CN118301658B

CN118301658B - Common site detection method, apparatus, device, storage medium and program product

Info

Publication number: CN118301658B
Application number: CN202410720865.9A
Authority: CN
Inventors: 尹志颖; 鲁继; 柴福仁; 欧阳晔
Original assignee: Asiainfo Technologies China Inc
Current assignee: Asiainfo Technologies China Inc
Priority date: 2024-06-05
Filing date: 2024-06-05
Publication date: 2024-07-30
Anticipated expiration: 2044-06-05
Also published as: CN118301658A

Abstract

The embodiment of the application discloses a common station address detection method, a device, equipment, a storage medium and a program product, which relate to the technical field of communication and comprise the following steps: acquiring a first minimization of drive tests data set of a first base station and a second minimization of drive tests data set of a second base station, wherein the first minimization of drive tests data set of the first base station and the second minimization of drive tests data set of the second base station are acquired in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network; if the first base station and the second base station are determined to have common coverage based on the first minimization drive test data set and the second minimization drive test data set, a first Gaussian mixture model is built based on the position information in the first minimization drive test data set, a second Gaussian mixture model is built based on the position information in the second minimization drive test data set, and if the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model is larger than the target overlapping degree, the common site of the first base station and the second base station is determined. The application provides an unsupervised co-site detection scheme based on big data.

Description

Common site detection method, apparatus, device, storage medium and program product

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method, an apparatus, a device, a storage medium, and a program product for co-sited detection.

Background

Common site refers to the practice of sharing base station facilities in the same geographic location by different communication systems (such as a 4G network and a 5G network) of the same operator, and is a common resource sharing mode in the communication industry, which is helpful for improving the resource utilization efficiency, accelerating the network deployment and reducing the cost of the operator to a certain extent.

Co-sited, while providing cost-effectiveness and rapid deployment advantages, may also introduce interference problems between different systems. In order to ensure that the different communication systems at the co-site do not negatively affect each other's performance, it is necessary to detect the base station at the co-site and further optimize the base station at the co-site to avoid that the different communication systems at the co-site negatively affect each other's performance.

Disclosure of Invention

In view of this, the present application provides a co-sited detection method, apparatus, device, storage medium and program product for detecting a co-sited base station.

In order to achieve the above object, the following solutions have been proposed:

A co-sited detection method comprising:

Acquiring a first minimization of drive tests data set of a first base station and a second minimization of drive tests data set of a second base station, wherein the first minimization of drive tests data set of the first base station and the second minimization of drive tests data set of the second base station are acquired in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network;

Determining whether there is co-coverage for the first base station and the second base station based on the first and second minimization of drive tests data sets;

if the first base station and the second base station have common coverage, a first Gaussian mixture model is built based on the position information in the first minimization drive tests data set, and a second Gaussian mixture model is built based on the position information in the second minimization drive tests data set;

obtaining the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model;

And if the overlapping degree is larger than the target overlapping degree, determining that the first base station and the second base station are co-located, otherwise, determining that the first base station and the second base station are not co-located.

The method, optionally, includes a process of determining whether there is co-coverage between the first base station and the second base station, including:

Preprocessing the first and second minimization drive tests data sets respectively to remove repeated data and abnormal data in the first minimization drive tests data set and repeated data and abnormal data in the second minimization drive tests data set;

Clustering the preprocessed first minimization of drive tests data set to obtain three first type clustering centers corresponding to the first base station; clustering the preprocessed second minimization drive tests data set to obtain three second class clustering centers corresponding to the second base station;

And if the distance between the geographic positions corresponding to any first type of clustering center and any second type of clustering center is smaller than a distance threshold, determining that the first base station and the second base station have co-coverage.

The method, optionally, of preprocessing the first and second minimization drive test data sets respectively, includes:

Dividing any one of the first and second minimization drive tests data sets into a plurality of minimization drive tests data sets for the any one of the minimization drive tests data sets; the method comprises the steps of acquiring the minimum drive test data belonging to the same minimum drive test data group, wherein the minimum drive test data belonging to the same minimum drive test data group are the minimum drive test data with the same target information, which are acquired in the same sub-period of the target period; the target information comprises position information and associated information of a mobility management entity;

for each minimization drive test data set, reserving the minimization drive test data with the reference signal receiving power in the minimization drive test data set closest to the median of the reference signal receiving power in the minimization drive test data set, and deleting other minimization drive test data in the minimization drive test data set;

And filtering the abnormal minimization drive test data of the minimization drive test data reserved in each minimization drive test data set to obtain a preprocessing result corresponding to any minimization drive test data set.

The method, optionally, the process of constructing the first gaussian mixture model and the second gaussian mixture model, includes:

constructing the first Gaussian mixture model based on the position information in the preprocessed first minimization of drive tests data set;

and constructing the second Gaussian mixture model based on the position information in the preprocessed second minimization drive test data set.

