Cross-network communication service control method and system and communication server

Abstract

The invention provides a cross-network communication service control method and system and a communication server, and the method comprises the steps: collecting the network state data of a first network and a second network in a historical time period when the communication between the first network and the second network is established; performing prediction processing on the network state data set by using an intelligent model to obtain network service parameters, the network service parameters including a first network service parameter and a second network service parameter; and adjusting the communication service state of the first network based on the first network service parameter, and adjusting the communication service state of the second network based on the second network service parameter so as to reduce the service difference between the first network and the second network at the current moment. According to the method, a response can be made immediately when the network state changes, and the two related network parameters are dynamically adjusted, so that the two networks with different performances establish stable and smooth communication connection, the conditions of burst flow or network congestion and the like can be dealt with, and the quality and stability of network service are improved.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a cross-network communication service control method, system and communication server. Background Art

[0002] With the popularization of fifth-generation mobile communication (5G) technology, network operators are facing unprecedented challenges, especially in ensuring network quality of service (QoS). 5G networks must not only provide higher data transmission rates than ever before, but also have ultra-low latency characteristics and support the access of large-scale Internet of Things (IoT) devices. However, in the actual deployment process, especially in terms of interoperability between different operators, problems such as inconsistent CSFB (Circuit Switched FallBack) mechanisms and inconsistent billing periods have emerged, which have seriously affected user experience and service quality.

[0003] The CSFB mechanism is a key technology that ensures that 4G / LTE users can smoothly fall back to the 2G / 3G network during voice calls. Its main purpose is to provide a reliable voice service solution when the 4G network does not yet support direct voice calls. However, in 5G networks, although CSFB is no longer needed in theory, in practical applications, CSFB still plays an important role because VoNR (Voice over New Radio) technology is not yet fully mature or popular. The differences in CSFB strategies among different operators, such as choosing different fallback networks (2G GSM or 3G UMTS), lead to inconsistent user experience when switching between different networks.

[0004] Currently, the main way to deal with the above problems is still the traditional method based on manual debugging. This approach usually includes the following steps: first, collect network performance data over a period of time through the network management system; second, analyze the network status based on the collected data and identify the key factors that may affect QoS; then, experienced engineers manually adjust relevant network parameters, such as QCI (Quality of Service Class Identifier) ​​and ARP (Allocation and Retention Priority); finally, continuously monitor the effect of the adjustment and further optimize according to the actual situation.

[0005] Although this approach has proven to be effective in the past, it also has obvious shortcomings. First, the manual debugging process is time-consuming and error-prone. Especially when faced with the huge and complex 5G network architecture, it is difficult to ensure the accuracy and timeliness of parameter adjustment by manpower alone. Due to the lack of real-time and automation capabilities, when the network encounters bursts of traffic or unexpected congestion, traditional manual debugging methods often cannot respond quickly, resulting in a decline in service quality. In addition, this optimization strategy based on post-analysis is difficult to form a complete and sustainable network optimization mechanism, making each problem solving like "firefighting" rather than solving the problem fundamentally. Summary of the invention

[0006] The object of the present invention is to provide a cross-network communication service control method, system and communication server which can automatically and dynamically adjust the parameters of two networks in real time during cross-network communication.

[0007] In order to achieve the above object, the present invention provides a cross-network communication service control method, which includes: When a first network establishes communication with a second network, collecting network status data of the first network and the second network in a historical time period, the network status data including network traffic characteristic data and network service quality indicator data, the first network and the second network being two different types of communication networks; Using an intelligent model to perform prediction processing on the network status data set to obtain network service parameters, the network service parameters including a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, the intelligent model being constructed based on a reinforcement learning model; The communication service state of the first network is adjusted based on the first network service parameter, and the communication service state of the second network is adjusted based on the second network service parameter to reduce the service difference between the first network and the second network at the current moment.

[0008] Preferably, the network service parameters output by the intelligent model and the network status data collected in the next cycle are constructed into a data pair, and the data pair is stored; when the intelligent model is in an idle state, the intelligent model is trained based on the data pair.

[0009] Preferably, after acquiring the network status data, the network status data is cleaned, feature extracted and standardized so as to produce the network status data into a tensor format compatible with the intelligent model.

[0010] Preferably, the traffic characteristic data includes any one or more of transmission delay, packet loss rate, and bandwidth utilization; the network service quality indicator data includes any one or more of service quality category identifier, allocation and retention priority, circuit switching fallback mechanism timer, circuit switching fallback mechanism success rate, call detail record generation time, and user priority setting.

