Low-power-consumption geomagnetic vehicle detection method and system

Abstract

The invention discloses a low-power-consumption geomagnetic vehicle detection method and system. The method comprises the following steps: acquiring triaxial geomagnetic data; performing first-order difference and filtering on the three-axis data to obtain a smooth first-order difference signal; in a vehicle-free state, a mean value and a standard deviation are obtained through calibration, a Gaussian probability density function is established, an integral interval is constructed based on a deviation amount, a single-axis condition probability item is calculated through integration, and three-axis fusion is performed in combination with a state-related correlation correction coefficient to obtain a vehicle existence probability; in a state machine model including no vehicle, arrival detection, delay confirmation, and vehicle passing and leaving detection, outputting a vehicle arrival moment and a vehicle leaving moment according to a vehicle existence probability, a hysteresis threshold, a continuous point criterion and a delay confirmation mechanism, and compensating and correcting group delay introduced by filtering; and reporting detection information when the vehicle arrival event or departure event is output. According to the method, low power consumption and baseline drift resistance are both considered under the condition of a relatively low sampling rate, and the vehicle detection robustness is improved.

Description

A low-power geomagnetic vehicle detection method and system Technical Field

[0001] This invention belongs to the field of intelligent transportation and Internet of Things traffic sensing technology, specifically relating to a low-power geomagnetic vehicle detection method and system. Background Technology

[0002] Vehicle arrival / departure time and traffic flow parameters are important data foundations for applications such as parking management, road toll collection, traffic operation monitoring and signal control. Existing vehicle detection methods include video detection, microwave / millimeter-wave radar detection, and induction coil detection. Among them, video detection is easily affected by changes in lighting, obstruction and weather conditions such as rain and fog, and there are certain maintenance costs when deployed on a large scale; radar detection has advantages under all-weather conditions, but equipment costs and deployment conditions impose certain constraints on large-scale promotion; induction coil detection requires road construction, and subsequent maintenance and replacement are inconvenient. Geomagnetic sensors can detect vehicles by sensing local magnetic field disturbances caused by the metal structure of the vehicle. They have the characteristics of being buried in a concealed manner, having relatively low node cost and power consumption, and being insensitive to lighting and general weather conditions, making them suitable for large-scale deployment of roadside sensing nodes. However, existing geomagnetic vehicle detection methods still face the following problems in engineering applications: (1) In order to obtain higher time resolution and clearer vehicle disturbance waveforms, some schemes use higher sampling frequencies (e.g., not less than 100Hz). In battery-powered scenarios or scenarios requiring long-term independent operation, high sampling frequency increases energy consumption for sampling and processing, shortens battery life, and increases maintenance frequency.

[0003] (2) Some schemes are based on environmental baseline and threshold judgment. The environmental baseline may drift due to factors such as temperature changes, sensor temperature drift and aging, electromagnetic interference from surrounding electrical facilities and slow changes in the environmental magnetic field. In situations such as heavy traffic, low speed or continuous vehicle passage, the baseline may not be updated in time or may be updated incorrectly, which may lead to false detection or missed detection and affect long-term operational stability.

[0004] (3) In order to suppress baseline drift, some schemes use first-order difference of geomagnetic signal to highlight the relative changes of adjacent sampling points, but the difference operation may amplify high-frequency noise; at the same time, when reducing sampling frequency to save energy, the decrease in time resolution will make it more difficult to stably determine vehicle event boundaries.

[0005] Therefore, a geomagnetic vehicle detection technology solution that balances low power consumption, anti-drift, and stable decision-making capabilities under lower sampling frequency conditions is still needed. Summary of the Invention In order to solve the above-mentioned problems existing in the prior art, the present invention provides a low-power geomagnetic vehicle detection method and system.

