Voice service quality improvement method and device, electronic equipment and storage medium

Abstract

The invention discloses a voice service quality improving method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring communication data corresponding to a voice call of a user terminal, the communication data comprising a coverage quality evaluation parameter, a resource occupation evaluation parameter and a voice quality evaluation parameter; a feature vector value corresponding to the communication data is determined, linear regression processing is carried out according to the feature vector value, a quality index value corresponding to the voice call is obtained, and the quality index value is used for representing the service quality of the voice call; and under the condition that the quality index value is lower than a preset quality threshold value, switching a service node corresponding to the voice call so as to improve the service quality of the voice call. The technical problem of poor voice service experience caused by single consideration factor of a voice escape strategy in related technologies is solved.

Description

Technical Field

[0001] The present application relates to the technical field of mobile wireless communications, and in particular to a method, device, electronic device and storage medium for improving voice service quality. Background Art

[0002] The voice escape strategy in the related technology takes single factors into consideration. For example, the 5G (5th Generation Mobile Communication Technology) voice escape strategy in the related technology adopts a frequency / system switching (or redirection) method based on coverage / voice quality (packet loss). Most of them only consider the current signal strength or voice packet loss rate, and do not comprehensively consider the impact of latency, resources, etc. on voice perception, resulting in problems such as poor voice service experience.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method, device, electronic device and storage medium for improving the quality of voice service, so as to at least solve the technical problem of poor voice service experience caused by the single consideration factor of the voice escape strategy in the related art.

[0005] According to one aspect of an embodiment of the present application, a method for improving the quality of voice service is provided, comprising: obtaining communication data corresponding to a voice call of a user terminal, wherein the communication data comprises: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making a voice call; determining a eigenvector value corresponding to the communication data, and performing linear regression processing based on the eigenvector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; when the quality index value is lower than a preset quality threshold value, switching the service node corresponding to the voice call to improve the service quality of the voice call.

[0006] Optionally, obtaining communication data corresponding to the voice call of the user terminal includes: obtaining coverage quality assessment parameters, wherein the coverage quality assessment parameters include: a measurement report reported by the user terminal, the measurement report including at least one of the following: a reference signal receiving power, a signal interference plus noise ratio; obtaining a resource occupancy assessment parameter, wherein the resource occupancy assessment parameter includes at least one of the following: a cell uplink physical resource block utilization rate, a cell downlink physical resource pool utilization rate; obtaining a voice quality assessment parameter, wherein the voice quality assessment parameter includes at least one of the following: a wireless resource control reconstruction number, a real-time transport protocol packet loss rate, and a jitter delay of a real-time transport protocol packet.

[0007] Optionally, performing linear regression processing based on the eigenvector value to obtain the quality index value corresponding to the voice call includes: performing data cleaning processing on the communication data corresponding to the voice call, wherein the data cleaning processing includes: outlier processing, missing value processing, and outlier detection; constructing a eigenvector value corresponding to the communication data after the data cleaning processing, and normalizing the eigenvector value; using a speech perception prediction model to determine the quality index value corresponding to the voice call based on the eigenvector value after the normalization processing.

[0008] Optionally, the steps of constructing a speech perception prediction model include: obtaining multiple groups of historical communication data corresponding to voice calls, and an average opinion score corresponding to each group of historical communication data, wherein the average opinion score is used to characterize the user's perceived experience satisfaction with the voice call service; constructing an initial linear regression model, wherein the initial linear regression model is used to predict the quality index value corresponding to the voice call based on the feature vector value corresponding to the historical communication data; using a gradient descent algorithm, the initial linear regression model is trained based on the historical communication data and the average opinion score corresponding to the historical communication data to obtain a speech perception prediction model.

[0009] Optionally, training the initial linear regression model includes: determining the mean opinion score corresponding to the historical communication data as the true value of the quality index value corresponding to the voice call, and determining the loss function based on the true value and the predicted value predicted by the initial linear regression model; determining the partial derivative of the loss function with respect to the model vector weight parameter to obtain the gradient equation corresponding to the loss function; using the gradient descent acceleration parameter to solve the gradient equation to update the model vector weight parameter in the initial linear regression model, wherein the gradient descent acceleration parameter is used to accelerate the convergence speed of the gradient descent process and reduce oscillations.

