CN114666423A

CN114666423A - TCP stream adjusting method and system

Info

Publication number: CN114666423A
Application number: CN202011408143.8A
Authority: CN
Inventors: 江舟; 连超; 赵军锋; 张平荣
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2020-12-04
Filing date: 2020-12-04
Publication date: 2022-06-24
Also published as: WO2022116665A1

Abstract

The embodiment of the application relates to the field of communication and discloses a method and a system for adjusting a TCP stream. In the application, through intelligent analysis statistics and trend prediction by using the NWDAF, SMF data provided by the SMF and UPF data provided by the UPF are analyzed, further adjustment strategies of TCP streams generated when various APPs access the network are estimated, if the TCP streams of certain types of APPs are accelerated, speed limitation is performed on certain types of APPs, finally the estimated TCP stream adjustment strategies are pushed to a TCP accelerating device which is accessed between the UPF and an external gateway in advance, so that the TCP accelerating device can reasonably adjust the TCP streams from the UPF, and further under limited network resources, the experience of user network access of the mobile terminal is improved as much as possible.

Description

TCP stream adjusting method and system

Technical Field

The embodiment of the application relates to the field of communication, in particular to a method and a system for adjusting a TCP stream.

Background

The 5G communication network has the characteristics of high speed (enhanced mobile broadband, eMBB), large capacity (large-scale machine communication, mtc) and low time delay (high reliability and low time delay communication, urrllc), so that with the development of the 5G communication network, the network access experience of users is greatly improved.

However, as people's internet habits are changed from a PC terminal to a mobile terminal, and in addition, with the ever-burning videos, live broadcasts and the like, the mobile network bandwidth requirements of users also increase rapidly. In order to improve the competitiveness of operators, it is important how to utilize limited network resources in this situation to improve the experience of network access of users of mobile terminals as much as possible.

Disclosure of Invention

An object of the embodiments of the present application is to provide a method and a system for adjusting a TCP (Transmission Control Protocol) stream, which aim to solve the above technical problems.

In order to solve the foregoing technical problem, an embodiment of the present application provides a method for adjusting a TCP flow, where the method is applied to an NWDAF (Network Data analysis Function), and the method includes:

SMF (Session management Function, respectively to Session management Function) and UPF (User Plane Function) send log subscription instructions;

receiving SMF data fed back by the SMF according to the log subscription instruction and UDF data fed back by the UPF according to the log subscription instruction;

analyzing the SMF data and the UDF data, and estimating a TCP stream adjustment strategy;

and pushing the TCP stream adjustment strategy to a TCP acceleration device which is accessed between the UPF and an external gateway in advance, so that the TCP acceleration device can adjust the TCP streams generated when each APP (Application) accesses the network according to the TCP stream adjustment strategy.

In order to achieve the above object, an embodiment of the present application further provides a TCP flow adjusting method, which is applied to a TCP acceleration device, where the TCP acceleration device is communicatively connected to a network data analysis function NWDAF, a user plane function UPF, and an external gateway, respectively, and the method includes:

sending a TCP stream adjustment policy subscription instruction to the NWDAF;

receiving a TCP flow adjustment strategy fed back by the NWDAF according to the TCP flow adjustment strategy subscription instruction, wherein the TCP flow adjustment strategy is determined by SMF data fed back by the NWDAF based on a Session Management Function (SMF) and UPF data fed back by the UPF;

and adjusting the TCP stream generated when each application program APP accesses the network according to the TCP flow adjustment strategy.

In order to achieve the above object, an embodiment of the present application further provides an apparatus for adjusting a TCP flow, including:

the log subscription module is used for respectively sending log subscription instructions to the session management function SMF and the user plane function UPF;

a data receiving module, configured to receive SMF data fed back by the SMF according to the log subscription instruction, and UDF data fed back by the UPF according to the log subscription instruction;

the data analysis module is used for analyzing the SMF data and the UDF data and estimating a TCP flow adjustment strategy;

and the strategy pushing module is used for pushing the TCP stream adjustment strategy to a TCP acceleration device which is pre-accessed between the UPF and the external gateway, so that the TCP acceleration device can adjust the TCP stream generated when each application program APP accesses the network according to the TCP stream adjustment strategy.

a policy subscription module, configured to send a TCP flow adjustment policy subscription instruction to the NWDAF;

a policy receiving module, configured to receive a TCP traffic adjustment policy fed back by the NWDAF according to the TCP traffic adjustment policy subscription instruction, where the TCP traffic adjustment policy is determined by SMF data fed back by the NWDAF based on a session management function SMF and UPF data fed back by the UPF;

and the TCP flow adjusting module is used for adjusting the TCP flow generated when each application program APP accesses the network according to the TCP flow adjusting strategy.

