determination condition in step S1 is: the trend component exceeds the environmental baseline tolerance threshold, and the numerical change of the residual component in a single sampling period exceeds a preset threshold.

4. The control method according to claim 1, characterized in that: The dynamic selection logic in step S2 is as follows: if the projection area ratio reaches or exceeds the first occlusion threshold, the primary protection action is maintained; otherwise, the signal degradation strategy is executed.

5. The control method according to claim 1, characterized in that: The abnormal data discarding mode is as follows: when the radar component parses the saturation flag bit, it executes the abnormal frame discarding logic and stops performing fast Fourier transform processing on the original data of the current frame.

6. The control method according to claim 1, characterized in that, Step S3, which involves using a preset recursive weight matrix to perform a weighted fusion calculation to fit the overall escape velocity of hand withdrawal, includes: The attenuation slope of the mutual capacitance signal and the area shrinkage rate of the contact area are obtained within the preset time period before the touch effective level jumps to the touch release level. The attenuation slope of the mutual capacitance signal and the area shrinkage rate are weighted and fused using a preset recursive weight matrix to calculate the spatial displacement vector that characterizes the hand withdrawal speed, which is used as the comprehensive escape velocity.

7. The control method according to claim 1, characterized in that, Step S4, which involves calculating the theoretical safe retreat time using the comprehensive escape velocity, includes: Using the full-scale input voltage value of the analog-to-digital converter of the radar component receiver as a reference upper limit, numerical simulations for the expected value of the echo voltage at the delay time are performed cyclically. When the expected value of the echo voltage is less than the full-scale input voltage value, and the absolute value of the numerical difference between two adjacent simulation cycles is lower than the preset convergence threshold, the sum of the step times accumulated in the current cycle is determined as the theoretical safe backoff time.

8. A radar near-field saturation anti-false alarm control device based on touch physical interruption, characterized in that, The control device includes a touch-sensitive data acquisition unit, a radar assembly, and a main control microprocessor. The control device implements the method described in any one of claims 1 to 7 through logical cooperation between the components, wherein: The touch acquisition unit is configured as the data source of the control device, used to provide the mutual capacitance signal and the touch status bit signal to the main control microprocessor, and includes a storage unit for temporarily storing the saturation flag bit; The radar component is configured as a controlled terminal of the control device, used to perform data frame discarding according to the state of the saturation flag bit, or to cut off beam transmission by responding to the physical pin level of the main control microprocessor through its hardware enable terminal. The main control microprocessor is configured as the control center of the control device and is electrically connected to the touch acquisition unit and the radar component, respectively.

9. The control device according to claim 8, characterized in that: The main control microprocessor is equipped with a coordinate mapping logic unit, which is used to store the relative position offset parameters between the touch acquisition unit and the radar component, and to map the two-dimensional coordinates in the original touch data to the mask area in the field of view of the radar component, so that the main control microprocessor can calculate the projection area ratio.

10. The control device according to claim 8, characterized in that: The hardware pins of the main control microprocessor and the enable pins of the radar component form a hardware interrupt link. When the primary protection action is executed, the main control microprocessor outputs the blocking level state through the hardware pin to cut off the power supply to the transmitting circuit of the radar component within a microsecond time. When the signal degradation strategy is executed or the power supply restoration conditions are met, the main control microprocessor outputs the enable level state through the hardware pin to restore the power supply to the transmitting circuit.

Images