[0011] The present invention also provides a cross-network communication service control system, which includes: a data collection module, which is used to collect network status data of the first network and the second network in a historical time period when the first network establishes communication with the second network, wherein the network status data includes network traffic characteristic data and network service quality indicator data, and the first network and the second network are two different types of communication networks; an intelligent model, which is used to perform prediction processing on the network status data set to obtain network service parameters, wherein the network service parameters include a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, and the intelligent model is constructed based on a reinforcement learning model; An adjustment module is used to adjust the communication service state of the first network based on the first network service parameter, and adjust the communication service state of the second network based on the second network service parameter, so as to reduce the service difference between the first network and the second network at the current moment.

[0012] Preferably, it also includes a database, which is used to construct data pairs from the network service parameters output by the intelligent model and the network status data collected in the next cycle, and store the data pairs; the data pairs are used to train the intelligent model.

[0013] Preferably, it also includes a data processing module, which is used to clean, extract features and standardize the network status data to produce the network status data into a tensor format compatible with the intelligent model.

[0014] Preferably, the traffic characteristic data includes any one or more of transmission delay, packet loss rate, and bandwidth utilization; the network service quality indicator data includes any one or more of service quality category identifier, allocation and retention priority, circuit switching fallback mechanism timer, circuit switching fallback mechanism success rate, call detail record generation time, and user priority setting.

[0015] The present invention also provides a communication server, comprising: one or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including instructions for executing the cross-network communication service control method as described above.

[0016] The present invention also provides a computer-readable storage medium, characterized in that it includes a computer program, and the computer program can be executed by a processor to implement the cross-network communication service control method as described above.

[0017] Compared with the prior art, the cross-network communication service control method provided by the above technical solution of the present invention uses an intelligent model to process the network status data of two networks of different types communicating and generate network service parameters. It can respond immediately when the network status changes, and dynamically adjust the relevant parameters of the two networks, so that the two networks with different performances can establish a stable and smooth communication connection, and can cope with situations such as sudden traffic or network congestion, thereby improving the quality and stability of network services. Secondly, the control method can dynamically adjust the behavior of the two related networks according to the changes in the network environment and the network service parameters, and has a high degree of flexibility and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of cross-network communication status in an embodiment of the present invention.

[0019] Figure 2 The figure is a flow chart of a network communication service control method in an embodiment of the present invention.

[0020] Figure 3 Schematic diagram of the principle architecture of the intelligent model in the embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to explain the technical content, structural features, achieved objectives and effects of the present invention in detail, the following is a detailed description in conjunction with the implementation methods and the accompanying drawings.

[0022] This embodiment discloses a cross-network communication service control method, which is intended to meet the roaming and multi-network access needs of users. This scenario involves two different telecommunication core networks, such as Figure 1 , CN-A and CN-B respectively, where CN-A is the network currently accessed by terminal device A, and CN-B is the network currently accessed by terminal device B. Terminal device A and terminal device B communicate through CN-A and CN-B.

[0023] For example, CN-A is a 5G network and CN-B is a 3G network. The 5G network needs to lower some performance parameters, while the 3G network needs to increase some performance parameters, so that terminal device A and terminal device B can establish stable and smooth communication, minimize the difference in experience between the two communicating users, and improve the user experience.

[0024] Please refer to Figure 1 and Figure 2 , the control method includes the following process: S10: When the first network establishes communication with the second network, network status data of the first network and the second network within a historical time period is collected, the network status data includes network traffic characteristic data and network service quality indicator data, the first network and the second network are two different types of communication networks, such as the first network is a 5G network, and the second network is a 3G network.

[0025] S11: Use an intelligent model to predict and process the network status data set to obtain network service parameters, where the network service parameters include a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, and the intelligent model is constructed based on a reinforcement learning model.

[0026] S12: adjusting the communication service state of the first network based on the first network service parameter, and adjusting the communication service state of the second network based on the second network service parameter, so as to reduce the service difference between the first network and the second network at the current moment.

[0027] In the network, different users may have different priorities. For users with high priority, data is transmitted first. However, the priority division is inconsistent for each network, and the preemption performance is inconsistent. For example, the highest priority division in one network is level 5, and the highest priority division in another network is level 8. Then it is necessary to adjust the priorities of users of different levels according to the above control method to make them consistent.

