Anomaly point determination method, device positioning method, positioning processing method and device
By constructing a reference signal received power prediction model and identifying signal amplification devices for abnormal coverage areas, abnormal data is eliminated, thus solving the positioning error problem caused by signal amplification devices in wireless communication networks and improving positioning accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA MOBILE GRP FUJIAN CO LTD
- Filing Date
- 2026-03-24
- Publication Date
- 2026-06-30
AI Technical Summary
Existing technologies struggle to accurately identify and eliminate abnormal data points caused by signal amplification devices in wireless communication networks, leading to a decrease in positioning accuracy.
By acquiring the timing advance and reference signal received power measurements of multiple sample points in the same serving cell, a reference signal received power prediction model is constructed. Abnormal sample points are identified based on the differences, and abnormal coverage areas are determined through location information to remove abnormal data from the fingerprint database.
It improves the accuracy of identifying abnormal sample points that affect signal amplification equipment, thereby increasing the accuracy of signal amplification equipment positioning and the positioning precision of user equipment.
Smart Images
Figure CN122317880A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of wireless communication network technology, and in particular to an anomaly point determination method, a device positioning method, a positioning processing method, and an apparatus. Background Technology
[0002] In wireless communication networks, fingerprint-based positioning technology is one of the key means to achieve user equipment (UE) location. This technology establishes a mapping relationship between geographical location and wireless signal characteristics. During positioning, it matches the wireless signal characteristics currently measured by the UE with records in the fingerprint database to determine the UE's location. Minimization of Drive Test (MDT) data provides wireless signal measurement data with precise latitude and longitude, which can serve as the basis for fingerprint database construction. MDT data samples are typically sparse, requiring the integration of massive Measurement Report (MR) data to backfill locations through signal feature matching, thereby expanding the fingerprint database.
[0003] In practical network deployments, the presence of signal amplification devices (such as repeaters or relays) can cause abnormally high reference signal reception power in areas far from the base station. Once this anomalous data is mixed into the fingerprint database, it disrupts the normal correspondence between location and signal characteristics, leading to significant errors in positioning algorithms based on this fingerprint database. Current technologies struggle to accurately identify anomalous sample points affected by signal amplification devices, resulting in the inability to effectively remove anomalous data and impacting positioning accuracy. Summary of the Invention
[0004] This application addresses some of the deficiencies mentioned in the background art by providing an anomaly point determination method, a device positioning method, a positioning processing method, and a device.
[0005] In a first aspect, embodiments of this application provide a method for determining anomalies, including: Acquire wireless measurement data from multiple first sample points in the same serving cell, wherein the wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value; Based on the time advance measurement values of each of the first sample points, the predicted value of the reference signal received power under normal propagation conditions is determined; Based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points, abnormal sample points are identified among the plurality of first sample points.
[0006] In one embodiment of the first aspect, the predicted value of the reference signal received power is determined by a pre-built prediction model, which is a fading model of the reference signal received power relative to the time advance.
[0007] In one embodiment of the first aspect, the predicted value of the reference signal received power is determined by a pre-built prediction model; The prediction model is constructed based on wireless measurement data from multiple second sample points that are not affected by signal anomalous enhancement. The wireless measurement data of each second sample point includes a time advance measurement and a reference signal received power measurement.
[0008] In one embodiment of the first aspect, constructing the prediction model includes: Logarithmic transformation is performed on the time advance measurement values of each of the second sample points to obtain the logarithmic value of the time advance measurement values; Perform a linear regression between the logarithmic value and the corresponding reference signal received power measurement value.
[0009] In one embodiment of the first aspect, determining the abnormal sample points among the plurality of first sample points includes: Calculate the residual between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points; Calculate the mean and standard deviation of the residuals for all the first sample points; The first sample point whose residual is greater than the sum of the mean and the standard deviation by a preset multiple is identified as an outlier sample point.
[0010] In one embodiment of the first aspect, the wireless measurement data of the first sample point is minimized drive test data.
[0011] Secondly, embodiments of this application provide a device positioning method, including: The method for determining abnormal sample points according to any one of the first aspects, wherein the wireless measurement data of each first sample point further includes location information; Based on the location information corresponding to each of the abnormal sample points, the abnormal coverage area is determined; Based on the abnormal coverage area, the location of the signal amplification device is determined.
[0012] Thirdly, a positioning processing method includes: The method for determining abnormal sample points according to any one of the first aspects, wherein the wireless measurement data of each first sample point further includes location information; Based on the location information corresponding to each of the abnormal sample points, the abnormal coverage area is determined; Based on the abnormal coverage area, fingerprint data in the fingerprint database is removed. Based on the fingerprint database after removing fingerprint data, a location process is performed.
[0013] In one embodiment of the third aspect, determining the abnormal coverage area includes: performing clustering processing on the abnormal sample points according to their corresponding location information, and determining each dense region formed by the clustering as the abnormal coverage area.
