A new positioning algorithm based on Bluetooth, system and Bluetooth digital key

Abstract

This invention proposes a novel Bluetooth-based positioning algorithm, system, and Bluetooth digital key. The method includes: creating a real-time FiFo sample recorder to record real-time data S of all samples within time T. 1‑n ; Retrieve the latest sample data S X X∈1-n; based on the positioning database, the latest sample data S X The system performs proximity or distance motion recognition; based on the recognition results, it sends unlock or lock control commands to the vehicle. This application proposes a novel Bluetooth-based positioning algorithm that, through a core logic of real-time FiFo sample recorder + motion recognition + control command triggering, specifically addresses the inaccurate positioning and operational anomalies caused by signal fluctuations, poor environmental adaptation, and response lag in existing Bluetooth keyless entry systems.

Description

Technical Field

[0001] This invention belongs to the field of Bluetooth digital key technology, specifically relating to a novel Bluetooth positioning algorithm, system, and Bluetooth digital key. Background Technology

[0002] With the continuous development of automotive electronics technology and the increasing demand for convenient travel, keyless entry systems have become one of the core configurations of modern automobiles. Among them, keyless entry solutions based on Bluetooth technology are widely used in vehicle locking and unlocking control scenarios due to their low power consumption and short-range communication characteristics. By establishing wireless communication between the user's smartphone and the vehicle's Bluetooth module, the vehicle can sense the user's location and automatically lock and unlock, eliminating the need for the user to manually operate the key and greatly improving convenience.

[0003] Currently, the core principle of mainstream vehicle Bluetooth positioning solutions is to pre-calibrate the distribution characteristics of Bluetooth signals around the vehicle and then identify the user's location based on real-time signal matching. Existing solutions perform signal calibration in ideal, unobstructed environments, and the phone's orientation and posture remain fixed during calibration, without considering the impact of human obstruction on the signal. However, in actual use, the phone inevitably moves with the user, and the human body acts as a strong obstruction to the Bluetooth signal, significantly attenuating the signal strength. This results in a significant deviation between the actual collected signal and the calibration model, directly affecting the accuracy of location identification.

[0004] Furthermore, due to differences in the performance of Bluetooth modules (such as transmit power, antenna gain, and signal modulation methods) among different brands and models of smartphones, the Bluetooth signal performance of different phones varies at the same location. To ensure system compatibility, car manufacturers need to perform the aforementioned calibration process for dozens or even hundreds of mainstream mobile phones on the market one by one. This not only requires a significant investment of manpower and resources for purchasing data collection equipment, setting up sites, and manual operations, but also consumes a considerable amount of time to complete data processing and model building, significantly increasing production costs and development cycles. Summary of the Invention

[0005] To address the limitations of existing technologies, such as idealized calibration environments, high calibration costs, signal distance characteristics deficiencies, and insufficient adaptability to real-world scenarios, this invention provides a novel Bluetooth-based positioning algorithm, system, and Bluetooth digital key. This system employs real-time multi-point data learning and calculation methods to enhance the product's flexible application range, reduce calibration steps, and save manpower and time costs.

[0006] In a first aspect, the present invention proposes a novel Bluetooth-based positioning algorithm, characterized by comprising:

[0007] S1: Create a real-time FiFo sample recorder to record real-time data S of all samples within time T.1-n ;

[0008] S2: Retrieve the latest sample data S X X∈1-n; based on the positioning database, the latest sample data S X Perform proximity or distance motion recognition;

[0009] S3: Based on the recognition result, send an unlock control command or a lock control command to the vehicle.

[0010] The novel Bluetooth-based positioning algorithm proposed in this application addresses the problems of inaccurate positioning and abnormal operation caused by signal fluctuations, poor environmental adaptation, and delayed response in existing Bluetooth keyless entry systems through the core logic of real-time FiFo sample recorder + action recognition + control command triggering.

[0011] Preferably, the location database includes:

[0012] When collecting the signal X at the distance from the first position of the vehicle without obstruction. 1-n ;

[0013] When the signal Y is obstructed, the signal at the second position of the vehicle is collected. 1-n ;

[0014] Acquire signal A when moving from the third to the fourth position from the vehicle 1-n And from the moving signal A 1-n Weak signal platform A in China i and strong signal platform A j And the magnitude of the increase ΔA = A j -A i ;

[0015] Collect signal B as the vehicle moves from the fourth position to the third position. 1-n And from the moving signal B 1-n Strong signal platform B was identified in the middle. i and weak signal platform B j The magnitude of the decrease is ΔB = B. i -B j ;

[0016] Where 1-n represents n devices that collect signals; the positions from farthest to closest to the vehicle are the first position, the third position, the fourth position, and the second position.

[0017] Preferably, the proximity action recognition in S2 includes:

[0018] Calculate the latest nearest sample data S X With signal X 1-n The data similarity is used to obtain multiple similarity values;

[0019] The current location is determined based on the multiple similarity values;

[0020] Starting from the current location S0 X Continuing to update the latest sample data S during the movement X Calculate the latest nearest sample data S after the update in a loop. X With weak signal platform A i Strong Signal Platform A j The similarity is calculated by determining the position after the move relative to the starting point S0. X The positional difference ΔSx is determined when the difference ΔSx is equal to or exceeds the increase magnitude ΔA, and only when the most recent closest sample data S... X Strong Signal Platform A j If the similarity reaches the threshold, the approach action is considered successful, and unlocking and welcoming commands are sent to the vehicle.

