CN113038501B - Network link optimization method and device and electronic equipment - Google Patents
Network link optimization method and device and electronic equipment Download PDFInfo
- Publication number
- CN113038501B CN113038501B CN201911345708.XA CN201911345708A CN113038501B CN 113038501 B CN113038501 B CN 113038501B CN 201911345708 A CN201911345708 A CN 201911345708A CN 113038501 B CN113038501 B CN 113038501B
- Authority
- CN
- China
- Prior art keywords
- network
- parameter
- parameters
- preset
- correlation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The application discloses a network link optimization method, a network link optimization device and electronic equipment, and relates to the technical field of wireless communication. The method comprises the steps of obtaining target network parameters of a target cell to be optimized, wherein the target network parameters comprise at least one designated parameter; adjusting the parameter value of each designated parameter according to a preset rule until the parameter value meets a first preset requirement; and inputting the specified parameters meeting the first preset requirement and other parameters except the specified parameters in the target network parameters into the parameter configuration model, and calculating to obtain parameter values of the target configuration parameters for optimizing the network link of the target cell, so that the accuracy of the network link optimization parameters is improved.
Description
Technical Field
The present application relates to the field of wireless communication technologies, and in particular, to a method and an apparatus for optimizing a network link, and an electronic device.
Background
In the related art, a wireless network signal is divided into two directions of uplink and downlink according to a propagation direction, for example, a terminal side transmits information to a network side (base station) through an uplink, and the network side transmits information to the terminal side through a downlink. In an ideal communication situation, in any area within the network coverage, the terminal side and the network side may receive the signals of the other party at the same time or cannot receive the signals of the other party at the same time, that is, the uplink and the downlink are balanced. However, in an actual network, the difference between uplink and downlink signals may deviate from a normal allowable range, which causes a serious uplink and downlink imbalance, and further causes a series of network problems such as a network signal being unable to be connected, a call being dropped, and a poor communication quality.
Disclosure of Invention
To solve the above problem, embodiments of the present application provide a method and an apparatus for optimizing a network link, and an electronic device, which are as follows.
In a first aspect, an embodiment of the present application provides a network link optimization method, including:
acquiring target network parameters of a target cell to be optimized, wherein the target network parameters comprise at least one designated parameter, and the designated parameter is a parameter with the correlation degree meeting a preset condition, which is selected from a plurality of network parameters influencing the balance of uplink/downlink links of a network;
adjusting the parameter value of each designated parameter according to a preset rule until a first preset requirement is met;
inputting the specified parameters meeting the first preset requirement and other parameters except the specified parameters in the target network parameters into a parameter configuration model, and calculating to obtain parameter values of the target configuration parameters for performing network link optimization on the target cell;
wherein the parameter configuration model is trained based on network parameters of a plurality of designated cells, and network uplink/downlink balance of the designated cells is obtained.
Further, as a possible implementation method, the step of adjusting the parameter value of each of the designated parameters to meet a first preset requirement according to a preset rule includes:
constructing a linear regression model based on the parameter value of each designated parameter and a plurality of preset balance degrees for representing the link balance, wherein each designated parameter is an independent variable, each balance degree is a model predicted value, and 1> balance degree > 0;
setting the output of the linear regression model as a first preset value based on the constructed linear regression model;
reversely calculating to obtain a target parameter value based on the linear regression model with the output as the first preset value;
and adjusting the parameter value of the designated parameter based on the target parameter value to obtain the designated parameter meeting the first preset requirement.
Further, as a possible implementation method, before the step of obtaining the target network parameter of the target cell to be optimized is performed, the method further includes a parameter determining step for determining the specified parameter, where the parameter determining step includes:
acquiring network parameters of a plurality of current network cells and a network label of each current network cell, wherein the network label is used for representing the network uplink/downlink balance of the current network cell corresponding to the network label;
determining a first correlation matrix based on the network parameters of each existing network cell and the network tags, wherein the first correlation matrix is used for representing the correlation between each network parameter and the network tags;
and selecting the network parameters of which the correlation degree with the balance of uplink/downlink of the network meets a preset condition from the network parameters based on the first correlation matrix as specified parameters.
Further, as a possible implementation method, the network link optimization method further includes a training process of the parameter configuration model, where the training process includes:
determining a second correlation matrix based on the network parameters of the current network cells, wherein the second correlation matrix is used for representing the correlation among the network parameters;
selecting a network parameter meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix, and selecting an existing network cell with balanced network uplink/downlink from a plurality of existing network cells based on the network label as a designated cell;
forming a training data set based on the network parameters meeting a second preset requirement in the designated cell;
and inputting the training data set into a preset machine learning model for feature learning to obtain the parameter configuration model.
Further, as a possible implementation method, the step of selecting a network parameter meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix includes:
determining the network parameters with the correlation smaller than a second preset value in the network parameters based on the first correlation matrix as first parameters;
determining a network parameter pair with correlation larger than a third preset value in the network parameters based on the second correlation matrix;
for each network parameter pair, selecting any network parameter from the network parameter pair as a second parameter;
and removing the first parameter and the second parameter from the network parameters to obtain the network parameters meeting a second preset requirement.
Further, as a possible implementation method, the training process further includes:
acquiring a test data set;
inputting the test data set into the parameter configuration model obtained by training, and calculating to-be-tested parameter values of to-be-tested configuration parameters;
and when detecting that the parameter value to be measured does not meet a third preset requirement, performing parameter tuning processing on the parameter configuration model until the parameter value to be measured calculated based on the parameter configuration model after tuning processing meets the third preset requirement.
Further, as a possible implementation method, the specified parameters at least include a ratio of PUSCH less than minus 120dbm, a co-frequency handover out success rate of voice service in eNodeB, and a co-frequency handover out success rate in eNodeB;
the target configuration parameter at least comprises one of the system threshold-SRVCC, the different system threshold-SRVCC, the reference signal power, the uplink/downlink voice packet loss rate exceeding the voice quality difference threshold and the voice quality packet loss evaluation period.