[0028] On the other hand, after obtaining the network status data, the network status data is also cleaned, feature extracted and standardized to make the network status data into a tensor format that is compatible with the intelligent model.

[0029] Specifically, Figure 3 The intelligent model includes a proxy server, a central controller, a training unit, an intelligent agent, and a database. The proxy server is connected to the first network through an interface X1, connected to the second network through an interface X2, and connected to the central controller through an interface X3. The central controller is connected to the intelligent agent through an interface X4 and connected to the training unit through an interface X7. The intelligent agent is connected to the database through an interface X5 and connected to the training unit through an interface X6. The training unit is connected to the database through an interface X8.

[0030] Based on this, when two different types of networks communicate, the intelligent model is connected to the two networks involved through interface X1 and interface X2 respectively.

[0031] Then, the intelligent model uses its interfaces X1 and X2 to perform data packet capture operations on the first network and the second network to collect network status data over a period of time in a historical period.

[0032] The collected network status data is then sent to the central controller via interface X3. This process ensures that all necessary data can be centrally managed and processed to support higher-level decision making.

[0033] Next, the central controller is responsible for preprocessing the received data, including data cleaning, feature extraction, and standardization. The preprocessed data is converted into a tensor format suitable for intelligent model processing. If there are missing values, the central controller will use zero padding to ensure the consistency of the tensor format. After processing, these prepared data are transmitted to the intelligent agent through interface X4, providing a basis for the next step of intelligent decision-making.

[0034] The agent then receives standardized data from the central controller and inputs it into a pre-trained neural network model for calculation.

[0035] Then, the agent saves the current network status data to the database through the X5 interface.

[0036] In order to further improve the decision-making process under the reinforcement learning framework, it is also possible to determine whether the current network state is the initial state; if not, the current state is regarded as the result of the previous decision and is associated with the previous decision state to form a complete state sequence. This helps to better track the changing trend of the network state and improve the accuracy and efficiency of decision-making.

[0037] Finally, the intelligent agent transmits the determined network service parameters back to the first network and the second network one by one through the interface X4, and instructs them to communicate according to the new parameter settings.

[0038] It should be noted that the network service parameters are a set of policy vectors, the content of which is shown in Table 1 below.

[0039] Table 1

[0040] On the other hand, the network service parameters output by the intelligent model and the network status data collected in the next cycle are constructed into data pairs, and the data pairs are stored. When the intelligent model is in an idle state, the intelligent model is trained based on the data pairs, so that the performance of the intelligent model can be adaptively improved.

[0041] On the other hand, the traffic characteristic data includes any one or more of transmission delay, packet loss rate, and bandwidth utilization; the network service quality indicator data includes any one or more of service quality category identifier, allocation and retention priority, circuit switching fallback mechanism timer, circuit switching fallback mechanism success rate, call detail record generation time, and user priority setting.

[0042] In summary, the present invention discloses a cross-network communication service control method. First, the network status data of two networks of different types communicating are processed using an intelligent model and network service parameters are generated. When the network status changes, the two related network parameters can be immediately responded to and dynamically adjusted, so that the two networks with different performances can establish a stable and smooth communication connection, and can cope with situations such as sudden traffic or network congestion, thereby improving the quality and stability of network services. Secondly, the control method can dynamically adjust the behavior of the two related networks according to the changes in the network environment and the network service parameters, and has a high degree of flexibility and adaptability.

[0043] In another preferred embodiment of the present invention, a cross-network communication service control system is also disclosed, which includes a data acquisition module, an intelligent model and an adjustment module.

[0044] The data acquisition module is used to collect network status data of the first network and the second network within a historical time period when the first network establishes communication with the second network. The network status data includes network traffic characteristic data and network service quality indicator data. The first network and the second network are two different types of communication networks.

[0045] An intelligent model is used to predict and process a network status data set to obtain network service parameters, wherein the network service parameters include a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, and the intelligent model is constructed based on a reinforcement learning model.

[0046] The adjustment module is used to adjust the communication service state of the first network based on the first network service parameter and adjust the communication service state of the second network based on the second network service parameter to reduce the service difference between the first network and the second network at the current moment.

[0047] Furthermore, the control system also includes a database, which is used to construct data pairs from the network service parameters output by the intelligent model and the network status data collected in the next cycle, and store the data pairs; the data pairs are used to train the intelligent model.

