Signaling data processing method and device and electronic equipment

Abstract

The embodiment of the invention discloses a multi-signaling data processing method and device and electronic equipment. The method comprises the steps that a task processing request for a target task is received; in response to the task processing request, first feature data and second feature data corresponding to the target task in pre-stored feature data are obtained, and the first feature data and the second feature data are obtained by compressing signaling data through different compression dimensions; performing feature fusion processing on the first feature data and the second feature data to determine third feature data; performing decoding processing on the third feature data to obtain target signaling data corresponding to the target task; and processing the target task according to the target signaling data.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a signaling data processing method, device and electronic equipment. Background Art

[0002] With the rapid development of mobile communication technology, mobile operators provide more and more mobile communication services to users, which leads to a surge in the amount of signaling data.

[0003] In order to handle tasks such as business perception analysis and user complaint backtracking, operators need to store historical signaling data and process tasks based on historical signaling data. However, due to the large amount of signaling data, the storage cost of signaling data is high, and during task processing, the efficiency of task processing is low due to the low efficiency of signaling data acquisition. Therefore, a technical solution is needed to improve the efficiency and accuracy of task processing while saving the storage cost of signaling data. Summary of the invention

[0004] The purpose of the embodiments of this specification is to provide a technical solution for improving task processing efficiency and accuracy while saving signaling data storage costs.

[0005] In order to implement the above technical solution, the embodiments of this specification are implemented as follows:

[0006] A signaling data processing method provided in an embodiment of the present specification includes: receiving a task processing request for a target task; in response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data with different compression dimensions; performing feature fusion processing on the first feature data and the second feature data to determine third feature data; decoding the third feature data to obtain target signaling data corresponding to the target task; and processing the target task according to the target signaling data.

[0007] A signaling data processing device provided in an embodiment of the present specification includes: a request receiving module, used to receive a task processing request for a target task; a data acquisition module, used to respond to the task processing request and obtain first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions; a feature fusion module, used to perform feature fusion processing on the first feature data and the second feature data to determine third feature data; a data decoding module, used to decode the third feature data to obtain target signaling data corresponding to the target task; and a task processing module, used to process the target task according to the target signaling data.

[0008] A signaling data processing device provided in an embodiment of the present specification comprises: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to: receive a task processing request for a target task; in response to the task processing request, obtain first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data with different compression dimensions; perform feature fusion processing on the first feature data and the second feature data to determine third feature data; perform decoding processing on the third feature data to obtain target signaling data corresponding to the target task; and process the target task according to the target signaling data.

[0009] An embodiment of the present specification also provides a storage medium, which is used to store computer-executable instructions, and the executable instructions implement the following process when executed by a processor: receiving a task processing request for a target task; in response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data with different compression dimensions; performing feature fusion processing on the first feature data and the second feature data to determine third feature data; decoding the third feature data to obtain target signaling data corresponding to the target task; and processing the target task according to the target signaling data.

[0010] The embodiments of the present specification also provide a computer program product, including a computer program, which implements the following process when executed by a processor: receiving a task processing request for a target task; in response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data with different compression dimensions; performing feature fusion processing on the first feature data and the second feature data to determine third feature data; decoding the third feature data to obtain target signaling data corresponding to the target task; and processing the target task according to the target signaling data.

[0011] A signaling data processing method provided in an embodiment of the present specification includes: receiving signaling data to be stored; performing feature extraction processing on the signaling data according to a first compression dimension, and storing the extracted feature data in a primary database; performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a backup database; wherein the second compression dimension is smaller than the first compression dimension; and upon receiving a task processing request for a target task, processing the target task based on the feature data stored in the primary database and / or the backup database.

[0012] A signaling data processing device provided in an embodiment of the present specification includes: a data acquisition module for receiving signaling data to be stored; a first compression module for performing feature extraction processing on the signaling data according to a first compression dimension, and storing the extracted feature data in a primary database; a second compression module for performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a backup database; wherein the second compression dimension is smaller than the first compression dimension; and a task processing module for processing the target task based on the feature data stored in the primary database and / or the backup database when a task processing request for the target task is received.

[0013] A signaling data processing device is provided in an embodiment of the present specification, and the signaling data processing device includes: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to: receive signaling data to be stored; perform feature extraction processing on the signaling data according to a first compression dimension, and store the extracted feature data in a primary database; perform feature extraction processing on the signaling data according to a second compression dimension, and store the extracted feature data in a backup database; wherein the second compression dimension is smaller than the first compression dimension; and upon receiving a task processing request for a target task, process the target task based on the feature data stored in the primary database and / or the backup database.

[0014] An embodiment of the present specification also provides a storage medium, which is used to store computer-executable instructions, and the executable instructions implement the following process when executed by a processor: receiving signaling data to be stored; performing feature extraction processing on the signaling data according to a first compression dimension, and storing the extracted feature data in a primary database; performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a backup database; wherein the second compression dimension is smaller than the first compression dimension; and upon receiving a task processing request for a target task, processing the target task based on the feature data stored in the primary database and / or the backup database.

[0015] An embodiment of the present specification also provides a computer program product, including a computer program, which implements the following process when executed by a processor: receiving signaling data to be stored; performing feature extraction processing on the signaling data according to a first compression dimension, and storing the extracted feature data in a primary database; performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a backup database; wherein the second compression dimension is smaller than the first compression dimension; and upon receiving a task processing request for a target task, processing the target task based on the feature data stored in the primary database and / or the backup database. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor.