In the above method, optionally, the constructing the first gaussian mixture model based on the location information in the preprocessed first minimization of drive tests data set, and constructing the second gaussian mixture model based on the location information in the preprocessed second minimization of drive tests data set includes:

Intercepting the longitude and latitude in the preprocessed first minimization drive test data set to a preset bit number after decimal point;

Constructing a first Gaussian mixture model by utilizing the intercepted longitude and latitude in the preprocessed first minimization of drive test data set;

and constructing a second Gaussian mixture model by using the intercepted longitude and latitude in the preprocessed second minimization of drive test data set.

Constructing a first Gaussian mixture model by utilizing the reference signal receiving power in the preprocessed first minimization of drive test data set and the intercepted longitude and latitude;

and constructing a second Gaussian mixture model by utilizing the reference signal receiving power in the preprocessed second minimization drive test data set and the intercepted longitude and latitude.

A co-sited detection apparatus comprising:

The acquisition module is used for acquiring a first minimization of drive test data set of the first base station and a second minimization of drive test data set of the second base station, which are acquired in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network;

A determining module configured to determine whether there is co-coverage of the first base station and the second base station based on the first and second minimization of drive tests data sets;

the construction module is used for constructing a first Gaussian mixture model based on the position information in the first minimization drive test data set and constructing a second Gaussian mixture model based on the position information in the second minimization drive test data set if the first base station and the second base station have co-coverage;

And the detection module is used for obtaining the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model, determining that the first base station and the second base station share the site if the overlapping degree is larger than the target overlapping degree, and otherwise, determining that the first base station and the second base station do not share the site.

A co-sited detection device comprising a memory and a processor;

The memory is used for storing programs;

the processor is configured to execute the program to implement the steps of the co-sited detection method according to any one of the above claims.

A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the co-sited detection method according to any one of the preceding claims.

A computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the co-sited detection method according to any one of the preceding claims.

As can be seen from the above technical solution, the common-site detection method, apparatus, device, storage medium and program product provided by the embodiments of the present application obtain a first minimization of drive test data set of a first base station and a second minimization of drive test data set of a second base station, which are collected in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network; if the first base station and the second base station are determined to have co-coverage based on the first minimization drive test data set and the second minimization drive test data set, a first Gaussian mixture model is built based on the position information in the first minimization drive test data set, a second Gaussian mixture model is built based on the position information in the second minimization drive test data set, if the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model is larger than the target overlapping degree, the co-sited of the first base station and the second base station is determined, otherwise, the co-sited of the first base station and the second base station is determined. The application provides an unsupervised co-site detection scheme based on big data.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of one implementation of a method for co-sited detection disclosed in an embodiment of the present application;

FIG. 2 is an exemplary diagram of the different cases when two Gaussian distributions are mixed as disclosed in the embodiments of the present application;

FIG. 3 is a flowchart of one implementation of determining whether there is co-coverage between a first base station and a second base station according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a co-sited detection device according to an embodiment of the present application;

fig. 5 is a block diagram of a hardware structure of a common site detection apparatus according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Taking a 4G network and a 5G network as an example, the goal of co-sited planning is to deploy 4G and 5G devices on the same site to achieve sharing and optimization of infrastructure. This means that 4G and 5G devices will share physical devices and resources like antennas, transmission lines etc. However, there are some differences in technical specifications between the 4G and 5G networks, including frequency bands, modulation schemes, etc., and thus their coverage areas may be different. In the case of co-sited, operators can adjust the coverage of 4G and 5G according to actual needs and network planning.

In order to realize co-site detection, one implementation manner is to input multidimensional features (such as RSRP mean values of a main cell and neighbor cells, TA, correlation coefficients of the main neighbor cells, and the like) in measurement reports (Measurement Report, MR) of two base stations into a pre-trained co-site detection model to obtain a detection result output by the co-site detection model, wherein the detection result represents whether the two base stations share the site. The essence of the co-sited detection model is a classification model, namely the classification model judges whether two base stations co-sited according to the multidimensional features of the measurement report data of the two base stations input into the classification model.

Each training sample in the training data set for training the co-sited detection model is a multidimensional feature in measurement report data of two base stations, and a label of the training sample characterizes whether the two base stations involved in the training sample are co-sited. The labels of the training samples may be manually labeled or may be determined based on base station location information in the two base station's ginseng, where if the base station location information in the two base station's ginseng is the same, the two base stations are determined to be co-sited, otherwise, the two base stations are determined to be not co-sited.

As an example, the co-site detection model may be any of the following: random Forest (Random Forest), gradient lifting decision tree (Gradient Boosting Decision Tree, GBDT), xgboost (eXtreme Gradient Boosting), etc.

The application discovers that the supervised co-sited detection scheme can cause inaccurate label of training samples due to inaccurate industrial parameters in the actual engineering implementation process. In addition, due to the limitation of the training samples, the prediction result of the co-sited detection model may be inaccurate due to factors such as the difference of wireless environments or coverage scenes of the base station to be subjected to co-sited detection and the base station related to the training samples, and insufficient coverage of the samples. Moreover, the supervised model has the problem of low learning efficiency.

