CN114390605B

CN114390605B - Switching method, device, equipment and storage medium

Info

Publication number: CN114390605B
Application number: CN202011133860.4A
Authority: CN
Inventors: 常世元; 李高盛; 李玉诗; 张斌
Original assignee: Datang Mobile Communications Equipment Co Ltd
Current assignee: Datang Mobile Communications Equipment Co Ltd
Priority date: 2020-10-21
Filing date: 2020-10-21
Publication date: 2023-07-14
Anticipated expiration: 2040-10-21
Also published as: CN114390605A

Abstract

The embodiment of the application provides a switching method, a device, equipment and a storage medium, wherein the method comprises the following steps: respectively acquiring RSRP measurement data samples of terminals in each cluster; determining a target cluster with highest RSRP measurement data similarity with the target terminal based on the RSRP measurement data sample of the terminal in each cluster; configuring a switching parameter threshold of a target terminal based on a switching parameter threshold corresponding to the target cluster, and completing switching of the target terminal; wherein, the switching parameter thresholds corresponding to different clusters are different. According to the embodiment of the application, the user is clustered, the switching parameter thresholds corresponding to different clusters are different, the cluster to which the target terminal belongs is determined, and the switching is performed after the corresponding switching parameter threshold is determined, so that different switching parameter configurations are performed on the terminals in different scenes, ping-pong switching is reduced, switching when the terminal signals are poor is avoided, and system performance is improved through advanced switching or delayed switching when dropped calls are reduced.

Description

Switching method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a switching method, device, apparatus, and storage medium.

Background

In the 5G system, handover is used to ensure link quality of a UE (User Equipment) in a mobile situation.

The existing switching method sets a unified threshold for switching parameters of all user terminals in the same cell, and the base station performs switching control according to the unified parameter set threshold according to the measurement report reported by the UE. However, when the user terminal scene covered by the same cell is complex, the scheme of using the unified switching parameters easily causes that part of the user terminals are switched too late or too early, thereby affecting the user perception and the system performance.

Therefore, how to propose a handover method suitable for the same cell coverage with complex user terminal scenario becomes a problem to be solved

Disclosure of Invention

The embodiment of the application provides a switching method, a device, equipment and a storage medium, which are used for solving the problems that in the prior art, the switching method is not suitable for the situation that the scene of a user terminal covered by the same cell is complex, and part of user terminals are easy to switch too late or too early, so that the complex switching of the scene of the user terminal suitable for the coverage of the same cell is realized.

In a first aspect, an embodiment of the present application provides a handover method, including:

Acquiring RSRP (Reference Signal Receiving Power, reference signal received power) measurement data samples of terminals in each cluster respectively;

determining a target cluster with highest RSRP measurement data similarity with the target terminal based on the RSRP measurement data samples of the terminals in each cluster;

configuring a switching parameter threshold of the target terminal based on the switching parameter threshold corresponding to the target cluster, and completing switching of the target terminal;

wherein, the switching parameter thresholds corresponding to different clusters are different.

Optionally, according to a handover method of one embodiment of the present application, before the acquiring RSRP measurement data samples of the terminals in each cluster, the method further includes:

acquiring Reference Signal Received Power (RSRP) measurement data of a terminal to be clustered for all cells covering the terminal to be clustered;

clustering all terminals to be clustered based on RSRP measurement data of all the terminals to be clustered in a preset area;

and determining a switching parameter threshold corresponding to each cluster based on the receiving level quantity of the terminal in each cluster.

Optionally, according to a handover method of an embodiment of the present application, the determining, based on the received level quantity of the terminal in each cluster, a handover parameter threshold corresponding to each cluster specifically includes:

Acquiring the receiving level quantity of all terminals in each cluster;

comparing the receiving level amounts of all the terminals of each cluster, and determining a switching parameter threshold corresponding to each cluster; the handover parameter threshold is inversely proportional to the amount of reception level of all terminals.

Optionally, according to a handover method of an embodiment of the present application, the obtaining reference signal received power RSRP measurement data of a terminal to be clustered for all cells covering the terminal to be clustered includes:

acquiring a first matrix formed by Reference Signal Received Power (RSRP) measurement data of a terminal to be clustered on all cells covering the terminal to be clustered;

based on the first matrix, obtaining a second matrix according to a TSNE (T-Stochastic neighbour Embedding, T distribution and random neighbor embedding) algorithm, wherein the second matrix comprises sample data of each terminal to be clustered; and the sample data of the terminals to be clustered are used for describing RSRP measurement data of the terminals to be clustered.

Optionally, according to a handover method of an embodiment of the present application, the obtaining, based on the first matrix, a second matrix according to a TSNE algorithm includes:

initializing sample data of each terminal to be clustered in the second matrix;

Calculating to obtain probability distribution p of the first matrix based on RSRP measurement data of each terminal to be clustered in the first matrix;

iteratively updating the second matrix according to a TSNE algorithm based on the probability distribution p;

and after the second matrix is determined to be iteratively updated, acquiring the second matrix in the last updating process.

Optionally, according to a handover method of an embodiment of the present application, the iteratively updating the second matrix according to a TSNE algorithm based on the probability distribution p includes:

in each updating process, calculating and obtaining probability distribution q of the second matrix based on sample data of each terminal in the second matrix obtained in the previous updating process;

calculating to obtain KL divergence between the probability distribution p and the probability distribution q;

the second matrix is updated based on the KL (Kullback-Leibler Divergence, relative entropy) divergence.