[0006] The technical problem to be solved by this invention is achieved through the following technical solution: This invention provides a low-power geomagnetic vehicle detection method, comprising: S1: using a triaxial geomagnetic sensor to acquire X-axis, Y-axis, and Z-axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data; S2: performing first-order difference operations on the triaxial geomagnetic data to obtain first-order difference signals for each axis, and using a digital low-pass filter to smooth the first-order difference signals for each axis to obtain smoothed first-order difference signals for each axis. , Indicates the sampling time. S3: Based on the smoothed first-order differential signal The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. The calculation and fusion steps of the single-axis conditional probability term include: (1) collecting calibration data in a vehicle-free state to obtain the mean of the smoothed first-order difference signal of each axis. with standard deviation And establish the Gaussian probability density function in the car-free state. (2) In Calculate deviation at sampling time And construct the integration interval For the Gaussian probability density function Integrate within the integration interval to obtain the single-axle vehicle-present conditional probability term, and use the complementary quantity of the single-axle vehicle-present conditional probability term as the single-axle vehicle-free conditional probability term; (3) Introduce a correlation correction coefficient. and and adopt and The product of the conditional probability terms of each axis is fused to obtain the fusion score. ,in, And based on the fusion score The probability of the vehicle's existence is obtained. S4: Using a state machine model, based on the probability of the vehicle's existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and preset delay confirmation points It determines vehicle arrival and departure times and records the corresponding arrival times. With departure time And employing the group delay introduced by the digital low-pass filter. Regarding the arrival time and the departure time S5: When the state machine model outputs a vehicle arrival event or a vehicle departure event, it sends the corresponding arrival time information or departure time information to the base station or server.

[0007] This invention also provides a low-power geomagnetic vehicle detection system for implementing the aforementioned low-power geomagnetic vehicle detection method. The system includes at least one sensing device and a base station communicatively connected to the sensing device. The sensing device includes: a triaxial geomagnetic sensor configured to acquire X, Y, and Z-axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data; and a data processing module connected to the triaxial geomagnetic sensor, configured to perform first-order differential operations on the triaxial geomagnetic data to obtain first-order differential signals for each axis, and to smooth the first-order differential signals for each axis using a digital low-pass filter to obtain smoothed first-order differential signals for each axis. , Indicates the sampling time. Based on the smoothed first-order differential signal The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. Using a state machine model, based on the probability of the vehicle's existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and preset delay confirmation points It determines vehicle arrival and departure times and records the corresponding arrival times. With departure time And employing the group delay introduced by the digital low-pass filter. Regarding the arrival time and the departure time Compensation and correction are performed to output vehicle arrival or vehicle departure events; a communication module, connected to the data processing module, is configured to output the arrival time when outputting vehicle arrival or vehicle departure events. or the departure time The arrival time is sent to the base station; the base station is configured to receive and forward the arrival time. or the departure time To the server or cloud platform.

[0008] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By combining first-order differential with digital low-pass filtering, baseline drift and high-frequency noise are suppressed under lower sampling frequency conditions, which improves long-term operation stability; (2) By combining probability fusion and state machine decision, the risk of false detection / missed detection caused by short-term interference is reduced, and more reliable arrival / departure events are output; (3) While satisfying detection stability, the sampling and calculation load is reduced, which is conducive to reducing terminal power consumption and maintenance costs.

[0009] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0010] Figure 1 is a flowchart illustrating a low-power geomagnetic vehicle detection method according to an embodiment of the present invention; Figure 2 is a schematic diagram illustrating the deployment of a geomagnetic sensor in a single-lane scenario according to an embodiment of the present invention; Figure 3 is a schematic diagram illustrating the state transition of a vehicle detection state machine according to an embodiment of the present invention; Figure 4 is a schematic diagram illustrating the structure of a vehicle detection system according to an embodiment of the present invention. Detailed Implementation

[0011] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0012] Existing geomagnetic vehicle detection methods are susceptible to baseline drift due to temperature drift, slow changes in the ambient magnetic field, and surrounding electromagnetic interference during long-term operation. This leads to decreased stability of decisions based on fixed thresholds or baseline updates, resulting in the risk of false positives and false negatives. Furthermore, using higher sampling frequencies to ensure detection accuracy increases terminal power consumption, which is unfavorable for battery-powered or long-term independent operation scenarios. Directly reducing the sampling frequency to lower power consumption reduces the temporal resolution of vehicle disturbance signals, and differential processing may amplify high-frequency noise, leading to decreased detection stability under low sampling rate conditions. Therefore, a geomagnetic vehicle detection method and system that can balance low power consumption and anti-drift / anti-interference capabilities while stably outputting vehicle arrival and departure events under lower sampling rate conditions is needed. To address this issue, this invention provides a low-power geomagnetic vehicle detection method and system.