[0014] Fourthly, embodiments of this application provide an anomaly point determination device, comprising: The acquisition module is used to acquire wireless measurement data of multiple first sample points in the same serving cell. The wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value. The prediction module is used to determine the predicted value of the reference signal received power under normal propagation conditions based on the time advance measurement value of each of the first sample points. The determination module is used to determine abnormal sample points among the plurality of first sample points based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points.
[0015] Fifthly, embodiments of this application provide a device positioning apparatus, including: An anomaly determination module is used to determine anomaly sample points according to the method described in any one of the first aspects, wherein the wireless measurement data of each first sample point further includes location information; The region determination module is used to determine the abnormal coverage area based on the location information corresponding to each of the abnormal sample points; The device positioning module is used to determine the location of the signal amplification device based on the abnormal coverage area.
[0016] Sixthly, embodiments of this application provide a positioning processing device, including: An anomaly determination module is used to determine anomaly sample points according to the method described in any one of the first aspects, wherein the wireless measurement data of each first sample point further includes location information; The region determination module is used to determine the abnormal coverage area based on the location information corresponding to each of the abnormal sample points; The data removal module is used to remove fingerprint data from the fingerprint database based on the abnormal coverage area. The positioning processing module is used to perform positioning processing based on the fingerprint database after removing fingerprint data.
[0017] In a seventh aspect, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any of the methods described in the first, second, or third aspects.
[0018] Eighthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described in the first, second, or third aspects.
[0019] Ninthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the methods described in the first, second, or third aspects.
[0020] According to the anomaly point determination method, device positioning method, positioning processing method and apparatus of the present application embodiments, the anomaly point determination method obtains the time advance measurement value and the reference signal received power measurement value of multiple first sample points in the same serving cell, determines the reference signal received power prediction value under normal propagation conditions based on the time advance measurement value, and determines the anomaly sample point based on the difference between the measurement value and the prediction value, thereby improving the accuracy of anomaly sample point identification.
[0021] The device positioning method is based on the abnormal sample points determined by the above-mentioned abnormal point determination method. Based on the location information of each abnormal sample point, the abnormal coverage area is determined, and the location of the signal amplification device is determined based on the abnormal coverage area. This improves the accuracy of abnormal sample point identification and thus improves the accuracy of signal amplification device positioning.
[0022] The location processing method is based on the abnormal sample points determined by the above-mentioned abnormal point determination method, determines the abnormal coverage area based on the location information of each abnormal sample point, removes fingerprint data from the fingerprint database based on the abnormal coverage area, and performs location processing based on the removed fingerprint database to improve the recognition accuracy of abnormal sample points, thereby improving the positioning accuracy of user equipment. Attached Figure Description
[0023] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0024] Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application.
[0025] Figure 2 This is a flowchart of an anomaly point determination method provided in an embodiment of this application.
[0026] Figure 3 This is a sub-flowchart of an anomaly point determination method provided in an embodiment of this application.
[0027] Figure 4 This is a flowchart of a device positioning method provided in an embodiment of this application.
[0028] Figure 5 This is a flowchart of a positioning processing method provided in an embodiment of this application.
[0029] Figure 6 This is another flowchart of a positioning processing method provided in an embodiment of this application.
[0030] Figure 7 This is a block diagram of an anomaly point determination device provided in an embodiment of this application.
[0031] Figure 8 This is a block diagram of a device positioning apparatus provided in an embodiment of this application.
[0032] Figure 9 This is a block diagram of a positioning processing device provided in an embodiment of this application.
[0033] Figure 10 This is a schematic diagram of a computer program product provided in an embodiment of this application.
[0034] Figure 11 This is a hardware block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0035] To make the objectives, technical solutions, and advantages of this application more apparent, exemplary embodiments according to this application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.
[0036] refer to Figure 1This diagram illustrates an application scenario according to an embodiment of this application. In this scenario, multiple User Equipment (UE) devices access the same serving cell and are distributed at different locations within the coverage area 101 of the serving cell. Signal amplification equipment is deployed within the serving cell, and some UEs are located within the coverage area 102 of the signal amplification equipment. The radio measurement data reported by each UE can be used as radio measurement data for sample points. Sample points corresponding to UEs located outside the coverage area 102 of the signal amplification equipment can be considered normal sample points, while sample points corresponding to UEs located within the coverage area 102 of the signal amplification equipment can be considered abnormal sample points.
[0037] The embodiments are described in detail below.
[0038] Example 1: Reference Figure 2 This application provides a method for determining anomalies, including: S201, acquire wireless measurement data of multiple first sample points in the same serving cell, wherein the wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value.
[0039] In wireless communication networks, Timing Advance (TA) is a parameter used to measure the distance between a user equipment (UE) and a base station. A larger TA value indicates a greater distance between the UE and the base station. Reference Signal Received Power (RSRP) is a parameter used to measure the power of the reference signal received by the UE from the base station.