In a second aspect, an embodiment of the present application further provides a network link optimization apparatus, including
A parameter obtaining module configured to perform obtaining of a target network parameter of a target cell to be optimized, where the target network parameter includes at least one specified parameter, and the specified parameter is a parameter whose correlation degree selected from a plurality of network parameters affecting network uplink/downlink balance meets a preset condition;
the parameter adjusting module is configured to adjust the parameter value of each designated parameter to meet a first preset requirement according to a preset rule;
the parameter calculation module is configured to input the specified parameters meeting the first preset requirements and other parameters except the specified parameters in the target network parameters into a parameter configuration model, and calculate parameter values of the target configuration parameters for network link optimization of the target cell; wherein the parameter configuration model is trained based on network parameters of a plurality of designated cells, and network uplink/downlink balance of the designated cells is obtained.
In a third aspect, an embodiment of the present application further provides an electronic device, including:
at least one processor;
at least one memory coupled to the processor;
wherein the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method as described above.
In a fourth aspect, the present application further provides a computer-readable storage medium, where the storage medium stores computer instructions, and the computer instructions cause the computer to execute the method described above.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects:
and analyzing and calculating the target network parameters of the target cell to be optimized based on a parameter configuration model for learning the network characteristics of the existing network to obtain the parameter values of the target configuration parameters for optimizing the network link of the target cell, so as to improve the network link optimization efficiency and ensure the accuracy of the network link optimization.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flowchart of a network link optimization method according to an embodiment of the present disclosure.
Fig. 2 is another schematic flow chart of a network link optimization method according to an embodiment of the present disclosure.
Fig. 3 is a flowchart illustrating a regression tree analysis process in the GDBT according to an embodiment of the present disclosure.
Fig. 4 is a block diagram of a network link optimization apparatus according to an embodiment of the present disclosure.
Fig. 5 is a block diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
Research shows that, taking an LTE (Long Term Evolution) network as an example, the following two optimization processing methods are mainly adopted for the uplink/downlink imbalance problem related in the LTE network.
(1) Based on historical optimization experience, performance indexes and test data are used as auxiliary data, the performance indexes and the test data are checked one by one according to an optimization flow, and finally an optimization scheme is output. The optimization scheme has high accuracy, but has the defects of high labor cost and low optimization efficiency.
(2) Based on the assistance of an expert system, historical optimization experience and various optimization problem check rules are solidified in the expert system. However, the essence of the expert system is to realize the flow optimization check of each item by combining the performance index and the related statistical data, and the expert system is a problem analysis idea of causal relationship. Although the scheme can improve certain optimization efficiency, the optimization accuracy is low due to the variability of the current network data and the wireless environment.
As can be seen from the optimization schemes given in (1) and (2), the optimization scheme currently involved is mainly based on network KPI (Key Performance indicator) to perform auxiliary analysis and problem positioning, and finally, the optimization scheme is output, which depends on the experience and capability of optimization engineers, and is easy to consider the failure after parameter adjustment, and needs to be adjusted repeatedly, so that the accuracy and efficiency are low.
In view of the problems of the related optimization schemes, the application provides a network link optimization method, a network link optimization device and electronic equipment, so as to improve optimization efficiency, ensure accuracy of an optimization result and reduce optimization cost.
Example one
As shown in fig. 1, a flowchart of a network link optimization method provided in the embodiment of the present application is schematically illustrated, where the network link optimization method may be executed by, but is not limited to, an electronic device, and specifically may be executed by hardware or/and software in the electronic device. Alternatively, the electronic device may be, but is not limited to, a terminal such as a smart phone, a computer, a server, and the like. Referring to fig. 1, the network link optimization method provided by the present application may include the following steps.
S1, obtaining the target network parameter of the target cell to be optimized.
The target cell to be optimized is a cell with a network/downlink imbalance problem, and target network parameters are different according to different network types (such as 3G/4G/5G) of a wireless network arranged in the target cell. Assume that the wireless network deployed in the target cell is LTE.
Optionally, the target network parameter refers to an index affecting uplink/downlink balance of the cell, and various types of collected data, for example: the collected data may include MR (measurement report) total sampling points, below-110 sampling points, below-115 sampling points, integrity, E-RAB (service provided by the intervention layer protocol of the user plane to the non-access layer) establishment success rate _ NEW (%), etc., and OTT data of the cell: the LTE downlink video traffic, the number of LTE micro credit subscribers, the LTE downlink web browsing traffic, the LTE downlink QQ communication traffic, the number of LTE web browsing requests, the LTE uplink QQ communication traffic and the like. OTT refers to the speed at which the service is too dependent on the network.
Further, the target network parameter may include at least one specific parameter, and the specific parameter is a parameter with a correlation satisfying a preset condition, which is selected from a plurality of network parameters affecting uplink/downlink balance of the network. For example, the specified parameters may include at least a ratio of PUSCH (Physical Uplink Shared CHannel) less than minus 120dbm, intra-eNodeB voice traffic co-frequency handover out success rate, intra-eNodeB co-frequency handover out success rate, and the like.
It should be noted that the specified parameter is selected according to historical network data in the current network cell, and the condition that the correlation degree satisfies the predetermined condition may be the parameter with the strongest correlation selected from a plurality of network parameters affecting the uplink/downlink balance of the network. As a possible implementation manner, the determination of the specified parameters may be performed before S1, that is, the network link optimization method provided by the present application may further include a parameter determination step for determining the specified parameters, and the parameter determination step S0 may include the following steps.
And S01, acquiring the network parameters of a plurality of current network cells and the network label of each current network cell.