[0013] To avoid ambiguity, the notation conventions in this embodiment are as follows: express The data collected by the geomagnetic sensor at the sampling time Axial geomagnetic data, ; express time First-order differential signal of the axis: ; Indicates to The filtered differential signal after digital low-pass filtering, i.e. Smoothed first-order differential signal of the axis; Indicates vehicle status events, where This indicates that there is a car. Indicates no vehicle is available; This represents the probability of vehicle existence obtained by fusing three-axis information. and These represent the preset arrival threshold and the preset departure threshold, respectively. and These represent the preset arrival confirmation point and the preset departure confirmation point, respectively. Indicates the preset delay confirmation points; and These represent the vehicle's arrival time and departure time, respectively. This indicates the duration the vehicle is occupied.

[0014] Figure 1 is a flowchart illustrating a low-power geomagnetic vehicle detection method provided by the present invention. As shown in Figure 1, the method includes: S1, using a triaxial geomagnetic sensor to collect X-axis, Y-axis, and Z-axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data.

[0015] S2. Perform first-order difference operations on the three-axis geomagnetic data to obtain the first-order difference signals for each axis. Then, use a digital low-pass filter to smooth the first-order difference signals for each axis to obtain smoothed first-order difference signals for each axis. , Indicates the sampling time. .

[0016] S3, Based on smoothing first-order difference signals The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. .

[0017] Here, the calculation and fusion steps of the single-axis conditional probability term include: (1) collecting calibration data in the vehicle-free state to obtain the mean of the smoothed first-order difference signal of each axis. with standard deviation And establish the Gaussian probability density function in the car-free state. (2) In Calculate deviation at sampling time And construct the integration interval For Gaussian probability density function Integrate within the integration interval to obtain the single-axle vehicle conditional probability term, and use the complementary quantity of the single-axle vehicle conditional probability term as the single-axle no-vehicle conditional probability term; (3) Introduce the correlation correction coefficient. and and adopt and The product of the conditional probability terms of each axis is fused to obtain the fusion score. ,in, And based on the fusion score Obtain the probability of vehicle existence .

[0018] S4. Using a state machine model, based on the probability of vehicle existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and delayed confirmation points It determines vehicle arrival and departure times and records the corresponding arrival times. With departure time And employs a group delay introduced by a digital low-pass filter. For arrival time and departure time Perform compensation corrections to output vehicle arrival or vehicle departure events.

[0019] S5. When the state machine model outputs a vehicle arrival event or a vehicle departure event, it sends the corresponding arrival time information or departure time information to the base station or server.

[0020] The specific implementation principle of the method provided by the present invention will be described in detail below.

[0021] Step 1: Deployment and Data Acquisition. As shown in Figure 2, the triaxial geomagnetic sensor is deployed at a preset location in a single-lane scenario (e.g., near or below the lane line). Preferably, the sensor is installed in the same direction as the vehicle's travel; however, in other embodiments, other installation locations and orientations can be selected based on road construction conditions, and unified to the vehicle's travel coordinate system through coordinate transformation. The triaxial geomagnetic sensor is controlled at a sampling frequency... Collect X, Y, Z axis geomagnetic data ,in , The sampling time number represents the sampling time sequence. Each sampling time. Preferably, In other embodiments, It can also be lower than The preset sampling frequency.

[0022] Step 2: First-order difference processing. Perform first-order difference operations on the raw geomagnetic data for each axis to obtain the first-order difference signal for each axis: Differential processing is used to highlight the changes between adjacent samples and reduce the impact of slowly changing background components on detection. In implementation, this can be achieved by... Start outputting the difference results, or let... .