[0017] Figure 1This is an embodiment of a signaling data processing method of this specification;

[0018] Figure 2 This is another embodiment of a signaling data processing method of this specification;

[0019] Figure 3 A schematic diagram of a signaling data processing process in this specification;

[0020] Figure 4 This is another embodiment of a signaling data processing method of this specification;

[0021] Figure 5 This is a schematic diagram of another signaling data processing process of this specification;

[0022] Figure 6 This is an embodiment of a signaling data processing device of this specification;

[0023] Figure 7 This is another embodiment of a signaling data processing device of the present specification;

[0024] Figure 8 This is an embodiment of a signaling data processing device in this specification. DETAILED DESCRIPTION

[0025] The embodiments of this specification provide a signaling data processing method, device and equipment.

[0026] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0027] The embodiments of this specification provide a signaling data processing method, device and equipment. With the rapid development of mobile communication technology, mobile operators provide more and more mobile communication services to users, which leads to a surge in the amount of signaling data. In order to process tasks such as business perception analysis and user complaint backtracking, operators need to store historical signaling data and perform task processing based on historical signaling data. However, due to the large amount of signaling data, the storage cost of signaling data is high, and during task processing, the task processing efficiency is low due to the low efficiency of signaling data acquisition. Therefore, a technical solution is needed to improve the efficiency and accuracy of task processing while saving the cost of signaling data storage. In this solution, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) is obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data of two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task. For specific processing, please refer to the specific content in the following embodiment.

[0028] like Figure 1 As shown, the embodiment of this specification provides a signaling data processing method, the execution subject of the method may be a server, wherein the server may be an independent server, or a server cluster composed of multiple servers, etc. The method may specifically include the following steps:

[0029] In step S102 , a task processing request for a target task is received.

[0030] Among them, the target tasks can be business perception analysis, user complaint backtracking, network status analysis, signaling analysis, network attack protection and other tasks that need to be processed on signaling data.

[0031] During implementation, taking the target task as a network attack protection task as an example, in order to ensure the security of the communication network, the server can trigger a task processing request for the network attack protection task based on a preset detection period (such as ten minutes, half an hour, etc.), that is, in the preset detection period, the server can receive a task processing request for the network attack protection task.

[0032] In step S104, in response to the task processing request, first feature data and second feature data corresponding to the target task are acquired from pre-stored feature data.

[0033] The first characteristic data and the second characteristic data may be obtained by compressing the signaling data in different compression dimensions.

[0034] In implementation, the server may compress the signaling data using encoders of different compression dimensions to obtain feature data of different dimensions, and store the feature data obtained by the compression process.

[0035] For example, the feature data matrix of the signaling data may be 200*1024 (the excess part may be truncated, and the insufficient part may be filled with 0), that is, the length of the signaling data may be 200 words, and the dimension of each word may be 1024. The server may obtain the signaling data 1, and compress the signaling data 1 through the pre-trained encoder 1 and encoder 2, respectively, to obtain the feature data 1 and feature data 2 corresponding to the signaling data 1. Among them, the compression dimension of the encoder 1 may be 200*256, that is, the dimension of the feature data 1 may be 200*256, indicating that the feature data 1 contains 200 words, and the dimension of each word is 256, the compression dimension of the encoder 2 may be 200*128, that is, the dimension of the feature data 2 may be 200*128, indicating that the feature data 2 contains 200 words, and the dimension of each word is 128, the compression ratio of the encoder 2 may be higher than the compression ratio of the encoder 1, so that the feature data compressed by the encoder 2 may occupy less storage space than the feature data compressed by the encoder 1.

[0036] In this way, the server can use two encoders with different compression dimensions to perform dimensionality reduction compression on the signaling data to store the compressed feature data, saving the storage cost of the signaling data and improving the accuracy of signaling data restoration through feature data of two different dimensions.

[0037] Encoder 1 and Encoder 2 may have different structural designs to meet different compression requirements. For example, Encoder 1 may focus on preserving the integrity of signaling data, while Encoder 2 may emphasize compression efficiency. Compared with Encoder 1, Encoder 2 may use a deeper network structure and fewer output layer neurons, so that signaling data can be converted into a smaller potential space dimension, thereby improving the compression ratio.

[0038] The server may obtain the first feature data and the second feature data corresponding to the target task from the pre-stored feature data according to the task identifier of the target task.

[0039] In step S106, feature fusion processing is performed on the first feature data and the second feature data to determine third feature data.

[0040] In implementation, the server may perform feature fusion processing on the first feature data and the second feature data through a preset feature fusion strategy to determine the third feature data.

[0041] Among them, the preset feature fusion strategy may include feature splicing strategy, feature addition strategy, feature multiplication strategy, etc.

[0042] In addition, the server may also perform feature fusion processing on the first feature data and the second feature data through a pre-trained feature fusion model to determine the third feature data. The feature fusion model may be obtained by training a model constructed by a machine learning algorithm based on the historical first feature data and the historical second feature data.

[0043] There are many methods for determining the third feature data through feature fusion processing. Different methods can be selected according to different actual application scenarios. The embodiments of this specification do not make specific limitations on this.

[0044] In addition, the server may also compress the signaling data through multiple different compression dimensions to obtain multiple different feature data, and store the multiple different feature data. For example, the pre-stored feature data may include multiple feature data such as first feature data, second feature data, and seventh feature data obtained by compressing the signaling data through different compression dimensions.