The network label is used for representing the balance of the network link of the current network cell corresponding to the network label; for example, if the network links of the current network cell are balanced, the network tag may be 0; conversely, the network tag may be 1, etc. When judging whether the network link of the current network cell is balanced, the method can be implemented according to the experience of the staff, and can also be implemented by means of an expert system, and the embodiment does not limit the method.
It should be noted that, in order to ensure the accuracy of the subsequent optimization result, when determining whether the network uplink/downlink of the current network cell is balanced, the network uplink/downlink unbalanced cell caused by the base station hardware fault, feeder installation, mobile phone user behavior, etc. may be eliminated.
S02, a first correlation matrix is determined based on the network parameters and the network labels of each of the present network cells.
The first correlation matrix is used for representing the correlation between each network parameter and the network label. It should be noted that each correlation coefficient in the correlation matrix is a statistical index used to reflect the closeness of the correlation between variables (e.g., between the network parameter and the cell balance, or between the network parameter and the network parameter, etc.), and the calculation formula of the correlation coefficient is as follows.
Wherein Cov (X, Y) is the covariance of X and Y, Var (X) is the variance of X, Var (Y) is the variance of Y,
the mean values of X and Y, respectively.
Further, the correlation matrix is defined as: is provided with (X) 1 ,X 2 ,X 3 ,...,X n ) Is an n-dimensional random vector, arbitrary X i And X j Is denoted as ρ ij (i, j ═ 1, 2.. times, n), then the correlation matrix R for the random vector can be expressed as:
it should be noted that X and Y are a network parameter and a network tag, respectively, when calculating the first correlation matrix, and X and Y are a network parameter, respectively, and the like, when calculating the second correlation matrix described below, which are set specifically as needed.
Taking LTE network as an example, the correlation coefficients of some network parameters and the uplink/downlink imbalance of the network can be shown in table 1. It should be noted that each index shown in table 1 is a network parameter involved in a target cell or an existing cell and a corresponding correlation coefficient, but the number of network parameters involved in an actual cell and the magnitude of the correlation coefficient are not limited to those shown in table 1. QCI (QoS Class Identifier) shown in table 1 is a scale value for measuring the packet forwarding behavior (e.g. packet loss rate, packet Delay budget) of a specific SDF (standard Delay Format), and SDU (service data unit).
S03, selecting a network parameter whose correlation with the balance of uplink/downlink of the network satisfies a predetermined condition from the network parameters based on the first correlation matrix as a designated parameter.
Based on the first correlation matrix between the network parameters and the network tags calculated in the foregoing S01 and S02, an index having a correlation degree greater than a preset value may be selected as a designated parameter, and the network parameters may also be sorted according to the magnitude of the correlation degree, so that one or more network parameters with the correlation degree arranged in front are selected as the designated parameter according to the sorting result, which is not limited in this embodiment. In this embodiment, three parameters, namely, the ratio of PUSCH less than minus 120dbm, the intra-eNodeB voice service co-frequency handover success rate, and the intra-eNodeB co-frequency handover success rate, may be selected as specified parameters, and may be specifically selected according to requirements.
TABLE 1
| Number of successful intra-eNode on-channel handover out executions by re-establishing back-to-source cell | 0.000451637 |
| Service PACP layer downlink discarded service SDU number (number) of cell QCI | 0.000970753 |
| SDU downlink empty port packet loss rate of DRB service PDCP layer with QCI of cell 2 | 0.00149151 |
| DRB service downlink data packet processing total time delay (millisecond-calculating time delay) (millisecond) of cell QCI | 0.00174511 |
| Service PDCP downlink layer downlink discarded service SDU number (packet) with QCI of 2 | 0.00215399 |
| eNodeB triggered service E-RAB abnormal release times with QCI of 2 | 0.00544603 |
| Number of times paging message is received by cell S1 interface | 0.00588669 |
| Switching out the number of times of abnormal release of E-RAB with QCI of 1 | 0.00603604 |
| Cell user plane downlink packet loss rate-HWLTE (%) | 0.00622891 |
| inter-eNodeB inter-pilot frequency handover out success rate | 0.00692379 |
| Number of paging messages of MME to user discarded by cell | 0.00943353 |
| Number of legacy E-RAB with QCI of 2 | 0.0113609 |
| inter-eNodeB same-frequency handover success rate | 0.0151564 |
| The cell switches out the execution success times by the voice service pilot frequency in the eNodeB of the reestablished back source cell | 0.0155749 |
And S2, adjusting the parameter value of each designated parameter to meet the first preset requirement according to a preset rule.
When the parameter value of the designated parameter is adjusted, the parameter value of each designated parameter can be directly adjusted to a preset value to obtain the designated parameter meeting the first preset requirement, for example, the ratio of PUSCH less than minus 120dbm can be directly adjusted to 0, the success rate of co-frequency switching out of voice service in eNodeB can be adjusted to 1, and the like.
In addition, when the parameter value of the designated parameter is adjusted, each designated parameter may be adjusted to meet the first preset requirement according to the preset rule given in S21 to S24 based on the actual situation of the target cell to be optimized, so as to ensure the accuracy of the subsequent calculation result, please refer to fig. 2 as follows.
And S21, constructing a linear regression model based on the parameter values of the designated parameters and a plurality of preset balance degrees for representing the link balance.
Wherein, each designated parameter is an independent variable, each balance degree is a model predicted value, 1>Degree of balance>0. It should be noted that the parameter value of each specific parameter refers to the corresponding actual parameter value of the target cell. For example, the ratio of the PUSCH less than minus 120dbm, the co-frequency handover out success rate of voice service in eNodeB, and the co-frequency handover out success rate in eNodeB may be used as arguments, and then according to the uplink/downlink balance determination standard: assuming that the deviation range of the uplink/downlink coverage rate is + -10 dB, samples of uplink/downlink balance are taken, the balance state (i.e. balance degree) with the deviation absolute value of 0dB is set as 1, the balance state with the deviation absolute value of 10dB is set as 0, and the formula is used
And setting the equilibrium state as a degree value of 0-1, wherein alpha is an absolute value of deviation.