Optionally, according to a handover method of an embodiment of the present application, the determining that the iterative updating of the second matrix ends includes:

the KL divergence is lower than a preset value; or the number of iterative updates exceeds the preset number of iterative updates.

Optionally, according to a handover method of an embodiment of the present application, the method further includes:

Clustering all the terminals to be clustered again at intervals of a first preset time, and determining a switching parameter threshold corresponding to each cluster; and/or

And re-determining the switching parameter threshold corresponding to each cluster at intervals of a second preset time.

In a second aspect, embodiments of the present application further provide a network device, including a memory, a transceiver, and a processor:

a memory for storing a computer program; a transceiver for transceiving data under control of the processor; a processor for reading the computer program in the memory and performing the following operations:

respectively acquiring RSRP measurement data samples of terminals in each cluster;

Optionally, before the network device according to an embodiment of the present application obtains RSRP measurement data samples of the terminals in each cluster, the operations further include:

Optionally, according to the network device of one embodiment of the present application, the determining, based on the amount of reception level of the terminal in each cluster, a handover parameter threshold corresponding to each cluster specifically includes:

acquiring the receiving level quantity of all terminals in each cluster;

Optionally, according to the network device of one embodiment of the present application, the obtaining the reference signal received power RSRP measurement data of the terminal to be clustered for all cells covering the terminal to be clustered includes:

Based on the first matrix, obtaining a second matrix according to a TSNE algorithm, wherein the second matrix comprises sample data of each terminal to be clustered; and the sample data of the terminals to be clustered are used for describing RSRP measurement data of the terminals to be clustered.

Optionally, according to an embodiment of the present application, the obtaining, based on the first matrix, the second matrix according to the TSNE algorithm includes:

initializing sample data of each terminal to be clustered in the second matrix;

Optionally, according to an embodiment of the present application, the iteratively updating the second matrix according to the TSNE algorithm based on the probability distribution p includes:

updating the second matrix based on the KL divergence.

Optionally, according to the network device of an embodiment of the present application, the determining that the iteration updating the second matrix ends includes:

Optionally, according to an embodiment of the present application, the operations further include:

In a third aspect, an embodiment of the present application further provides a switching device, including:

the acquisition module is used for respectively acquiring RSRP measurement data samples of the terminals in each cluster;

the determining module is used for determining a target cluster with highest RSRP measurement data similarity with the target terminal based on the RSRP measurement data samples of the terminals in each cluster;

the switching module is used for configuring the switching parameter threshold of the target terminal based on the switching parameter threshold corresponding to the target cluster, and completing the switching of the target terminal;

In a fourth aspect, embodiments of the present application further provide a processor-readable storage medium storing a computer program for causing the processor to perform the steps of the handover method according to the first aspect as described above.

According to the switching method, the device, the equipment and the storage medium, through clustering users, the switching parameter thresholds corresponding to different clusters are different, determining the cluster to which the target terminal belongs and determining the corresponding switching parameter threshold and then switching, different switching parameter configurations can be carried out on the terminal under different scenes, ping-pong switching is reduced, switching when the terminal signal is poor is avoided, and system performance is improved through advanced switching or delayed switching when dropped calls are reduced.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief description will be given below of the drawings that are needed in the embodiments or the prior art descriptions, and it is obvious that the drawings in the following description are some embodiments of the present application, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a switching method provided in an embodiment of the present application;

fig. 2 is a schematic flow chart of another switching method provided in the embodiment of the present application;

fig. 3 is a schematic structural diagram of a switching device according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a network device according to an embodiment of the present application.

Detailed Description

In the embodiment of the application, the term "and/or" describes the association relationship of the association objects, which means that three relationships may exist, for example, a and/or B may be represented: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship.

The term "plurality" in the embodiments of the present application means two or more, and other adjectives are similar thereto.

The following description of the technical solutions in the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

The embodiment of the application provides a switching method and a switching device, which are used for reducing ping-pong switching, avoiding switching when signals are poor, and improving user perception through advanced switching or delayed switching when dropped calls are reduced.

The method and the device are based on the same application, and because the principles of solving the problems by the method and the device are similar, the implementation of the device and the method can be referred to each other, and the repetition is not repeated.

The NR (New Radio) system switching algorithm adopts a mode of UE auxiliary network control, a base station configures measurement parameters for the UE, the UE executes measurement according to the parameters configured by the base station and sends a measurement report, and the base station decides when to execute switching and to which target cell. The switching scheme sets a unified threshold for the switching parameters of all user terminals in the same cell, and does not carry out distinguishing control for users in the cell. In practice, the user terminal covered by the same cell is complex, for example, in the case that a certain cell covers both the trunk and the surrounding villages, the terminal moving rapidly in the trunk should be configured to switch as soon as possible, so as to avoid switching after more signal degradation and increase the risk of radio link failure; meanwhile, for access terminals covering the edges of surrounding residential areas, the switching parameters are expected to be more conservative, so that ping-pong switching is avoided frequently, and the two requirements are contradictory. Similar switching scenes are also many, and for schemes using uniform switching parameters, partial user terminals are easy to switch too late or too early in some scenes, so that user perception is influenced and system performance is influenced.