[0010] Optionally, when the quality indicator value is lower than a preset quality threshold value, switching the service node corresponding to the voice call includes: sending a first control instruction to the user terminal, wherein the first control instruction is used to instruct the user terminal to measure the corresponding service cell signal quality and the inter-frequency cell signal quality; obtaining a first measurement report returned by the user terminal in response to the first control instruction, and judging whether the inter-frequency cell meets the preset switching conditions based on the first measurement report; when the inter-frequency cell meets the preset switching conditions, performing inter-frequency switching to switch the voice call of the user terminal from the current corresponding service cell to the inter-frequency cell.

[0011] Optionally, when the heterofrequency cell does not meet the preset switching conditions, the method also includes: sending a second control instruction to the user terminal, wherein the second control instruction is used to instruct the user terminal to measure the signal quality of the service cell and the signal quality of the heterofrequency cell; obtaining a second measurement report returned by the user terminal in response to the second control instruction, and judging whether the heterofrequency cell meets the preset switching conditions based on the second measurement report; when the heterofrequency cell meets the preset switching conditions, performing heterofrequency switching to switch the voice call of the user terminal from the service cell in the current communication network system to the heterofrequency cell in other communication network systems.

[0012] According to another aspect of an embodiment of the present application, a voice service quality improvement device is also provided, including: a data acquisition module, used to obtain communication data corresponding to a voice call of a user terminal, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality of the user terminal's location, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making a voice call; a quality prediction module, used to determine a eigenvector value corresponding to the communication data, and perform linear regression processing based on the eigenvector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; a switching processing module, used to switch the service node corresponding to the voice call when the quality index value is lower than a preset quality threshold value, so as to improve the service quality of the voice call.

[0013] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the method for improving voice service quality is executed when the program is run.

[0014] According to another aspect of the embodiments of the present application, a non-volatile storage medium is provided, the non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for improving voice service quality by running the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, which implements the steps of the method for improving voice service quality when the computer program is executed by a processor.

[0016] In an embodiment of the present application, communication data corresponding to a voice call of a user terminal is obtained, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making a voice call; a feature vector value corresponding to the communication data is determined, and a linear regression process is performed based on the feature vector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; when the quality index value is lower than a preset quality threshold value, the service node corresponding to the voice call is switched to improve the service quality of the voice call, by performing linear regression on multiple factors such as the received signal quality, the service cell resource occupancy, and the voice quality during the 5G user call process, the 5G user voice call perception within the evaluation period, and performing hetero-frequency / hetero-system switching or redirection based on the voice perception, thereby achieving the purpose of improving 5G voice perception, thereby solving the technical problem of poor voice service experience caused by the single consideration of the voice escape strategy in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal (or electronic device) for implementing a method for improving voice service quality provided in an embodiment of the present application;

[0019] Figure 2 It is a schematic diagram of a method flow for improving voice service quality provided according to an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of a method flow for predicting speech perception during a 5G voice call provided in an embodiment of the present application;

[0021] Figure 4 It is a schematic diagram of an escape strategy for performing 5G voice perception improvement according to an embodiment of the present application;

[0022] Figure 5 It is a structural diagram of a voice service quality improvement device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] The 5G voice escape strategy in related technologies adopts an inter-frequency / inter-system switching (or redirection) method based on coverage / voice quality (packet loss). It either only considers the current signal strength RSRP (Reference Signal Received Power) or only considers the current voice packet loss rate. It does not comprehensively consider the impact of latency, resources, etc. on voice perception, and ignores the problem of 5G voice perception.

[0026] In order to solve the above problems, relevant solutions are provided in the embodiments of the present application, which comprehensively consider the impact of signal strength, cell resource occupancy, voice packet loss, and voice delay on voice perception, predict and start 5G voice escape in advance, and improve 5G user voice perception, which is described in detail below.

[0027] According to an embodiment of the present application, a method embodiment for improving the quality of voice service is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or electronic device) for implementing a method for improving voice service quality is shown. Figure 1 As shown, the computer terminal 10 (or electronic device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0029] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or electronic device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0030] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the voice service quality improvement method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned voice service quality improvement method is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or electronic device).