In order to avoid being influenced by inaccurate industrial parameters, industrial parameter data are not used any more, and an unsupervised co-site detection scheme based on big data is provided.

The common-station address detection scheme of the application belongs to a detection method in the aspect of a mobile communication wireless access side, and can be used for common-station address detection of a 4G base station and a 5G base station which are planned as common-station addresses in a local network by an operator. In the practical application scenario, although two base stations are planned to be co-located, in the actual site building process, the two base stations which are well built are not co-located due to the influence of factors such as construction or site environment, and for general services (i.e. the co-located services are not considered), the co-located services can be deployed regardless of whether the base stations are co-located or not, but when the co-located related services are to be deployed later, the co-located detection of the base stations is needed, and when the co-located services of the base stations are detected, the related services are deployed again, otherwise, the co-located related tasks cannot be deployed.

As shown in fig. 1, a flowchart for implementing a method for detecting a co-sited according to an embodiment of the present application may include:

Step S101: a first minimization of drive tests dataset of a first base station acquired during a target period of time and a second minimization of drive tests dataset of a second base station are obtained. The first base station belongs to a first communication network and the second base station belongs to a second communication network.

Minimization of drive tests (Minimization DRIVE TEST, MDT) is data reported to the base station by the mobile communication terminal, and thus, the Minimization of drive tests is data without human participation. Each mobile communication terminal which establishes connection with the base station can collect and report the minimization of drive test data to the base station at regular or irregular time.

The first communication network and the second communication network are different communication networks. For example, the first communication network is a 4G network and the second communication network is a 5G network; or the first communication network is a 5G network, the second communication network is a 6G network, etc.

If the mobile communication terminal accesses the first communication network through the first base station, the mobile communication terminal reports the minimization of drive test data of the first communication network to the first base station. And if the mobile communication terminal accesses the second communication network through the second base station, reporting the minimization of drive test data of the second communication network to the second base station by the mobile communication terminal.

Taking 4G networks and 5G networks as examples, the minimization of drive tests data in the 4G networks includes, but is not limited to, the following information: acquisition time, base station ID, own cell timing advance (TIMING ADVANCE, TA), own cell reference signal received Power (REFERENCE SIGNAL RECEIVING Power, RSRP), mobility management entity application process identifier (MME Application Identifier, MMEAPID), mobility management entity (Mobility MANAGEMENT ENTITY, MME) packet identity, MME number, UE (i.e. mobile communication terminal) longitude, UE latitude.

The minimization of drive tests data in the 5G network includes, but is not limited to, the following information: acquisition time, base station ID, own cell timing advance (TIMING ADVANCE, TA), own cell reference signal received Power (Synchronization SIGNAL REFERENCE SIGNAL RECEIVING Power, SSRSRP), access and mobility management function identifier (AMFUENGAPID), AMF area identity (AMFRegionID), AMF pointer (AMFPointer), UE longitude, UE latitude.

As an example, the target period may be a period of time nearest to the current time (time at which co-sited detection starts), for example, may be one month, or half month, or two months, or the like.

Step S102: a determination is made as to whether there is co-coverage of the first base station and the second base station based on the first and second MDT data sets.

The existence of the first base station and the second base station is that the signal coverage of the first base station and the signal coverage of the second base station have an overlapping area.

As an example, a statistical analysis may be performed on the first and second minimization of drive tests data sets to determine if there is co-coverage of the first and second base stations.

Step S103: if the first base station and the second base station have common coverage, a first Gaussian mixture model is built based on the position information in the first minimization of drive test data set, and a second Gaussian mixture model is built based on the position information in the second minimization of drive test data set.

The inventor of the present application has found that if only the coverage situation of two base stations is seen, it is impossible to determine whether the two base stations share the site.

In the case where there is co-coverage between the first base station and the second base station, the first base station and the second base station may or may not co-sited; if there is no co-coverage between the first base station and the second base station, it is not necessarily possible for the first base station and the second base station to co-site.

Step S104: and obtaining the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model.

The degree of overlap of the two gaussian mixture models may be determined based on the saddle point formed by mixing the two gaussian mixture models and the lower peak of the two peaks associated with the saddle point.

Taking two-dimensional gaussian distribution as an example, referring to fig. 2, an exemplary diagram of different situations when two gaussian distributions are mixed is provided in an embodiment of the present application. When two Gaussian distributions overlap, a saddle shape is formed (as shown in the case a and the case b in fig. 2), and two peaks and a saddle point appear when two-dimensional Gaussian distributions are mixed; when the two gaussian distributions are very close, the saddle point disappears, thereby forming a gaussian distribution (as shown in fig. 2, c).

Each gaussian mixture model may have only one peak or may have multiple peaks.