[0040] Wireless measurement data can be derived from Minimization of Drive Test (MDT) data. MDT data is measurement data automatically collected and reported by user equipment during normal communication. MDT data records typically include information such as serving cell identifier, timing advance, reference signal received power, and location. Location information can be expressed in latitude and longitude.
[0041] In this step, the wireless measurement data can be grouped by serving cell for subsequent analysis. Specifically, for each serving cell, the timing advance measurement and reference signal received power measurement of multiple first sample points belonging to that cell are obtained. It should be noted that the source of the wireless measurement data is not limited to MDT data; it can also come from other data that includes timing advance and reference signal received power.
[0042] S202, Based on the time advance measurement values of each first sample point, determine the predicted value of the reference signal received power under normal propagation conditions.
[0043] Under normal propagation conditions, there is a predictable fading relationship between the received reference signal power and the time lead. As the time lead increases, the received reference signal power gradually decreases, which conforms to the basic physical laws of wireless signal propagation. When a signal amplification device is deployed in the network, this device amplifies and forwards the wireless signal, resulting in abnormally high received reference signal power in areas with a large time lead, i.e., areas far from the base station, thus disrupting the above-mentioned normal fading relationship.
[0044] Therefore, based on the above fading relationship, given a time advance measurement, the reference signal received power value that should be received under normal propagation conditions can be determined, i.e., the reference signal received power prediction value.
[0045] In one alternative implementation, the predicted reference signal received power is determined using a pre-built prediction model, which is a fading model of the reference signal received power relative to time advance. Normal signal attenuation follows a logarithmic distance path loss model; therefore, using the fading model, the predicted reference signal received power can be determined based on the time advance measurement.
[0046] In one alternative implementation, the predicted reference signal received power is determined by a pre-built prediction model based on radio measurement data from multiple second sample points unaffected by signal amplification. The second sample points can be radio measurement data from sample points within the same serving cell that are unaffected by signal amplification equipment, or radio measurement data from sample points in a serving cell without signal amplifiers. Each second sample point includes a time advance measurement and a reference signal received power measurement. Using radio measurement data from multiple second sample points unaffected by signal amplification ensures that the prediction model is more accurate in determining the predicted reference signal received power under normal propagation conditions. It should be understood that the aforementioned fading model of reference signal received power relative to time advance can be built based on radio measurement data from multiple second sample points unaffected by signal amplification.
[0047] In one alternative implementation, the above prediction model is constructed in the following manner: Logarithmically transform the timing advance measurements for each second sample point to obtain the logarithmic value of the timing advance measurement. Then, perform linear regression (LR) on the logarithmic value and the corresponding reference signal received power measurement. After linear regression, the model parameters are obtained, leading to the prediction model. It should be understood that the corresponding reference signal received power measurement refers to the reference signal received power measurement corresponding to the timing advance measurement.
[0048] There is an approximately linear relationship between the received power of the reference signal and the logarithm of the time advance, that is: ; in, For reference signal received power prediction value, α is the logarithmic value of the time lead, used to perform a logarithmic transformation on the time lead TA, converting it to a decibel scale to conform to the linear characteristics of path loss. α and β are model parameters, i.e., the model parameters obtained after the above linear regression.
[0049] It should be noted that the linear regression method described above is only one exemplary implementation. In other implementations, other methods such as multinomial regression, support vector regression, and neural networks can also be used to construct the prediction model, as long as it can reflect the correlation between the time lead and the received power of the reference signal under normal propagation conditions, and the received power of the reference signal under normal propagation conditions can be determined based on the time lead. Similarly, the method for determining the predicted value of the received power of the reference signal is not limited to using a prediction model; it can also be determined by looking up tables, using empirical formulas, etc.
[0050] S203, based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power of each first sample point, abnormal sample points are identified among the multiple first sample points.
[0051] For each first sample point, calculate the difference between its measured reference signal received power and its predicted reference signal received power. The larger the difference, the higher the reference signal received power at that point is compared to the normal expected value at that distance, and the greater the likelihood that the signal is being amplified by a signal amplification device.
[0052] Specifically, whether the first sample is an abnormal sample point can be determined by whether the absolute value of the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power is greater than a threshold. If it is greater than the threshold, the first sample is determined to be an abnormal sample point; otherwise, the first sample is determined to be a normal sample point.
[0053] In one optional implementation, it can be determined whether the corresponding first sample point is an abnormal sample point based on whether the residual between the reference signal received power measurement value and the corresponding reference signal received power prediction value of the first sample point is greater than a dynamic threshold, thereby identifying abnormal sample points among multiple first sample points.
[0054] For example, see Figure 3 Identify outlier sample points among multiple first sample points, including: S301, calculate the residual between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each first sample point.