And S22, setting the output of the linear regression model as a first preset value based on the constructed linear regression model.
Alternatively, the linear regression model may be constructed as, but is not limited to:
h θ (x 1 、x 2 、x 3 )＝θ 0 +θ 1 x 1 +θ 2 x 2 +θ 3 x 3
wherein (x) 1 、x 2 、x 3 ) For the three specified parameters mentioned above, (θ) 0 、θ 1 、θ 2 、θ 4 ) Are regression coefficients.
And S23, reversely calculating the target parameter value based on the linear regression model with the output as the first preset value.
And S24, adjusting the parameter value of the designated parameter based on the target parameter value to obtain the designated parameter meeting the first preset requirement.
In S22 to S24, the first preset value may be 1, that is, the equilibrium state is set to 1. And solving the values of the optimal ratio of the uplink and downlink balance data to the PUSCH which is less than minus 120dbm, the success rate of the voice service co-frequency switching-out in the eNodeB and the success rate of the co-frequency switching-out in the eNodeB. According to the solution and analysis, the concrete result is finally obtained as follows: the ratio of PUSCH less than minus 120dbm is set as 0, the success rate of co-frequency switching-out of voice service in eNodeB is set as 1, and the success rate of co-frequency switching-out in eNodeB is set as 1.
That is, when the ratio of the PUSCH smaller than minus 120dbm is set to 0, the intra-eNodeB voice service co-frequency handover success rate is set to 1, and the intra-eNodeB co-frequency handover success rate is set to 1, the parameter configuration model can output a parameter configuration scheme that is most similar to the network requirement of the target cell and has excellent index characteristics, so as to obtain a parameter value of the target configuration parameter of the target cell.
It should be noted that the foregoing S21-S24 are implemented based on linear regression and gradient descent methods to adjust the designated parameters, and the specific adjustment process is not described herein again.
And S3, inputting the specified parameters meeting the first preset requirements and other parameters except the specified parameters in the target network parameters into a parameter configuration model, and calculating to obtain parameter values of the target configuration parameters for optimizing the network link of the target cell.
Wherein the parameter configuration model is trained based on network parameters (historical network data) of a plurality of designated cells, and network uplink/downlink balance of the designated cells is obtained. Optionally, the target configuration parameter at least includes one of a threshold of the system, SRVCC (single Radio Voice Call continuity), a threshold of the different system, SRVCC, a reference signal power, an uplink/downlink Voice packet loss rate exceeding a Voice quality difference threshold, and a Voice quality packet loss evaluation period. It should be noted that the reference signal power is designed according to the specifications of the communication standard. It should be noted that, taking LTE as an example, the system refers to the interoperation between LTE cells, including LTE cells with the same frequency and different frequencies, and the different system refers to the interoperation between LTE and wcdma, gsm, cdma 2000.
Further, before performing the network link optimization method given in the foregoing S1 to S3, the present application may further include a training process of performing model training on the parameter configuration model, where the training process includes S31 to S34.
S31, a second correlation matrix is determined based on the network parameters of each of the present network cells.
Wherein the second correlation matrix is used for characterizing the correlation between the network parameters. For the calculation process of the second correlation matrix, reference may be made to the foregoing equations 1to 3, which are not described herein again.
And S32, selecting network parameters meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix, and selecting an existing network cell with balanced network uplink/downlink from a plurality of existing network cells based on the network label as a designated cell.
Alternatively, as a possible implementation manner, the process of selecting the network parameter meeting the second preset requirement and designating the cell in S32 may be implemented by the following steps.
S321, determining the network parameters with the correlation degree smaller than a second preset value in the network parameters based on the first correlation matrix as first parameters;
s322, determining a network parameter pair with the correlation degree larger than a third preset value in the network parameters based on the second correlation matrix;
s323, for each network parameter pair, selecting any network parameter from the network parameter pair as a second parameter;
s324, the first parameter and the second parameter are removed from the network parameters, and the network parameters meeting the second preset requirement are obtained.
In the foregoing S321 to S321, when selecting the network parameter, the correlation matrix between the network parameter and the network tag, and between the network parameter and the network parameter, is calculated for the current network cell provided with the network tag (balanced or unbalanced). For example, for deleting a first parameter having a small correlation with the network tag (e.g., the absolute value of the correlation is less than 0.001), the correlation between the network parameters is large (e.g., the absolute value of the correlation is greater than 0.8), and in order to reduce redundant information, the analysis-by-synthesis retains one of them, and deletes a second parameter.
S33, forming a training data set based on the network parameters meeting the second predetermined requirement in the designated cell.
And S34, inputting the training data set into a preset machine learning model for feature learning to obtain a parameter configuration model.
The machine learning model may be selected from, but not limited to, CNN (Convolutional Neural Networks), classification Decision Tree, random forest, GBDT (Gradient Boosting Decision Tree), and the like. The machine learning model is used as the GBDT for example, and considering the characteristics of the GBDT including both discrete data and continuous data, the GBDT can be used to flexibly process both discrete and continuous data. Meanwhile, the GBDT has strong robustness on the abnormal values, even if the missing values exist in the original data, the missing values can be subjected to data filling, and a good effect can still be obtained by selecting a proper loss function.