Alternatively, in the case where there is only one peak per gaussian mixture model, the degree of overlap of the two gaussian mixture models is positively correlated with the saddle point, and the lower peak (i.e., peak 1) of the two peaks associated with the saddle point (peak 1 and peak 2 as shown in case b in fig. 2) is negatively correlated. As an example, the degree of overlap of two gaussian mixture models may be: the ratio of the probability density at the saddle point to the probability density at the lower peak. By way of example, if two gaussian mixture models have no saddle points after mixing, and only one peak (as shown in the case c in fig. 2), then the overlap of the two gaussian mixture models is 1 (i.e., the two gaussian mixture models are substantially identical or the two gaussian mixture models are very similar).

Alternatively, where there are multiple peaks for each gaussian mixture model, at least one saddle point may be formed. Assuming that each gaussian mixture model has N (N is equal to or greater than 2) peaks, after mixing the two gaussian mixture models, for any first peak value in the first gaussian mixture model and any second peak value in the second gaussian mixture model, calculating an overlapping degree of the two gaussian mixture models based on saddle points corresponding to any first peak value and any second peak value and lower peak value in any first peak value and any second peak value, obtaining N ² overlapping degrees in total by the two gaussian mixture models based on the saddle points, selecting N nearest overlapping degrees in the N ² overlapping degrees, and determining the average value of the N nearest overlapping degrees as the overlapping degree of the two gaussian mixture models.

Step S105: and if the overlapping degree is larger than the target overlapping degree, determining that the first base station and the second base station are co-sited, otherwise, determining that the first base station and the second base station are not co-sited.

As an example, the target overlap is 0.8, i.e. if the overlap is greater than 0.8, it is determined that the first base station and the second base station are co-sited, otherwise it is determined that the first base station and the second base station are not co-sited.

Optionally, if the overlapping degree is greater than the first target overlapping degree, determining that the first base station and the second base station share the site, otherwise, if the overlapping degree is greater than the second target overlapping degree, and is less than or equal to the first target overlapping degree, determining that the first base station and the second base station may share the site, and if the overlapping degree is less than or equal to the second target overlapping degree, determining that the first base station and the second base station have no correlation, that is, have no site. The second target overlap is less than the first target overlap.

As an example, the first target overlap is 0.8 and the second target overlap is 0.6, the first base station and the second base station are determined to be co-sited if the overlap is greater than 0.8, the first base station and the second base station are determined to be possibly co-sited if the overlap is greater than 0.6 and less than or equal to 0.8, and the first base station and the second base station are determined to be not co-sited if the overlap is less than or equal to 0.6.

The common station address detection method provided by the embodiment of the application obtains a first minimization of drive test data set of a first base station and a second minimization of drive test data set of a second base station, wherein the first minimization of drive test data set of the first base station and the second minimization of drive test data set of the second base station are acquired in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network; if the first base station and the second base station are determined to have co-coverage based on the first minimization drive test data set and the second minimization drive test data set, a first Gaussian mixture model is built based on the position information in the first minimization drive test data set, a second Gaussian mixture model is built based on the position information in the second minimization drive test data set, if the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model is larger than the target overlapping degree, the co-sited of the first base station and the second base station is determined, otherwise, the co-sited of the first base station and the second base station is determined. The application provides an unsupervised co-site detection scheme based on big data, which analyzes the authenticity of the co-sites of base stations of different networks in a co-coverage scene from an objective angle, and avoids the negative influence of human participation on the co-site detection.

In an alternative embodiment, a flowchart of an implementation of determining whether there is co-coverage between the first base station and the second base station is shown in fig. 3, may include:

Step S301: and respectively preprocessing the first and second minimization drive tests data sets to remove repeated data and abnormal data in the first and second minimization drive tests data sets.

The method comprises the steps of preprocessing a first minimization drive test data set to remove repeated data and abnormal data in the first minimization drive test data set; preprocessing the second minimization drive tests data set to remove repeated data and abnormal data in the second minimization drive tests data set. By removing repeated data and abnormal data in the minimized drive test data, the data processing amount can be reduced, and the efficiency and accuracy of common station address detection can be improved.

Step S302: clustering the preprocessed first minimization drive tests data set to obtain three first type clustering centers corresponding to the first base station; and clustering the preprocessed second MDT data set to obtain three second class clustering centers corresponding to the second base station.

As an example, the first preprocessed set of minimization drive tests data may be clustered using a K-means algorithm, and the second preprocessed set of minimization drive tests data may be clustered using a K-means algorithm. The specific clustering process can refer to the existing scheme, and is not described in detail here.

In order to completely cover the signal radiation range of the base station, the category number of the application is 3 when the preprocessed first minimization drive test data set is clustered, and the category number of the application is 3 when the preprocessed second minimization drive test data set is clustered.

Step S303: and if the distance between the geographic positions corresponding to any first type of clustering center and any second type of clustering center is smaller than the distance threshold, determining that the first base station and the second base station have co-coverage.

That is, the application calculates the geographic distance between the longitude and latitude in the minimized drive test data as the center of any first type of cluster and the longitude and latitude in the minimized drive test data as the center of any second type of cluster.