[0033] In the above operating environment, the embodiment of the present application provides a method for improving voice service quality. Figure 2 is a schematic diagram of a method flow for improving voice service quality provided in accordance with an embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0034] Step S202, obtaining communication data corresponding to the voice call of the user terminal, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making a voice call;

[0035] Step S204, determining a eigenvector value corresponding to the communication data, and performing linear regression processing based on the eigenvector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call;

[0036] Step S206: When the quality indicator value is lower than the preset quality threshold value, the service node corresponding to the voice call is switched to improve the service quality of the voice call.

[0037] Through the above steps, linear regression is performed to generate a 5G voice perception prediction model for multiple factors such as the received signal quality, service cell resource occupancy, and voice quality during a 5G user call, and the 5G user voice call perception within the evaluation period is calculated based on the voice perception prediction model (i.e., the quality index value is determined). When the 5G user voice call perception (quality index value) within the evaluation period is lower than the threshold, inter-frequency / inter-system switching or redirection based on voice perception is performed, thereby achieving the purpose of improving 5G voice perception, and further solving the technical problem of poor voice service experience caused by the single consideration of voice escape strategies in related technologies.

[0038] The following is a further introduction to the method for improving voice service quality in steps S202 to S206 of the embodiment of the present application.

[0039] First, the data collection and modeling process in the embodiments of the present application is introduced.

[0040] In this embodiment, the steps for obtaining the communication data reported by the 5G user are as follows.

[0041] In some embodiments of the present application, obtaining communication data corresponding to a voice call of a user terminal includes the following steps: obtaining coverage quality assessment parameters, wherein the coverage quality assessment parameters include: a measurement report reported by the user terminal, the measurement report including at least one of the following: a reference signal received power, a signal interference plus noise ratio; obtaining resource occupancy assessment parameters, wherein the resource occupancy assessment parameters include at least one of the following: a cell uplink physical resource block utilization rate, a cell downlink physical resource pool utilization rate; obtaining voice quality assessment parameters, wherein the voice quality assessment parameters include at least one of the following: a wireless resource control reconstruction number, a real-time transport protocol packet loss rate, and a jitter delay of a real-time transport protocol packet.

[0042] Specifically, the communication data reported by users can be obtained through the 5G professional network management, which mainly includes three aspects of communication data, including: 1) Coverage quality evaluation parameters used to evaluate the coverage quality of the 5G user's location, mainly including: Reference signal received power RSRP and signal interference plus noise ratio (Signal-to-Interference-plus-Noise Ratio, SINR) in the 5G MR (Measurement Report) reported by the user; 2) Resource occupancy evaluation parameters used to evaluate the resource occupancy of the 5G user's service cell, mainly including: uplink and downlink PRB (Physical Resource Block) utilization; 3) Voice quality evaluation parameters used to evaluate the voice quality of 5G users during calls, mainly including: Radio Resource Control (RRC) reconstruction times, Real-time Transport Protocol (RTP) packet loss rate, and RTP packet jitter delay.

[0043] In addition, after obtaining the RSRP, SINR, and communication data such as the cell PRB, RTP packet, and RRC reconstruction of the 5G MR reported by the user, it can also be parsed into the big data lake and numbered, and the communication data can be cleaned and pre-processed.

[0044] The cleaned communication data can then be used as training data to build a speech perception prediction model, which is described in detail below.

[0045] In some embodiments of the present application, the steps of constructing a speech perception prediction model include: obtaining multiple groups of historical communication data corresponding to voice calls, and the average opinion score corresponding to each group of historical communication data, wherein the average opinion score is used to characterize the user's perceived experience satisfaction with the voice call service; constructing an initial linear regression model, wherein the initial linear regression model is used to predict the quality index value corresponding to the voice call based on the feature vector value corresponding to the historical communication data; using a gradient descent algorithm, the initial linear regression model is trained based on the historical communication data and the average opinion score corresponding to the historical communication data to obtain a speech perception prediction model.