[0045] The server may select feature data corresponding to the task processing accuracy level of the target task from pre-stored feature data according to the task processing accuracy level of the target task. For example, when the task processing accuracy level of the target task is low (such as lower than a preset level threshold), the first feature data and the second feature data corresponding to the target task may be selected, and the third feature data may be determined based on the first feature data and the second feature data.

[0046] When the task processing accuracy level of the target task is high (e.g., higher than a preset level threshold), the first feature data, the second feature data, and the seventh feature data corresponding to the target task may be selected, and the third feature data may be determined based on the first feature data, the second feature data, and the seventh feature data. The data dimension of the seventh feature data may be greater than the data dimension of the first feature data and the data dimension of the second feature data.

[0047] In step S108, the third characteristic data is decoded to obtain target signaling data corresponding to the target task.

[0048] In implementation, the server may decode the third feature data through a pre-trained decoder to obtain target signaling data corresponding to the target task.

[0049] Among them, the above-mentioned encoders (i.e., encoder 1 and encoder 2) and decoders can be composed of deep neural networks such as Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTM), Gated Recurrent Unit (GRU), Transformer, etc., which can be used to capture patterns and dependencies in time series data and can be used for processing signaling data.

[0050] In step S110, the target task is processed according to the target signaling data.

[0051] The embodiment of the present specification provides a signaling data processing method, by receiving a task processing request for a target task, responding to the task processing request, obtaining the first feature data and the second feature data corresponding to the target task in the pre-stored feature data, the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions, performing feature fusion processing on the first feature data and the second feature data, determining the third feature data, decoding the third feature data, obtaining the target signaling data corresponding to the target task, and processing the target task according to the target signaling data. In this way, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) are obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data with two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task.

[0052] In practical applications, the first feature data may be obtained by extracting features from signaling data using a pre-trained first extraction model, and the second feature data may be obtained by extracting features from signaling data using a pre-trained second extraction model. In the above step S108, the third feature data is decoded to obtain target signaling data corresponding to the target task. There are many different processing methods. An optional processing method is provided below, such as Figure 2 , specifically it may include the processing of the following step S1082.

[0053] In step S1082, the third feature data is decoded using a pre-trained second extraction model to obtain target signaling data.

[0054] In practical applications, before the above step S104, the signaling data may be subjected to feature extraction processing, and the extracted feature data may be stored. An optional processing method is provided below, such as Figure 2 As shown, the process may specifically include the following steps S202 to S206.

[0055] In step S202, signaling data to be stored is received.

[0056] In implementation, taking signaling data generated by the core network as an example, the core network can be responsible for the management and control of the communication network to generate core network signaling, and the deep packet inspection (DPI) system can collect, identify and classify the core network traffic to parse the core network signaling into the original signaling document (Extended Data Record, XDR), which can be used as the signaling data to be stored.

[0057] For example, the server can deploy a DPI system in the core network, and the probe devices distributed on the network traffic path can capture the signaling data packets passing through the core network in real time. The captured signaling data packets can be deeply analyzed through DPI technology to convert complex network traffic data into structured XDR. Specifically, through DPI technologies such as protocol identification and content analysis, complex network traffic data can be converted into structured XDR as shown in Table 1 below.

[0058] Table 1

[0059]

[0060]

[0061] In step S204, feature extraction is performed on the signaling data using a pre-trained first extraction model, and the extracted feature data is stored in a primary database.

[0062] In step S206, feature extraction is performed on the signaling data using the pre-trained second extraction model, and the extracted feature data is stored in a standby database.

[0063] The compression dimension of the second extraction model is smaller than the compression dimension of the first extraction model.

[0064] In implementation, the first extraction model and the second extraction model can be constructed based on the same or different deep neural network algorithms, and the compression dimensions of the first extraction model and the second extraction model can be different, that is, the compression dimension of the second extraction model can be smaller than the compression dimension of the first extraction model.

[0065] In this way, the first extraction model can be used to ensure that the characteristic data stored in the main database can retain the complete information of the signaling data, and the second extraction model can be used to improve the compression efficiency of the signaling data and save the storage cost of the standby database.

[0066] In practical applications, in the above step S104, there are various ways to process the first feature data and the second feature data corresponding to the target task in the pre-stored feature data. An optional processing method is provided below, such as Figure 2 , specifically, the processing of the following steps S1042 may include:

[0067] In step S1042, in response to the task processing request, first feature data corresponding to the target task is obtained from the feature data stored in the active database, and second feature data corresponding to the target task is obtained from the feature data stored in the standby database.

[0068] In practical applications, before the above step S304, the second extraction model may be trained. An optional processing method is provided below, such as Figure 2 As shown, the processing may specifically include the following steps S308 to S318.

[0069] In step S208, a pre-trained first extraction model and historical signaling data are obtained.

[0070] The historical signaling data may be any signaling data acquired within a preset storage period (eg, the past half month, the past three months, etc.).

[0071] In implementation, the first extraction model may be obtained by training a model constructed by a deep learning algorithm based on sample signaling data.

[0072] In step S210, feature extraction processing is performed on the historical signaling data using a pre-trained first extraction model to obtain fourth feature data.

[0073] In step S212, feature extraction is performed on the historical signaling data using the second extraction model to obtain fifth feature data.

[0074] In step S214, feature fusion processing is performed on the fourth feature data and the fifth feature data to obtain sixth feature data.

[0075] In step S216, the sixth feature data is decoded using the second extraction model to obtain first signaling data.