In order to solve the above-mentioned problems, various embodiments of the present application provide a multi-scenario recognition method, which is based on that RSRP values (Reference Signal Receiving Power, reference signal received power) of all cells are received in a user measurement report, and can separate out all terminal clusters covered by the same cell but having different handover requirements in a target area. And then, the characteristics of RSRP in each cluster and between different clusters are statistically analyzed to determine the switching requirement level of the terminal in different clusters under the coverage of a certain cell. Thus, the network device can perform different switching parameter configurations for terminals with different types of characteristics. The method not only can reduce ping-pong switching, avoid switching when the signal is poor, but also can improve user perception through advanced switching or delayed switching when dropped calls are reduced.

The present application is described in detail below in connection with a number of embodiments:

fig. 1 is a schematic flow chart of a switching method provided in an embodiment of the present application, as shown in fig. 1, the method includes the following steps:

step 100, respectively acquiring RSRP measurement data samples of terminals in each cluster;

specifically, in this embodiment, by clustering users in a target area, all users covered by a common cell and having different switching requirements are separated, and then the characteristics of RSRP of terminals in each cluster are statistically analyzed, so as to configure different switching parameter thresholds for terminals in different clusters.

Therefore, when the target terminal newly accesses the network, in order to determine the handover parameter threshold of the target terminal, RSRP measurement data samples of the terminals in each cluster may be first obtained respectively;

specifically, the extraction of RSRP measurement data samples can be performed according to the number proportion of terminals in each cluster; for example, if the ratio of the number of terminals in clusters a, B, C and D is 2:3:4:5, the ratio of RSRP measurement data samples extracted in clusters a, B, C and D may be 2:3:4:5, respectively.

Specifically, the RSRP measurement data samples extracted from each cluster may be kept unchanged for a certain time, and when a new target terminal needs to be switched in an access manner, the subsequent steps may be completed directly based on the existing RSRP measurement data samples, and the RSRP measurement data samples extracted from each cluster may also be updated irregularly.

Specifically, the embodiment does not limit the acquisition mode and the number of RSRP measurement data samples of the terminals in each cluster.

Step 110, determining a target cluster with highest RSRP measurement data similarity with the target terminal based on the RSRP measurement data samples of the terminals in each cluster;

specifically, after acquiring the RSRP measurement data samples of the terminals in each cluster, comparing the RSRP measurement data samples of the target terminals with the RSRP measurement data samples of the terminals in each cluster, and determining the cluster with the highest similarity to the RSRP measurement data of the target terminals as the target cluster, wherein the switching parameter threshold of the corresponding acquired target cluster is used as the switching parameter threshold of the target terminal.

Specifically, when comparing the RSRP measurement data based on the target terminal with the RSRP measurement data samples of the terminals in each cluster, the RSRP measurement data of the target terminal and the RSRP measurement data samples of the terminals in each cluster may be calculated based on the average similarity of the gaussian similarity or the euclidean distance, and the cluster with the highest similarity with the RSRP measurement data of the target terminal is determined as the target cluster.

In this embodiment, a linear or nonlinear similarity calculation method is used, including but not limited to a gaussian similarity method and a euclidean distance method, to calculate the difference between RSRP measurement data of the target terminal and RSRP measurement data samples of the terminals in each cluster, and determine the attribute of the samples, thereby completing personalized switching of the target terminal.

And 120, configuring the switching parameter threshold of the target terminal based on the switching parameter threshold corresponding to the target cluster, and completing the switching of the target terminal.

Specifically, after determining that the cluster with the highest RSRP measurement data similarity with the target terminal is the target cluster, the switching parameter threshold of the target cluster can be correspondingly acquired as the switching parameter threshold of the target terminal, the switching parameter threshold of the target terminal is configured, and the switching work of the target terminal is completed based on the switching parameter threshold of the target terminal.

The embodiment can distinguish multiple scenes to perform flexible personalized switching operation on the access terminal, and improves user perception.

Taking an access terminal as an example, the terminal triggers an event A3 and reports the event in the process of moving from a main cell to an adjacent cell, and the network side completes the switching work according to preset switching parameter thresholds corresponding to each cluster by extracting RSRP characteristic data and comparing with RSRP measurement data samples of the terminals in each cluster. Therefore, ping-pong switching can be reduced, switching when signals are poor is avoided, and user perception can be improved through advanced switching or delayed switching while call drop is reduced.

According to the switching method provided by the embodiment of the application, the user is clustered, the switching parameter thresholds corresponding to different clusters are different, the cluster to which the target terminal belongs is determined, and the switching is performed after the corresponding switching parameter threshold is determined, so that different switching parameter configurations can be performed on the terminal under different scenes, ping-pong switching is reduced, switching when the terminal signal is poor is avoided, switching in advance or delayed switching is performed while call drop is reduced, and system performance is improved.

Optionally, based on any one of the foregoing embodiments, before the acquiring RSRP measurement data samples of the terminals in each cluster, the method further includes:

Specifically, before determining the cluster to which the target terminal belongs, all clusters and the handover parameter threshold corresponding to each cluster need to be determined first.

Therefore, reference Signal Received Power (RSRP) measurement data of terminals to be clustered for covering all cells of the terminals to be clustered can be firstly obtained, and based on the RSRP measurement data of all the terminals to be clustered in a preset area, the terminals to be clustered are clustered, so that users covered by all the same cell and having different switching requirements are separated.