It should be understood that GBDT is to establish a plurality of interrelated decision trees, and the residuals of the previous decision tree are learned through iteration after the previous decision tree is generated, so as to obtain the subsequent decision tree, which can be used for classification and regression. As shown in fig. 3, which is a flow chart of a regression tree analysis process, the calculation of the later trees is performed based on the residual error of the first regression tree, for example, in fig. 3, the probability that the sample belongs to each category is obtained at the final leaf node, where the probability that the sample belongs to category 6 is the largest, so the sample is classified into category six. The residual error of the sample is 1 minus the probability that the sample belongs to the sixth class, and the construction steps of the following tree are similar to those in fig. 3.
As one possible implementation, the process of training the GBDT with the training data set may include:
inputting: training data set T { (x) 1 ,y 1 ),(x 2 ,y 2 ),...,(x N ,y N ) }, iteration times M;
initialization:
n represents the sample size;
and (3) loop iteration: form ═ 1to M do ():
for a multi-classification problem, when constructing a machine learning model, a regression tree, such as F, needs to be created for each class k k (x) K, K is the total number of classes, m for each iteration, the current residual is calculated probabilistically:
according to
The mth regression tree is obtained by learning, and the corresponding leaf node areas are
Compared with the expert system auxiliary optimization scheme in the related technology, the characteristic analysis and classification are carried out on the large sample data of the existing network through the machine learning model to establish the parameter configuration model, the model coefficient of the parameter configuration model is obtained according to the machine learning training, and the accuracy and the optimization efficiency of the network link optimization result can be further improved through the more accurate output parameter optimization scheme.
Further, in order to ensure the performance of the parameter configuration model and improve the reliability of the output result of the model, as a possible implementation manner, the network link optimization method provided by the present application may further include a process of performing model test on the trained model, where the process may include the following steps.
S35, acquiring a test data set;
s36, inputting the test data set into the trained parameter configuration model, and calculating to obtain the to-be-tested parameter value of the to-be-tested configuration parameter;
and S37, when the parameter value to be measured does not meet the third preset requirement, performing parameter tuning processing on the parameter configuration model until the parameter value to be measured calculated based on the parameter configuration model after tuning processing meets the third preset requirement.
In the above-mentioned S35 to S37, the obtaining method of the test data set may refer to the obtaining method of the training data set, which is not described herein again. Alternatively, when the training data set is obtained, after a plurality of network parameters satisfying the second preset requirement are obtained based on S31 and S32, the plurality of network parameters may be divided according to a preset ratio (e.g., 7: 3, etc.), wherein 70% of the training data set is used for model training and 30% of the testing data set is used for model testing.
When the parameter tuning is performed, the parameters of the model can be tuned by using a grid search algorithm and cross validation, and finally, an optimal parameter configuration model is obtained. In addition, model parameter tuning may also be performed by expanding the number of samples or replacing the training data set, which is not limited herein.
It can be understood that, when determining whether the parameter value to be measured meets the third preset requirement, the parameter value to be measured and the preset test value may be compared, and when the parameter value to be measured and the preset test value are consistent, the parameter value to be measured can be considered to meet the third preset requirement, otherwise, the model parameter tuning processing is performed. Wherein the preset test value corresponds to the test data set.
Further, as can be seen from the foregoing network link optimization method, the present application has at least the following technical effects:
and analyzing and calculating the target network parameters of the target cell to be optimized based on the parameter configuration model for learning the network characteristics of the existing network to obtain the parameter values of the target configuration parameters for optimizing the network link of the target cell, thereby improving the network link optimization efficiency and ensuring the accuracy of the network link optimization.
Example two
Fig. 4 is a block diagram illustrating a network link optimization device 100 according to an exemplary embodiment, where the network link optimization device 100 is applicable to a first client. Referring to fig. 4, the network link optimization apparatus 100 includes a parameter acquisition module 110, a parameter adjustment module 120, and a parameter calculation module 130.
A parameter obtaining module 110 configured to perform obtaining a target network parameter of a target cell to be optimized, where the target network parameter includes at least one specified parameter, and the specified parameter is a parameter whose correlation degree selected from a plurality of network parameters affecting network uplink/downlink balance meets a preset condition;
the parameter adjusting module 120 is configured to perform adjustment of parameter values of the designated parameters according to preset rules until the first preset requirements are met;
the parameter calculation module 130 is configured to perform input of the specified parameter meeting the first preset requirement and other parameters except the specified parameter in the target network parameter into the parameter configuration model, and calculate a parameter value of the target configuration parameter for performing network link optimization on the target cell; the parameter configuration model is obtained by training based on network parameters of a plurality of specified cells, and the network uplink/downlink balance of the specified cells is balanced.
The specific manner in which the respective modules perform operations has been described in detail in the embodiment of the method with respect to the apparatus 100 in the present embodiment, and will not be elaborated here. For example, the detailed description about the parameter obtaining module 110 may refer to the description about S1 in the first embodiment, the detailed description about the parameter adjusting module 120 may refer to the description about S2 in the first embodiment, and so on.
EXAMPLE III
Referring to fig. 5, a block diagram of an electronic device 10 according to an exemplary embodiment is provided, where the electronic device 10 may at least include a processor 11 and a memory 12 for storing instructions executable by the processor 11. Wherein the processor 11 is configured to execute the instructions to implement all or part of the steps of the network link optimization method as in the above embodiments.
The processor 11 and the memory 12 are electrically connected directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines.
Wherein the processor 11 is adapted to read/write data or programs stored in the memory and to perform corresponding functions.
The memory 12 is used to store programs or data, such as instructions executable by the processor 110. The Memory 12 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
Further, as a possible implementation, the electronic device 10 may also include power components, multimedia components, audio components, input/output (I/O) interfaces, sensor components, and communication components, among others.
The power supply components provide power to the various components of the electronic device 10. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the electronic device 10.
The multimedia components include a screen that provides an output interface between the electronic device 10 and the user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component includes a front facing camera and/or a rear facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the electronic device 10 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component is configured to output and/or input an audio signal. For example, the audio component may include a Microphone (MIC) configured to receive an external audio signal when the electronic device 10 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in the memory 12 or transmitted via the communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals.