[0023] Step 3: FIR low-pass filtering is used to suppress high-frequency noise introduced or amplified by differential processing for the first-order differential signals of each axis. Digital low-pass filtering is performed to obtain smoothed first-order differential signals for each axis. : ,in These are the filter coefficients. The above filter is an FIR low-pass filter, and it is preferable to use a certain number of taps. (corresponding order) sampling frequency .

[0024] In some embodiments, the X / Y axis The cutoff frequency is approximately Z-axis The cutoff frequency is approximately The minimum stopband attenuation is approximately Fixed group delay introduced by FIR filtering It can be obtained through offline calculation and in the output. , Time compensation correction; for example, in the case of linear phase symmetry coefficients. One sampling point.

[0025] To facilitate direct implementation, in some selected embodiments, the values ​​of the filter coefficients are shown in Table 1, where the X-axis and Y-axis share the same set of coefficients (i.e., The Z-axis uses a different set of coefficients ( ).

[0026] Table 1. FIR Low-Pass Filter Coefficients

[0027] In other embodiments, the number of taps can be maintained. Alternatively, the cutoff frequency parameters can be kept unchanged and the coefficients redesigned; or the same set of coefficients can be used for all three axes to simplify implementation; none of the above constitutes a limitation.

[0028] Step 4: Calculate the single-axis conditional probability term and define the event. ,in This indicates that there is a car. This indicates no vehicle is present. Preferably, under vehicle-free conditions, a segment of calibration data is collected, and the differential signals for each axle are filtered. Statistical analysis was performed to obtain the mean. with standard deviation And using Gaussian probability density function Fitting the distribution under no-car conditions: ,in, express The independent variable in the equation. At time... Define deviation and construct the central interval To ensure that the probability of a vehicle on a single axle increases accordingly as the deviation increases, this embodiment defines the probability mass in the center interval as the conditional probability term of a vehicle on a single axle: And its complementary quantity is defined as the single-axle no-vehicle conditional probability term: It should be noted that this integral operation can be performed using the error function under a Gaussian distribution. The calculation can be performed, or approximated by a pre-defined lookup table, to reduce the computational overhead on the embedded side.

[0029] Step 5: Three-axis data fusion. Since there may be statistical correlation between the three-axis filtered differential signals under vehicle-free and vehicle-present conditions, this embodiment introduces a correlation correction coefficient. and ,in, Corresponding to no vehicle status , Corresponding vehicle status At that moment For each axis, we obtain the single-axis conditional probability term. (See step 4 above), where , Define the fusion score: ,in, The prior probability can be given by historical traffic flow statistics; in a simplified implementation, it can be taken as... And this is incorporated into the constant term. Then the probability of the merged vehicle existing is: It should be noted that, and It can be obtained through offline calibration; it can also be used in simplified implementations. .

[0030] Step 6: The state machine determines the arrival / departure event, as shown in Figure 3. The state machine includes: no vehicle state. , Reaching the detection state Delayed status (i.e., delayed confirmation status) Vehicle passage status Leaving the detection state The state machine is based on... With reaching the threshold , leaving threshold The comparison results, combined with the number of consecutive points , and delay points To achieve state transitions, which typically satisfy... This creates hysteresis and suppresses jitter.

[0031] Arrival Detection: When When it was first established, it was by Enter And start counting; if the condition is met consecutively achieve sampling points (i.e.) If so, confirm the vehicle's arrival and record it. Then enter Delayed confirmation After sampling points, proceed to... (Right now ); if in Appeared in Then count Clear and return .

[0032] Leaving the test: duped Entering upon initial establishment And start counting; if the condition is met consecutively achieve Each sampling point confirms the vehicle's departure and records it. ,return If in Appeared in Then count Clear and return Continue monitoring. Event times can be converted to actual times according to the sampling sequence number (e.g., ...). If a fixed group delay is introduced by the preceding filter, it can be used to... , Compensation and corrections will be made.