As an example, the distance threshold may be 100 meters, 200 meters, or any value between 100 meters and 200 meters.

And if the distance between any first type of clustering center and the geographic position corresponding to the second type of clustering center is larger than or equal to a distance threshold, namely, the distance between the first type of clustering center and the geographic position corresponding to the second type of clustering center is not smaller than the distance threshold, determining that the first base station and the second base station are not covered together.

In an alternative embodiment, one implementation of the preprocessing of the first mdt data set may be:

Dividing the first minimization of drive tests data set into a plurality of minimization of drive tests data sets; the method comprises the steps that the minimization drive test data belonging to the same minimization drive test data group are the minimization drive test data with the same target information, wherein the minimization drive test data are collected in the same sub-period of the target period; the target information includes location information and association information of a Mobility Management Entity (MME), wherein the association information of the MME includes at least one of: the mobility management entity applies a process identifier (MMEAPID), a mobility management entity group identity (MME group identity), and a mobility management entity number (MME number).

Specifically, the target period may be divided into a plurality of sub-periods (for example, one sub-period every 5 minutes, or one sub-period every 15 minutes, etc.), and the minimization of drive test data with the collection time within the same sub-period is grouped according to the target information, that is: and dividing the minimized drive test data with the same target information into a group in the minimized drive test data with the acquisition time within the same subperiod. Based on this, assuming that the above-described target period is divided equally into M sub-periods, the present application divides the minimized drive test data acquired in the i-th sub-period into P _i (i=1, 2,3, … …, M) minimized drive test data sets based on the target information, and divides the first minimized drive test data set into Q minimized drive test data sets, where q=p ₁+ P₂+ P₃+……,P_M.

The application determines each piece of minimization drive test data belonging to the same minimization drive test data group as the repeated minimization drive test data.

For each minimization drive test data set, the minimization drive test data with the Reference Signal Received Power (RSRP) in the minimization drive test data set closest to the median of the reference signal received power in the minimization drive test data set is reserved, and other minimization drive test data in the minimization drive test data set is deleted.

For the j (j=1, 2,3, … …, Q) th minimization drive test data set, a statistical analysis may be performed on RSRP of each minimization drive test data in the j (minimization) th minimization drive test data set to determine the RSRP median in the j (minimization) th minimization drive test data set, the minimization drive test data closest to the RSRP median is reserved as a representative of the j (minimization) th minimization drive test data set, and other minimization drive test data in the j (minimization) th minimization drive test data set is deleted as repeated data.

The RSRP median in the jth minimization drive test data set may be the RSRP in a certain minimization drive test data in the jth minimization drive test data set, where the minimization drive test data with the RSRP and the RSRP median closest to each other is the minimization drive test data with the RSRP as the median.

The RSRP median in the jth minimization drive test data set may also be the average of two RSRPs with RSRP ranks located at the middle, where the minimization drive test data with RSRP closest to the RSRP median may be any one of the two minimization drive test data with RSRP ranks located at the middle.

And filtering the abnormal minimization drive test data of the minimization drive test data reserved in each minimization drive test data set to obtain a preprocessing result (namely a preprocessed first minimization drive test data set) corresponding to the first minimization drive test data set.

Taking the foregoing partitioning of the first set of minimization drive tests into Q sets of minimization drive tests as an example, Q sets of minimization drive tests remain by deleting repeated minimization drive tests in each set of minimization drive tests. The remaining Q pieces of minimization drive test data may have abnormal minimization drive test data, and the abnormal minimization drive test data needs to be filtered out, so as to avoid negative effects of the abnormal minimization drive test data on co-sited detection.

Based on the application, the compressed data volume is prepared for the subsequent flow under the condition of ensuring that the position data is not lost.

In an alternative embodiment, an implementation manner of preprocessing the second minimization of drive test data set may be:

Dividing the second minimization of drive tests data set into a plurality of minimization of drive tests data sets; the method comprises the steps that the minimization drive test data belonging to the same minimization drive test data group are the minimization drive test data with the same target information, wherein the minimization drive test data are collected in the same sub-period of the target period; the target information includes location information and association information of a mobility management entity (AMF), the association information of the AMF including at least one of: an access and mobility management function identifier (AMFUENGAPID), an AMF area identification (AMFRegionID) and an AMF pointer (AMFPointer).

Specifically, the target period may be divided into a plurality of sub-periods (for example, one sub-period every 5 minutes, or one sub-period every 15 minutes, etc.), and the minimization of drive test data with the collection time within the same sub-period is grouped according to the target information, that is: and dividing the minimized drive test data with the same target information into a group in the minimized drive test data with the acquisition time within the same subperiod. Based on this, assuming that the above-described target period is divided equally into M sub-periods, the present application divides the minimized drive test data acquired in the i-th sub-period into P _i (i=1, 2,3, … …, M) minimized drive test data sets based on the target information, and divides the second minimized drive test data set into Q minimized drive test data sets, where q=p ₁+ P₂+ P₃+……,P_M.