The I/O interface provides an interface between the processing component and a peripheral interface module, which may be a keyboard, click wheel, button, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor assembly includes one or more sensors for providing various aspects of status assessment for the electronic device 10. For example, the sensor assembly may detect an open/closed state of the electronic device 10, the relative positioning of the components, such as a display and keypad of the electronic device 10, the sensor assembly may also detect a change in the position of the electronic device 10 or a component of the electronic device 10, the presence or absence of user contact with the electronic device 10, orientation or acceleration/deceleration of the electronic device 10, and a change in the temperature of the electronic device 10. The sensor assembly may include a proximity sensor configured to detect the presence of a nearby object in the absence of any physical contact. The sensor assembly may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly may further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component is configured to facilitate wired or wireless communication between the electronic device 10 and other devices. The electronic device 10 may access a wireless network based on a communication standard, such as WiFi, a carrier network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the electronic device 10 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.
It should be understood that the configuration shown in fig. 5 is merely a schematic diagram of the configuration of the electronic device 10, and that the electronic device 10 may include more or fewer components than shown in fig. 5, or have a different configuration than shown in fig. 5. The components shown in fig. 5 may be implemented in hardware, software, or a combination thereof.
Example four
In an exemplary embodiment, a non-transitory computer readable storage medium comprising instructions, such as the memory 12 comprising instructions, executable by the processor 11 of the electronic device 10 to perform the network link optimization method described above is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (6)
1. A method for network link optimization, comprising:
acquiring target network parameters of a target cell to be optimized, wherein the target network parameters comprise at least one designated parameter, and the designated parameter is a parameter with the correlation degree meeting a preset condition, which is selected from a plurality of network parameters influencing the balance of uplink/downlink links of a network; the parameters of which the correlation degrees meet the preset condition comprise parameters of which the correlation degrees are larger than a preset value, or one or more parameters of which the correlation degrees are arranged in front under the condition that the plurality of network parameters are sorted according to the magnitude of the correlation degrees;
adjusting the parameter value of each designated parameter according to a preset rule until a first preset requirement is met;
inputting the specified parameters meeting the first preset requirement and other parameters except the specified parameters in the target network parameters into a parameter configuration model, and calculating to obtain parameter values of the target configuration parameters for performing network link optimization on the target cell;
wherein the parameter configuration model is trained based on network parameters of a plurality of designated cells, and network uplink/downlink balance of the designated cells is obtained;
the step of adjusting the parameter value of each designated parameter to meet a first preset requirement according to a preset rule comprises the following steps:
constructing a linear regression model based on the parameter value of each designated parameter and a plurality of preset balance degrees for representing the link balance, wherein each designated parameter is an independent variable, each balance degree is a model predicted value, and 1> balance degree > 0;
setting the output of the linear regression model as a first preset value based on the constructed linear regression model;
reversely calculating to obtain a target parameter value based on the linear regression model with the output as the first preset value;
adjusting the parameter value of the designated parameter based on the target parameter value to obtain the designated parameter meeting a first preset requirement;
before performing the step of obtaining target network parameters of the target cell to be optimized, the method further comprises a parameter determining step for the specified parameter determination, the parameter determining step comprising:
acquiring network parameters of a plurality of current network cells and a network label of each current network cell, wherein the network label is used for representing the network uplink/downlink balance of the current network cell corresponding to the network label;
determining a first correlation matrix based on the network parameters of each existing network cell and the network tags, wherein the first correlation matrix is used for representing the correlation between each network parameter and the network tags;
selecting a network parameter, the correlation degree of which with the balance of uplink/downlink of the network meets a preset condition, from the network parameters based on the first correlation matrix as a designated parameter;
the network link optimization method further comprises a training process of the parameter configuration model, wherein the training process comprises the following steps:
determining a second correlation matrix based on the network parameters of the current network cells, wherein the second correlation matrix is used for representing the correlation among the network parameters;
selecting a network parameter meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix, and selecting an existing network cell with balanced network uplink/downlink from a plurality of existing network cells based on the network label as a designated cell;
forming a training data set based on the network parameters meeting second preset requirements in the designated cell;
inputting the training data set into a preset machine learning model for feature learning to obtain the parameter configuration model;
the step of selecting a network parameter meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix includes:
determining network parameters with the correlation degree smaller than a second preset value in the network parameters based on the first correlation matrix as first parameters;
determining a network parameter pair with the correlation degree larger than a third preset value in the network parameters based on the second correlation matrix;
for each network parameter pair, selecting any network parameter from the network parameter pair as a second parameter;
and removing the first parameter and the second parameter from the network parameters to obtain the network parameters meeting a second preset requirement.
2. The network link optimization method of claim 1, wherein the training process further comprises:
acquiring a test data set;
inputting the test data set into the parameter configuration model obtained by training, and calculating to-be-tested parameter values of to-be-tested configuration parameters;
when the parameter value to be measured does not meet a third preset requirement, performing parameter tuning processing on the parameter configuration model until the parameter value to be measured calculated based on the parameter configuration model after tuning processing meets the third preset requirement;
the parameter value to be tested meeting the third preset requirement comprises that the parameter value to be tested is consistent with a preset test value.
3. The network link optimization method according to any of claims 1-2, wherein the specified parameters at least include a ratio of PUSCH less than minus 120dbm, intra-eNodeB voice traffic co-frequency handover out success rate, intra-eNodeB co-frequency handover out success rate;
the target configuration parameter at least comprises one of the system threshold-SRVCC, the different system threshold-SRVCC, the reference signal power, the uplink/downlink voice packet loss rate exceeding the voice quality difference threshold and the voice quality packet loss evaluation period.