Specifically, in order to obtain the handover parameter threshold corresponding to each cluster, different handover requirements of the terminals in different clusters can be obtained based on the relationship of the change of the terminal receiving level in each cluster along with time and the comparison of the level strength change dynamics between the clusters, and different handover parameter thresholds (such as different delay times and/or different cell offsets) are further set.

Specifically, when clustering all the terminals to be clustered, clustering modes such as DBSCAN clustering and K-means clustering can be adopted, and all the modes capable of implementing clustering on the terminals to be clustered in this embodiment are applicable to this embodiment, which is not limited.

Taking DBSCAN (Density-Based Spatial Clustering of Applications with Noise-based clustering algorithm) clustering as an example, the principle of DBSCAN clustering is a cluster of high-Density population. And if the r adjacent area of the data point corresponding to a certain terminal to be clustered has more than or equal to epsilon data points corresponding to the terminal to be clustered, the data point corresponding to the terminal to be clustered is called as a core object. The parameter r describes the neighborhood distance threshold for a certain data point and epsilon describes the threshold for the number of samples in a neighborhood of distance r for a certain sample. All points within the r-neighborhood of the core object continue looping in this way until there are no data points that satisfy the condition.

Optionally, based on any one of the foregoing embodiments, the determining, based on the received level amount of the terminal in each cluster, a handover parameter threshold corresponding to each cluster specifically includes:

acquiring the receiving level quantity of all terminals in each cluster;

Specifically, after the terminal clusters of various different switching requirements are determined, corresponding data characteristics can also be calculated. Through the comparison of terminal level variation of different clusters, the OMC (Operation and Maintenance Center, operation maintenance center) can flexibly set different switching parameter threshold values for different scenes.

Specifically, in a fixed time range, the change of the level of the terminal in each cluster along with time can be counted, the change amount of the level is in direct proportion to the switching requirement of the terminal in the cluster, so that the switching requirement degree of the access terminal in each cluster can be determined, and the switching parameter threshold can be distributed to different clusters according to the corresponding proportion by comparing the change amount of the level, such as parameters of switching delay time and the like.

For example, the larger the level change amount, the shorter the required hysteresis time, and the faster the switching needs to be completed.

For example, the larger the level change amount, the smaller the cell offset amount, and the faster the handover needs to be completed.

Therefore, the reception level amounts of all the terminals in each cluster can be acquired first; and determining a switching parameter threshold corresponding to each cluster based on the receiving level amounts of all the terminals of each cluster.

It can be understood that when determining the threshold of the handover parameter corresponding to each cluster, the condition that the larger the level change is satisfied, the shorter the required hysteresis time is, and the smaller the cell offset is.

For example, the handover parameter threshold may be set inversely proportional to the amount of reception level of all terminals, i.e., the larger the amount of reception level, the smaller the handover parameter threshold; the handover parameter threshold may be set to be inversely proportional to the received level amounts of all terminals, for example, the ratio of the received level amounts corresponding to the clusters a, B, C and D is 1:2:3:4, and the handover parameter threshold corresponding to the clusters a, B, C and D may be set to be 12:6:4:3.

Optionally, based on any one of the foregoing embodiments, the acquiring reference signal received power RSRP measurement data of a terminal to be clustered for all cells covering the terminal to be clustered includes:

Specifically, due to the complexity of the wireless communication environment, when the terminals to be clustered measure the reference signal received power RSRP of all cells covering the terminals to be clustered, the terminals to be clustered are interfered by different degrees, so that data generated by the same terminals to be clustered at different times at the same place are different. In order to remove redundant information and reduce the interference of complex environments, the collected measurement data of the terminals to be clustered need to be preprocessed. The purpose of preprocessing not only can remove noise points, but also can enable the data of the same type in the measurement data of the terminal to be clustered to be more close after the data preprocessing, and the data of different types are far away from each other.

In this embodiment, the sample data in the high-dimensional space may be mapped to the low-dimensional space through the TSNE algorithm with minimum data structure loss, and meanwhile, the data of the same type in the sample data of the terminal to be clustered is enabled to be closer, and the data of different types are far away from each other.

Specifically, a first matrix formed by reference signal received power RSRP measurement data of a terminal to be clustered on all cells covering the terminal to be clustered, that is, sample data of a high-dimensional space, may be obtained first, and then a second matrix, that is, measurement data of the terminal to be clustered after preprocessing, may be obtained according to a TSNE algorithm.

It can be understood that the elements in the first matrix and the second matrix are both used for describing RSRP measurement data of the terminals to be clustered, the first matrix is directly composed of RSRP measurement data of the terminals to be clustered, and the second matrix is used for describing the preprocessed RSRP measurement data.

In this embodiment, after the measurement data is preprocessed by the TSNE algorithm, the data structure can still be maintained after the measurement data in the high-dimensional space is mapped to the low-dimensional space, and meanwhile, the data of the same type is closer to each other, and the heterogeneous data is further away from each other. Further, clustering analysis is carried out on the data in the low-dimensional space, so that clear multi-scene classification can be given.

Specifically, in this embodiment, when preprocessing measurement data, algorithms or models other than the TSNE algorithm may be used, so long as preprocessing processes or algorithms that can achieve similar dimension compression or redundant information removal are applicable to this embodiment, which is not limited in this embodiment.