[0076] In step S218, it is determined whether the second feature extraction model has converged based on the historical signaling data and the first signaling data. If the second feature extraction model has not converged, the second feature extraction model is continued to be trained based on the historical signaling data and the pre-trained first extraction model until the second feature extraction model converges to obtain the trained second feature extraction model.

[0077] In implementation, Figure 3 As shown, it is taken as an example that the first extraction model includes encoder 1 and decoder 1, and the second extraction model includes encoder 2 and decoder 2.

[0078] Among them, encoder 1 can use a convolutional neural network, which can include a preset number of convolutional layers, pooling layers and fully connected layers. Encoder 1 can be used to extract the core features of signaling data and obtain signaling latent space coding 1, that is, the core features of the signaling data can be compressed based on the preset first compression dimension to ensure high restoration accuracy. The convolution layer used by encoder 1 can use a smaller convolution kernel (such as 3*3) and a smaller step size (such as the step size can be 1). The pooling layer of encoder 1 can use maximum pooling to retain the main feature information of the signaling data, and the fully connected layer can be used to extract high-level abstract features.

[0079] Decoder 1 may also use a convolutional neural network. The structure of decoder 1 is similar to that of encoder 1, but the reverse operation is adopted. Decoder 1 can be used to reversely map the compressed signaling latent space encoding 1 back to the original signaling vector space, reconstruct the original signaling, and obtain signaling data.

[0080] Encoder 2 may use a long short-term memory network, and may include a preset number of LSTM layers and fully connected layers. Encoder 2 may be used to extract the core features of the signaling data, that is, it may perform a higher compression based on the second compression dimension (the second compression dimension may be smaller than the first compression dimension), sacrificing a certain degree of restoration accuracy to achieve a lower storage cost. Among them, multiple LSTM layers in encoder 2 may be superimposed to process sequence data to capture deep-level time series features, and the fully connected layer may be used to extract high-level abstract features.

[0081] Decoder 2 can also use a long short-term memory network, which is similar to the structure of encoder 2, but takes the reverse operation. Decoder 2 can first fuse the signaling latent space code 1 output by encoder 1 and the signaling latent space code 2 output by the encoder, and then reversely map the fused signaling latent space code back to the original signaling vector space to more accurately reconstruct the original signaling.

[0082] By adopting deep learning networks such as CNN and LSTM, the patterns and dependencies in signaling data can be effectively captured, thereby achieving refined redundancy elimination and reducing storage requirements and costs.

[0083] Among them, Figure 3 As shown, the sample signaling data used to train the first extraction model and the historical signaling data used to train the second extraction model can be constructed from historical original signaling documents. The loss functions of the two extraction models can be constructed as mean square error (MSE). The optimization algorithm can use optimization algorithms such as Adam or SGD to update parameters. Before model training, the network parameters in the model can be determined by random initialization.

[0084] First, the first extraction model can be trained. Figure 3 As shown, the acquired sample signaling data can be input into encoder 1, encoder 1 outputs signaling latent space code 1 to decoder 1, and decoder 1 reconstructs signaling latent space code 1 into first signaling data and then outputs it. The difference between the sample signaling data and the first signaling data is calculated as the gradient back propagation to update the network parameters of encoder 1 and decoder 1. Repeat the above forward propagation and back propagation process until the network loss function converges or reaches the preset number of training rounds, and finally obtain the trained encoder 1 and decoder 1, that is, obtain the trained first extraction model.

[0085] Then, the parameters of encoder 1 and decoder 1 can be frozen and the second extraction model can be trained. Figure 3 As shown, the acquired historical signaling data (i.e., the original signaling document XDR) can be input into encoder 1 and encoder 2 respectively, and encoder 1 and encoder 2 respectively output signaling latent space code 1 and signaling latent space code 2 to decoder 2, and decoder 2 performs feature fusion on signaling latent space code 1 and signaling latent space code 2, and then reconstructs the second signaling data (i.e., the reconstructed original signaling document XDR) based on the fused feature data and then outputs it. The difference between the historical signaling data and the second signaling data is calculated as the gradient back propagation to update the network parameters of encoder 2 and decoder 2. Repeat the above forward propagation and back propagation process until the network loss function converges or reaches the preset number of training rounds, and finally obtain the trained encoder 2 and decoder 2, that is, obtain the trained second extraction model.

[0086] The signaling data to be stored can be compressed by encoder 1 of the trained first extraction model and encoder 2 of the trained second extraction model. Both encoders can use deep learning algorithms to extract key features from the signaling data and map them to signaling latent space coding for compression. Encoder 1 and encoder 2 can use deep neural networks with different structural designs to meet different compression requirements. Among them, encoder 1 can focus on preserving the integrity of the signaling data, while encoder 2 can emphasize compression efficiency. Compared with encoder 1, encoder 2 can use a deeper network structure and fewer output layer neurons, so that the signaling data can be converted into a smaller latent space dimension, thereby improving the compression ratio.

[0087] The signaling latent space coding (i.e., feature data) output by encoder 1 can be stored in the primary database, and the signaling latent space coding (i.e., feature data) output by encoder 2 can be stored in the backup database. In this way, the primary database can maintain a high data restoration quality, while the backup database can accept more precision loss while performing disaster recovery backup to achieve lower storage costs. At the same time, the backup database can provide feature enhancement functions in high-quality scenarios to further improve the accuracy of signaling restoration.