For each of the MDT data sets, reserving the MDT data with the reference signal received power (SSRSRP) in the MDT data set closest to the median of the reference signal received power in the MDT data set, and deleting other MDT data in the MDT data set.

For the j (j=1, 2,3, … …, Q) th minimization drive test data set, a statistical analysis may be performed on SSRSRP in each of the j-th minimization drive test data set to determine the SSRSRP median in the j-th minimization drive test data set, the minimization drive test data closest to the SSRSRP and SSRSRP median being retained as representative of the j-th minimization drive test data set, the other minimization drive test data in the j-th minimization drive test data set being deleted as recurring data.

The SSRSRP median of the jth minimization drive test data set may be SSRSRP of one of the jth minimization drive test data sets, and at this time, the minimization drive test data with the closest median between SSRSRP and SSRSRP is the minimization drive test data with SSRSRP as the median.

The SSRSRP median in the j-th set of minimization drive tests data may also be the average of the SSRSRP order of the two SSRSRP median, at which time the closest minimization drive tests data in SSRSRP and SSRSRP median may be any one of the SSRSRP order of the two median minimization drive tests data.

And filtering the abnormal minimization drive test data of the minimization drive test data reserved in each minimization drive test data set to obtain a preprocessing result (namely a preprocessed second minimization drive test data set) corresponding to the second minimization drive test data set.

Taking the foregoing division of the second set of minimization drive tests into Q sets of minimization drive tests as an example, Q sets of minimization drive tests remain by deleting repeated minimization drive tests in each set of minimization drive tests. The remaining Q pieces of minimization drive test data may have abnormal minimization drive test data, and the abnormal minimization drive test data needs to be filtered out, so as to avoid negative effects of the abnormal minimization drive test data on co-sited detection.

As described above, the present application pre-processes the first and second minimization of drive tests data sets in the same manner.

In an alternative embodiment, when the first and second sets of minimization drive tests are preprocessed, the ratio of the abnormal minimization drive tests to all the minimization drive tests retained in each set of minimization drive tests is within a target range, which may be [5%,10% ] as an example. For example, the above-mentioned ratio may be 5% or 10% or 7% or the like.

In an alternative embodiment, an implementation manner of filtering the abnormal minimization of drive test data of the minimization of drive test data reserved in each minimization of drive test data set may be as follows:

Based on the reference signal received power in the minimization drive test data, an isolated Forest (Isolation Forest) algorithm is adopted to detect abnormal reference signal received power in the minimization drive test data reserved in each minimization drive test data set, and the minimization drive test data containing the abnormal reference signal received power is determined to be abnormal minimization drive test data.

The overall implementation flow of this step can refer to the existing scheme, and will not be described here again, and some differences between the present application and the existing scheme will be mainly described here:

The orphan forest algorithm identifies abnormal reference signal received power by constructing a plurality of orphan trees (also referred to as decision trees) based on the reference signal received power in the retained minimization of drive test data. In the present application, the number of the isolated trees has a value in the range of [50,150]. As an example, the number of orphaned trees may be 50 or 100 or 150, etc.

The isolated forest algorithm calculates abnormal scores of the reference signal received power in each piece of minimization drive test data, and sorts the pieces of minimization drive test data according to the order from small to large of the abnormal scores of the reference signal received power. In the application, the first H pieces of minimization drive test data are determined to be the abnormal minimization drive test data according to the target duty ratio of the abnormal minimization drive test data. That is, the duty ratio of the first H pieces of minimization drive tests data among all the minimization drive tests data retained in each minimization drive test data set is the target duty ratio.

The detected anomaly minimization of drive tests data is deleted.

In an alternative embodiment, an implementation manner of constructing the first gaussian mixture model and the second gaussian mixture model may be:

And constructing a first Gaussian mixture model based on the position information in the preprocessed first minimization drive test data set.

And constructing a second Gaussian mixture model based on the position information in the preprocessed second minimization drive tests data set.

That is, the present application pre-processes the first and second minimization drive tests data sets, and then constructs a gaussian mixture model using the location information in the pre-processed minimization drive tests data sets. Therefore, the constructed Gaussian mixture model can be ensured to more accurately represent the probability distribution condition of the position of the mobile communication terminal.

In constructing the gaussian mixture model, the original position information (i.e., the original latitude and longitude data) in the minimization of drive tests data can be used to construct the gaussian mixture model. In order to further improve the accuracy of co-sited detection, the method and the device rasterize original position information in the minimization of drive test data, and construct a first Gaussian mixture model and a second Gaussian mixture model based on the rasterized position information.

Based on this, an implementation manner of constructing the first gaussian mixture model and the second gaussian mixture model provided by the embodiment of the present application may be:

intercepting the longitude and latitude in the preprocessed first minimization drive test data set and the longitude and latitude in the preprocessed second minimization drive test data set to a preset bit number after decimal point.