Optionally, based on any one of the embodiments above, the obtaining, based on the first matrix, a second matrix according to a TSNE algorithm includes:

initializing sample data of each terminal to be clustered in the second matrix;

Specifically, sample data of each terminal to be clustered in the second matrix may be initialized first;

specifically, the probability distribution p of the first matrix can be obtained by calculation based on the RSRP measurement data of each terminal to be clustered in the first matrix; and based on the probability distribution p, iteratively updating the second matrix according to a TSNE algorithm, and acquiring the second matrix in the last updating process after the process of determining the iterative updating of the second matrix is finished, and performing subsequent clustering processing by taking the second matrix as preprocessed measurement data.

Specifically, it is assumed that the input data of the TSNE algorithm is expressed as x= (x) ₁ ,x ₂ ,x ₃ ,...,x _n ) I.e. a first matrix, wherein the ith measurement data is denoted as x _i ＝(x _i1 ,x _i2 ,...,x _im ). In a high-dimensional space, the data structure is represented using a joint probability distribution of the individual measurement data, as follows:

wherein p is _ij Representing data x in a high-dimensional space _i And x _j The probability of nearest neighbors, sigma, is determined according to the principle of maximum entropy,

representing a given data point x _i Probability distribution of all other data. Sigma centered on each sample point is such that the entropy of the final distribution is small, usually with log (k) as the upper bound, k being the number of neighbor points determined. In TSNE, a binary search is used to iteratively find an optimal σ with confusion as a measure.

Wherein x is ₁ ,x ₂ ,x ₃ ,...,x _n Represents the 1 st terminal, the 2 nd terminal, the 3 rd terminal, …, the n-th terminal; the i-th measurement data represents measurement data of the i-th terminal, and m represents the overlay terminal x _i Number of cells of x _i1 ,x _i2 ,...,x _im Finger terminal x _i RSRP measurement data in 1 st cell, terminal x _i RSRP measurement data in cell 2, …, terminal x _i RSRP measurement data at the mth cell.

The embodiment can divide and discharge the switching requirement degree of the terminals in each cluster by using, but not limited to, a time sequence statistical method, namely, according to the time sequence data combined with the scene recognition result, the switching requirement degree is ordered for the clustered data features, and corresponding switching parameter thresholds are configured, so that the user perception is improved.

Optionally, based on any one of the above embodiments, iteratively updating the second matrix according to a TSNE algorithm based on the probability distribution p includes:

updating the second matrix based on the KL divergence.

Specifically, the data is represented as y= (y) in the low-dimensional space ₁ ,y ₂ ,y ₃ ,...,y _n ) I.e. a second matrix, in which y _i ＝(y _i1 ,y _i2 ,...,y _it ) T is the dimension of the sample i after being processed; y is ₁ ,y ₂ ,y ₃ ,...,y _n Represents the 1 st terminal, the 2 nd terminal, the 3 rd terminal, …, the n-th terminal; y is _i1 ,y _i2 ,...,y _it Finger terminal y _i Sample data in cell 1, terminal y _i Sample data in cell 2, …, terminal y _i Sample data at the t-th cell.

Specifically, in each update process, the probability distribution q of the second matrix may be calculated based on the sample data of each terminal in the second matrix obtained in the previous update process; the low dimensional spatial distribution is represented by a more general T distribution as follows:

wherein q _ij Representing data y in a low dimensional space _i And y _j A neighbor probability between;

specifically, in the current updating process, after determining the probability distribution q of the second matrix, the KL divergence between the probability distribution p and the probability distribution q may be calculated and obtained.

Specifically, in this embodiment, the TSNE algorithm may measure the degree of similarity between the probability distributions p and q using KL divergence, as follows:

wherein Q is _i Representing a given data point y _i Probability distribution with all other data; in this embodiment, in each updating process, the above-mentioned optimization objective loss function needs to be calculated, and it can be understood that the smaller the KL divergence, the closer the data structures of the high-dimensional space and the low-dimensional space are, and the better the data processing effect is.

Specifically, in each updating process, after the KL divergence, which is the similarity between p and q, is obtained by calculating the optimization objective loss function, a gradient descent method may be adopted in the optimization process, where the gradient is represented as follows:

in the low-dimensional space, the data updating mode is as follows:

wherein y is ^t And representing data in a low-dimensional space at the time t, wherein alpha is the learning rate.

In this embodiment, y is updated in each update process, and an updated second matrix is obtained.

Optionally, based on any one of the above embodiments, the determining that the iteration updating the second matrix ends includes:

Specifically, when the iterative update process of the second matrix is determined to be ended, the iterative update process of the second matrix may be determined based on the KL divergence, for example, if the KL divergence is determined to be lower than a preset value, the iterative update process of the second matrix is determined to be ended;

Specifically, when the iterative update process of the second matrix is determined to be finished, the iterative update process may be determined based on a preset update number, for example, the preset update number is 50, and then the update is finished after the second matrix is updated 50 times, and the second matrix in the 50 th update process is based on the preprocessed measurement data for the subsequent clustering process.

Optionally, based on any one of the above embodiments, the method further includes:

Specifically, the clusters may be updated periodically, and correspondingly, the switching parameter threshold may also be updated periodically; when the clusters are not updated, the switching parameter thresholds corresponding to the clusters can be updated irregularly. The goal of flexible switching is met, and accordingly, RSRP measurement data samples extracted from each cluster can also be updated each time.