[0033] For example, a set of preferred parameters ( )for: At sampling frequency Below, the time scale corresponding to the above point parameters is: the time to confirmation is approximately The departure confirmation time is approximately The delayed confirmation time is approximately In other embodiments, when the sampling frequency changes, the above time scale is adjusted according to the number of points. Automatic scaling inversely proportional to the property.

[0034] Step 7: Reporting Detection Results. After the state machine outputs the vehicle arrival and departure events, the vehicle arrival time is uploaded via the wireless communication module. With departure time The wireless communication module can use low-power wide-area network (LPWAN) communication or cellular communication; preferably, it can use at least one of LoRa / LoRaWAN, NB-IoT, 4G / 5G, etc. These communication methods are optional and not limiting. The reported information includes at least: device identifier, lane / parking space identifier, and vehicle arrival time. Vehicle departure time Optionally, it also includes vehicle occupancy time. Event sequence number and quality identifier, etc.

[0035] This invention also provides a low-power geomagnetic vehicle detection system, which includes at least one sensing device and a base station communicatively connected to the sensing device. The sensing device includes: a triaxial geomagnetic sensor configured to acquire X, Y, and Z-axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data; and a data processing module connected to the triaxial geomagnetic sensor, configured to perform first-order differential operations on the triaxial geomagnetic data to obtain first-order differential signals for each axis, and to smooth the first-order differential signals for each axis using a digital low-pass filter to obtain smoothed first-order differential signals for each axis. , Indicates the sampling time. Based on smoothing first-order differential signals The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. Using a state machine model, based on the probability of the vehicle's existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and preset delay confirmation points It determines vehicle arrival and departure times and records the corresponding arrival times. With departure time And employing the group delay introduced by the digital low-pass filter. Regarding the arrival time and the departure time Compensation and correction are performed to output vehicle arrival or departure events; the communication module, connected to the data processing module, is configured to output the arrival time when outputting vehicle arrival or departure events. or departure time Send to the base station; the base station is configured to receive and forward arrival times. or departure time To the server or cloud platform.

[0036] For example, Figure 4 is a schematic diagram of a low-power geomagnetic vehicle detection system. As shown in Figure 4, the system includes: a triaxial geomagnetic sensor, a processor / MCU, a memory, a wireless communication module, and a base station / gateway and a server / cloud platform that are connected to the wireless communication module.

[0037] The triaxial geomagnetic sensor is used to collect geomagnetic data along the X, Y, and Z axes. The triaxial geomagnetic sensor may be a magnetoresistive geomagnetic sensor (such as AMR or TMR), and optionally a model such as VCM5883L. The above models are only preferred examples and do not constitute a limitation.

[0038] The processor / MCU is connected to the triaxial geomagnetic sensor and is used to sequentially perform first-order difference, FIR low-pass filtering, single-axis conditional probability term calculation, triaxial fusion, and state machine decision processing on the collected data, and output the vehicle arrival time. departure time and optional duration Event information; the memory is used to store program instructions and parameter configurations, and to perform read and write interactions with the processor / MCU.

[0039] The wireless communication module is connected to the processor / MCU and is used to upload vehicle event information to the base station / gateway or server / cloud platform. The communication method can be low power wide area network communication or cellular communication, preferably supporting at least one of LoRa / LoRaWAN, NB-IoT, 4G / 5G and other communication methods. The above communication methods are only optional and do not constitute a limitation.

[0040] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0041] In this specification, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. While different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce a good effect.