[0046] Specifically, in this embodiment, after the predictive linear regression algorithm is used for modeling, the improved gradient descent tuning algorithm is used for iterative tuning, the feature parameters are iteratively learned, and the optimal parameter group with the smallest loss is obtained quickly, so as to obtain the final model of 5G voice perception prediction. For example, the training data can be shown in the following table, with MR (RSRP, SINR), PRB utilization (uplink, downlink), RRC reconstruction times, RTP packet loss rate and jitter delay as linear regression feature values, and 5G voice mean opinion score (Mean Opinion Score, MOS) as the linear regression true value.

[0047]

[0048]

[0049] The historical communication data in the training data is vectorized to obtain the corresponding feature vector value. Then, based on the feature vector value, a correlation model between 5G voice perception and signal quality, network resources, and voice quality is constructed, as shown below.

[0050] First, in order to speed up the convergence of the algorithm, reduce data complexity, avoid analysis errors caused by differences in data value ranges, and improve the accuracy and credibility of data analysis, before iterative learning, the data range can be simplified and the feature vector values ​​of multiple dimensions can be normalized, as shown in the following formula:

[0051]

[0052] Among them, X N is the normalized eigenvector value, X i is the value of the ith original eigenvector, is the mean of the eigenvector values, S i is the mean square error. After normalization, all eigenvector values ​​are in the range of [-1, 1].

[0053] Afterwards, the initial linear regression model is constructed based on the normalized eigenvector value, as shown in the following formula:

[0054] h w (x)=w0+w1 x1+w2 x2+w3 x3+…+w n x n

[0055] Among them, h w (x) is the predicted value of 5G speech perception (i.e., the predicted quality index value), w is the model vector weight parameter, w0 is the deviation, x is the normalized feature vector value, and the feature variable x nThere are n (7 in this embodiment), and the matrix representation of the multivariate 5G speech perception prediction linear regression model (ie, the initial linear regression model) is shown as follows:

[0056]

[0057] Where X = [x0, x1, x2, x 3， …,x n ],W=[w0,w1,w2,w 3， …,w n ],W T Represents the transposed vector, x0 is 1.

[0058] Then, the initial linear regression model can be iteratively tuned using the improved gradient descent tuning algorithm for 5G speech perception. The specific steps are as follows.

[0059] In some embodiments of the present application, training the initial linear regression model includes the following steps: determining the average opinion score corresponding to the historical communication data as the true value of the quality index value corresponding to the voice call, and determining the loss function based on the true value and the predicted value predicted by the initial linear regression model; determining the partial derivative of the loss function with respect to the model vector weight parameter to obtain the gradient equation corresponding to the loss function; using the gradient descent acceleration parameter to solve the gradient equation to update the model vector weight parameter in the initial linear regression model, wherein the gradient descent acceleration parameter is used to speed up the convergence speed of the gradient descent process and reduce oscillations.

[0060] Specifically, we first construct the loss function of multivariate linear regression, as shown in the following formula:

[0061]

[0062] Among them, y i is the true value of the quality index value corresponding to the i-th group of data, that is, the MOS value corresponding to the historical communication data, and m is the number of data groups.

[0063] Taking partial derivatives of all model vector weight parameters w, we get n sets of equations, as shown below:

[0064]

[0065] Among them, (x i ) k is a certain eigenvector value in the i-th row and k-th column.

[0066] Then the gradient descent method is used to find the minimum value of the loss function of multivariate linear regression:

[0067]

[0068] Among them, α is the learning rate.

[0069] In the implementation of this application, a gradient descent acceleration parameter μ can be introduced so that in the process of gradient descent solution, when the direction of the gradient descent vector is consistent, the descent is accelerated by μ times; when the direction of the gradient descent vector is opposite, the descent is decelerated by μ times. It is used to overcome the problem of large swings in machine iterative learning caused by large random changes in signal quality, network resource occupancy, and voice quality, and improve learning efficiency. The specific formula is as follows:

[0070]

[0071] By continuously iterating and learning the feature parameters through the improved gradient descent algorithm, the optimal parameter group with the minimum loss can be obtained, that is, the optimal model vector weight parameters in the initial linear regression model can be determined. For example, the model parameters obtained using the training data in the above table are as follows:

[0072] w=[1.5, 1.2, 1, 0.3, 1, 2, 0.8]

[0073] w0=0.2

[0074] After the speech perception prediction model is trained, the model can be used to predict the quality index value M corresponding to the voice call based on the newly collected communication data. The specific steps are as follows.

[0075] In some embodiments of the present application, linear regression processing is performed based on the feature vector value to obtain the quality index value corresponding to the voice call, which includes the following steps: performing data cleaning processing on the communication data corresponding to the voice call, wherein the data cleaning processing includes: outlier processing, missing value processing, and outlier detection; constructing a feature vector value corresponding to the communication data after the data cleaning processing, and normalizing the feature vector value; using a speech perception prediction model to determine the quality index value corresponding to the voice call based on the feature vector value after the normalization processing.

[0076] Specifically, during a 5G user voice call, the base station can monitor the uplink and downlink PRB utilization rates of the user's service cell, the MR reported by the user, the number of user RRC reconstruction times, the RTP packet loss rate, the RTP jitter delay and other communication data every 2s evaluation period, and then use the voice perception prediction model to calculate the 5G user's voice perception prediction value, that is, the quality indicator value M. For example, the final determined quality indicator value M is shown in the following table.

[0077]

[0078] When the predicted speech perception prediction value (i.e., quality index value) M is lower than the preset quality threshold value, the speech escape measurement and control can be initiated by triggering the A2+A5 and A2+B1 events, such as Figure 3 As shown, the specific steps are as follows.

[0079] In some embodiments of the present application, when the quality indicator value is lower than a preset quality threshold value, switching the service node corresponding to the voice call includes the following steps: sending a first control instruction to the user terminal, wherein the first control instruction is used to instruct the user terminal to measure the corresponding service cell signal quality and the inter-frequency cell signal quality; obtaining a first measurement report returned by the user terminal in response to the first control instruction, and judging whether the inter-frequency cell meets the preset switching conditions based on the first measurement report; when the inter-frequency cell meets the preset switching conditions, performing an inter-frequency switching to switch the voice call of the user terminal from the current corresponding service cell to the inter-frequency cell.

[0080] In some embodiments of the present application, when the heterofrequency cell does not meet the preset switching conditions, the method also includes the following steps: sending a second control instruction to the user terminal, wherein the second control instruction is used to instruct the user terminal to measure the signal quality of the service cell and the signal quality of the heterofrequency cell; obtaining a second measurement report returned by the user terminal in response to the second control instruction, and judging whether the heterofrequency cell meets the preset switching conditions based on the second measurement report; when the heterofrequency cell meets the preset switching conditions, performing heterofrequency switching to switch the voice call of the user terminal from the service cell in the current communication network system to the heterofrequency cell in other communication network systems.

[0081] In this embodiment, the above-mentioned preset quality threshold value is illustrated by taking 3 as an example, as shown in the above table, where the quality index value M=2.27<3 predicted by the user: "189*******3", it is determined that the voice perception quality of the 5G user is poor at the frequency point of the current service cell. In this case, the A2+A5 and A2+B1 events are triggered, and voice escape measurement and control are initiated.

[0082] Specifically, based on 5G voice perception M, the perception-based 5G voice A2+A5 and A2+B1 measurement and control events can be defined separately. Figure 4 As shown, the first control instruction is triggered first, that is, the A2+A5 measurement control is issued. When ② the A2+A5 measurement report meets the preset conditions, the intra-system frequency switching is performed. Otherwise, the second control instruction is triggered, that is, the A2+B1 measurement control is issued. When the A2+B1 measurement report meets the preset conditions, the inter-system switching is performed. After the inter-frequency / inter-system switching is successful, the perception-based voice successfully escapes, thereby improving the 5G voice satisfaction.