Fig. 2 is a schematic flow chart of another switching method provided in the embodiment of the present application, as shown in fig. 2, the method includes the following steps:

step 200, data preprocessing;

specifically, after RSRP measurement data of the terminals to be clustered are obtained, the sample data of the high-dimensional space can be mapped to the low-dimensional space through a TSNE algorithm with minimum data structure loss, and meanwhile, the situation that the same type of data in the sample data of the terminals to be clustered is more similar and different types of data are far away from each other is achieved.

Specifically, a first matrix formed by Reference Signal Received Power (RSRP) measurement data of a terminal to be clustered on all cells covering the terminal to be clustered, namely sample data of a high-dimensional space, can be obtained first, and then a second matrix can be obtained according to a TSNE algorithm, so that noise points can be removed, and meanwhile, after data preprocessing, the same type of data in the measurement data of the terminal to be clustered can be more close, and different types of data are far away from each other.

P _i representing a given data point x _i Probability distribution of all other data. Sigma centered on each sample point is such that the entropy of the final distribution is small, usually with log (k) as the upper bound, k being the number of neighbor points determined. In TSNE, a binary search is used to iteratively find an optimal σ with confusion as a measure.

And then, based on the probability distribution p of the first matrix, iteratively updating the second matrix according to a TSNE algorithm, and after the process of determining that the iterative updating of the second matrix is finished, acquiring the second matrix in the last updating process as preprocessed measurement data to carry out subsequent clustering processing.

Specifically, in each update process of y, the probability distribution q of the second matrix may be calculated and obtained based on the sample data of each terminal in the second matrix obtained in the previous update process; the low dimensional spatial distribution is represented by a more general T distribution as follows:

wherein q _ij Representing data y in a low dimensional space _i And y _j And the probability of neighbor between. Specifically, in the current updating process, after determining the probability distribution q of the second matrix, the KL divergence between the probability distribution p and the probability distribution q may be calculated and obtained.

wherein Q is _i Representing a given data point y _i With all other dataIs a probability distribution of (c). In this embodiment, in each updating process, the above-mentioned optimization objective loss function needs to be calculated, and it can be understood that the smaller the KL divergence, the closer the data structures of the high-dimensional space and the low-dimensional space are, and the better the data processing effect is.

in the low-dimensional space, the data updating mode is as follows:

/>

Step 210, sample clustering;

specifically, the preprocessed measurement data samples may be clustered.

Taking DBSCAN clustering as an example, the principle of DBSCAN clustering is cluster-like of a high-density population. And if the r adjacent area of the data point corresponding to a certain terminal to be clustered has more than or equal to epsilon data points corresponding to the terminal to be clustered, the data point corresponding to the terminal to be clustered is called as a core object. The parameter r describes the neighborhood distance threshold for a certain data point and epsilon describes the threshold for the number of samples in a neighborhood of distance r for a certain sample. All points within the r-neighborhood of the core object continue looping in this way until there are no data points that satisfy the condition.

Step 220, multi-scene analysis;

specifically, after obtaining clusters of multiple different scenes, different switching requirements of the terminals in different clusters can be obtained based on the relation of the level change of the terminals in each cluster along with time and through the comparison between the level strength change dynamics between the clusters, and different switching parameter thresholds (such as different delay times and/or different cell offsets) are further set.

Step 230, switching management;

specifically, when a target terminal is newly added, after acquiring the RSRP measurement data sample of the terminal in each cluster, the RSRP measurement data sample of the target terminal can be compared with the RSRP measurement data sample of the terminal in each cluster, and the cluster with the highest similarity with the RSRP measurement data of the target terminal is determined to be the target cluster, and the handover parameter threshold of the corresponding acquired target cluster is used as the handover parameter threshold of the target terminal. The base station completes its switching operation according to the switching parameter threshold.

Step 240, personalizing the handover.

Specifically, in the process that an access terminal moves from a main cell to a neighboring cell, an event A3 is triggered and reported, and the network completes switching work according to switching parameter thresholds corresponding to each cluster by extracting RSRP characteristic data of the access terminal and comparing the RSRP characteristic data with RSRP measurement data samples of the terminals in each cluster.

According to the multi-scene recognition method based on TSNE combined with DBSCAN, which is provided by the embodiment of the application, based on the characteristic that RSRP values (Reference Signal Receiving Power, reference signal received power) of all cells are received in a user measurement report, all terminals covered by the same cell and having different switching requirements in a target area can be separated. And then determining the switching requirement level of different clusters under the coverage of a certain cell by statistically analyzing the variation of RSRP between each cluster and different clusters along with time through time sequence data. Thus, the base station can perform different switching parameter configuration on the terminals with different types of characteristics. The method not only can reduce ping-pong switching, avoid switching when the signal is poor, but also can improve user perception through advanced switching or delayed switching when dropped calls are reduced.

The technical scheme provided by the embodiment of the application can be suitable for various systems, in particular to a 5G system. For example, suitable systems may be global system for mobile communications (global system of mobile communication, GSM), code division multiple access (code division multiple access, CDMA), wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA) universal packet Radio service (general packet Radio service, GPRS), long term evolution (long term evolution, LTE), LTE frequency division duplex (frequency division duplex, FDD), LTE time division duplex (time division duplex, TDD), long term evolution-advanced (long term evolution advanced, LTE-a), universal mobile system (universal mobile telecommunication system, UMTS), worldwide interoperability for microwave access (worldwide interoperability for microwave access, wiMAX), 5G New air interface (New Radio, NR), and the like. Terminal devices and network devices are included in these various systems. Core network parts such as evolved packet system (Evloved Packet System, EPS), 5G system (5 GS) etc. may also be included in the system.