[0042] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A low-power geomagnetic vehicle detection method, characterized in that, include: S1: Use a triaxial geomagnetic sensor to collect X-axis, Y-axis and Z-axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data; S2: Perform first-order difference operations on the three-axis geomagnetic data to obtain the first-order difference signals for each axis. Then, use a digital low-pass filter to smooth the first-order difference signals for each axis to obtain smoothed first-order difference signals for each axis. ， Indicates the sampling time. S3: Based on the smoothed first-order differential signal The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. The calculation and fusion steps of the single-axis conditional probability term include: (1) collecting calibration data in a vehicle-free state to obtain the mean of the smoothed first-order difference signal of each axis. with standard deviation And establish the Gaussian probability density function in the car-free state. (2) In Calculate deviation at sampling time And construct the integration interval For the Gaussian probability density function Integrate within the integration interval to obtain the single-axle vehicle-present conditional probability term, and use the complementary quantity of the single-axle vehicle-present conditional probability term as the single-axle vehicle-free conditional probability term; (3) Introduce a correlation correction coefficient. and and adopt and The product of the conditional probability terms of each axis is fused to obtain the fusion score. ,in, And based on the fusion score The probability of the vehicle's existence is obtained. S4: Using a state machine model, based on the probability of the vehicle's existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and preset delay confirmation points It determines vehicle arrival and departure and records the corresponding arrival times. With departure time And employing the group delay introduced by the digital low-pass filter. Regarding the arrival time and the departure time S5: When the state machine model outputs a vehicle arrival event or a vehicle departure event, it sends the corresponding arrival time information or departure time information to the base station or server.

2. The method according to claim 1, characterized in that, The preset sampling rate is 50Hz.

3. The method according to claim 1, characterized in that, The digital low-pass filter is a finite impulse response (FIR) low-pass filter.

4. The method according to claim 1, characterized in that, The calculation formulas for the first-order differential signals of each axis are as follows: ;in, for The data collected by the geomagnetic sensor at the sampling time Axial geomagnetic data, for The data collected by the geomagnetic sensor at the sampling time Axial geomagnetic data, for time The first-order differential signal of the axis.

5. The method according to claim 1, characterized in that, The state machine model is used to manage vehicle detection states, which include: no vehicle state, arrival detection state, delay state, vehicle passing state, and departure detection state. The transition time from the arrival detection state to the delay state is defined as the vehicle arrival time, and the transition time from the departure detection state to the no vehicle state is defined as the vehicle departure time.

6. The method according to claim 5, characterized in that, The confirmation condition for reaching the detection state is: the probability of the vehicle existing. The preset number of arrival confirmation points Each sampling period is greater than or equal to the preset threshold. 。 7. The method according to claim 5, characterized in that, The confirmation condition for leaving the detection state is: the probability of the vehicle's presence. The number of consecutive preset departure confirmation points Each sampling period is less than or equal to the preset departure threshold. 。 8. The method according to claim 1, characterized in that, The Gaussian probability density function The expression is as follows: ;in, express The independent variable in the equation.

9. The method according to claim 8, characterized in that, The calculation formulas for the single-axle vehicle presence conditional probability term and the single-axle no-vehicle conditional probability term are as follows: in, This represents the conditional probability term for the presence of a vehicle on the single axle. This represents the single-axle vehicle-free conditional probability term.

10. A low-power geomagnetic vehicle detection system, characterized in that, To implement the method according to any one of claims 1 to 9, the system includes at least one sensing device and a base station communicatively connected to the sensing device; wherein the sensing device includes: a triaxial geomagnetic sensor configured to acquire X, Y, and Z axis geomagnetic data at a preset sampling rate to obtain triaxial geomagnetic data; and a data processing module connected to the triaxial geomagnetic sensor, configured to perform first-order difference operations on the triaxial geomagnetic data respectively to obtain first-order difference signals for each axis, and to smooth the first-order difference signals for each axis using a digital low-pass filter to obtain smoothed first-order difference signals for each axis. ， Indicates the sampling time. Based on the smoothed first-order differential signal The single-axis conditional probability term is calculated separately and then fused across the three axes to obtain the vehicle's existence probability. Using a state machine model, based on the probability of the vehicle's existence... and preset threshold Preset departure threshold And combined with the preset arrival confirmation points Preset departure confirmation points and preset delay confirmation points It determines vehicle arrival and departure and records the corresponding arrival times. With departure time And employing the group delay introduced by the digital low-pass filter. Regarding the arrival time and the departure time Compensation and correction are performed to output vehicle arrival or vehicle departure events; a communication module, connected to the data processing module, is configured to output the arrival time when outputting vehicle arrival or vehicle departure events. or the departure time The arrival time is sent to the base station; the base station is configured to receive and forward the arrival time. or the departure time To the server or cloud platform.

Images