[0088] The embodiment of the present specification provides a signaling data processing method, by receiving a task processing request for a target task, responding to the task processing request, obtaining the first feature data and the second feature data corresponding to the target task in the pre-stored feature data, the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions, performing feature fusion processing on the first feature data and the second feature data, determining the third feature data, decoding the third feature data, obtaining the target signaling data corresponding to the target task, and processing the target task according to the target signaling data. In this way, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) are obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data with two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task.

[0089] like Figure 4 As shown, the embodiment of this specification provides a signaling data processing method, the execution subject of the method may be a server, wherein the server may be an independent server, or a server cluster composed of multiple servers, etc. The method may specifically include the following steps:

[0090] In step S402, signaling data to be stored is received.

[0091] In implementation, Figure 5 As shown, the core network can collect traffic, and the DPI system can parse the core network signaling into the original signaling document XDR, and the original signaling document XDR is the signaling data to be stored.

[0092] In step S404, feature extraction processing is performed on the signaling data according to the first compression dimension, and the extracted feature data is stored in the main database.

[0093] In step S406, feature extraction processing is performed on the signaling data according to the second compression dimension, and the extracted feature data is stored in the standby database.

[0094] The second compression dimension may be smaller than the first compression dimension.

[0095] In implementation, Figure 5 As shown, the signaling compression module can be integrated with encoder 1 and encoder 2, the signaling compression module can be deployed at the front end of the signaling storage module, the signaling storage module can be composed of a main database and a backup database, the signaling compression module can be used to compress the signaling data to be stored, and the signaling storage module can be used to store compressed signaling data (i.e., feature data).

[0096] Encoder 1 can compress the original signaling document XDR into signaling latent space code according to the first compression dimension, and store the extracted signaling latent space code (i.e., feature data) in the main database. Encoder 2 can compress the original signaling document XDR into signaling latent space code according to the second compression dimension, and store the extracted signaling latent space code (i.e., feature data) in the standby database.

[0097] Among them, encoder 1 and encoder 2 can be encoder 1 and encoder 2 in the above embodiment. The structure and training process of the encoder can refer to the above related content and will not be repeated here.

[0098] In addition, the signaling compression module may further include multiple encoders with different compression dimensions. Similarly, the signaling storage module may also include multiple databases, that is, in addition to the main database, there may also be multiple backup databases.

[0099] In step S408, when a task processing request for a target task is received, the target task is processed based on the characteristic data stored in the primary database and / or the standby database.

[0100] In implementation, feature data stored in different databases can be extracted according to different target tasks, and the feature data can be restored through the decoder in the signaling restoration module, so as to process the target task through the signaling data obtained through the restoration process. Among them, the signaling restoration module can be integrated with decoder 1 and decoder 2, and deployed at the back end of the signaling storage module. The back end of the signaling restoration module can be deployed with an application system, and the application system can use the reconstructed signaling data to process tasks such as business perception analysis and online log detail query.

[0101] For example, for target tasks in common scenarios (such as business perception analysis, network status analysis, and other tasks), the target tasks can be processed through the feature data stored in the main database, that is, the feature data corresponding to the target task extracted from the main database can be reverse mapped through decoder 1 to obtain signaling data corresponding to the target task, and the target task can be processed based on the signaling data.

[0102] For target tasks in high-demand scenarios (such as Internet log detail query, network attack analysis and other tasks), decoder 2 can simultaneously extract feature data corresponding to the target task from the main database and the backup database, and perform feature fusion processing on the extracted feature data. Then, the fused feature data is reverse mapped to obtain more accurate reconstructed original signaling data, thereby improving the processing accuracy of the target task.

[0103] In this way, through feature fusion, more information can be extracted from the primary database and the backup database to ensure the accuracy of signaling data restoration, and then the encoding information of the two databases is processed by decoder 2 to improve the restoration quality of the signaling.

[0104] Among them, the processing process of the target task in the above-mentioned high-demand scenario, as well as the composition and training process of decoder 1 and decoder 2 can be referred to the aforementioned related content and will not be repeated here.

[0105] In addition, when the signaling compression module includes multiple encoders, the signaling restoration module may also include multiple decoders.

[0106] In this way, by adding signaling compression module and signaling restoration module, the intelligent characteristics of deep learning algorithm can be used to automatically identify and remove redundant information in signaling data, improve the storage efficiency of signaling data and reduce storage cost. At the same time, the role of standby database in data backup and feature enhancement can be fully utilized, which not only ensures the security and availability of signaling data, but also improves the accuracy of signaling restoration. It provides operators with efficient and economical signaling data storage solutions. By storing and analyzing core network signaling data, it can effectively help operators to manage and optimize the network, monitor security and troubleshoot, protect user rights and interests, and promote the stable operation of the network.

[0107] In addition, the dual encoder and dual decoder design can adjust the encoder compression ratio and decoder restoration accuracy according to scene requirements, improving storage efficiency and signaling restoration flexibility. At the same time, in addition to the backup function, the standby database can also be used for feature enhancement. The combination of the standby database and the main database can further improve the accuracy of signaling data restoration.

[0108] With the popularization of 5G technology and the Internet of Things, the amount of network signaling data will increase dramatically, and the demand for efficient storage and fast restoration will become more urgent. Therefore, this technical solution can be applied to operators, cloud service providers and large enterprises.