[0048] Furthermore, the control system also includes a data processing module, which is used to clean, extract features and standardize the network status data to produce the network status data into a tensor format that is compatible with the intelligent model.

[0049] The working principle and process of the control system in this embodiment are detailed in the cross-network communication service control method in the above embodiment, which will not be repeated here.

[0050] The present invention also discloses a communication server, which includes one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the program includes instructions for executing the cross-network communication service control method as described above. The processor can adopt a general central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits to execute related programs to implement the functions required to be executed by the module in the communication server of the embodiment of the present application, or to execute the cross-network communication service control method of the method embodiment of the present application.

[0051] The present invention also discloses a computer-readable storage medium, which includes a computer program, and the computer program can be executed by a processor to complete the cross-network communication service control method as described above. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a read-only memory (ROM), or a random access memory (RAM), or a magnetic medium, such as a floppy disk, a hard disk, a tape, a magnetic disk, or an optical medium, such as a digital versatile disc (DVD), or a semiconductor medium, such as a solid state disk (SSD), etc.

[0052] The present application also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the electronic device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the electronic device executes the above-mentioned cross-network communication service control method. The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. Therefore, equivalent changes made according to the scope of the patent application of the present invention are still within the scope covered by the present invention.

Claims

1. A cross-network communication service control method, characterized in that: include: When a first network establishes communication with a second network, collecting network status data of the first network and the second network in a historical time period, the network status data including network traffic characteristic data and network service quality indicator data, the first network and the second network being two different types of communication networks; Using an intelligent model to perform prediction processing on the network status data set to obtain network service parameters, the network service parameters including a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, the intelligent model being constructed based on a reinforcement learning model; The communication service state of the first network is adjusted based on the first network service parameter, and the communication service state of the second network is adjusted based on the second network service parameter to reduce the service difference between the first network and the second network at the current moment.

2. The inter-network communication service control method according to claim 1, characterized in that: The network service parameters output by the intelligent model and the network status data collected in the next cycle are constructed into a data pair, and the data pair is stored; when the intelligent model is in an idle state, the intelligent model is trained based on the data pair.

3. The inter-network communication service control method according to claim 1, characterized in that: After the network status data is acquired, the network status data is cleaned, feature extracted, and standardized so as to make the network status data into a tensor format compatible with the intelligent model.

4. The inter-network communication service control method according to claim 1, characterized in that: The traffic characteristic data includes any one or more of transmission delay, packet loss rate, and bandwidth utilization; the network service quality indicator data includes any one or more of service quality category identifier, allocation and retention priority, circuit switching fallback mechanism timer, circuit switching fallback mechanism success rate, call detail record generation time, and user priority setting.

5. A cross-network communication service control system, characterized in that: include: a data collection module, which is used to collect network status data of the first network and the second network in a historical time period when the first network establishes communication with the second network, wherein the network status data includes network traffic characteristic data and network service quality indicator data, and the first network and the second network are two different types of communication networks; an intelligent model, which is used to perform prediction processing on the network status data set to obtain network service parameters, wherein the network service parameters include a first network service parameter corresponding to the first network and a second network service parameter corresponding to the second network, and the intelligent model is constructed based on a reinforcement learning model; An adjustment module is used to adjust the communication service state of the first network based on the first network service parameter, and adjust the communication service state of the second network based on the second network service parameter, so as to reduce the service difference between the first network and the second network at the current moment.

6. The inter-network communication service control system according to claim 5, characterized in that: It also includes a database, which is used to construct data pairs from the network service parameters output by the intelligent model and the network status data collected in the next cycle, and store the data pairs; the data pairs are used to train the intelligent model.

7. The inter-network communication service control system according to claim 5, characterized in that: It also includes a data processing module, which is used to clean, extract features and standardize the network status data to make the network status data into a tensor format that is compatible with the intelligent model.

8. The inter-network communication service control system according to claim 5, characterized in that: The traffic characteristic data includes any one or more of transmission delay, packet loss rate, and bandwidth utilization; the network service quality indicator data includes any one or more of service quality category identifier, allocation and retention priority, circuit switching fallback mechanism timer, circuit switching fallback mechanism success rate, call detail record generation time, and user priority setting.

9. A communication server, characterized in that: include: one or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs including instructions for executing the cross-network communication service control method as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: It comprises a computer program which can be executed by a processor to implement the inter-network communication service control method as claimed in any one of claims 1 to 4.

Images