4. A network link optimization device, comprising
A parameter obtaining module configured to perform obtaining of a target network parameter of a target cell to be optimized, where the target network parameter includes at least one specified parameter, and the specified parameter is a parameter whose correlation degree selected from a plurality of network parameters affecting network uplink/downlink balance meets a preset condition; the parameters of which the correlation degrees meet the preset condition comprise parameters of which the correlation degrees are larger than a preset value, or one or more parameters of which the correlation degrees are arranged in front under the condition that the plurality of network parameters are sorted according to the magnitude of the correlation degrees;
the parameter adjusting module is configured to adjust the parameter value of each designated parameter to meet a first preset requirement according to a preset rule;
the parameter calculation module is configured to input the specified parameters meeting the first preset requirements and other parameters except the specified parameters in the target network parameters into a parameter configuration model, and calculate parameter values of the target configuration parameters for network link optimization of the target cell; wherein the parameter configuration model is trained based on network parameters of a plurality of designated cells, and network uplink/downlink balance of the designated cells is obtained;
the parameter adjusting module is specifically configured to:
constructing a linear regression model based on the parameter value of each designated parameter and a plurality of preset balance degrees for representing the link balance, wherein each designated parameter is an independent variable, each balance degree is a model predicted value, and 1> balance degree > 0;
setting the output of the linear regression model as a first preset value based on the constructed linear regression model;
reversely calculating to obtain a target parameter value based on the linear regression model with the output as the first preset value;
adjusting the parameter value of the designated parameter based on the target parameter value to obtain the designated parameter meeting a first preset requirement;
the apparatus further includes a parameter determination module configured to determine the specified parameter, where the parameter determination module is specifically configured to:
before the step of obtaining the target network parameters of the target cell to be optimized is executed, obtaining the network parameters of a plurality of existing network cells and the network label of each existing network cell, wherein the network label is used for representing the network uplink/downlink balance of the existing network cell corresponding to the network label;
determining a first correlation matrix based on the network parameters of each existing network cell and the network tags, wherein the first correlation matrix is used for representing the correlation between each network parameter and the network tags;
selecting a network parameter, the correlation degree of which with the balance of uplink/downlink of the network meets a preset condition, from the network parameters based on the first correlation matrix as a designated parameter;
the device further comprises a parameter configuration model training module for training the parameter configuration model, wherein the parameter configuration model training module is specifically configured to:
determining a second correlation matrix based on the network parameters of the current network cells, wherein the second correlation matrix is used for representing the correlation among the network parameters;
selecting a network parameter meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix, and selecting an existing network cell with balanced network uplink/downlink from a plurality of existing network cells based on the network label as a designated cell;
forming a training data set based on the network parameters meeting a second preset requirement in the designated cell;
inputting the training data set into a preset machine learning model for feature learning to obtain the parameter configuration model;
selecting the network parameters meeting a second preset requirement from the network parameters based on the first correlation matrix and the second correlation matrix specifically comprises:
determining network parameters with the correlation degree smaller than a second preset value in the network parameters based on the first correlation matrix as first parameters;
determining a network parameter pair with the correlation degree larger than a third preset value in the network parameters based on the second correlation matrix;
for each network parameter pair, selecting any network parameter from the network parameter pair as a second parameter;
and removing the first parameter and the second parameter from the network parameters to obtain the network parameters meeting a second preset requirement.
5. An electronic device, comprising:
at least one processor;
at least one memory coupled to the processor;
wherein the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any of claims 1to 3.
6. A computer-readable storage medium, wherein the storage medium stores computer instructions that cause the computer to perform the method of any of claims 1to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911345708.XA CN113038501B (en) | 2019-12-24 | 2019-12-24 | Network link optimization method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911345708.XA CN113038501B (en) | 2019-12-24 | 2019-12-24 | Network link optimization method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113038501A CN113038501A (en) | 2021-06-25 |
| CN113038501B true CN113038501B (en) | 2022-08-05 |
Family
ID=76451589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911345708.XA Active CN113038501B (en) | 2019-12-24 | 2019-12-24 | Network link optimization method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113038501B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114245407A (en) * | 2021-12-30 | 2022-03-25 | 中国联合网络通信集团有限公司 | Network optimization method, device, equipment and storage medium |
| CN115134240B (en) * | 2022-05-12 | 2023-11-14 | 芜湖美的厨卫电器制造有限公司 | Intelligent scene optimization method and device, storage medium and electrical equipment |
| CN115048885B (en) * | 2022-08-12 | 2022-11-15 | 阿里巴巴（中国）有限公司 | Circuit design parameter adjusting method and device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6134444A (en) * | 1998-03-30 | 2000-10-17 | Motorola, Inc. | Method and apparatus for balancing uplink and downlink transmissions in a communication system |