[0055] S302, calculate the mean and standard deviation of the residuals for all first sample points.
[0056] S303: The first sample point whose residual is greater than the sum of a preset multiple of the mean and the standard deviation is identified as an outlier sample point.
[0057] In this example, the sum of preset multiples of the mean and standard deviation is used as a dynamic threshold. The first sample point with a residual greater than this dynamic threshold is identified as an outlier. The preset multiple can be adjusted based on the network's tolerance for false positives and false negatives. For example, a preset multiple of 3 can cover approximately 99.7% of normal data based on the normal distribution characteristics; a preset multiple of 2.5 can be used in scenarios with stricter requirements; and a preset multiple of 3.5 can be used in scenarios with more lenient requirements.
[0058] It should be noted that the dynamic threshold method described above is only one exemplary implementation. In other implementations, other methods such as fixed thresholds, percentiles, and box plots can also be used to determine outlier sample points.
[0059] For example, for each sample point i within the cell, its RSRP measurement value is RSRP. _i The TA measurement value is TA. _i Based on the above prediction model, calculate its predicted value RSRP. predict_i =α 10log10(TA _i The residual for sample point i is calculated as follows: )+β. ; in, Represents sample points i The residual, Represents sample points i The reference signal received power measurement value, This represents the predicted received power of the reference signal. The larger the residual, the higher the RSRP value of the sample point is compared to the expected value, and the greater the possibility that the signal is amplified by an amplifier.
[0060] Calculate the mean μ and standard deviation σ of the residuals for all samples. Set a dynamic threshold Threshold = μ + n. σ and n can be set according to requirements, for example, to a value between 2.5 and 3.5. Residual _i The sample points in Threshold are marked as abnormal sample points.
[0061] In one optional implementation, after identifying anomalous sample points among a plurality of first sample points, clustering can be performed on the anomalous sample points among the plurality of first sample points to obtain dense regions, which can be used as anomalous coverage areas affected by the signal amplification device, and each sample point in the anomalous coverage area can be identified as an anomalous sample point.
[0062] The anomaly point determination method according to the embodiments of this application obtains the time advance measurement value and the reference signal received power measurement value of multiple first sample points in the same serving cell, determines the reference signal received power prediction value under normal propagation conditions based on the time advance measurement value, and determines the anomaly sample point based on the difference between the measurement value and the prediction value, thereby improving the accuracy of identifying anomaly sample points affected by signal amplification equipment.
[0063] Example 2: Reference Figure 4 This application provides a device positioning method, including: S401, abnormal sample points are determined according to the abnormal point determination method, wherein the wireless measurement data of each first sample point also includes location information. The abnormal point determination method is any of the abnormal point determination methods in the embodiments of this application.
[0064] Location information can be the latitude and longitude coordinates of the first sample point. In MDT data, latitude and longitude information is usually collected by the positioning module of the user equipment and reported along with the measurement data.
[0065] S402, Based on the location information corresponding to each abnormal sample point, determine the abnormal coverage area.
[0066] In one optional implementation, outlier sample points are clustered according to their corresponding location information, and the dense regions formed by the clusters are identified as outlier coverage areas. The clustering algorithm can be a density-based clustering algorithm such as Density-Based Spatial Clustering of Applications with Noise (DBSCAN), a partition-based clustering algorithm such as K-means Clustering Algorithm (K-means), or a hierarchical clustering algorithm, a grid-based clustering algorithm, or a model-based clustering algorithm.
[0067] It should be noted that the method for determining abnormal coverage areas is not limited to clustering; it can also be determined through other methods such as grid partitioning and convex hull calculation.
[0068] It should be noted that a serving cell may contain one or more signal amplification devices. When multiple signal amplification devices are present, clustering will generate multiple abnormal coverage areas, each corresponding to the coverage area of a different signal amplification device.
[0069] S403, determine the location of the signal amplification device based on the abnormal coverage area.
[0070] In one alternative implementation, the coordinates of the center point of the abnormal coverage area can be calculated and used as the location of the signal amplification device.
[0071] The methods for calculating the center point coordinates include, but are not limited to: the median of the latitude and longitude of all outlier sample points within the calculation area; the geometric centroid of the convex hull formed by all outlier sample points within the calculation area; determining the peak point with the highest density through kernel density estimation; and the arithmetic mean of the latitude and longitude of all outlier sample points within the calculation area.
[0072] The device positioning method according to the embodiments of this application determines the abnormal coverage area based on the location information of abnormal sample points, and determines the location of the signal amplification device based on the abnormal coverage area, thereby improving the accuracy and operability of the signal amplification device positioning.
[0073] Example 3: Reference Figure 5 This application provides a positioning processing method, including the following steps: S501, abnormal sample points are determined according to the abnormal point determination method, wherein the wireless measurement data of each first sample point also includes location information, and the abnormal point determination method is any abnormal point determination method in the embodiments of this application.
[0074] This step is the same as step S401 in the device positioning method, and will not be repeated here.
[0075] S502, based on the location information corresponding to each abnormal sample point, determine the abnormal coverage area.
[0076] This step is the same as step S402 in the device positioning method. Abnormal coverage areas can be determined by methods such as clustering, which will not be described in detail here.
[0077] S503 removes fingerprint data from the fingerprint database based on abnormal coverage areas.
[0078] A fingerprint database is a record of the mapping relationship between location and wireless signal characteristics, used to locate user equipment through signal feature matching. In this step, fingerprint data in abnormal coverage areas is removed from the fingerprint database to eliminate contamination of the database by signal amplification devices.
[0079] In one alternative implementation, the abnormal coverage area can be set as a geofence, and the fingerprint database can be traversed to delete all fingerprint data records located within the geofence. In another alternative implementation, the center location and coverage radius can be determined based on the abnormal coverage area, and then the removal range can be set accordingly.
[0080] S504, perform location processing based on the fingerprint database after removing fingerprint data.
[0081] After removing fingerprint data from the fingerprint database, relevant positioning processing methods can be used to perform positioning processing on the fingerprint data in the database. Positioning processing may include, but is not limited to: training a positioning model based on the fingerprint database after removing fingerprint data, using the positioning model to backfill the location of measurement report data without location information, and positioning the user equipment based on the backfilled data.
[0082] By using a fingerprint database after removing fingerprint data to perform positioning processing, it is possible to avoid interference from abnormal data contaminated by signal amplification devices, thereby obtaining higher positioning accuracy.
[0083] The positioning processing method according to the embodiments of this application improves the positioning accuracy of user equipment by removing fingerprint data from the fingerprint database based on abnormal coverage areas and performing positioning processing based on the removed fingerprint database.
[0084] Example 4: The technical solution of this application embodiment will be described in detail below with reference to an implementation method. (Refer to...) Figure 6 The location processing method includes the following steps: S601, Input MDT data.
[0085] In this step, you can input MDT data to obtain minimized drive test data that includes serving cell identity (CI), reference signal received power, timing advance, and latitude and longitude.
[0086] S602, Cell Data Packet, acquires wireless measurement data from multiple first sample points of the serving cell.
[0087] In this step, the MDT data can be grouped according to the serving cell identifier and analyzed on a cell-by-cell basis. That is, for each serving cell, the MDT data of multiple first sample points belonging to that serving cell or the MDT data of multiple required first sample points from the MDT data of that serving cell are extracted as the wireless measurement data of multiple first sample points of that serving cell.
[0088] S603, construct the TA-RSRP normal fading model.
[0089] In this step, wireless measurement data from multiple second sample points unaffected by signal amplification devices can be selected to construct a fading model of the reference signal received power relative to the time lead. The second sample points can come from sample data within the same serving cell that are unaffected by signal amplification devices. In another implementation, the second sample points can also come from sample data from other serving cells known to be free of signal amplification devices.
[0090] Specifically, the TA-RSRP normal fading model can be: ; in, For reference signal received power prediction value, α is the logarithm of the time advance, and β are model parameters that can be calculated using wireless measurement data from the second sample point.
[0091] S604, calculate the residual of the first sample point.
[0092] For each first sample point within the serving cell, its time advance measurement is substituted into the fading model established in step S603 to calculate the predicted reference signal received power. Then, the residual between the measured and predicted reference signal received power of that sample point is calculated. The larger the residual, the higher the reference signal received power at that point is compared to the normal expected value at that distance, and the greater the likelihood that the signal is amplified by signal amplification equipment.
[0093] S605 determines whether the residual is greater than the dynamic threshold.
[0094] If the residual is greater than the dynamic threshold, proceed to step S606; if the residual is not greater than the dynamic threshold, proceed to step S611.
[0095] The dynamic threshold can be set according to requirements. It can calculate the mean μ and standard deviation σ of the residuals of all first sample points. The dynamic threshold is set to μ + n × σ, where n is a preset parameter that can be set according to requirements, such as 2.5, 3, 3.5, etc.
[0096] S606 has been identified as an abnormal sample point.
[0097] The first sample point whose residual is greater than the dynamic threshold is identified as an outlier sample point.
[0098] The received power of the reference signal at these anomalous sample points is abnormally high relative to their time lead, which does not conform to the fading pattern under normal propagation conditions, and is believed to be affected by the signal amplification equipment.
[0099] Step S607: Perform geographic clustering.
[0100] All outlier sample points are clustered according to their location information such as latitude and longitude using a clustering algorithm. Density-based clustering algorithms, as well as partition-based, hierarchical, grid-based, or model-based clustering algorithms, can be used.
[0101] S608, clustering anomaly coverage area.
[0102] The densely clustered areas are the abnormal coverage areas, which are the areas affected by signal amplification equipment.
[0103] S609 outputs the center's latitude and longitude.
[0104] Calculate the latitude and longitude of the center point of each anomalous coverage area, which will be used as the location output of the signal amplification device. The calculation methods for the center point coordinates include, but are not limited to: calculating the median of the latitude and longitude of all anomalous sample points in the area; calculating the geometric centroid of the convex hull formed by all anomalous sample points in the area; calculating the geographic density distribution of the area using a kernel density estimation algorithm, and taking the peak point with the highest density as the center point; and calculating the arithmetic mean of the latitude and longitude of all anomalous sample points in the area.
[0105] S610, deletes MR data of abnormally covered areas from the fingerprint database.
[0106] The abnormal coverage area determined in step S609 is set as a geofence. The fingerprint database is traversed, and all fingerprint data records located within the geofence are deleted to eliminate the contamination of the fingerprint database by the signal amplification device.
[0107] S611 was identified as a normal sample point.
[0108] The first sample point whose residual is no greater than the dynamic threshold is identified as a normal sample point. These sample points maintain a normal fading relationship between the received power of the reference signal and the timing advance, and are not affected by the signal amplification equipment.
[0109] S612, retained in the fingerprint database.
[0110] Wireless measurement data from normal sample points are retained in the fingerprint database as valid data for subsequent positioning processing.
[0111] S613, Positioning Training and Backfilling.
[0112] Location processing is performed using only fingerprint database data that has not been contaminated by signal amplification devices. This processing includes, but is not limited to: training a location model based on the processed fingerprint database; using the model to backfill location information from measurement reports lacking location data; and locating the user equipment based on the backfilled data. By using a fingerprint database, location errors caused by signal amplification devices can be eliminated at the source, improving the location accuracy of user equipment.
[0113] In one technical solution of this application, time advance is introduced as a distance dimension and correlated with the received power of a reference signal. Under normal propagation conditions, there is an approximately linear fading relationship between the received power of the reference signal and the logarithm of the measured time advance. By establishing a prediction model through linear regression of this relationship, and comparing the actual measured values with the model predictions, abnormal sample points that violate the normal fading law can be identified from a physical mechanism perspective. Compared with technical solutions that only analyze based on the single dimension of received power of the reference signal, the technical solution of this application can distinguish between signal anomalies caused by signal amplification equipment and signal fluctuations caused by natural factors such as terrain, thus improving the accuracy of detection.
[0114] In one technical solution of this application embodiment, a statistically based dynamic threshold is used for anomaly detection. By calculating the mean and standard deviation of the residuals of all sample points, the dynamic threshold is set as the sum of preset multiples of the mean and standard deviation. When the preset multiple is 3, it can cover approximately 99.7% of normal data based on the normal distribution characteristics; 2.5 can be used in scenarios with stricter requirements, and 3.5 can be used in scenarios with more lenient requirements. Compared to technical solutions that rely on fixed thresholds or empirical rules, the dynamic threshold in the technical solution of this application embodiment can adapt to the signal environment of different cells, achieving accurate quantitative identification and reducing false alarm and false negative rates.
[0115] In one technical solution of this application embodiment, precise positioning of the signal amplification device is achieved by clustering abnormal sample points. Abnormal sample points are clustered according to their location information, and the densely clustered areas are determined as abnormal coverage areas. The coordinates of the center point of each abnormal coverage area are calculated as the location of the signal amplification device. Compared to technical solutions that only provide a vague range of abnormal areas, the technical solution of this application embodiment can directly output the latitude and longitude coordinates of the signal amplification device, providing clear and operable positioning information for network operation and maintenance.
[0116] In one technical solution of this application embodiment, by removing fingerprint data in abnormal coverage areas from the fingerprint database and performing positioning processing based on the removed fingerprint database, the contamination of the fingerprint database by the signal amplification device is eliminated from the source, thereby improving the positioning accuracy of subsequent user equipment.
[0117] Example 5: Reference Figure 7 This application provides an anomaly point determination device, including: The acquisition module 701 is used to acquire wireless measurement data of multiple first sample points in the same serving cell. The wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value. Prediction module 702 is used to determine the predicted value of the reference signal received power under normal propagation conditions based on the time advance measurement value of each first sample point. The determination module 703 is used to determine abnormal sample points among multiple first sample points based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each first sample point.
[0118] In an optional implementation, the predicted value of the reference signal received power is determined by a pre-built prediction model, which is a fading model of the reference signal received power relative to the time advance.
[0119] In one alternative implementation, the reference signal received power prediction is determined by a pre-built prediction model. The prediction model is built based on wireless measurement data from multiple second sample points that are not affected by signal anomalous enhancement. The wireless measurement data from each second sample point includes a timing advance measurement and a reference signal received power measurement.
[0120] In one optional implementation, the apparatus includes a model building module for building a prediction model, comprising: performing a logarithmic transformation on the time advance measurements of each second sample point to obtain a logarithmic value of the time advance measurement; and performing a linear regression of the logarithmic value with the corresponding reference signal received power measurement.
[0121] In an optional implementation, the determining module 703, when determining abnormal sample points among multiple first sample points, specifically performs the following: calculating the residual between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each first sample point; calculating the mean and standard deviation of the residuals of all first sample points; and determining the first sample points whose residuals are greater than the sum of a preset multiple of the mean and the standard deviation as abnormal sample points.
[0122] In one alternative implementation, the wireless measurement data for the first sample point is minimized drive test data.
[0123] The specific implementation of each module in the anomaly point determination device can be referred to the corresponding description in the above-described anomaly point determination method embodiment, and will not be repeated here.
[0124] Example 6: Reference Figure 8 This application provides a device positioning apparatus, comprising: The anomaly determination module 801 is used to determine anomaly sample points according to any anomaly point determination method in the embodiments of this application, wherein the wireless measurement data of each first sample point also includes location information.
[0125] The region determination module 802 is used to determine the abnormal coverage area based on the location information corresponding to each abnormal sample point.
[0126] The device positioning module 803 is used to determine the location of the signal amplification device based on abnormal coverage areas.
[0127] In one optional implementation, the region determination module 802, when determining abnormal coverage areas, specifically performs clustering processing on the abnormal points according to their corresponding location information, and determines each dense region formed by the clustering as the abnormal coverage area. The clustering algorithm can be a density-based clustering algorithm, or a partition-based clustering algorithm, hierarchical clustering algorithm, etc.
[0128] In one optional implementation, the device positioning module 803, when determining the location of the signal amplification device, specifically performs the following: calculating the center point coordinates of each abnormal coverage area, and using the center point coordinates as the location of the signal amplification device. The calculation methods for the center point coordinates include, but are not limited to: calculating the median of the location information of all abnormal points within the area; calculating the geometric centroid of the convex hull formed by all abnormal points within the area; determining the peak point with the highest density through kernel density estimation; and calculating the arithmetic mean of the location information of all abnormal points within the area.
[0129] It should be noted that there may be one or more signal amplification devices within a serving cell. When there are multiple signal amplification devices, the area determination module 802 will form multiple abnormal coverage areas through clustering. The device positioning module 803 will calculate the center point coordinates of each abnormal coverage area and use them as the location output of each signal amplification device.
[0130] The specific implementation methods of each module in the device positioning device can be referred to the corresponding descriptions in the above device positioning method embodiments, and will not be repeated here.
[0131] Example 7: Reference Figure 9 This application provides a positioning processing device, including: The anomaly determination module 901 is used to determine anomaly sample points according to any anomaly point determination method in the embodiments of this application, wherein the wireless measurement data of each first sample point also includes location information.
[0132] The region determination module 902 is used to determine the abnormal coverage area based on the location information corresponding to each abnormal sample point.
[0133] The data removal module 903 removes fingerprint data from the fingerprint database based on abnormal coverage areas.
[0134] The positioning processing module 904 is used to perform positioning processing based on the fingerprint database after removing fingerprint data.
[0135] In an optional implementation, the region determination module 902 is used to determine abnormal coverage areas by: performing clustering processing on abnormal sample points according to their corresponding location information, and determining each dense area formed by the clustering as an abnormal coverage area.
[0136] In one optional implementation, the data removal module 903, when removing fingerprint data from the fingerprint database, specifically: sets the abnormal coverage area as a geofence, traverses the fingerprint database, and deletes all fingerprint data records located within the geofence. In another optional implementation, the data removal module 903 can also determine the center location and coverage radius based on the abnormal coverage area, and then set the removal range accordingly.
[0137] In one optional implementation, the positioning processing module 904, when performing positioning processing, specifically performs at least one of the following processes: training a positioning model based on a fingerprint database after removing fingerprint data; using the positioning model to backfill the location of measurement report data without location information; and positioning the user equipment based on the backfilled data. By using a fingerprint database after removing fingerprint data to perform positioning processing, interference from abnormal data contaminated by signal amplification devices can be avoided in the positioning results, thereby achieving higher positioning accuracy.
[0138] The specific implementation methods of each module in the positioning processing device can be referred to the corresponding descriptions in the above-described positioning processing method embodiments, and will not be repeated here.
[0139] An exemplary embodiment of this application also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any method of the embodiments of this application.
[0140] Exemplary embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of any method of the embodiments of this application.
[0141] refer to Figure 10 An exemplary embodiment of this application also provides a computer program product 1000, including a computer program 1001, wherein the computer program, when executed by a processor, implements the steps of any method of the embodiments of this application.
[0142] refer to Figure 11The present invention describes a structural block diagram of an electronic device 1100 that can serve as a server or client of this application, which is an example of a hardware device that can be applied to various aspects of this application. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.