[0083] This application constructs a 5G voice perception prediction model based on linear regression and improved gradient descent algorithms, takes the signal quality (RSRP, SINR), service cell resource occupancy and voice quality (RTP packet loss rate and RTP packet jitter) during 5G voice calls as feature values, and the actual voice MOS as the true value, outputs the 5G voice perception prediction model, and based on the 5G voice perception inter-frequency / inter-system switching or redirection, during the evaluation period of the 5G voice call process, when the 5G voice perception prediction value is lower than the threshold, triggers and starts the inter-frequency / inter-system switching or redirection solution based on 5G voice perception, which can be applied to the escape strategy of new mobile network voice calls. The current mobile communication technologies all adopt standard specifications, which are applicable to the mobile networks of all domestic operators, and can provide technical support for the improvement of voice perception of various operators; at the same time, this solution has the ability to predict 5G voice call perception, and can support 5G base station planning and 5G network optimization based on perception improvement. It is applicable to the current 5G base stations of various manufacturers and frequency bands, has good scalability, and can be applied to other multi-frequency networks; it can be fully applied to 5G perception network optimization, and by predicting 5G user voice perception, it can initiate voice escape of voice perception, improve industry customer satisfaction, and enhance market competitiveness.

[0084] According to an embodiment of the present application, an embodiment of a device for improving voice service quality is also provided. Figure 5 Schematic diagram of a device for improving voice service quality provided according to an embodiment of the present application. Figure 5 As shown, the device comprises:

[0085] The data acquisition module 50 is used to acquire the communication data corresponding to the voice call of the user terminal, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter. The coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal. The resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal. The voice quality assessment parameter is used to characterize the voice quality of the user terminal when making a voice call;

[0086] The quality prediction module 52 is used to determine the eigenvector value corresponding to the communication data, and perform linear regression processing based on the eigenvector value to obtain the quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call;

[0087] The switching processing module 54 is used to switch the service node corresponding to the voice call when the quality indicator value is lower than the preset quality threshold value, so as to improve the service quality of the voice call.

[0088] Optionally, obtaining communication data corresponding to the voice call of the user terminal includes: obtaining coverage quality assessment parameters, wherein the coverage quality assessment parameters include: a measurement report reported by the user terminal, the measurement report including at least one of the following: a reference signal receiving power, a signal interference plus noise ratio; obtaining a resource occupancy assessment parameter, wherein the resource occupancy assessment parameter includes at least one of the following: a cell uplink physical resource block utilization rate, a cell downlink physical resource pool utilization rate; obtaining a voice quality assessment parameter, wherein the voice quality assessment parameter includes at least one of the following: a wireless resource control reconstruction number, a real-time transport protocol packet loss rate, and a jitter delay of a real-time transport protocol packet.

[0089] Optionally, performing linear regression processing based on the eigenvector value to obtain the quality index value corresponding to the voice call includes: performing data cleaning processing on the communication data corresponding to the voice call, wherein the data cleaning processing includes: outlier processing, missing value processing, and outlier detection; constructing a eigenvector value corresponding to the communication data after the data cleaning processing, and normalizing the eigenvector value; using a speech perception prediction model to determine the quality index value corresponding to the voice call based on the eigenvector value after the normalization processing.

[0090] Optionally, the steps of constructing a speech perception prediction model include: obtaining multiple groups of historical communication data corresponding to voice calls, and an average opinion score corresponding to each group of historical communication data, wherein the average opinion score is used to characterize the user's perceived experience satisfaction with the voice call service; constructing an initial linear regression model, wherein the initial linear regression model is used to predict the quality index value corresponding to the voice call based on the feature vector value corresponding to the historical communication data; using a gradient descent algorithm, the initial linear regression model is trained based on the historical communication data and the average opinion score corresponding to the historical communication data to obtain a speech perception prediction model.

[0091] Optionally, training the initial linear regression model includes: determining the mean opinion score corresponding to the historical communication data as the true value of the quality index value corresponding to the voice call, and determining the loss function based on the true value and the predicted value predicted by the initial linear regression model; determining the partial derivative of the loss function with respect to the model vector weight parameter to obtain the gradient equation corresponding to the loss function; using the gradient descent acceleration parameter to solve the gradient equation to update the model vector weight parameter in the initial linear regression model, wherein the gradient descent acceleration parameter is used to accelerate the convergence speed of the gradient descent process and reduce oscillations.

[0092] Optionally, when the quality indicator value is lower than a preset quality threshold value, switching the service node corresponding to the voice call includes: sending a first control instruction to the user terminal, wherein the first control instruction is used to instruct the user terminal to measure the corresponding service cell signal quality and the inter-frequency cell signal quality; obtaining a first measurement report returned by the user terminal in response to the first control instruction, and judging whether the inter-frequency cell meets the preset switching conditions based on the first measurement report; when the inter-frequency cell meets the preset switching conditions, performing inter-frequency switching to switch the voice call of the user terminal from the current corresponding service cell to the inter-frequency cell.

[0093] Optionally, when the heterofrequency cell does not meet the preset switching conditions, it also includes: sending a second control instruction to the user terminal, wherein the second control instruction is used to instruct the user terminal to measure the signal quality of the service cell and the signal quality of the heterofrequency cell; obtaining a second measurement report returned by the user terminal in response to the second control instruction, and judging whether the heterofrequency cell meets the preset switching conditions based on the second measurement report; when the heterofrequency cell meets the preset switching conditions, performing heterofrequency switching to switch the voice call of the user terminal from the service cell in the current communication network system to the heterofrequency cell in other communication network systems.

[0094] It should be noted that the various modules in the above-mentioned voice service quality improvement device can be program modules (for example, a set of program instructions that implement a certain specific function) or hardware modules. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0095] It should be noted that the voice service quality improvement device provided in this embodiment can be used to perform Figure 2 The voice service quality improvement method shown, therefore, the relevant explanations and instructions on the above-mentioned voice service quality improvement method are also applicable to the embodiments of the present application and will not be repeated here.

[0096] An embodiment of the present application also provides a non-volatile storage medium, the non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following voice service quality improvement method by running the computer program: obtaining communication data corresponding to the voice call of the user terminal, wherein the communication data includes: coverage quality assessment parameters, resource occupancy assessment parameters, and voice quality assessment parameters, the coverage quality assessment parameters are used to characterize the network signal coverage strength and quality of the user terminal's location, the resource occupancy assessment parameters are used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameters are used to characterize the voice quality of the user terminal when making a voice call; determining a eigenvector value corresponding to the communication data, and performing linear regression processing based on the eigenvector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; when the quality index value is lower than a preset quality threshold value, switching the service node corresponding to the voice call to improve the service quality of the voice call.

[0097] The embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for improving the quality of voice service described in each embodiment of the present application: obtaining communication data corresponding to the voice call of the user terminal, wherein the communication data includes: coverage quality assessment parameters, resource occupancy assessment parameters, and voice quality assessment parameters, the coverage quality assessment parameters are used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameters are used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameters are used to characterize the voice quality of the user terminal when making a voice call; determining a feature vector value corresponding to the communication data, and performing linear regression processing based on the feature vector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; when the quality index value is lower than a preset quality threshold value, switching the service node corresponding to the voice call to improve the service quality of the voice call.

[0098] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0099] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0101] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0102] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0103] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0104] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for improving voice service quality, characterized in that: include: Obtain communication data corresponding to a voice call of a user terminal, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making the voice call; Determine a feature vector value corresponding to the communication data, and perform linear regression processing according to the feature vector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; When the quality indicator value is lower than a preset quality threshold value, the service node corresponding to the voice call is switched to improve the service quality of the voice call.

2. The method for improving voice service quality according to claim 1, characterized in that: Acquiring communication data corresponding to the voice call of the user terminal includes: Acquire the coverage quality assessment parameter, wherein the coverage quality assessment parameter includes: a measurement report reported by the user terminal, the measurement report including at least one of the following: a reference signal received power and a signal interference plus noise ratio; Acquire the resource occupancy evaluation parameter, wherein the resource occupancy evaluation parameter includes at least one of the following: cell uplink physical resource block utilization rate, cell downlink physical resource pool utilization rate; The voice quality evaluation parameter is obtained, wherein the voice quality evaluation parameter includes at least one of the following: the number of radio resource control reestablishment times, the real-time transport protocol packet loss rate, and the jitter delay of the real-time transport protocol packet.

3. The method for improving voice service quality according to claim 1, characterized in that: Performing linear regression processing according to the feature vector value to obtain a quality index value corresponding to the voice call includes: Performing data cleaning processing on the communication data corresponding to the voice call, wherein the data cleaning processing includes: abnormal value processing, missing value processing, and outlier detection; Constructing the feature vector value corresponding to the communication data after the data cleaning process, and normalizing the feature vector value; A speech perception prediction model is used to determine the quality indicator value corresponding to the voice call according to the feature vector value after the normalization process.

4. The method for improving voice service quality according to claim 3, characterized in that: The steps to build a speech perception prediction model include: Acquire multiple groups of historical communication data corresponding to voice calls, and an average opinion score corresponding to each group of the historical communication data, wherein the average opinion score is used to represent the user's perceived experience satisfaction with the voice call service; Constructing an initial linear regression model, wherein the initial linear regression model is used to predict the quality indicator value corresponding to the voice call according to the feature vector value corresponding to the historical communication data; The initial linear regression model is trained using a gradient descent algorithm according to the historical communication data and the mean opinion score corresponding to the historical communication data to obtain the speech perception prediction model.

5. The method for improving voice service quality according to claim 4, characterized in that: Training the initial linear regression model includes: Determine the mean opinion score corresponding to the historical communication data as the true value of the quality indicator value corresponding to the voice call, and determine a loss function based on the true value and the predicted value obtained by the initial linear regression model; Determine the partial derivative of the loss function with respect to the model vector weight parameter to obtain a gradient equation corresponding to the loss function; The gradient equation is solved by using a gradient descent acceleration parameter to update the model vector weight parameter in the initial linear regression model, wherein the gradient descent acceleration parameter is used to speed up the convergence speed of the gradient descent process and reduce oscillation.

6. The method for improving voice service quality according to claim 1, characterized in that: When the quality indicator value is lower than a preset quality threshold value, switching the service node corresponding to the voice call includes: Sending a first control instruction to the user terminal, wherein the first control instruction is used to instruct the user terminal to measure the corresponding serving cell signal quality and inter-frequency cell signal quality; Acquire a first measurement report returned by the user terminal in response to the first control instruction, and determine whether the inter-frequency cell meets a preset switching condition according to the first measurement report; When the inter-frequency cell meets the preset switching condition, inter-frequency switching is performed to switch the voice call of the user terminal from the currently corresponding serving cell to the inter-frequency cell.

7. The method for improving voice service quality according to claim 6, characterized in that: In a case where the inter-frequency cell does not meet the preset switching condition, the method further includes: Sending a second control instruction to the user terminal, wherein the second control instruction is used to instruct the user terminal to measure the signal quality of the serving cell and the signal quality of the alien system cell; Acquire a second measurement report returned by the user terminal in response to the second control instruction, and determine whether the heterogeneous system cell meets a preset switching condition according to the second measurement report; When the heterogeneous cell satisfies the preset switching condition, heterogeneous cell switching is performed to switch the voice call of the user terminal from the serving cell in the current communication network system to the heterogeneous cell in another communication network system.

8. A device for improving voice service quality, characterized in that: include: A data acquisition module, used to acquire communication data corresponding to a voice call of a user terminal, wherein the communication data includes: a coverage quality assessment parameter, a resource occupancy assessment parameter, and a voice quality assessment parameter, wherein the coverage quality assessment parameter is used to characterize the network signal coverage strength and quality at the location of the user terminal, the resource occupancy assessment parameter is used to characterize the network resource usage of the service cell corresponding to the user terminal, and the voice quality assessment parameter is used to characterize the voice quality of the user terminal when making the voice call; A quality prediction module, used to determine a eigenvector value corresponding to the communication data, and perform linear regression processing according to the eigenvector value to obtain a quality index value corresponding to the voice call, wherein the quality index value is used to characterize the service quality of the voice call; The switching processing module is used to switch the service node corresponding to the voice call when the quality indicator value is lower than the preset quality threshold value, so as to improve the service quality of the voice call.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the method for improving the voice service quality described in any one of claims 1 to 7 is executed when the program is run.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for improving voice service quality as described in any one of claims 1 to 7 by running the computer program.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for improving voice service quality described in any one of claims 1 to 7 are implemented.

Images