[0109] The embodiment of the present specification provides a signaling data processing method, which receives signaling data to be stored, performs feature extraction processing on the signaling data according to a first compression dimension, and stores the extracted feature data in a primary database, performs feature extraction processing on the signaling data according to a second compression dimension, and stores the extracted feature data in a standby database, wherein the second compression dimension is smaller than the first compression dimension, and in the case of receiving a task processing request for a target task, processes the target task based on the feature data stored in the primary database and / or the standby database. In this way, since the feature data stored in the primary database and the standby database is the feature data obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. In addition, the restoration accuracy of the signaling data corresponding to the target task can be improved through the feature data stored in the primary database and / or the standby database, so as to accurately process the target task and improve the processing effect of the target task.

[0110] The above is a signaling data processing method provided in the embodiment of this specification. Based on the same idea, the embodiment of this specification also provides a signaling data processing device, such as Figure 6 shown.

[0111] The signaling data processing device includes: a request receiving module 601, a data acquisition module 602, a feature fusion module 603, a data decoding module 604 and a task processing module 605, wherein:

[0112] The request receiving module 601 is used to receive a task processing request for a target task;

[0113] A data acquisition module 602 is used to respond to the task processing request and acquire, from pre-stored feature data, first feature data and second feature data corresponding to the target task, where the first feature data and the second feature data are obtained by compressing the signaling data at different compression dimensions;

[0114] A feature fusion module 603 is used to perform feature fusion processing on the first feature data and the second feature data to determine third feature data;

[0115] A data decoding module 604 is used to decode the third characteristic data to obtain target signaling data corresponding to the target task;

[0116] The task processing module 605 is used to process the target task according to the target signaling data.

[0117] In the embodiment of the present specification, the first feature data is obtained by extracting features from the signaling data using a pre-trained first extraction model, and the second feature data is obtained by extracting features from the signaling data using a pre-trained second extraction model.

[0118] The data decoding module 604 is used to decode the third feature data through the pre-trained second extraction model to obtain the target signaling data.

[0119] In the embodiment of this specification, the device further includes:

[0120] A data receiving module, used for receiving signaling data to be stored;

[0121] A first extraction module, configured to perform feature extraction processing on the signaling data by using the pre-trained first extraction model, and store the extracted feature data in a primary database;

[0122] A second extraction module, used to perform feature extraction processing on the signaling data through the pre-trained second extraction model, and store the extracted feature data in a standby database, wherein the compression dimension of the second extraction model is smaller than the compression dimension of the first extraction model;

[0123] The data acquisition module 602 is used to acquire first characteristic data corresponding to the target task from the characteristic data stored in the primary database, and to acquire second characteristic data corresponding to the target task from the characteristic data stored in the standby database.

[0124] In the embodiment of this specification, the device further includes:

[0125] A first acquisition module, used to acquire the pre-trained first extraction model and historical signaling data;

[0126] The first processing module is used to perform feature extraction processing on the historical signaling data through the pre-trained first extraction model to obtain fourth feature data

[0127] A second processing module, configured to perform feature extraction processing on the historical signaling data by using the second extraction model to obtain fifth feature data;

[0128] a feature determination module, configured to perform feature fusion processing on the fourth feature data and the fifth feature data to obtain sixth feature data;

[0129] A decoding module, used for decoding the sixth feature data through the second extraction model to obtain first signaling data;

[0130] A model training module is used to determine whether the second feature extraction model converges based on the historical signaling data and the first signaling data. If the second feature extraction model does not converge, continue to train the second feature extraction model based on the historical signaling data and the pre-trained first extraction model until the second feature extraction model converges to obtain a trained second feature extraction model.

[0131] The embodiment of the present specification provides a signaling data processing device, which receives a task processing request for a target task, responds to the task processing request, obtains the first feature data and the second feature data corresponding to the target task in the pre-stored feature data, the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions, performs feature fusion processing on the first feature data and the second feature data, determines the third feature data, decodes the third feature data, obtains the target signaling data corresponding to the target task, and processes the target task according to the target signaling data. In this way, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) are obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data with two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task.

[0132] The above is a signaling data processing method provided in the embodiment of this specification. Based on the same idea, the embodiment of this specification also provides a signaling data processing device, such as Figure 7 shown.

[0133] The signaling data processing device includes: a data acquisition module 701, a first compression module 702, a second compression module 703 and a task processing module 704, wherein:

[0134] The data acquisition module 701 is used to receive signaling data to be stored;

[0135] A first compression module 702, configured to perform feature extraction processing on the signaling data according to a first compression dimension, and store the extracted feature data in a primary database;

[0136] A second compression module 703 is used to perform feature extraction processing on the signaling data according to a second compression dimension, and store the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension;

[0137] The task processing module 704 is used to process the target task based on the characteristic data stored in the primary database and / or the standby database when receiving a task processing request for the target task.

[0138] The embodiment of the present specification provides a signaling data processing device, which receives signaling data to be stored, performs feature extraction processing on the signaling data according to a first compression dimension, and stores the extracted feature data in a primary database, performs feature extraction processing on the signaling data according to a second compression dimension, and stores the extracted feature data in a backup database, wherein the second compression dimension is smaller than the first compression dimension, and in the case of receiving a task processing request for a target task, processes the target task based on the feature data stored in the primary database and / or the backup database. In this way, since the feature data stored in the primary database and the backup database is the feature data obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. In addition, the restoration accuracy of the signaling data corresponding to the target task can be improved through the feature data stored in the primary database and / or the backup database, so as to accurately process the target task and improve the processing effect of the target task.

[0139] The above is a signaling data processing device provided in the embodiment of this specification. Based on the same idea, the embodiment of this specification also provides a signaling data processing device, such as Figure 8 shown.

[0140] The signaling data processing device may provide a terminal device or a server, etc. for the above-mentioned embodiment.