As an example, the preset number of bits is four, i.e., four bits after the longitude is truncated to the decimal point, and four bits after the latitude is truncated to the decimal point. That is, regardless of the longitude or latitude, the fifth digit following the decimal point and the digit following the decimal point are deleted directly, and only the four digits following the decimal point are reserved.

And constructing a first Gaussian mixture model by using the intercepted longitude and latitude in the preprocessed first minimization of drive test data set.

The first gaussian mixture model is a two-dimensional gaussian mixture model with latitude and longitude as variables. Optionally, the first gaussian mixture model has three peaks, i.e. the first gaussian mixture model is weighted by three single gaussian models, and the sum of the weights of the single gaussian models is 1.

The second gaussian mixture model is also a two-dimensional gaussian mixture model with latitude and longitude as variables. Optionally, the second gaussian mixture model has three peaks, that is, the second gaussian mixture model is weighted by three single gaussian models, and the sum of the weights of the single gaussian models is 1.

The construction process of the first gaussian mixture model and the construction process of the second gaussian mixture model can refer to the existing scheme, and the application is not described in detail.

By intercepting longitude and latitude data, the probability density of the preprocessed first and second minimization drive test data sets is calculated more closely to that of the first and second minimization drive test data sets on a small area, and thus the accuracy of co-sited detection is further improved.

Another implementation manner of constructing the first gaussian mixture model and the second gaussian mixture model provided by the embodiment of the present application may be:

And constructing a first Gaussian mixture model by utilizing the Reference Signal Received Power (RSRP) in the preprocessed first minimization of drive test data set and the intercepted longitude and latitude.

The first Gaussian mixture model is a three-dimensional Gaussian mixture model taking RSRP and longitude and latitude after interception as variables. Optionally, the first gaussian mixture model has three peaks, that is, the first gaussian mixture model is weighted by three single gaussian models, and the sum of weights of the single gaussian models is 1, and each single gaussian model corresponds to one peak.

Alternatively, the RSRP may be normalized to obtain a normalized RSRP, and the three-dimensional gaussian mixture model is constructed by using the normalized RSRP and the longitude and latitude after interception as variables, as the first gaussian mixture model.

For any RSRP, a difference between the any RSRP and the minimum RSRP in the preprocessed first minimization of drive test data set (denoted as a first difference) and a difference between the maximum RSRP and the minimum RSRP in the preprocessed first minimization of drive test data set (denoted as a second difference) may be obtained, and a ratio of the first difference to the first difference is determined as a normalized result of the any RSRP.

And constructing a second Gaussian mixture model by utilizing the reference signal received power (SSRSRP) in the preprocessed second minimization of drive test dataset and the intercepted longitude and latitude.

The second Gaussian mixture model is also a three-dimensional Gaussian mixture model taking SSRSRP and the intercepted longitude and latitude as variables. Optionally, the second gaussian mixture model has three peaks, that is, the second gaussian mixture model is weighted by three single gaussian models, the sum of weights of the single gaussian models is 1, and each single gaussian model corresponds to one peak.

Optionally, the SSRSRP may be normalized to obtain normalized SSRSRP, and the three-dimensional gaussian mixture model is constructed with normalized SSRSRP and the truncated longitude and latitude as variables to be used as the second gaussian mixture model.

By normalizing RSRP and SSRSRP, the two gaussian mixture models are made comparable.

For any one SSRSRP, a difference between the any one SSRSRP and the minimum SSRSRP in the preprocessed second minimization of drive tests data set (denoted as a first difference) and a difference between the maximum SSRSRP and the minimum SSRSRP in the preprocessed second minimization of drive tests data set (denoted as a second difference) may be obtained, and a ratio of the first difference to the first difference is determined as a normalized result of the any one SSRSRP.

Unlike the previous embodiment of building the gaussian mixture model, the present embodiment further improves the accuracy of co-site detection by taking reference signal received power (RSRP/SSRSRP) into account when building the gaussian mixture model.

Assuming that the single-Gaussian model forming the first Gaussian mixture model is a first-type single-Gaussian model and the single-Gaussian model forming the second Gaussian mixture model is a second-type single-Gaussian model, the probability density of saddle points obtained by mixing any one of the first-type single-Gaussian model and any one of the second-type single-Gaussian model is related to the weight of any one of the first-type single-Gaussian models and the weight of any one of the second-type single-Gaussian models. The degree of overlap of the two gaussian mixture models is not a proportion of the data (i.e., position data) falling within the overlap region, and thus is irrelevant in terms of the amount of data, and is only related to the distribution of the data.