The terminal device according to the embodiments of the present application may be a device that provides voice and/or data connectivity to a user, a handheld device with a wireless connection function, or other processing device connected to a wireless modem, etc. The names of the terminal devices may also be different in different systems, for example in a 5G system, the terminal devices may be referred to as User Equipment (UE). The wireless terminal device may communicate with one or more Core Networks (CNs) via a radio access Network (Radio Access Network, RAN), which may be mobile terminal devices such as mobile phones (or "cellular" phones) and computers with mobile terminal devices, e.g., portable, pocket, hand-held, computer-built-in or vehicle-mounted mobile devices that exchange voice and/or data with the radio access Network. Such as personal communication services (Personal Communication Service, PCS) phones, cordless phones, session initiation protocol (Session Initiated Protocol, SIP) phones, wireless local loop (Wireless Local Loop, WLL) stations, personal digital assistants (Personal Digital Assistant, PDAs), and the like. The wireless terminal device may also be referred to as a system, subscriber unit (subscriber unit), subscriber station (subscriber station), mobile station (mobile), remote station (remote station), access point (access point), remote terminal device (remote terminal), access terminal device (access terminal), user terminal device (user terminal), user agent (user agent), user equipment (user device), and the embodiments of the present application are not limited.

The network device according to the embodiment of the present application may be a base station, where the base station may include a plurality of cells for providing services for a terminal. A base station may also be called an access point or may be a device in an access network that communicates over the air-interface, through one or more sectors, with wireless terminal devices, or other names, depending on the particular application. The network device may be operable to exchange received air frames with internet protocol (Internet Protocol, IP) packets as a router between the wireless terminal device and the rest of the access network, which may include an Internet Protocol (IP) communication network. The network device may also coordinate attribute management for the air interface. For example, the network device according to the embodiments of the present application may be a network device (Base Transceiver Station, BTS) in a global system for mobile communications (Global System for Mobile communications, GSM) or code division multiple access (Code Division Multiple Access, CDMA), a network device (NodeB) in a wideband code division multiple access (Wide-band Code Division Multiple Access, WCDMA), an evolved network device (evolutional Node B, eNB or e-NodeB) in a long term evolution (long term evolution, LTE) system, a 5G base station (gNB) in a 5G network architecture (next generation system), a home evolved base station (Home evolved Node B, heNB), a relay node (relay node), a home base station (femto), a pico base station (pico), and the like. In some network structures, the network device may include a Centralized Unit (CU) node and a Distributed Unit (DU) node, which may also be geographically separated.

Fig. 3 is a schematic structural diagram of a switching device according to an embodiment of the present application, as shown in fig. 3, where the device includes: an acquisition module 310, a determination module 320 and a switching module 330; wherein, the liquid crystal display device comprises a liquid crystal display device,

the acquiring module 310 is configured to acquire RSRP measurement data samples of the terminals in each cluster respectively;

the determining module 320 is configured to determine, based on the RSRP measurement data samples of the terminals in each cluster, a target cluster having the highest RSRP measurement data similarity with the target terminal;

the switching module 330 is configured to configure a switching parameter threshold of the target terminal based on the switching parameter threshold corresponding to the target cluster, so as to complete switching of the target terminal;

Specifically, after the switching device obtains the RSRP measurement data samples of the terminals in each cluster through the obtaining module 310, the determining module 320 determines the target cluster with the highest RSRP measurement data similarity with the target terminal based on the RSRP measurement data samples of the terminals in each cluster; and finally, configuring a switching parameter threshold of the target terminal based on the switching parameter threshold corresponding to the target cluster through a switching module, and completing switching of the target terminal.

It should be noted that, the above device provided in this embodiment of the present application can implement all the method steps implemented in the method embodiment and achieve the same technical effects, and detailed descriptions of the same parts and beneficial effects as those of the method embodiment in this embodiment are omitted.

According to the switching device provided by the embodiment of the application, the user is clustered, the switching parameter thresholds corresponding to different clusters are different, the cluster to which the target terminal belongs is determined, and the switching is performed after the corresponding switching parameter threshold is determined, so that different switching parameter configurations can be performed on the terminals in different scenes, ping-pong switching is reduced, switching when terminal signals are poor is avoided, switching in advance or delay switching is performed while call drop is reduced, and system performance is improved.

Fig. 4 is a schematic structural diagram of a network device according to an embodiment of the present application, and as shown in fig. 4, the network device includes a memory 420, a transceiver 400, and a processor 410:

a memory 420 for storing a computer program; a transceiver 400 for transceiving data under the control of the processor 410; a processor 410 for reading the computer program in the memory and performing the following operations:

A transceiver 400 for receiving and transmitting data under the control of a processor 410.

Wherein in fig. 4, a bus architecture may comprise any number of interconnected buses and bridges, and in particular one or more processors represented by processor 410 and various circuits of memory represented by memory 420, linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., which are well known in the art and, therefore, will not be described further herein. The bus interface provides an interface. Transceiver 400 may be a number of elements, including a transmitter and a receiver, providing a single 410 responsible for managing the bus architecture and general processing for communicating with various other devices over a transmission medium, and memory 420 may store data used by processor 410 in performing operations.