[0021] Preferably, the distance action recognition in S2 includes:

[0022] Calculate the latest sample data S Y With signal Y 1-n The data similarity is used to obtain multiple similarity values;

[0023] The current location is determined based on the multiple similarity values;

[0024] Starting from the current location S0 Y During the movement, continue to update the latest sample data S that is farther away. Y Calculate the latest farthest sample data S after the update in a loop. Y With strong signal platform B i And weak signal platform B j The similarity is calculated by determining the position after the move relative to the starting point S0. Y The difference in position ΔS Y When the difference ΔS Y The value is equal to or exceeds the decrease magnitude ΔB, and only if the latest value is farthest from the sample data S. Y And weak signal platform B j If the similarity reaches the threshold, the distancing action is considered successful, and a locking command is issued to the vehicle.

[0025] Preferably, the novel Bluetooth-based positioning algorithm described in this application further includes: identification of abnormal proximity or distance, including:

[0026] Determine the difference ΔSx or the difference ΔS Y If either the approach or movement time ΔT fails to meet the threshold, it is determined to be an abnormal approach or departure situation, and no vehicle command is issued.

[0027] Based on the similarity calculated between the device that collects the signal and the sample data collected, and based on the signal change patterns of the device collecting the signal from different directions, it is determined to be a vehicle-around behavior.

[0028] Based on the same inventive concept, this application also proposes a novel Bluetooth-based positioning system, which further includes:

[0029] FiFo sample recorder records real-time data S of all samples within time T. 1-n ;

[0030] The identification unit, based on the latest sample data S acquired... X X∈1-n; based on the positioning database, the latest sample data S X Perform proximity or distance motion recognition;

[0031] The control unit is used to send unlock control commands or lock control commands to the vehicle based on the recognition results.

[0032] Furthermore, the system also includes: a location database;

[0033] The positioning database stores the signal X at the distance from the vehicle's first position when there are no obstructions. 1-n When there is an obstruction, the signal Y at the second position of the vehicle 1-n Signal A moves from the third to the fourth position relative to the vehicle. 1-n And from the moving signal A 1-n Weak signal platform A in China i and strong signal platform A j And the magnitude of the increase ΔA = A j -A i Signal B moves from the fourth position to the third position relative to the vehicle. 1-n And from the moving signal B 1-n Strong signal platform B was identified in the middle. i and weak signal platform B j The magnitude of the decrease is ΔB = B. i -B j .

[0034] Based on the same inventive concept, this application also proposes a Bluetooth digital key, characterized in that the Bluetooth digital key is applicable to the novel Bluetooth positioning algorithm as described in the first aspect, to control the contactless unlocking or locking of a vehicle.

[0035] Based on the same inventive concept, this application also proposes a vehicle equipped with multiple Bluetooth sensors for use in conjunction with a Bluetooth digital key, wherein the vehicle is controlled to unlock or lock according to the Bluetooth digital key as described above.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] This invention proposes a novel Bluetooth-based positioning algorithm, system, and Bluetooth digital key. It simultaneously acquires signals X1-n and Y1-n from an unobstructed first position (distant, unobstructed) and an occluded second position (near-range human occlusion). This accurately captures the signal baseline features under different occlusion conditions, solving the problem of signal discrepancies with the calibration model in scenarios where the user carries a mobile phone in a pocket or bag, thus improving the accuracy of static position determination to over 95%. For two core dynamic scenarios—approaching (third → fourth position) and moving away (fourth → third position)—the algorithm extracts the weak signal platform Ai, strong signal platform Aj, and the rise amplitude ΔA of signal A1-n, and the strong signal platform Bi, weak signal platform Bj, and the fall amplitude ΔB of signal B1-n. This transforms the gradual change process of the dynamic signal into quantifiable feature parameters, providing a clear basis for subsequent action trend recognition and avoiding the misjudgment of instantaneous signal fluctuations by traditional algorithms.

[0038] By using the FiFo sample recorder of this invention to cache the latest data in real time, the recognition unit can start the judgment without waiting for the full data collection, which shortens the approach unlock response time to less than 0.8 seconds and the distance lock response time to less than 1 second. This enables users to unlock when they are close to the phone and lock when they are far away. No matter whether the user puts the phone in the left or right pocket, shirt pocket, or bag, or holds it or the screen is facing different directions, the algorithm can accurately recognize the action by using the occlusion scene signal Y1-n and dynamic amplitude ΔA / ΔB, completely solving the problem of unstable recognition caused by different carrying methods. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the principle of a novel Bluetooth-based positioning algorithm according to an embodiment of the present invention.

[0040] Figure 2 This is the weak signal platform A1 shown in the embodiment of the present invention. n Strong signal platform A2 n .

[0041] Figure 3 This is the remote time-strength signal platform B1 shown in the embodiment of the present invention. n Weak signal platform B2 n .

[0042] Figure 4 These are other abnormal signal situations illustrated in the embodiments of the present invention.

[0043] Figure 5 This is the vehicle surround signal shown in the embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0045] Example 1: As Figure 1 As shown, this invention proposes a novel Bluetooth-based positioning algorithm, comprising:

[0046] S1: Create a real-time FiFo sample recorder to record real-time data S of all samples within time T. 1-n ;

[0047] Preferably, the location database includes:

[0048] When collecting the signal X at the distance from the first position of the vehicle without obstruction. 1-n ;

[0049] When the signal Y is obstructed, the signal at the second position of the vehicle is collected. 1-n ;

[0050] like Figure 2 As shown, signal A is collected when the vehicle moves from the third position to the fourth position. 1-n And from the moving signal A 1-n Weak signal platform A in China i and strong signal platform A j And the magnitude of the increase ΔA = A j -A i ;

[0051] like Figure 3 The signal B shown is acquired when the vehicle moves from the fourth position to the third position. 1-n And from the moving signal B 1-n Strong signal platform B was identified in the middle. i and weak signal platform B j The magnitude of the decrease is ΔB = B. i -B j ;

[0052] Where 1-n represents n devices that collect signals; the positions from farthest to closest to the vehicle are the first position, the third position, the fourth position, and the second position.

[0053] This embodiment uses a family car equipped with a Bluetooth keyless entry system as the application object. The user carries a smartphone with an integrated Bluetooth module as the signal acquisition and interaction terminal. Four Bluetooth sensors, i.e., n signal acquisition devices, are deployed around the vehicle, with n=4, and are respectively installed at the front, rear, left door, and right door. The automatic locking and unlocking control of the vehicle is realized through the novel Bluetooth positioning algorithm of this invention. In this embodiment, the time T is set to 5 seconds, and the first, third, fourth, and second positions, from farthest to closest to the vehicle, correspond to 10M, 3M, 1.5M, and 0.5M, respectively. The specific parameters and execution flow are as follows.

[0054] FiFo sample recorder initialization: A sample recorder based on the FiFo (First-In, First-Out) mechanism is built in the vehicle controller, with a buffer capacity of 50 records, meaning one data record is collected every 0.1 seconds within 5 seconds, for a total of 50 sample data records S1-S50. Data fields include the collection timestamp, signal strength of the four Bluetooth sensors (unit: dBm), and terminal device identifier, such as the mobile phone's Bluetooth MAC address. When the number of samples reaches 50, the oldest data record is automatically discarded to ensure that the recorder always stores the latest real-time sample data from the last 5 seconds.

[0055] When a user moves around the vehicle with a smartphone, four Bluetooth sensors receive Bluetooth broadcast signals from the phone in real time. The signal strength value of each sensor is collected every 0.1 seconds, and the collection timestamp, the four signal strength values, and the phone's MAC address are packaged into a single sample data entry, stored in the FiFo sample recorder in the order of collection. For example, sample data S20 at a certain moment might be: 2025-09-10 08:30:00.200; Front sensor: -65dBm; Rear sensor: -72dBm; Left door sensor: -68dBm; Right door sensor: -70dBm; MAC: AA:BB:CC:DD:EE:FF.

[0056] In this embodiment, the signal X1-X4 at the first position (10M) without obstruction is collected as follows: In an open parking lot (without buildings or people blocking the way), the user places the mobile phone at 10M (first position), keeping the screen facing upward and without people blocking the way. The four Bluetooth sensors continuously collect 100 sets of signal strength, and the average value is taken as X1-X4 (X1 corresponds to the front sensor, X2 corresponds to the rear sensor, X3 corresponds to the left door sensor, and X4 corresponds to the right door sensor).

[0057] In this embodiment, signal Y1-Y4 is collected at the second position (0.5M) when there is obstruction: the user puts the mobile phone in the right trouser pocket (human body obstruction state), stands 0.5M from the right car door (second position), and the four Bluetooth sensors continuously collect 100 sets of signal strength, and the average value is taken as Y1-Y4.

[0058] In this embodiment, the user, carrying a mobile phone, moves at a constant speed of 0.5 m / s from 3M (third position) to 1.5M (fourth position) along the right door. Four Bluetooth sensors simultaneously collect signals, resulting in a dynamic signal sequence A1-A4 (one data set every 0.1 seconds). Signal analysis software is then used to process A1-A4 and extract feature parameters.

[0059] The weak signal platform Ai: the average signal value at 3M (starting point), Ai1 = -70dBm (front of the vehicle), Ai2 = -73dBm (rear of the vehicle), Ai3 = -71dBm (left door), Ai4 = -69dBm (right door);

[0060] Strong signal platform Aj: The average signal strength at 1.5M (end point), Aj1 = -60dBm (front of the vehicle), Aj2 = -63dBm (rear of the vehicle), Aj3 = -61dBm (left door), Aj4 = -59dBm (right door);

[0061] The increase in amplitude ΔA: ΔA1=Aj1-Ai1=10dBm, ΔA2=Aj2-Ai2=10dBm, ΔA3=Aj3-Ai3=10dBm, ΔA4=Aj4-Ai4=10dBm, and the average value ΔA=10dBm is taken as the amplitude judgment threshold for the approaching action.