[0143] Electronic device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1102 or a computer program loaded into random access memory (RAM) 1103 from storage unit 1108. The RAM 1103 may also store various programs and data required for device operation. The computing unit 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. An input / output (I / O) interface 1105 is also connected to bus 1104.
[0144] Multiple components in electronic device 1100 are connected to I / O interface 1105, including: input unit 1106, output unit 1107, storage unit 1108, and communication unit 1109. Input unit 1106 can be any type of device capable of inputting information to electronic device 1100. Input unit 1106 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device. Output unit 1107 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 1108 may include, but is not limited to, disk and optical disk. Communication unit 1109 allows electronic device 1100 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.
[0145] The computing unit 1101 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above. For example, in some embodiments, the methods of the embodiments of this application can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 1100 via ROM 1102 and / or communication unit 1109. In some embodiments, the computing unit 1101 can be configured to perform the methods of the embodiments of this application by any other suitable means (e.g., by means of firmware).
[0146] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. An outlier determination method characterized by, include: Acquire wireless measurement data from multiple first sample points in the same serving cell, wherein the wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value; Based on the time advance measurement values of each of the first sample points, the predicted value of the reference signal received power under normal propagation conditions is determined; Based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points, abnormal sample points are identified among the plurality of first sample points.
2. The method of claim 1, wherein, The predicted value of the reference signal received power is determined by a pre-built prediction model, which is a fading model of the reference signal received power relative to the time advance.
3. The method according to claim 1 or 2, characterized in that, The predicted value of the reference signal received power is determined by a pre-built prediction model; The prediction model is constructed based on wireless measurement data from multiple second sample points that are not affected by signal anomalous enhancement. The wireless measurement data of each second sample point includes a time advance measurement and a reference signal received power measurement.
4. The method of claim 3, wherein, Constructing the prediction model includes: Logarithmic transformation is performed on the time advance measurement values of each of the second sample points to obtain the logarithmic value of the time advance measurement values; Perform a linear regression between the logarithmic value and the corresponding reference signal received power measurement value.
5. The method of claim 1, wherein, The step of determining the abnormal sample points among the plurality of first sample points includes: Calculate the residual between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points; Calculate the mean and standard deviation of the residuals for all the first sample points; The first sample point whose residual is greater than the sum of the mean and the standard deviation by a preset multiple is identified as an outlier sample point.
6. The method of claim 1, wherein, The wireless measurement data of the first sample point is the minimized drive test data.
7. A device positioning method characterized by, include: The method for determining abnormal sample points according to any one of claims 1 to 6, wherein the wireless measurement data of each first sample point further includes location information; Based on the location information corresponding to each of the abnormal sample points, the abnormal coverage area is determined; Based on the abnormal coverage area, the location of the signal amplification device is determined.
8. A positioning processing method characterized by comprising: include: The method for determining abnormal sample points according to any one of claims 1 to 6, wherein the wireless measurement data of each first sample point further includes location information; Based on the location information corresponding to each of the abnormal sample points, the abnormal coverage area is determined; Based on the abnormal coverage area, fingerprint data in the fingerprint database is removed. Based on the fingerprint database after removing fingerprint data, a location process is performed.
9. The method of claim 8, wherein, The process of determining the abnormal coverage area includes: performing clustering processing on the abnormal sample points according to their corresponding location information, and determining each dense region formed by the clustering as the abnormal coverage area.
10. An outlier determining apparatus characterized by comprising: include: The acquisition module is used to acquire wireless measurement data of multiple first sample points in the same serving cell. The wireless measurement data of each first sample point includes a time advance measurement value and a reference signal received power measurement value. The prediction module is used to determine the predicted value of the reference signal received power under normal propagation conditions based on the time advance measurement value of each of the first sample points. The determination module is used to determine abnormal sample points among the plurality of first sample points based on the difference between the measured value of the reference signal received power and the corresponding predicted value of the reference signal received power for each of the first sample points.
11. A device positioning apparatus, characterized by, include: An anomaly determination module is used to determine anomaly sample points according to any one of claims 1 to 6, wherein the wireless measurement data of each first sample point further includes location information; The region determination module is used to determine the abnormal coverage area based on the location information corresponding to each of the abnormal sample points; The device positioning module is used to determine the location of the signal amplification device based on the abnormal coverage area.
12. A positioning processing device, characterized by include: An anomaly determination module is used to determine anomaly sample points according to any one of claims 1 to 6, wherein the wireless measurement data of each first sample point further includes location information; The region determination module is used to determine the abnormal coverage area based on the location information corresponding to each of the abnormal sample points; The data removal module is used to remove fingerprint data from the fingerprint database based on the abnormal coverage area. The positioning processing module is used to perform positioning processing based on the fingerprint database after removing fingerprint data.
13. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.
14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method described in any one of claims 1 to 9.
15. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method described in any one of claims 1 to 9.