In order to achieve the above object, an embodiment of the present application further provides a network data analysis function, including:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the adaptation method applied to a TCP flow of an NWDAF as described above.

In order to achieve the above object, an embodiment of the present application further provides a TCP acceleration apparatus, including:

at least one processor; and the number of the first and second groups,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method of adjusting a TCP flow as applied to a TCP acceleration device as described above.

To achieve the above object, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the TCP flow adjusting method applied to the NWDAF as described above, or the TCP flow adjusting method applied to the TCP acceleration apparatus as described above.

In order to achieve the above object, an embodiment of the present application further provides a system for adjusting a TCP flow, including:

the TCP acceleration apparatus as described above; and the number of the first and second groups,

a network data analysis function NWDAF, a user plane function UPF, and an external gateway as described above communicatively coupled to the TCP acceleration device; and the number of the first and second groups,

a Session Management Function (SMF) communicatively coupled to the NWDAF; wherein the content of the first and second substances,

the NWDAF is also communicatively coupled with the UPF;

the NWDAF is to perform the adaptation method of the TCP flow as applied to the NWDAF as described above;

the TCP acceleration device is configured to execute the TCP flow adjustment method applied to the TCP acceleration device as described above.

The NWDAF is a data perception analysis network element, automatically perceives and analyzes the network based on network data, and participates in the whole life cycle of network planning, construction, operation and maintenance, network optimization and operation, so that the network is easy to maintain and control, the use efficiency of network resources is improved, and the service experience of users is improved. Based on this, the method and system for adjusting the TCP flow provided by the application analyze SMF data provided by SMF and UPF data provided by UPF by using NWDAF intelligent analysis statistics and trend prediction, and then estimate the adjustment policy of the TCP flow generated when each APP accesses the network, if accelerating the TCP flow of a certain type of APP, limiting the speed of a certain type, and finally pushing the estimated adjustment policy of the TCP flow to a TCP acceleration device pre-accessed between UPF and an external gateway, so that the TCP acceleration device can reasonably adjust the TCP flow from UPF, and further improve the experience of user network access of the mobile terminal as much as possible under limited network resources.

Drawings

One or more embodiments are illustrated by the corresponding figures in the drawings, which are not meant to be limiting.

Fig. 1 is a schematic structural diagram of a TCP flow adjustment system according to a first embodiment of the present application;

fig. 2 is a schematic structural diagram of a TCP flow adjustment system in a 5G network according to a first embodiment of the present application;

fig. 3 is a flowchart of a TCP flow adjustment method according to a second embodiment of the present application;

fig. 4 is an interaction diagram of functional entities involved in a TCP flow adjustment method according to a second embodiment of the present application;

fig. 5 is a flowchart of a TCP flow adjustment method according to a third embodiment of the present application;

fig. 6 is a schematic structural diagram of an apparatus for adjusting a TCP flow according to a fourth embodiment of the present application;

fig. 7 is a schematic structural diagram of an apparatus for adjusting a TCP flow according to a fifth embodiment of the present application;

fig. 8 is a schematic structural diagram of a network data analysis function according to a sixth embodiment of the present application;

fig. 9 is a schematic structural diagram of a TCP acceleration device according to a seventh embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. However, it will be appreciated by those of ordinary skill in the art that in the examples of the present application, numerous technical details are set forth in order to provide a better understanding of the present application. However, the technical solution claimed in the present application can be implemented without these technical details and various changes and modifications based on the following embodiments. The following embodiments are divided for convenience of description, and should not constitute any limitation to the specific implementation manner of the present application, and the embodiments may be mutually incorporated and referred to without contradiction.

A first embodiment of the present application relates to a TCP flow adjustment system, as shown in fig. 1. A TCP flow adaptation system, comprising: TCP acceleration device 10, network data analysis function 20, session management function 30, user plane function 40 and external gateway 50.

The TCP acceleration device 10 is in communication connection with the network data analysis function 20, the user plane function 40 and the external gateway 50, respectively, and the network data analysis function 20 is in communication connection with the session management function 30 and the user plane function 40.

It should be understood that the network Data analysis function 20 in this embodiment is an nwdaf (network Data analysis function) in the communication field. The network is a functional entity provided by a fifth Generation mobile communication technology (5th-Generation, 5G) system, can be regarded as a data perception analysis network element, automatically perceives and analyzes the network mainly based on network data, and participates in the whole life cycle of network planning, construction, operation and maintenance, network optimization and operation, so that the network is easy to maintain and control, the use efficiency of network resources is improved, and the user service experience is improved.