[0141] The signaling data processing device may have relatively large differences due to different configurations or performances, and may include one or more processors 801 and memory 802, and the memory 802 may store one or more storage applications or data. Among them, the memory 802 may be a short-term storage or a persistent storage. The application stored in the memory 802 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the signaling data processing device. Furthermore, the processor 801 may be configured to communicate with the memory 802 to execute a series of computer executable instructions in the memory 802 on the signaling data processing device. The signaling data processing device may also include one or more power supplies 803, one or more wired or wireless network interfaces 804, one or more input and output interfaces 805, and one or more keyboards 806.

[0142] Specifically, in this embodiment, the signaling data processing device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions in the signaling data processing device, and the one or more programs are configured to be executed by one or more processors, including computer executable instructions for performing the following:

[0143] receiving a task processing request for a target task;

[0144] In response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data at different compression dimensions;

[0145] Performing feature fusion processing on the first feature data and the second feature data to determine third feature data;

[0146] Decoding the third characteristic data to obtain target signaling data corresponding to the target task;

[0147] Process the target task according to the target signaling data.

[0148] Specifically, in this embodiment, the signaling data processing device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions in the signaling data processing device, and the one or more programs are configured to be executed by one or more processors, including computer executable instructions for performing the following:

[0149] receiving signaling data to be stored;

[0150] Performing feature extraction processing on the signaling data according to the first compression dimension, and storing the extracted feature data in a primary database;

[0151] Performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension;

[0152] When a task processing request for a target task is received, the target task is processed based on the characteristic data stored in the primary database and / or the standby database.

[0153] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the signaling data processing device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0154] The embodiment of the present specification provides a signaling data processing device, which receives a task processing request for a target task, responds to the task processing request, obtains the first feature data and the second feature data corresponding to the target task in the pre-stored feature data, the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions, performs feature fusion processing on the first feature data and the second feature data, determines the third feature data, decodes the third feature data, obtains the target signaling data corresponding to the target task, and processes the target task according to the target signaling data. In this way, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) are obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data with two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task.

[0155] Furthermore, based on the above Figures 1 to 5 In one embodiment, the present specification further provides a storage medium for storing computer executable instruction information. In a specific embodiment, the storage medium may be a USB flash drive, an optical disk, a hard disk, etc. When the computer executable instruction information stored in the storage medium is executed by the processor, the following process can be implemented:

[0156] receiving a task processing request for a target task;

[0157] In response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data at different compression dimensions;

[0158] Performing feature fusion processing on the first feature data and the second feature data to determine third feature data;

[0159] Decoding the third characteristic data to obtain target signaling data corresponding to the target task;

[0160] Process the target task according to the target signaling data.

[0161] Furthermore, based on the above Figures 1 to 5 In one embodiment, the present specification further provides a storage medium for storing computer executable instruction information. In a specific embodiment, the storage medium may be a USB flash drive, an optical disk, a hard disk, etc. When the computer executable instruction information stored in the storage medium is executed by the processor, the following process can be implemented:

[0162] receiving signaling data to be stored;

[0163] Performing feature extraction processing on the signaling data according to the first compression dimension, and storing the extracted feature data in a primary database;

[0164] Performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension;

[0165] When a task processing request for a target task is received, the target task is processed based on the characteristic data stored in the primary database and / or the standby database.

[0166] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the above-mentioned storage medium embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0167] The embodiment of the present specification provides a storage medium, which receives signaling data to be stored, performs feature extraction processing on the signaling data according to a first compression dimension, and stores the extracted feature data in a primary database, performs feature extraction processing on the signaling data according to a second compression dimension, and stores the extracted feature data in a backup database, wherein the second compression dimension is smaller than the first compression dimension, and in the case of receiving a task processing request for a target task, processes the target task based on the feature data stored in the primary database and / or the backup database. In this way, since the feature data stored in the primary database and the backup database is the feature data obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. In addition, the restoration accuracy of the signaling data corresponding to the target task can be improved through the feature data stored in the primary database and / or the backup database, so as to accurately process the target task and improve the processing effect of the target task.

[0168] Furthermore, based on the above Figures 1 to 5 In one or more embodiments of the present specification, a computer program product is provided, including a computer program. When the computer program in the computer program product is executed by a processor, the following process can be implemented:

[0169] receiving a task processing request for a target task;

[0170] In response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data at different compression dimensions;

[0171] Performing feature fusion processing on the first feature data and the second feature data to determine third feature data;

[0172] Decoding the third characteristic data to obtain target signaling data corresponding to the target task;

[0173] Process the target task according to the target signaling data.

[0174] Furthermore, based on the above Figures 1 to 5 In one or more embodiments of the present specification, a computer program product is provided, including a computer program. When the computer program in the computer program product is executed by a processor, the following process can be implemented:

[0175] receiving signaling data to be stored;

[0176] Performing feature extraction processing on the signaling data according to the first compression dimension, and storing the extracted feature data in a primary database;

[0177] Performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension;

[0178] When a task processing request for a target task is received, the target task is processed based on the characteristic data stored in the primary database and / or the standby database.