The processor 410 may be a Central Processing Unit (CPU), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a Field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable logic device (Complex Programmable Logic Device, CPLD), or it may employ a multi-core architecture.

It should be noted that, the above device provided in the embodiment of the present invention can implement all the method steps implemented in the method embodiment and achieve the same technical effects, and detailed descriptions of the same parts and beneficial effects as those of the method embodiment in the embodiment are omitted herein.

Optionally, before the acquiring RSRP measurement data samples of the terminals in each cluster respectively, the operations further include:

Specifically, the above device provided in the embodiment of the present application can implement all the method steps implemented in the embodiment of the method, and can achieve the same technical effects, and the parts and beneficial effects that are the same as those of the embodiment of the method in the embodiment are not described in detail herein.

acquiring the receiving level quantity of all terminals in each cluster;

initializing sample data of each terminal to be clustered in the second matrix;

updating the second matrix based on the KL divergence.

Optionally, based on any one of the above embodiments, the operations further include:

It should be noted that, in the embodiment of the present application, the division of the units is schematic, which is merely a logic function division, and other division manners may be implemented in actual practice. In addition, each functional unit in each embodiment 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 integrated units may be implemented in hardware or in software functional units.

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

It should be noted that, the above device provided in the embodiment of the present invention can implement all the method steps implemented in the method embodiment and achieve the same technical effects, and detailed descriptions of the same parts and beneficial effects as those in the method embodiment in this embodiment are omitted.

In another aspect, embodiments of the present application further provide a processor readable storage medium storing a computer program, where the computer program is configured to cause the processor to perform the method provided in the foregoing embodiments, where the method includes:

The computer program stored on the processor readable storage medium according to this embodiment enables the processor to implement all the method steps implemented by the method embodiment and achieve the same technical effects, and details of the same parts and advantages as those of the method embodiment in this embodiment are not described herein.

The processor-readable storage medium may be any available medium or data storage device that can be accessed by a processor, including, but not limited to, magnetic storage (e.g., floppy disks, hard disks, magnetic tape, magneto-optical disks (MOs), etc.), optical storage (e.g., CD, DVD, BD, HVD, etc.), semiconductor storage (e.g., ROM, EPROM, EEPROM, nonvolatile storage (NAND FLASH), solid State Disk (SSD)), and the like.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, magnetic disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These processor-executable instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims and the equivalents thereof, the present application is intended to cover such modifications and variations.

Claims

1. A method of handover, comprising:

wherein, the switching parameter thresholds corresponding to different clusters are different;

and calculating the average similarity between the RSRP measurement data of the target terminal and the RSRP measurement data samples of the terminals in each cluster based on the Gaussian similarity or Euclidean distance, and determining the cluster with the highest similarity with the RSRP measurement data of the target terminal as the target cluster.

2. The handover method according to claim 1, wherein before the acquiring RSRP measurement data samples of the terminals in each cluster, the method further comprises:

3. The handover method according to claim 2, wherein the determining the handover parameter threshold corresponding to each cluster based on the received level amount of the terminal in each cluster specifically includes:

acquiring the receiving level quantity of all terminals in each cluster;

4. The handover method according to claim 2, wherein the obtaining reference signal received power RSRP measurement data of the terminal to be clustered for all cells covering the terminal to be clustered includes:

5. The handover method according to claim 4, wherein the obtaining a second matrix according to a TSNE algorithm based on the first matrix comprises:

initializing sample data of each terminal to be clustered in the second matrix;

6. The handover method according to claim 5, wherein iteratively updating the second matrix according to a TSNE algorithm based on the probability distribution p comprises:

updating the second matrix based on the KL divergence.

7. The method of switching according to claim 5, wherein the determining that the iteration of updating the second matrix ends comprises:

8. The handover method according to claim 2, wherein the method further comprises:

9. A network device comprising a memory, a transceiver, and a processor:

10. The network device of claim 9, wherein prior to the separately obtaining RSRP measurement data samples for the terminals in each cluster, the operations further comprise:

11. The network device of claim 10, wherein the determining the handover parameter threshold corresponding to each cluster based on the received level of the terminal in each cluster specifically comprises:

acquiring the receiving level quantity of all terminals in each cluster;

12. The network device according to claim 10, wherein the obtaining reference signal received power RSRP measurement data of the terminals to be clustered for all cells covering the terminals to be clustered comprises:

13. The network device of claim 12, wherein the obtaining a second matrix from the TSNE algorithm based on the first matrix comprises:

initializing sample data of each terminal to be clustered in the second matrix;

14. The network device of claim 13, wherein iteratively updating the second matrix according to a TSNE algorithm based on the probability distribution p comprises:

updating the second matrix based on the KL divergence.

15. The network device of claim 13, wherein the determining that the iteration of updating the second matrix ends comprises:

16. The network device of claim 10, wherein the operations further comprise:

17. A switching device, comprising:

and the second determining module is used for calculating the average similarity between the RSRP measured data of the target terminal and the RSRP measured data samples of the terminals in each cluster based on the Gaussian similarity or the Euclidean distance, and determining the cluster with the highest similarity with the RSRP measured data of the target terminal as the target cluster.

18. A processor-readable storage medium, characterized in that the processor-readable storage medium stores a computer program for causing the processor to perform the method of any one of claims 1 to 8.