| US6167259A (en) * | 1998-06-19 | 2000-12-26 | Ericsson Inc. | System and method of quantifying the degree of balance on forward link and reverse link channels |
| EP2510450A1 (en) * | 2009-12-09 | 2012-10-17 | Reverb Networks | Self-optimizing network for fixed wireless access |
| CN104684025A (en) * | 2013-12-03 | 2015-06-03 | 富士通株式会社 | Network load balancing method, device and network |
| CN105557023A (en) * | 2013-09-26 | 2016-05-04 | 阿尔卡特朗讯 | Wireless telecommunications network nodes and methods |
| CN106686713A (en) * | 2015-11-10 | 2017-05-17 | 中兴通讯股份有限公司 | Method and device for controlling uplink interference |
| CN107222877A (en) * | 2016-03-21 | 2017-09-29 | 中兴通讯股份有限公司 | CIO methods of adjustment and device |
| CN107484237A (en) * | 2017-09-27 | 2017-12-15 | 四川省大见通信技术有限公司 | A kind of unbalanced Adaptable System of uplink downlink and its method |
| CN109548164A (en) * | 2019-01-11 | 2019-03-29 | 长沙学院 | A kind of adaptive scheduling switching method and system based on loading demand |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6785249B2 (en) * | 1998-10-05 | 2004-08-31 | Qualcomm, Incorporated | Method and apparatus for detecting forward and reverse link imbalance in digital cellular communication systems |
| US8606288B2 (en) * | 2007-06-11 | 2013-12-10 | Telefonaktiebolaget L M Ericsson (Publ) | Radio resource management based on parameters indicating imbalance between uplink and downlink |
| US9179363B2 (en) * | 2011-08-12 | 2015-11-03 | Telefonaktiebolaget Lm Ericsson (Publ) | Systems and methods for determining a coupling characteristic in a radio communications network |
| US9204356B2 (en) * | 2012-09-04 | 2015-12-01 | Apple Inc. | Reducing call drops in uplink power limited scenarios |
| TWI555360B (en) * | 2015-03-27 | 2016-10-21 | In the uplink transmission system to solve the radio frequency is not perfect joint estimation compensation method | |
| US10785101B2 (en) * | 2017-08-23 | 2020-09-22 | Futurewei Technologies, Inc. | Automatically optimize parameters via machine learning |
-
2019
- 2019-12-24 CN CN201911345708.XA patent/CN113038501B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6134444A (en) * | 1998-03-30 | 2000-10-17 | Motorola, Inc. | Method and apparatus for balancing uplink and downlink transmissions in a communication system |
| US6167259A (en) * | 1998-06-19 | 2000-12-26 | Ericsson Inc. | System and method of quantifying the degree of balance on forward link and reverse link channels |
| CN1305671A (en) * | 1998-06-19 | 2001-07-25 | 艾利森公司 | System and method of quantifying degree of balance between forward link and reverse link channel |
| EP2510450A1 (en) * | 2009-12-09 | 2012-10-17 | Reverb Networks | Self-optimizing network for fixed wireless access |
| CN105557023A (en) * | 2013-09-26 | 2016-05-04 | 阿尔卡特朗讯 | Wireless telecommunications network nodes and methods |
| CN104684025A (en) * | 2013-12-03 | 2015-06-03 | 富士通株式会社 | Network load balancing method, device and network |
| CN106686713A (en) * | 2015-11-10 | 2017-05-17 | 中兴通讯股份有限公司 | Method and device for controlling uplink interference |
| CN107222877A (en) * | 2016-03-21 | 2017-09-29 | 中兴通讯股份有限公司 | CIO methods of adjustment and device |
| CN107484237A (en) * | 2017-09-27 | 2017-12-15 | 四川省大见通信技术有限公司 | A kind of unbalanced Adaptable System of uplink downlink and its method |
| CN109548164A (en) * | 2019-01-11 | 2019-03-29 | 长沙学院 | A kind of adaptive scheduling switching method and system based on loading demand |
Non-Patent Citations (4)
| Title |
|---|
| 3G无线网络优化探析;曹昊江;《电子世界》;20151208(第23期);全文 * |
| R1-100044 "A/N transmission in the uplink for carrier aggregation";Ericsson等;《3GPP tsg_ran\WG1_RL1》;20100115;全文 * |
| S4-190628 "Assistance Information Procedures, Types and Parameters";Qualcomm Incorporated;《3GPP tsg_sa\wg4_codec》;20190701;全文 * |
| S4-190741 "Modifications to Use Case Descriptions under Clause 6.5";Qualcomm Incorporated;《3GPP tsg_sa\wg4_codec》;20190701;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113038501A (en) | 2021-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113038501B (en) | Network link optimization method and device and electronic equipment | |
| CN114521012A (en) | Positioning method, positioning device, terminal equipment, base station and position management server | |
| US9271175B2 (en) | Wireless quality collecting device, wireless quality collecting method, and computer-readable recording medium | |
| Aguilar-Garcia et al. | Location-aware self-organizing methods in femtocell networks | |
| WO2023143572A1 (en) | Positioning method based on artificial intelligence (ai) model, and communication device | |
| WO2024140442A1 (en) | Model updating method and apparatus, and device | |
| WO2024169796A1 (en) | Model supervision method and apparatus, and communication device | |
| WO2024008111A1 (en) | Data acquisition method and device | |
| WO2023186090A1 (en) | Verification method, apparatus and device | |
| WO2024067438A1 (en) | Ai model reasoning method, device and readable storage medium | |
| WO2024140445A1 (en) | Model information transmission method and apparatus, and device | |
| WO2023125855A1 (en) | Model updating method and communication device | |
| CN112312411A (en) | Traffic prediction method of VoLTE service and terminal equipment | |
| WO2024152940A1 (en) | Information transmission method and apparatus, and device | |
| WO2024140510A1 (en) | Model training processing method and apparatus, and related device | |
| CN111479280A (en) | Dynamic configuration of test chambers for wireless communication | |
| WO2024083004A1 (en) | Ai model configuration method, terminal, and network side device | |
| WO2023134653A1 (en) | Communication network prediction method, and terminal and network-side device | |
| WO2024208260A1 (en) | Data acquisition method for ai-based csi prediction and apparatus | |
| WO2024093997A1 (en) | Method and apparatus for determining model applicability, and communication device | |
| WO2024208211A1 (en) | Csi prediction performance monitoring method, apparatus, terminal and network side device | |
| CN118176681A (en) | Beam prediction method, device, equipment and storage medium | |
| WO2024140444A1 (en) | Data collection method and apparatus, terminal, and network-side device | |
| US20240323741A1 (en) | Measurement method and apparatus, device, and storage medium | |
| WO2023179617A1 (en) | Locating method and apparatus, terminal and network side device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |