Digital Key End Positioning Performance Classification Method and Device, Classification Equipment and Medium

Abstract

Embodiments of the present invention relate to the field of communication technologies, and disclose a method and apparatus for classifying the positioning performance of digital key terminals, a classification device, and a medium. The method includes obtaining positioning parameter data of N digital key terminals to be classified; N is a natural number greater than 0; and classifying the positioning parameter data of the N digital key terminals to be classified by using an unsupervised learning method to obtain positioning performance classification information of the N digital key terminals to be classified. The classification method according to the embodiments of the present invention classifies different types of digital key terminals with relatively consistent positioning performance, so that it is only necessary to calibrate one type of digital key terminal once, thereby doubling the digital key calibration task volume and greatly improving the calibration efficiency.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of communication technologies, and in particular, to a method and device for classifying the positioning performance of a digital key end, a classification device, and a medium. Background Art

[0002] With the development of technology, Bluetooth digital keys are rapidly applied in the Passive Entry Passive Start (PEPS) system. The positioning of Bluetooth digital keys is derived from traditional PEPS systems, so the calibration method of traditional PEPS keys is also referred to. Since the hardware of traditional PEPS keys is produced by tier 1, the consistency is very high. In the Bluetooth digital key positioning scheme, when a smart phone is used as a digital key, on the one hand, due to the large number of smart phone brands and types, there are differences in the radio frequency performance between mobile phones; on the other hand, for different vehicle models, the installation positions of in-vehicle Bluetooth modules may be different, and the in-vehicle sheet metal environment also varies. Therefore, when calibrating Bluetooth digital keys, in-vehicle calibration needs to be performed for different user mobile phones.

[0003] The existing calibration method is to perform in-vehicle calibration for each mobile phone model and each vehicle model as much as possible. Calculated by calibrating 100 mobile phones for each vehicle model and calibrating 20 vehicle models per year in the past, the number of mobile phones to be calibrated each year is 2,000, and the calibration task volume is extremely large. Therefore, there is an urgent need for a method to effectively reduce the calibration task volume. Summary of the Invention

[0004] Embodiments of the present invention provide a method and device for classifying the positioning performance of a digital key end, a classification device, and a medium to solve the technical problems of heavy calibration task volume and low calibration efficiency in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides a method for classifying the positioning performance of a digital key end, the method including: obtaining positioning parameter data of N digital key ends to be classified;

[0006] Classifying the N digital key ends to be classified by using an unsupervised learning method according to the positioning parameter data of the N digital key ends to be classified to obtain positioning performance classification information of the N digital key ends to be classified.

[0007] In addition, the positioning parameter data is RSSI data of each digital key end;

[0008] The RSSI data includes L RSSI data respectively collected at each sampling angle when the digital key end is at P distances from the communication peer and the digital key end is at Q sampling angles at each distance; P, Q, and L are all natural numbers greater than 0.

[0009] In addition, the digital key end is a smart phone, and each of the Q angles is an angle within the range of plus or minus 45° of rotation of the smart phone relative to its initial position along its own axis.

[0010] In addition, the distance differences between adjacent distances among the P distances are equal, and the angle differences between adjacent angles among the Q angles are equal.

[0011] In addition, the classification of the positioning performance classification information of the N types of digital key ends to be classified is obtained by using an unsupervised learning method according to the positioning parameter data of the N types of digital key ends to be classified, including:

[0012] Calculating the average value and / or variance of the L RSSI data collected at each angle at the same distance from the communication peer for each type of digital key end;

[0013] Classifying according to the average value and / or variance of the RSSI data at each angle at each distance of each type of digital key end.

[0014] In addition, the classification according to the average value and / or variance of the RSSI data at each angle at each distance of each type of digital key end includes:

[0015] Normalizing the average value and / or variance of the RSSI data at each angle at each distance of each type of digital key end;

[0016] Classifying according to the normalized RSSI data.

[0017] In addition, the classification according to the normalized RSSI data includes:

[0018] Classifying according to the normalized RSSI data by using the K-means clustering algorithm to obtain the positioning performance classification information of the N types of digital key ends to be classified.

[0019] In a second aspect, an embodiment of the present invention provides a device for classifying the positioning performance of a digital key end, including:

[0020] A data acquisition module, configured to acquire the positioning parameter data of N types of digital key ends to be classified;

[0021] A classification module, configured to classify the positioning performance classification information of the N types of digital key ends to be classified by using an unsupervised learning method according to the positioning parameter data of the N types of digital key ends to be classified.

[0022] In a third aspect, an embodiment of the present invention provides a classification device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the digital key end positioning performance classification method described in the first aspect is implemented.

[0023] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the digital key end positioning performance classification method described in the first aspect is implemented.

[0024] The technical solutions provided by the embodiments of the present invention have at least the following positive effects compared with the prior art:

[0025] In the digital key end positioning performance classification method of the embodiment of the present invention, by obtaining the positioning parameter data of N types of digital key ends to be classified, and using an unsupervised learning method to classify according to the positioning parameter data of N types of digital key ends to be classified to obtain the positioning performance classification information of N types of digital key ends to be classified, the digital key ends can be classified according to their positioning performance. Since the positioning performance of digital key ends of the same category is relatively close, when calibrating digital keys, only one digital key end of the same category needs to be selected for calibration, and the calibration parameters can be reused for the remaining digital key ends, so that the workload of digital key calibration is reduced by several times, and the calibration efficiency can also be greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0027] Figure 1 It is a schematic flowchart of the digital key end positioning performance classification method provided in Embodiment 1 of the present invention;

[0028] Figure 2 It is a schematic structural diagram of the acquisition device for the positioning parameter data of the digital key end provided by the embodiment of the present invention;

[0029] Figure 3 It is a schematic flowchart of classification using the K-means algorithm in the digital key end positioning performance classification method provided in Embodiment 1 of the present invention;

[0030] Figure 4 It is a schematic structural diagram of the digital key end positioning performance classification device provided in Embodiment 2 of the present invention;

[0031] Figure 5Schematic structural diagram of the classification device provided in Embodiment 3 of the present invention. Detailed implementation manners

[0032] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. In addition, it should be noted that for the convenience of description, only parts related to the present invention are shown in the drawings, rather than all the structures.

[0033] In the prior art, for mobile phones with a large number of brand types (i.e., the digital key end), Bluetooth digital key in-vehicle calibration needs to be carried out separately. Moreover, with the continuous increase in the number of vehicle models and mobile phone types, the Bluetooth digital key calibration task is very heavy and the calibration efficiency is low. In view of this problem, the inventors found in practice that the radio frequency performance differences of some mobile phones are relatively small, and they can be classified into one category and one of them can be selected for Bluetooth digital key calibration. Other mobile phones in the same category can reuse its calibration parameters and can meet the usage functions of Bluetooth digital key applications. Based on this, the inventors proposed a digital key end positioning performance classification method. By obtaining the positioning parameter data of multiple smartphones to be calibrated, such as RSSI data, and then accurately classifying smartphones with similar positioning performance into one category through unsupervised machine learning. Since only one calibration is required for multiple smartphones in the same category, the calibration workload can be reduced by several times, and the calibration efficiency can be greatly improved.

[0034] Embodiment 1

[0035] Figure 1 Schematic flowchart of the digital key end positioning performance classification method provided in Embodiment 1 of the present invention. This embodiment is applied to a classification device. This method can accurately classify digital key ends with similar positioning performance, such as smartphones, into one category. This method can be executed by a digital key end positioning performance classification device provided in the embodiments of the present invention. The device can be implemented in software and / or hardware and configured in the classification device. The classification device can be a computing device or system capable of executing classification algorithms, and no specific limitation is made here. The embodiments of the present invention specifically include the following steps:

[0036] Step 101: Obtain the positioning parameter data of N types of digital key ends to be classified.

[0037] Exemplarily, the digital key ends to be classified can be smartphones, but not limited thereto. The digital key ends to be classified can also be terminals with digital key functions such as smart bracelets.

[0038] The positioning parameter data can be the RSSI (Received Signal Strength Indicator) data for each digital key end. In the PEPS system, it is necessary to calibrate the PE / PS function, that is, set the RSSI value when setting the PE / PS. By calibrating the PE / PS, the control of the distance between the digital key end and the vehicle end during unlocking and locking is realized, that is, the positioning of the digital key end is achieved. When the Bluetooth connection between the digital key end and the vehicle end is established and authenticated, the PE / PS function can be realized according to the real-time RSSI data between the digital key end and the vehicle end. The digital key end and the vehicle end can also be positioned through other communication parameters such as time of flight. Correspondingly, the positioning parameter data of the digital key end can be time-of-flight data, which is not specifically limited here.

[0039] In step 101, a relatively large number of categories of digital key ends can be selected for classification. For example, mobile phones with a certain user group on the market can be used for classification. Therefore, the types of digital key ends to be classified are the number of categories of actually available digital key ends, such as 50 or 100.

[0040] Among them, the RSSI data of each digital key end to be classified can include L RSSI data collected at each sampling angle when the digital key end is at P distances from the communication peer and at each of the Q sampling angles at each distance. P, Q, and L are all natural numbers greater than 0.

[0041] Optionally, the distance difference between adjacent distances among the P distances is equal, which is convenient for obtaining and statistically analyzing the data, and can make the sample of the RSSI data more representative. The angle difference between adjacent angles among the Q angles is equal, which is convenient for statistically analyzing the obtained RSSI data and is also beneficial to ensuring the representativeness of the sample of the RSSI data.

[0042] Optionally, the digital key end can be a smart phone, and the Q angles can all be angles within the range of plus or minus 45° of the smart phone rotating along its own axis relative to its initial position. Among them, the communication peer can be a Bluetooth module applied to the vehicle end. When collecting the RSSI data of the smart phone, the initial position of the smart phone can be that its display screen is facing the Bluetooth antenna of the vehicle end Bluetooth module, so as to have better communication performance. When the smart phone rotates within the range of plus or minus 45° along its own axis, the smart phone can collect RSSI data samples in the posture of facing the vehicle end Bluetooth frontally, so that the collected RSSI sample data can better reflect the positioning performance of the intelligent terminal. Of course, this embodiment does not specifically limit the initial position angle of the smart phone. At the same time, this embodiment does not specifically limit the attitude angle of the intelligent terminal relative to the communication peer during RSSI data collection, as long as it is convenient to collect sufficient and effective RSSI data samples.

[0043] Please refer to Figure 2 As shown, RSSI data can be quickly collected by the RSSI data acquisition device. The test device includes a mobile phone holder 202 for placing the smart phone 200 to be sampled, and the mobile phone holder 202 can drive the smart phone 200 to rotate, for example, drive the smart phone to rotate along its own axis, so that the phone can be at Q different angles. The communication peer for Bluetooth ranging communication with the smart phone 200 can be set on the test bench 204. The communication peer 204 can be a Bluetooth module applied to the vehicle end. The height of the test bench can be 0.6 meters, which is close to the height of the Bluetooth module in the vehicle.

[0044] Exemplarily, the obtained RSSI data of the smart phone may include: at positions where the distances between the smart phone 200 and the Bluetooth module on the test bench 204 are 1 meter, 2 meters, and 3 meters respectively, the smart phone 200 is controlled to rotate respectively, and after each rotation of 15° angular step, 3 seconds are collected at this angle, and 10 RSSI values during the ranging communication between the smart phone 200 and the communication peer are collected per second, that is, 30 RSSI values are collected at each angle. For example, the smart phone 200 can collect RSSI data at 7 angular positions such as 45°, 60°, 75°, 90°, 105°, 120°, and 135°. Among them, 90° is the angle at which the smart phone screen faces the Bluetooth antenna of the vehicle-end Bluetooth module. In this embodiment, the size of the angular step is not specifically limited, and larger or smaller angular steps can also be used, such as 10°, 20°, etc., as long as the collected RSSI data sample size can meet the classification requirements.

[0045] It can be understood that in this embodiment, there is no specific limitation on the acquisition method of the RSSI data of the N types of digital key ends to be classified, as long as sufficient and effective RSSI data samples can be efficiently collected.

[0046] Step 102: Use an unsupervised learning method to classify according to the positioning parameter data of the N types of digital key ends to be classified to obtain the positioning performance classification information of the N types of digital key ends to be classified.

[0047] The unsupervised learning method is suitable for classifying unlabeled samples. Examples of unsupervised learning are clustering. Clustering algorithms mainly include two types: partitioning methods and hierarchical methods. Among them, the partitioning methods include the K-means clustering algorithm, the K-medoids algorithm, and the CLARANS algorithm. In this embodiment, the K-means algorithm is used for classification.

[0048] Step 102 classifies the positioning parameter data of N types of digital key terminals to be classified using unsupervised learning to obtain the positioning performance classification information of N types of digital key terminals to be classified, including: calculating the average value and / or variance of L RSSI data collected at each angle at the same distance from the communication peer for each digital key terminal; classifying according to the average value and / or variance of RSSI data at each angle at each distance of each digital key terminal. For example, calculating the average value, variance, or both the average value and variance of 30 RSSI data collected by a smartphone at an angle of 45° and a distance of 1m, which can effectively reduce the data sample size and improve the efficiency and accuracy of machine learning.

[0049] Further, in this embodiment, the average value and / or variance of RSSI data at each angle at each distance of each digital key terminal are normalized, and classification is performed according to the normalized RSSI data.

[0050] Please refer to Figure 3 , and the classification of the positioning performance classification information of N types of digital key terminals to be classified is obtained by using the K-means clustering algorithm according to the normalized RSSI data, including the following steps:

[0051] Step 301: Select the RSSI data of k digital key terminals from the normalized RSSI data of N types of digital key terminals to be classified as the initial clustering centers.

[0052] Among them, the value of K can be set according to experience and is not specifically limited here. For example, 50 types of smartphones may be suitable for being divided into 8 categories, and the value of K is 8.

[0053] Step 302: Calculate the distance from each clustering object to the clustering center, and select the centroid with the shortest distance as one category.

[0054] Among them, the RSSI data of one type of digital key terminal to be classified is a clustering object.

[0055] Step 303: Calculate the centroid in the new category.

[0056] The centroid in the new category can be calculated by the method of calculating the average value, which will not be elaborated here.

[0057] Step 304: Determine whether the end condition is satisfied. If the end condition is not satisfied, return to repeat steps 302 to 304 until the end condition is satisfied; if the end condition is satisfied, execute step 305.

[0058] The end condition can be that there is no change in the clustering result compared with the previous clustering, or the number of clustering times reaches a preset value. Specifically, the standard measure function can be calculated until the termination condition is reached.

[0059] Step 305: Obtain the positioning performance classification information of N digital key terminals to be classified.

[0060] That is, output the positioning performance classification information of N digital key terminals to be classified. Among them, the positioning performance classification information includes the categories to which each of the N digital key terminals to be classified belongs.

[0061] It can be understood that other suitable clustering methods in unsupervised machine learning can also be used for classification, and no specific restrictions are made here.

[0062] It can be understood that when it is difficult to determine the value of K, K can be made to take multiple values, classify separately for each K value, obtain multiple classification results, and then perform sampling verification for each classification result respectively to determine the best classification result.

[0063] Through the K-means clustering algorithm, the 50 selected smartphones in the above text are clustered into 8 groups. To verify the accuracy of the classification results, digital key PE / PS in-vehicle tests are carried out on the classification results. The verification method is as follows: In the above 8 groups of classification results, select one smartphone (calibration phone) from each group for actual calibration, and the remaining phones in the group copy the calibration parameters of the calibration phone. Then verify the PE / PS functions of the smartphones that copy the calibration parameters in the vehicle.

[0064] When performing calibration and PE / PS verification, the environment within at least 10 meters around the vehicle on the left, right, and rear is kept open and clear. The vehicle-end Bluetooth antennas are respectively arranged on the left and right B-pillars, the rear bumper, and under the central display screen beside the gearshift lever.

[0065] The calibration smartphones in each group calibrate PS, PE, welcome, and the connection area. Among them, the standard for PS calibration is that the maximum window movement is within 30 cm; the standard for PE calibration is at about 3 meters at the 90° of the antennas on the left, right, and rear. In case of inability to meet 3 m, calibrate according to the actual results.

[0066] The results of the PE / PS in-vehicle test are as follows:

[0067] 1) The PS coverage of smartphones within the same group using the calibration parameters of the same group meets the general usage requirements;

[0068] 2) The PE of some smartphones is in the range of 1 - 2 meters, and the PE range of the remaining smartphones is mostly in the range of 2 - 5 meters, all of which can ensure the normal use of the PE function.

[0069] Compared with the prior art, in the embodiment of the present invention, by obtaining the positioning parameter data of N types of digital key terminals to be classified, and classifying them in an unsupervised learning manner according to the positioning parameter data of the N types of digital key terminals to be classified to obtain the positioning performance classification information of the N types of digital key terminals to be classified, the digital key terminals can be classified according to their positioning performance. Since the positioning performances of digital key terminals of the same category are relatively close, when calibrating digital keys, only one digital key terminal of the same category needs to be selected for calibration, and the calibration parameters can be reused for the remaining digital key terminals, thus reducing the workload of digital key calibration by several times and significantly improving the calibration efficiency. Moreover, the classification result of this method is more accurate than that of the classification method based on experience. In addition, by introducing a machine learning algorithm to process and classify smartphone data, compared with some traditional rule-based algorithms, it does not require human supervision, and the classification result will be more accurate as the sample size increases.

[0070] Embodiment 2

[0071] An embodiment of the present invention provides a device configured in a classification device. As Figure 4 shown, the classification device includes: a data acquisition module 402 and a classification module 404.

[0072] The data acquisition module 402 is used to obtain the positioning parameter data of N types of digital key terminals to be classified;

[0073] The classification module 404 is used to classify the positioning parameter data of N types of digital key terminals to be classified in an unsupervised learning manner to obtain the positioning performance classification information of the N types of digital key terminals to be classified.

[0074] Optionally, the positioning parameter data is the RSSI data of each digital key terminal; the RSSI data includes L RSSI data collected at P distances from the communication peer of the digital key terminal and at each of the Q sampling angles at each distance, where P, Q, and L are all natural numbers greater than 0.

[0075] Optionally, the digital key terminal is a smartphone, and the Q angles are all angles within the range of plus or minus 45° of the smartphone rotating along its own axis relative to its initial position.

[0076] Optionally, the distance difference between adjacent distances among the P distances is equal, and the angle difference between adjacent angles among the Q angles is equal.

[0077] Optionally, the classification module 404 includes:

[0078] The RSSI data processing sub-module is used to calculate the average and / or variance of the L RSSI data collected at each angle at the same distance from the communication peer for each digital key terminal.

[0079] A classification sub-module, configured to classify according to the average value and / or variance of the RSSI data at each angle at each distance of each digital key end.

[0080] Optionally, the classification module 404 may further include:

[0081] A normalization sub-module, configured to normalize the average value and / or variance of the RSSI data at each angle at each distance of each digital key end; correspondingly, the classification sub-module is configured to classify according to the normalized RSSI data.

[0082] Optionally, the classification sub-module is specifically configured to classify according to the normalized RSSI data by using the K-means clustering algorithm to obtain the positioning performance classification information of N types of digital key ends to be classified.

[0083] Compared with the prior art, the classification device according to the embodiment of the present invention obtains the positioning parameter data of N types of digital key ends to be classified, and classifies them by using an unsupervised learning method according to the positioning parameter data of N types of digital key ends to be classified to obtain the positioning performance classification information of N types of digital key ends to be classified, so that the digital key ends can be classified according to their positioning performance. Since the positioning performances of digital key ends of the same category are relatively close, when calibrating digital keys, only one digital key end of the same category needs to be selected for calibration, and the calibration parameters can be reused for the remaining digital key ends, thereby reducing the digital key calibration workload by several times and greatly improving the calibration efficiency at the same time.

[0084] Embodiment III

[0085] Figure 5 It is a schematic structural diagram of a classification device provided in Embodiment III of the present invention. The classification device 50 includes a memory 51, a processor 52, and a computer program stored on the memory 51 and executable on the processor 52. When the processor 52 executes the program, the technical solution described in the foregoing Embodiment I is implemented.

[0086] Compared with the prior art, the classification device according to the embodiment of the present invention obtains the positioning parameter data of N types of digital key ends to be classified, and classifies them by using an unsupervised learning method according to the positioning parameter data of N types of digital key ends to be classified to obtain the positioning performance classification information of N types of digital key ends to be classified, so that the digital key ends can be classified according to their positioning performance. Since the positioning performances of digital key ends of the same category are relatively close, when calibrating digital keys, only one digital key end of the same category needs to be selected for calibration, and the calibration parameters can be reused for the remaining digital key ends, thereby reducing the digital key calibration workload by several times and greatly improving the calibration efficiency at the same time.

[0087] Embodiment 4

[0088] Embodiment 4 of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a computer processor, it is used to execute the technical solutions of any method embodiment.

[0089] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation manner. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, server, or grid device, etc.) to execute the methods described in various embodiments of the present invention.

[0090] It should be noted that in the embodiments of the above update device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0091] Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for classifying the positioning performance of a digital key end, characterized in that, including: obtaining positioning parameter data of N digital key terminals to be classified; N is a natural number greater than 0; classifying the N digital key terminals to be classified by using an unsupervised learning method according to the positioning parameter data of the N digital key terminals to be classified to obtain positioning performance classification information of the N digital key terminals to be classified; the positioning parameter data is RSSI data of each digital key terminal; the RSSI data includes L RSSI data collected at each sampling angle when the digital key terminal is at P distances from the communication peer and at Q sampling angles at each distance; P, Q, and L are all natural numbers greater than 0; the classifying the N digital key terminals to be classified by using an unsupervised learning method according to the positioning parameter data of the N digital key terminals to be classified to obtain the positioning performance classification information of the N digital key terminals to be classified includes: respectively calculating the average value and / or variance of the L RSSI data collected at each angle of each digital key terminal at the same distance from the communication peer; classifying according to the average value and / or variance of the RSSI data at each angle at each distance of each digital key terminal.

2. The digital key end positioning performance classification method according to claim 1, wherein the digital key terminal is a smart phone, and the Q angles are all angles within the range of positive and negative 45° rotation of the smart phone relative to its initial position along its own axis.

3. The digital key end positioning performance classification method according to claim 1, characterized in that the distance difference between adjacent distances among the P distances is equal, and the angle difference between adjacent angles among the Q angles is equal.

4. The digital key end positioning performance classification method according to claim 1, wherein the classifying according to the average value and / or variance of the RSSI data at each angle at each distance of each digital key terminal includes: normalizing the average value and / or variance of the RSSI data at each angle at each distance of each digital key terminal; classifying according to the normalized RSSI data.

5. The digital key end positioning performance classification method according to claim 4, wherein the classifying according to the normalized RSSI data includes: classifying the normalized RSSI data by using the K-means clustering algorithm to obtain the positioning performance classification information of the N digital key terminals to be classified.

6. A device for a method of classifying the positioning performance of a digital key end, characterized in that, including: a data acquisition module, configured to obtain positioning parameter data of N digital key terminals to be classified; a classification module, configured to classify the N digital key terminals to be classified by using an unsupervised learning method according to the positioning parameter data of the N digital key terminals to be classified to obtain the positioning performance classification information of the N digital key terminals to be classified; the positioning parameter data is RSSI data of each digital key terminal; the RSSI data includes L RSSI data collected at each sampling angle when the digital key terminal is at P distances from the communication peer and at Q sampling angles at each distance; P, Q, and L are all natural numbers greater than 0; the classification module includes: an RSSI data processing sub-module, configured to respectively calculate the average and / or variance of the L RSSI data collected at each angle of each digital key terminal at the same distance from the communication peer; a classification sub-module, configured to classify according to the average value and / or variance of the RSSI data at each angle at each distance of each digital key terminal.

7. A classification device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the digital key end positioning performance classification method device according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the digital key end positioning performance classification method device according to any one of claims 1-5.

Images