Corresponding to the method embodiment, the present application provides a common-site detection device, and a schematic structural diagram of the common-site detection device provided by the embodiment of the present application is shown in fig. 4, which may include:

the device comprises an acquisition module 401, a judgment module 402, a construction module 403 and a detection module 404;

The obtaining module 401 is configured to obtain a first minimization of drive test data set of a first base station and a second minimization of drive test data set of a second base station, where the first minimization of drive test data set is collected in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network;

A determining module 402 is configured to determine whether there is co-coverage between the first base station and the second base station based on the first and second minimization of drive tests data sets;

The building module 403 is configured to build a first gaussian mixture model based on the position information in the first minimization of drive tests data set and a second gaussian mixture model based on the position information in the second minimization of drive tests data set if there is co-coverage between the first base station and the second base station;

the detection module 404 is configured to obtain a degree of overlap of the first gaussian mixture model and the second gaussian mixture model, determine that the first base station and the second base station are co-sited if the degree of overlap is greater than a target degree of overlap, and otherwise determine that the first base station and the second base station are not co-sited.

The common station address detection device provided by the embodiment of the application obtains a first minimization of drive test data set of a first base station and a second minimization of drive test data set of a second base station, wherein the first minimization of drive test data set of the first base station and the second minimization of drive test data set of the second base station are acquired in a target period; the first base station belongs to a first communication network, and the second base station belongs to a second communication network; if the first base station and the second base station are determined to have co-coverage based on the first minimization drive test data set and the second minimization drive test data set, a first Gaussian mixture model is built based on the position information in the first minimization drive test data set, a second Gaussian mixture model is built based on the position information in the second minimization drive test data set, if the overlapping degree of the first Gaussian mixture model and the second Gaussian mixture model is larger than the target overlapping degree, the co-sited of the first base station and the second base station is determined, otherwise, the co-sited of the first base station and the second base station is determined. The application provides an unsupervised co-site detection scheme based on big data, which analyzes the authenticity of the co-sites of base stations of different networks in a co-coverage scene from an objective angle, and avoids the negative influence of human participation on the co-site detection.

In an alternative embodiment, the determining module 402 is configured to, when determining whether there is co-coverage between the first base station and the second base station:

In an alternative embodiment, the determining module 402 is configured to, when preprocessing the first and second minimization of drive tests data sets, respectively:

In an alternative embodiment, the building module 403 is configured to, when building the first gaussian mixture model and the second gaussian mixture model:

In an alternative embodiment, the building module 403 is configured to build the first gaussian mixture model based on the location information in the preprocessed first minimization of drive tests data set, and when building the second gaussian mixture model based on the location information in the preprocessed second minimization of drive tests data set, is configured to:

The common-site detection device provided by the embodiment of the application can be applied to common-site detection equipment, such as PC terminals, cloud platforms, servers, server clusters and the like. Alternatively, fig. 5 shows a block diagram of a hardware structure of the co-sited detection apparatus, and referring to fig. 5, the hardware structure of the co-sited detection apparatus may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

In the embodiment of the application, the number of the processor 1, the communication interface 2, the memory 3 and the communication bus 4 is at least one, and the processor 1, the communication interface 2 and the memory 3 complete the communication with each other through the communication bus 4;

the processor 1 may be a central processing unit CPU, or an Application-specific integrated Circuit ASIC (Application SPECIFIC INTEGRATED Circuit), or one or more integrated circuits configured to implement embodiments of the present invention, etc.;

The memory 3 may comprise a high-speed RAM memory, and may further comprise a non-volatile memory (non-volatile memory) or the like, such as at least one magnetic disk memory;

Wherein the memory stores a program, the processor is operable to invoke the program stored in the memory, the program operable to:

Alternatively, the refinement function and the extension function of the program stored in the memory may be described with reference to the above.

The embodiment of the present application also provides a storage medium storing a program adapted to be executed by a processor, the program being configured to:

Alternatively, the refinement function and the extension function of the program stored in the storage medium may be described with reference to the above.

Embodiments of the present application also provide a computer program product comprising a computer program/instruction which, when executed by a processor, is adapted to:

Alternatively, the refinement and expansion functions of the computer program/instructions may be as described above.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A co-sited detection method, comprising:

2. The method of claim 1, wherein determining whether there is co-coverage for the first base station and the second base station comprises:

3. The method of claim 2, wherein the separately preprocessing the first and second minimization of drive tests data sets comprises:

4. The method of claim 2, wherein the process of constructing the first gaussian mixture model and the second gaussian mixture model comprises:

5. The method of claim 4, wherein the constructing the first gaussian mixture model based on the location information in the preprocessed first minimization of drive tests data set and the second gaussian mixture model based on the location information in the preprocessed second minimization of drive tests data set comprises:

6. The method of claim 4, wherein the constructing the first gaussian mixture model based on the location information in the preprocessed first minimization of drive tests data set and the second gaussian mixture model based on the location information in the preprocessed second minimization of drive tests data set comprises:

7. A co-sited detection apparatus, comprising:

8. A co-sited detection apparatus comprising a memory and a processor;

The memory is used for storing programs;

The processor being configured to execute the program to implement the steps of the co-sited detection method according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the co-sited detection method according to one of claims 1-6.

10. A computer program product comprising computer program/instructions which, when executed by a processor, implement the steps of the co-sited detection method according to any one of claims 1-6.