[0062] In this embodiment, the signals B1-B4 and feature parameters of the moving signal (from the fourth position 1.5M to the third position 3M) ​​are extracted as follows: A user carrying a mobile phone moves at a constant speed of 0.5m / s from the 1.5M position (fourth position) along the right door towards the 3M position (third position). There is no human body obstructing the movement. Four Bluetooth sensors simultaneously collect signals, resulting in the dynamic signal sequence B1-B4. Feature parameters are extracted using signal analysis software.

[0063] Strong signal platform Bi: The average signal value at 1.5M (starting point), Bi1 = -60dBm (front of the vehicle), Bi2 = -63dBm (rear of the vehicle), Bi3 = -61dBm (left door), Bi4 = -59dBm (right door);

[0064] Weak signal platform Bj: The average signal value at 3M (end point), Bj1 = -70dBm (front of the vehicle), Bj2 = -73dBm (rear of the vehicle), Bj3 = -71dBm (left door), Bj4 = -69dBm (right door);

[0065] The decrease amplitude values ​​ΔB are: ΔB1 = Bi1 - Bj1 = 10dBm, ΔB2 = Bi2 - Bj2 = 10dBm, ΔB3 = Bi3 - Bj3 = 10dBm, ΔB4 = Bi4 - Bj4 = 10dBm. The average value ΔB = 10dBm is taken as the threshold for judging the amplitude of the moving away action.

[0066] S2: Take the latest sample data SX, X∈1-n; perform proximity action recognition or distance action recognition on the latest sample data SX according to the positioning database;

[0067] S3: Based on the recognition result, send an unlock control command or a lock control command to the vehicle.

[0068] In this embodiment, the proximity action recognition in S2 includes:

[0069] Calculate the latest nearest sample data S X With signal X 1-n The data similarity is used to obtain multiple similarity values;

[0070] The current location is determined based on the multiple similarity values;

[0071] Starting from the current location S0 X Continuing to update the latest sample data S during the movement X Calculate the latest nearest sample data S after the update in a loop. X With weak signal platform A i Strong Signal Platform A j The similarity is calculated by determining the position after the move relative to the starting point S0. X The positional difference ΔSx is determined when the difference ΔSx is equal to or exceeds the increase magnitude ΔA, and only when the most recent closest sample data S... X Strong Signal Platform A j If the similarity reaches the threshold, the approach action is considered successful, and unlocking and welcoming commands are sent to the vehicle.

[0072] In this embodiment, a user carrying a smartphone approaches the vehicle from 10 meters away. Once within the Bluetooth signal coverage area, the vehicle-mounted FiFo sample recorder has cached the latest real-time data from the past 5 seconds (one data entry is collected every 0.1 seconds). At this time, the latest approaching sample data SX (denoted as SX50, collection timestamp: 2025-09-15 09:10:05.000) is extracted from the recorder.

[0073] The signal strength data from its four sensors are as follows:

[0074]

[0075] The signal data X1-X4 (X1 corresponds to sensor 1, X2 corresponds to sensor 2, and so on) at the first unobstructed location (10M) in the positioning database are: X1 = -85dBm, X2 = -88dBm, X3 = -86dBm, X4 = -87dBm. A cosine similarity algorithm is used to calculate the similarity between SX50 and X1, X2, X3, and X4 respectively, to quantify the degree of matching of their signal features. The calculation formula is:

[0076]

[0077] Since the signal strength is negative, the sign is retained during calculation, and the cosine value is used to reflect the consistency of the signal trend.

[0078] The similarity calculation results for each sensor are as follows: four similarity values ​​were obtained: 0.998, 0.999, 0.998, and 0.999.

[0079] The preset similarity threshold is ≥0.95. All four similarity values ​​mentioned above far exceed the threshold, and the average value reaches 0.9985, indicating that the signal characteristics of SX50 highly match the signals X1-X4 at an unobstructed location 10M in the positioning database. Combining the physical locations corresponding to X1-X4 in the positioning database, the current user's location is confirmed as the first location (10M).

[0080] The confirmed first location (10M) is set as the starting point of the approach action, S0X = 10M. The user continues to approach the vehicle at a speed of 0.5m / s. The FiFo sample recorder automatically updates a new sample data every 0.1 seconds. The algorithm performs similarity calculation and location tracking in a loop according to the following logic:

[0081] First loop: Tracking to 3M (weak signal platform Ai region)

[0082] Six seconds after the user moves, the latest sample data SX110 (collection timestamp: 2025-09-15 09:10:11.000) is extracted from the FiFo recorder. Its signal data are: front of the vehicle -71dBm, rear of the vehicle -74dBm, left door -72dBm, right door -70dBm.

[0083] The similarity between SX110 and the weak signal platform Ai (at 3M, Ai1 = -70dBm, Ai2 = -73dBm, Ai3 = -71dBm, Ai4 = -69dBm) is calculated iteratively. The results are 0.98, 0.99, 0.98, and 0.99, respectively. The average value is 0.985 ≥ 0.95. It is determined that the user has moved to the third position (at 3M). At this time, the starting point is updated to S0X' = 3M.

[0084] Second cycle: Tracking towards 1.5M (strong signal platform Aj region).

[0085] After the user continues to move for 3 seconds, the latest sample data SX140 (collection timestamp: 2025-09-15 09:10:14.000) is extracted. The signal data are: front of the vehicle -61dBm, rear of the vehicle -64dBm, left door -62dBm, right door -60dBm.

[0086] Simultaneously, the similarity between SX140 and the weak signal platform Ai and the strong signal platform Aj (at 1.5M, Aj1 = -60dBm, Aj2 = -63dBm, Aj3 = -61dBm, Aj4 = -59dBm) was calculated: the average similarity with Ai was 0.92 (<0.95), and the average similarity with Aj was 0.97 (≥0.95), indicating that the user has entered the fourth position (1.5M) range corresponding to the strong signal platform Aj.

[0087] Calculate the difference ΔSx between the moved position and the starting point S0X';

[0088] The starting point S0X' is 3M, and the current position after moving is 1.5M. The position difference between the two is ΔSx = 3M - 1.5M = 1.5M.

[0089] Based on the location database, the physical distance difference corresponding to ΔA = 10dBm is 1.5M (the signal rise amplitude from 3M to 1.5M is 10dBm). Therefore, the distance difference corresponding to ΔSx = 1.5M is equal to that corresponding to ΔA, satisfying the condition that the difference ΔSx is equal to or exceeds the rise amplitude value ΔA.

[0090] The latest sample data SX140 and the strong signal platform Aj have four similarity values ​​of 0.97, 0.98, 0.96 and 0.97, respectively, with an average value of 0.97 ≥ 0.95, reaching the preset similarity threshold.

[0091] Since the distances corresponding to ΔSx and ΔA are equal and the similarity between SX140 and Aj meets the standard, the approach action is deemed successful. The vehicle control unit immediately sends two sets of commands to the vehicle:

[0092] Unlock command: Controls all four doors of the vehicle to unlock simultaneously, disabling the anti-theft system;

[0093] Welcome instructions: The exterior rearview mirrors automatically unfold, the interior reading lights illuminate, and the door handle ambient lights remain on, providing a welcome prompt to the user.

[0094] Preferably, the distance action recognition in S2 includes:

[0095] Calculate the latest sample data S Y With signal Y 1-nThe data similarity is used to obtain multiple similarity values;

[0096] The current location is determined based on the multiple similarity values;

[0097] Starting from the current location S0 Y During the movement, continue to update the latest sample data S that is farther away. Y Calculate the latest farthest sample data S after the update in a loop. Y With strong signal platform B i And weak signal platform B j The similarity is calculated by determining the position after the move relative to the starting point S0. Y The difference in position ΔS Y When the difference ΔS Y The value is equal to or exceeds the decrease magnitude ΔB, and only if the latest value is farthest from the sample data S. Y And weak signal platform B j If the similarity reaches the threshold, the distancing action is considered successful, and a locking command is issued to the vehicle.

[0098] In this embodiment, the application scenario of proximity action recognition continues: taking a family car equipped with a Bluetooth keyless entry system as the object, the vehicle has four Bluetooth sensors around it (sensors 1-4 correspond to the front, rear, left door, and right door respectively), and the user carries a smartphone as the signal terminal. The positioning database has been pre-stored: signals Y1-Y4 at the second obstructed position (0.5M), strong signal platform Bi (1.5M, corresponding to the fourth position), and weak signal platform Bj (3M, corresponding to the third position). The initial data collection determined the drop amplitude value ΔB = 10dBm, and the similarity judgment threshold was set to ≥0.95. The following is the complete execution flow of S2 distance action recognition:

[0099] The data similarity between the latest moving-away sample data SY and the signal Y1-n was calculated, and multiple similarity values ​​were obtained. After the user unlocked the vehicle, they carried their smartphone and started moving away from the vehicle from 0.5M away from the right door (second position). At this time, the vehicle's FiFo sample recorder had cached the latest real-time data within the last 5 seconds. The latest moving-away sample data SY (denoted as SY200, collection timestamp: 2025-09-15 09:15:00.000) was extracted, and the signal strength data of its four sensors are as follows:

[0100]

[0101] The location database contains signals Y1-Y4 (Y1 corresponds to sensor 1, Y2 to sensor 2, and so on) at the second occluded location (0.5M): Y1 = -55dBm, Y2 = -58dBm, Y3 = -60dBm, Y4 = -48dBm. Using the cosine similarity algorithm consistent with proximity action recognition, the similarity between SY200 and Y1, Y2, Y3, and Y4 is calculated. The formula is as follows:

[0102]

[0103] The sign of the negative signal strength value is preserved, and the signal feature matching degree is quantified by the cosine value;

[0104] Thus, four similarity values ​​were obtained: 0.998, 0.999, 0.998, and 0.999.

[0105] Since all four similarity values ​​meet the threshold of ≥0.95, with an average value of 0.9985, it indicates that the signal characteristics of SY200 highly match the signals Y1-Y4 at the location database, which are 0.5M away from obstruction. Based on the physical locations corresponding to Y1-Y4, the current user's location is confirmed as the second location (0.5M).

[0106] Starting from the current location (S0Y), dynamically track and iteratively calculate the similarity:

[0107] The confirmed second location (0.5M) is set as the starting point of the movement away from the vehicle, S0Y = 0.5M. The user moves away from the vehicle at a speed of 0.5m / s. The FiFo sample recorder updates the latest data every 0.1 seconds, and the algorithm cyclically performs similarity calculation and location tracking.

[0108] First cycle: Track to 1.5M (strong signal platform Bi region).

[0109] Two seconds after the user moves, the latest sample data SY220 (collection timestamp: 2025-09-15 09:15:02.000) is extracted from the FiFo recorder. The signal data are: front of the vehicle -60dBm, rear of the vehicle -63dBm, left door -61dBm, right door -59dBm.

[0110] The similarity between SY220 and the strong signal platform Bi (at 1.5M, Bi1 = -60dBm, Bi2 = -63dBm, Bi3 = -61dBm, Bi4 = -59dBm) is calculated iteratively. The results are 0.999, 0.999, 0.999, and 0.999, respectively. The average value is 1.0 ≥ 0.95. It is determined that the user has moved to the fourth position (at 1.5M). At this time, the starting point is updated to S0Y' = 1.5M.

[0111] Second cycle: Track to 3M (weak signal platform Bj area).

[0112] After the user continues to move for 3 seconds, the latest sample data SY250 (collection timestamp: 2025-09-15 09:15:05.000) is extracted. The signal data are: front of the vehicle -70dBm, rear of the vehicle -73dBm, left door -71dBm, right door -69dBm.

[0113] Simultaneously, the similarity between SY250 and the strong signal platform Bi and the weak signal platform Bj (at 3M, Bj1 = -70dBm, Bj2 = -73dBm, Bj3 = -71dBm, Bj4 = -69dBm) was calculated: the average similarity with Bi was 0.92 (<0.95), and the average similarity with Bj was 0.999 (≥0.95), indicating that the user has entered the third position (3M) range corresponding to the weak signal platform Bj.

[0114] Calculate the difference ΔSY between the moved position and the starting point S0Y':

[0115] The starting point S0Y' is 1.5M, and the current position after moving is 3M. ​​The position difference between the two is ΔSY = 3M - 1.5M = 1.5M.

[0116] Based on the location database, the physical distance difference corresponding to ΔB = 10dBm is 1.5M (the signal drop from 1.5M to 3M is 10dBm). Therefore, the distance difference corresponding to ΔSY = 1.5M is equal to that corresponding to ΔB, satisfying the condition that the difference ΔSY is equal to or exceeds the drop value ΔB.

[0117] Verify the similarity between SY and the weak signal platform Bj:

[0118] The latest sample data SY250 and the weak signal platform Bj have four similarity values ​​of 0.999, with an average value of 0.999 ≥ 0.95, reaching the preset similarity threshold.

[0119] Since the distances corresponding to ΔSY and ΔB are equal and the similarity between SY250 and Bj meets the standard, the distancing action is deemed successful. The vehicle control unit immediately sends a locking command to the vehicle: controlling the simultaneous locking of all four doors, activating the anti-theft system, turning off the interior lights, and folding the exterior rearview mirrors.

[0120] Preferably, the novel Bluetooth-based positioning algorithm described in this application further includes: identification of abnormal proximity or distance, including:

[0121] Determine the difference ΔSx or the difference ΔS Y If either the approach or movement time ΔT fails to meet the threshold, it is determined to be an abnormal approach or departure situation, and no vehicle command is issued.

[0122] Based on the similarity calculated between the device that collects the signal and the sample data collected, and based on the signal change patterns of the device collecting the signal from different directions, it is determined to be a vehicle-around behavior.

[0123] like Figure 4 As shown, other abnormal situations can be identified by using ΔSx, ΔTx, and signals obtained from some antenna devices that are inconsistent with the B data in this library. This allows for the determination of abnormal approaching and moving away actions. Furthermore, based on the installation position of the antenna devices on the vehicle body, similar actions such as circling the vehicle can be identified. Figure 5 As shown, actions such as turning around.

[0124] In this embodiment, when the user is at 3M (the third position, SOX' = 3M), a sudden Bluetooth signal interference occurs (such as the activation of Bluetooth devices in nearby shops), causing an abnormal and abrupt change in a certain data point in the FiFo sample recorder. The abnormal sample data SX160 (collection timestamp: 2025-09-15 09:20:08.000) is extracted, with signal data as follows: front of the car -60dBm, rear of the car -63dBm, left door -61dBm, right door -59dBm (completely consistent with the strong signal platform Aj at 1.5M). The location corresponding to SX160 is determined to be at 1.5M, ΔSx = 3M - 1.5M = 1.5M (satisfying the condition of being equal to ΔA); the previous normal sample data SX159 (collection timestamp: 2025-09-15 09:20:07.900) still corresponds to the location at 3M, ΔT = 0.1 seconds.

[0125] The preset time threshold for moving 1.5M is ≥2.5 seconds. However, the current ΔT = 0.1 seconds < 2.5 seconds, which meets the judgment condition that either the difference ΔSx or the movement time ΔT does not meet the threshold. Therefore, it is judged as an abnormal approach. The vehicle control unit does not send an unlock command and records the abnormal data for subsequent interference analysis.

[0126] As the user moved from 1.5M (S0Y'=1.5M) to 3M, they briefly turned back to 2M due to forgetting an item, and then continued to move away.

[0127] The sample data SY300 (collection timestamp: 2025-09-15 09:25:10.000) was extracted during the turnaround. The signal data were: front of the vehicle -65dBm, rear of the vehicle -68dBm, left door -66dBm, and right door -64dBm.

[0128] The position corresponding to SY300 is determined to be 2M, and ΔSY = 2M - 1.5M = 0.5M < 1.5M (which does not meet the condition of being equal to or exceeding ΔB).

[0129] ΔT = 1 second from S0Y' = 1.5M to 2M (satisfying part of the time threshold ≥ 2.5 seconds).

[0130] Since ΔSY = 0.5M < 1.5M (does not meet the difference threshold), and any condition that does not meet the threshold is met, it is judged as an abnormal distance situation. The vehicle control unit does not send a locking command. The normal judgment process will be triggered only after the user continues to move to 3M and both ΔSY and ΔT meet the threshold.

[0131] In this embodiment, the user, carrying a mobile phone, starts moving clockwise around the vehicle from 1.5M from the right door (fourth position) at a speed of 0.4m / s, without any obvious tendency to move closer or further away. The FiFo sample recorder collects data at 0.1-second intervals, and selects sample data from three key time points (SY400, SY450, SY500) to analyze signal change patterns: by calculating the similarity between each sample and Ai (3M) and Aj (1.5M) in the positioning database, and combining the signal strength change trends of different orientation sensors, the following pattern was found: the average similarity of the three samples with Ai and Aj is between 0.85 and 0.90 (not reaching the threshold of ≥0.95), and there is no obvious upward or downward trend;

[0132] Directional signal trend:

[0133] When moving from the right door to the rear of the vehicle: the right door sensor signal increases from -58dBm to -67dBm (weakening), while the rear sensor signal decreases from -70dBm to -61dBm (strengthening);

[0134] When moving from the rear of the vehicle to the left door: the rear sensor signal drops from -61dBm to -60dBm (slight increase), and the left door sensor signal drops from -69dBm to -62dBm (increase);

[0135] The signal changes exhibit a cyclical characteristic where the signal strengthens in one direction and weakens in the opposite direction, which is completely different from the trend of all sensors strengthening when the signal is close and weakening when the signal is far away.

[0136] Based on the fact that the signal similarity of the four acquisition devices did not reach the threshold and the cyclical pattern of alternating enhancement / weakening of signals from sensors in different directions, which is consistent with the logic of judging vehicle-around behavior based on the signal change pattern of multiple devices, it is determined to be vehicle-around behavior. The vehicle control unit does not send unlocking or locking commands, but only records the user's movement trajectory. At the same time, it can trigger auxiliary functions (such as the vehicle outline lights lighting up in sequence to remind the user of the vehicle boundary).

[0137] Verification through the above embodiments shows that the abnormal situation identification mechanism of this algorithm can effectively filter out false triggering scenarios such as signal interference and user backtracking. The vehicle walk-around behavior recognition can accurately distinguish movement that is not intended to unlock or lock, reducing the system's misoperation rate to below 0.5%, and significantly improving the reliability and security of the Bluetooth keyless entry system.

[0138] Example 2: This application also proposes a novel Bluetooth-based positioning system, the system further comprising:

[0139] FiFo sample recorder records real-time data S of all samples within time T. 1-n ;

[0140] The identification unit, based on the latest sample data S acquired... X X∈1-n; based on the positioning database, the latest sample data S X Perform proximity or distance motion recognition;

[0141] The control unit is used to send unlock control commands or lock control commands to the vehicle based on the recognition results.

[0142] Furthermore, the system also includes: a location database;

[0143] The positioning database stores the signal X at the distance from the vehicle's first position when there are no obstructions. 1-n When there is an obstruction, the signal Y at the second position of the vehicle 1-n Signal A moves from the third to the fourth position relative to the vehicle. 1-n And from the moving signal A 1-n Weak signal platform A in China i and strong signal platform A j And the magnitude of the increase ΔA = A j -A i Signal B moves from the fourth position to the third position relative to the vehicle. 1-n And from the moving signal B 1-n Strong signal platform B was identified in the middle. i and weak signal platform B j The magnitude of the decrease is ΔB = B. i -B j .

[0144] In a third embodiment, this application also proposes a Bluetooth digital key, characterized in that the Bluetooth digital key is applicable to the novel Bluetooth positioning algorithm described in the first aspect, for controlling contactless unlocking or locking of a vehicle.

[0145] In embodiment four, this application also proposes a vehicle equipped with multiple Bluetooth sensors for use in conjunction with a Bluetooth digital key, wherein the vehicle is controlled to unlock or lock according to the Bluetooth digital key as described above.

[0146] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0147] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the statement "comprising a…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0148] Although the description of the invention has been given in conjunction with the specific embodiments described above, it will be apparent to those skilled in the art that many substitutions, modifications, and variations can be made based on the above description. Therefore, all such substitutions, modifications, and variations are included within the spirit and scope of the appended claims.

Claims

1. A Bluetooth-based positioning method, characterized in that, include: S1 : Create a real-time FiFo sample logger, record all sample real-time data S1,...,S n ; S2: Retrieve the latest sample data S X According to the location database, the latest sample data S X Perform proximity or distance motion recognition; S3: Based on the recognition result, send an unlock control command or a lock control command to the vehicle; The location database includes: When collecting signals X1,...,X at a distance of 0 from the vehicle's first position without obstruction. n ; When the signal Y1,...,Y is collected at the second position of the vehicle with obstruction. n ; Collect signals A1,...,A as the vehicle moves from the third to the fourth position. n And from the moving signal A1,...,A n Determine the average signal value A of the weak signal platform. i The average signal value A of the strong signal platform j And the increase value ΔA=A j -A i ; Collect signals B1,...,B as the vehicle moves from the fourth position to the third position. n And from the moving signal B1,...,B n Determine the average signal value B of the strong signal platform. i And the average signal value B of the weak signal platform j ; and the magnitude of the decrease ΔB=B i -B j ; Where n represents the number of devices collecting signals; the positions from farthest to closest to the vehicle are the first position, the third position, the fourth position, and the second position. The proximity action recognition in S2 includes: Calculate the latest nearest sample data S X With signals X1,...,X n The data similarity is used to obtain multiple similarity values; The current location is determined based on the multiple similarity values; Starting from the current location S0 X Continuing to update the latest sample data S during the movement X Calculate the latest nearest sample data S after the update in a loop. X The average signal A of the weak signal platform i The average signal value A of the strong signal platform j The similarity is calculated by determining the position after the move relative to the starting point S0. X The position difference ΔSx, based on the mapping relationship between distance and signal strength, is determined when the difference ΔSx is equal to or exceeds the rise amplitude value ΔA, and only when the most recent nearest sample data S... X The average signal value A of the strong signal platform j If the similarity reaches the threshold, the approach action is considered successful, and unlocking and welcoming commands are sent to the vehicle. The distance action recognition in S2 includes: Calculate the latest sample data S Y With signals Y1,...,Y n The data similarity is used to obtain multiple similarity values; The current location is determined based on the multiple similarity values; Starting from the current location S0 Y During the movement, continue to update the latest sample data S that is farther away. Y Calculate the latest farthest sample data S after the update in a loop. Y The average signal B of the strong signal platform i The average signal value B of the weak signal platform j The similarity is calculated by determining the position after the move relative to the starting point S0. Y The difference in position ΔS Y Based on the mapping relationship between distance and signal strength, when the difference ΔS Y The value is equal to or exceeds the decrease magnitude ΔB, and only if the latest value is farthest from the sample data S. Y The average signal value B of the weak signal platform j If the similarity reaches the threshold, the distancing action is considered successful, and a locking command is issued to the vehicle.

2. The Bluetooth-based positioning method according to claim 1, characterized in that, Also includes: Identification of abnormal approach or distance situations, including: Determine the difference ΔS X or difference ΔS Y If either the approach or movement time ΔT fails to meet the threshold, it is determined to be an abnormal approach or departure situation, and no vehicle command is issued.

3. The Bluetooth-based positioning method according to claim 2, characterized in that, Also includes: Identification of abnormal approach or distance situations also includes: Based on the similarity calculated between the equipment that collects the signals and the real-time sample data, and based on the signal change patterns of the equipment that collects signals from different directions, it is determined to be a vehicle-around behavior.

4. A system based on Bluetooth positioning according to any one of claims 1-3, characterized in that, Also includes: The FiFo sample recorder records real-time data S1,...,S of all samples within time T. n ; The identification unit, based on the latest sample data S acquired... X According to the location database, the latest sample data S X Perform proximity or distance motion recognition; The control unit is used to send unlock control commands or lock control commands to the vehicle based on the recognition results.

5. The system according to claim 4, characterized in that, Also includes: Locating the database; The positioning database stores the signals X1,...,X1 at the distance from the vehicle's first position when there is no obstruction. n When there is an obstruction, the signal Y1,...,Y at the second position from the vehicle n Signals A1,...,A move from the third to the fourth position relative to the vehicle. n And from the moving signal A1,...,A n Determine the average signal value A of the weak signal platform. i The average signal value A of the strong signal platform j And the increase value ΔA=A j -A i Signals B1,...,B move from the fourth position to the third position relative to the vehicle. n And from the moving signal B1,...,B n Determine the average signal value B of the strong signal platform. i And the average signal value B of the weak signal platform j ; And the magnitude of the decrease ΔB=B i -B j .

6. A Bluetooth digital key, characterized in that, The Bluetooth digital key is applicable to the Bluetooth positioning method as described in any one of claims 1-3 to control contactless unlocking or locking of the vehicle.

7. A vehicle, characterized in that, The vehicle is equipped with multiple Bluetooth sensors for use with a Bluetooth digital key, and the vehicle is unlocked or locked according to the Bluetooth digital key as described in claim 6.

Images