[0179] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the above-mentioned computer program product embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0180] The embodiment of the present specification provides a computer program product, which receives a task processing request for a target task, responds to the task processing request, obtains the first feature data and the second feature data corresponding to the target task in the pre-stored feature data, the first feature data and the second feature data are obtained by compressing the signaling data with different compression dimensions, performs feature fusion processing on the first feature data and the second feature data, determines the third feature data, decodes the third feature data, obtains the target signaling data corresponding to the target task, and processes the target task according to the target signaling data. In this way, on the one hand, since the pre-stored feature data (i.e., the first feature data and the second feature data) are obtained by compressing the signaling data, when the amount of signaling data to be stored is large, the storage cost of the signaling data can be saved, and the acquisition efficiency of the signaling data corresponding to the target task can be improved. On the other hand, the feature data obtained by feature fusion of feature data with two different compression dimensions can improve the restoration accuracy of the target signaling data corresponding to the target task, so as to accurately process the target task according to the target signaling data and improve the processing effect of the target task.

[0181] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0182] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0183] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0184] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0185] For the convenience of description, the above devices are described in terms of functions and are divided into various units. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0186] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, one or more embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, one or more embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0187] The embodiments of this specification are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable fraud case serial and parallel device to produce a machine, so that the instructions executed by the processor of the computer or other programmable fraud case serial and parallel device generate instructions for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0188] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable fraud case serial and parallel device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0189] These computer program instructions may also be loaded onto a computer or other programmable device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0190] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0191] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0192] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0193] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0194] It should be understood by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, one or more embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, one or more embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0195] One or more embodiments of the present specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0196] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0197] The above description is only an embodiment of this specification and is not intended to limit this document. For those skilled in the art, this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the scope of the claims of this specification.

Claims

1. A signaling data processing method, characterized in that: The method comprises: receiving a task processing request for a target task; In response to the task processing request, obtaining first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing signaling data at different compression dimensions; Performing feature fusion processing on the first feature data and the second feature data to determine third feature data; Decoding the third characteristic data to obtain target signaling data corresponding to the target task; Process the target task according to the target signaling data.

2. The method according to claim 1, characterized in that: The first feature data is obtained by extracting features from signaling data using a pre-trained first extraction model, and the second feature data is obtained by extracting features from signaling data using a pre-trained second extraction model. The decoding process of the third characteristic data to obtain target signaling data corresponding to the target task includes: The third feature data is decoded by the pre-trained second extraction model to obtain the target signaling data.

3. The method according to claim 2, characterized in that In the acquiring of the pre-stored feature data, before the first feature data and the second feature data corresponding to the target task, the method further includes: receiving signaling data to be stored; Performing feature extraction processing on the signaling data by using the pre-trained first extraction model, and storing the extracted feature data in a primary database; Performing feature extraction processing on the signaling data by using the pre-trained second extraction model, and storing the extracted feature data in a standby database, wherein the compression dimension of the second extraction model is smaller than the compression dimension of the first extraction model; The first feature data and the second feature data corresponding to the target task in the pre-stored feature data acquisition include: The first characteristic data corresponding to the target task is acquired from the characteristic data stored in the primary database, and the second characteristic data corresponding to the target task is acquired from the characteristic data stored in the standby database.

4. The method according to claim 3, characterized in that Before performing feature extraction processing on the signaling data by using the pre-trained first extraction model and storing the extracted feature data in the main database, the method further includes: Acquire the pre-trained first extraction model and historical signaling data; Performing feature extraction processing on the historical signaling data by using the pre-trained first extraction model to obtain fourth feature data; Performing feature extraction processing on the historical signaling data by using the second extraction model to obtain fifth feature data; Performing feature fusion processing on the fourth feature data and the fifth feature data to obtain sixth feature data; Decoding the sixth feature data using the second extraction model to obtain first signaling data; Determine whether the second feature extraction model converges based on the historical signaling data and the first signaling data. If the second feature extraction model does not converge, continue to train the second feature extraction model based on the historical signaling data and the pre-trained first extraction model until the second feature extraction model converges to obtain a trained second feature extraction model.

5. A signaling data processing method, characterized in that: The method comprises: receiving signaling data to be stored; Performing feature extraction processing on the signaling data according to the first compression dimension, and storing the extracted feature data in a primary database; Performing feature extraction processing on the signaling data according to a second compression dimension, and storing the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension; When a task processing request for a target task is received, the target task is processed based on the characteristic data stored in the primary database and / or the standby database.

6. A signaling data processing device, characterized in that: The device comprises: A request receiving module, used for receiving a task processing request for a target task; a data acquisition module, configured to, in response to the task processing request, acquire first feature data and second feature data corresponding to the target task from pre-stored feature data, wherein the first feature data and the second feature data are obtained by compressing the signaling data at different compression dimensions; A feature fusion module, used for performing feature fusion processing on the first feature data and the second feature data to determine third feature data; A data decoding module, used for decoding the third characteristic data to obtain target signaling data corresponding to the target task; A task processing module is used to process the target task according to the target signaling data.

7. A signaling data processing device, characterized in that: The device comprises: A data acquisition module, used for receiving signaling data to be stored; A first compression module, configured to perform feature extraction processing on the signaling data according to a first compression dimension, and store the extracted feature data in a primary database; A second compression module, configured to perform feature extraction processing on the signaling data according to a second compression dimension, and store the extracted feature data in a standby database; wherein the second compression dimension is smaller than the first compression dimension; The task processing module is used to process the target task based on the characteristic data stored in the primary database and / or the standby database when receiving a task processing request for the target task.

8. An electronic device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the data processing method according to any one of claims 1 to 5 when executed by the processor.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the data processing method according to any one of claims 1 to 5 are implemented.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the steps of the data processing method according to any one of claims 1 to 5 when executed by a processor.

Images