Similarly, the session management function 30 is also a functional entity of the 5G service-based architecture, namely, the smf (session management function) in the communication field. Specifically, in practical applications, the SMF is mainly responsible for interacting with the classified Data plane, creating, updating, and deleting a Protocol Data Unit (PDU) session, and managing a session context (session context) with the user plane function 40, that is, in practical applications, there is also a communication connection between the session management function 30 and the user plane function 40.

Similarly, the user Plane function 40 is also an important functional entity in the 5G service-based architecture, namely, the upf (user Plane function) in the communication field.

The TCP acceleration device 10 is a functional entity newly added to the 5G service-based architecture in this embodiment, and is mainly used for accelerating or adjusting the speed limit of the TCP flow according to the TCP flow adjustment policy provided by the network data subfunction 20.

In order to better understand the adjustment strategy of the TCP flow provided by the present embodiment to adjust the TCP flow, the following description is made in conjunction with the architecture diagram in the 5G network given in fig. 2.

For convenience of explanation, the network element functional entities in the architecture shown in fig. 2 are briefly described as follows:

as shown in fig. 2, NSSF (Network Slice Selection Function), AUSF (Authentication Server Function), NEF (Network Exposure Function), UDM (Unified Data Management Function), NRF (Network Repository Function), PCE (Policy Control Function), AF (Application Function), which are Network Function entities that communicate with AMF (Access and Mobility Management Function), SMF (Session Management Function), and NWDAF (Network Data analysis Function) through Network bus, which communicate with AMF (Network Data analysis Function) through N1 (UE and UE interface), which is a User interface (AMF 2), similar to the s1-mme interface in 4G, the UE accesses (R) AN (Radio Access Network) through a wired or wireless manner, where (R) AN communicates with UPF through AN N3 interface (AN interface between (R) AN and UPF, mainly used for transferring 5G (R) uplink and downlink user plane Data between AN and UPF), and communicates with SMF through AN N4 interface (AN interface between SMF and UPF, used for transmitting control plane information between SMF and UPF), and TCPO (TCP acceleration device) communicates with UPF and DN (Data Network, such as operator service, internet Access or third-party service, etc.) through AN N6 interface (AN interface between UPF and DN, used for transferring uplink and downlink user Data between UPF and DN), and communicates with DN Network based on IP and routing protocol, respectively.

It should be noted that the DN is the external gateway in this embodiment, and in practical application, the external gateway may be a FW (firewall), or another network device in the core network that carries the role of the gateway, besides the DN.

In addition, it should be understood that the above is only a specific architecture in the 5G network, and the technical solution of the present application is not limited, and in practical applications, specific network element functional entities in the architecture may vary according to business needs, as long as the architecture is ensured to include the TCP acceleration device 10, the network data analysis function 20, the session management function 30, the user plane function 40, and the external gateway 50 in the embodiment, and communication connectivity between the 5 network element functional entities is ensured according to the connection relationship in the embodiment.

Based on the structure shown in fig. 2, when the adjustment of the TCP flow is implemented, the specific steps are:

first, a TCP flow adjustment policy subscription instruction is sent by the TCPO (i.e., the TCP acceleration device 10 in fig. 1) to the NWDAF (i.e., the network data analysis function 20 in fig. 1).

Next, after receiving a TCP flow adjustment policy subscription instruction sent by the TCPO, the NWDAF generates a log subscription instruction, and sends the generated log subscription instruction to the SMF (i.e., the session management function 30 in fig. 1) and the UPF (i.e., the user plane function 40 in fig. 1), respectively, so that when the SMF monitors that a creation, an update, or a PDU session is created, the SMF can push generated data (hereinafter referred to as SMF data for convenience of description) to the NWDAF; enabling the UF to push the resulting data (hereinafter referred to as UPF data for ease of description) to the NWDAF when a flow release operation is monitored.

And then, the NWDAF analyzes SMF data and UDF data based on the capability of analyzing and predicting the network data, so as to predict a TCP stream adjustment strategy suitable for the current network, and pushes the obtained TCP stream adjustment strategy to the TCPO.

And finally, the TCPO adjusts the TCP stream generated when each APP (Application) released by the UPF accesses the network according to the received TCP stream adjustment strategy, for example, the TCP stream corresponding to the high-value APP is accelerated and then sent to an external gateway DN, the TCP stream corresponding to the low-value APP is subjected to speed limitation, and then the TCP stream is sent to the external gateway DN.

Therefore, the TCP flow adjustment system provided in this embodiment analyzes SMF data provided by SMF and UPF data provided by UPF by using NWDAF intelligent analysis statistics and trend prediction, and then estimates an adjustment policy for TCP flow generated when each APP accesses a network, and if accelerating TCP flow of a certain type of APP, speed limiting is performed on a certain type, and finally the estimated TCP flow adjustment policy is pushed to a TCP acceleration device pre-accessed between UPF and an external gateway, so that the TCP acceleration device can reasonably adjust TCP flow from UPF, and further, under limited network resources, user network access experience of a mobile terminal is improved as much as possible.

A second embodiment of the present application relates to a TCP flow adjustment method, which is applied to a network data analysis function NWDAF.

Regarding NWDAF, other network element functional entities involved in implementing the TCP flow adjustment method, and connection relationships between these network element functional entities, see fig. 1 and fig. 2 corresponding to the TCP flow adjustment system provided in the first embodiment of the present application in detail.

The following describes implementation details of the TCP flow adjustment method of the present embodiment, and the following description is provided only for easy understanding and is not necessary to implement the present embodiment.

The specific flow of this embodiment is shown in fig. 3, and specifically includes the following steps:

step 301, sending a log subscription instruction to the session management function SMF and the user plane function UPF, respectively.

And step 302, receiving the SMF data fed back by the SMF according to the log subscription instruction and the UDF data fed back by the UPF according to the log subscription instruction.

And 303, analyzing the SMF data and the UDF data, and estimating a TCP stream adjustment strategy.

Step 304, pushing the TCP flow adjustment policy to a TCP acceleration device pre-accessed between the UPF and the external gateway, so that the TCP acceleration device adjusts the TCP flow generated when each application APP accesses the network according to the TCP flow adjustment policy.

For better understanding of the operations in step 301 to step 304, the following is described in detail with reference to fig. 4:

(1) the TCPO (TCP accelerator) sends a TCP flow adjustment policy subscription instruction to the NWDAF.

That is, in practical applications, the trigger NWDAF to perform the operations of step 301 to step 304 is based on the assumption that the NWDAF receives a TCP flow adjustment policy subscription instruction sent by a TCPO (TCP acceleration device).

(2) After receiving the TCP flow adjustment policy subscription instruction sent by the TCPO, the NWDAF generates a log subscription instruction, and sends the generated log subscription instruction to the SMF and the UPF, respectively.

In practical applications, the log subscription instruction is specifically an instruction for subscribing a Call History Record (CHR) log, that is, the instruction is extracted from the CHR log when the SMF and the UPF which are fed back by the SMF according to the log subscription instruction and received subsequently are based on the SMF data and the UPF data which are fed back by the log subscription instruction.

As can be seen from the description of the SMF in the first embodiment, the SMF is mainly responsible for interacting with the classified data plane, and creating, updating, and deleting PDU sessions. Therefore, when there is a PDU session generated by interactive operations such as creation, update, or deletion, the SMF performs an operation of collecting control plane creation information and pushes collected data (hereinafter referred to as SMF data) to the NWDAF.

Correspondingly, since the UPF is mainly responsible for the UE, the present embodiment mainly refers to a flow generated when the APP in the mobile user equipment accesses the network, and the present embodiment mainly refers to a TCP flow being sent to the external gateway, that is, a flow release is performed. Therefore, when a flow is released, the UPF performs flow information acquisition and pushes acquired data (hereinafter referred to as UPF data) to the NWDAF.

(3) After receiving the SMF data and the UPF data, the NWDAF analyzes the SMF data and the UDF data based on self analysis and prediction capabilities, and then predicts to obtain a TCP flow adjustment strategy.

Specifically, the operations of analyzing the SMF data and the UDF data with respect to the NWDAF and estimating the TCP flow adjustment policy are specifically:

firstly, the SMF data and the UDF data are merged according to preset field information, and then data to be processed are obtained.

Specifically, in this embodiment, the SEID (serial number of NFC) and the N4 interface are used as the preset field information, and the SMF data and the UDF data are merged, that is, the data provided by the N4 interface and having the same SEID are merged.

Based on the mode, the merged data to be processed mainly comprises the following steps: an International Mobile Subscriber Identity (IMSI), a private network address, a private network port, a protocol type, a destination address, a destination port, a packet number, a duration, a known APP type, a Uniform Resource Locator (URL), and the like.

And then, analyzing data in different time dimensions and different using areas in the data to be processed based on a preset trend prediction machine learning algorithm by taking a preset time granularity and a preset area as dimensions to obtain the TCP stream adjustment strategy.

Specifically, in this embodiment, the estimation of the TCP flow adjustment policy is to quickly identify an APP with a large occupied flow but low value and a high value, such as a hot APP.

Therefore, when a TCP flow adjustment strategy is estimated based on the above manner, specifically, a preset time granularity and a preset region are taken as dimensions, and source IP addresses of each user side, destination IP addresses corresponding to APPs in each user side, and historical traffic use information of the APPs in different time dimensions and different use regions in the data to be processed are counted; then, by traversing each APP, establishing a mapping relation between the source IP address and a destination IP address and historical traffic use information corresponding to the traversed APP, and obtaining a historical traffic use statistical table of each APP; and finally, analyzing each piece of data with mapping relation recorded in the historical flow use statistical table based on a preset trend prediction machine learning algorithm to obtain the TCP flow adjustment strategy.

The above-mentioned predetermined time granularity may be counted in minutes, hours, days, that is, every minute, or every hour, or once a day; the preset area may be a cell.

In addition, in practical applications, the TCP acceleration device for adjusting the TCP flow can only recognize the destination IP, and accelerate or adjust the speed limit of the corresponding TCP flow through the destination IP. Therefore, in the above-mentioned TCP flow adjustment policy, a destination IP address that needs acceleration and a destination IP address that needs speed limitation are recorded specifically.

Regarding the destination IP address requiring speed limiting, specifically, the NWDAF analyzes each piece of data having a mapping relationship recorded in the historical traffic usage statistical table based on a preset trend prediction machine learning algorithm, and then identifies the destination IP address corresponding to the low-value APP with large occupied traffic.

It should be noted that, in this embodiment, the low-value APP with large occupied traffic is specifically obtained by analyzing and counting the historical traffic usage of the previous N cells in which the network congestion phenomenon occurs, based on the data recorded in the to-be-historical traffic usage statistical table by the NWDAF, and when congestion occurs, the IMSI of the UE frequently occurring and the APP with large occupied traffic but low usage percentage are large.

For example, it is found through statistics that congestion occurs in the current statistical time, and 1000 users in a certain cell of the first N cells access the network, but only 10 users, that is, 10 UEs with different IMSIs, access the network using the APP of the P2P type, but occupy a large amount of traffic allocated to the cell. At this time, in order not to affect the internet experience of other user groups (other users 990), it is considered that the APP of P2P type used by the UEs with different IMSIs is a low-value APP with large occupied traffic, that is, speed limitation is required.

Correspondingly, the destination IP address to be accelerated is specifically a destination IP address corresponding to the high-value APP identified by analyzing each piece of data having a mapping relationship recorded in the historical traffic usage statistical table based on a preset trend prediction machine learning algorithm by the NWDAF.

Regarding the high-value APP, in practical applications, the high-value APP may be determined according to the user popularity, for example, an APP with a high user access amount may be considered as a popular APP. This type of APP can be considered a high value APP.

It should be understood that the above examples are only examples for better understanding of the technical solution of the present embodiment, and are not to be taken as the only limitation to the present embodiment.

Further, considering the influence of factors such as seasons and holidays on the prediction result, that is, high-frequency access to a certain type of APP may only occur in a specific season or holiday, for example, cold winter, most user groups like home-based pursuit, and this stage of video-type APP may become popular APP, and for example, online sales promotion activities performed by e-commerce at a specific time may be performed at a fixed time point of this time, live-type software may become popular APP due to online sales promotion. Therefore, in order to make the estimated TCP flow adjustment policy more consistent with the actual situation, when estimating the TCP flow adjustment policy, the NWDAF may first obtain the generation time of the historical flow usage information in the data having a mapping relationship in each piece recorded in the historical flow usage statistical table, and then analyze the data having a mapping relationship in each piece recorded in the historical flow usage statistical table based on the trend prediction machine learning algorithm and the generation time to obtain the TCP flow adjustment policy.

Therefore, when the TCP flow adjustment strategy is estimated, the generation time of the historical flow use information of each APP is introduced, so that the current season can be determined according to the generation time, whether the current season is a holiday, some specific activity days and the like, and further, when the data in the historical flow use statistical table are analyzed based on a trend prediction machine learning algorithm, the influence of factors such as the season and the holiday on the trend can be considered, and the estimated TCP flow adjustment strategy is more reasonable.

It should be noted that, in practical applications, the trend prediction machine learning algorithm may be an Autoregressive differential Moving Average Model (ARIMA), a prophet (time series prediction library) Model, a Long-Short-Term Memory Model (LSTM), and the like, which are not listed here, and this embodiment is not limited thereto.

For ease of understanding, the following describes the operation of predicting the TCP flow adjustment policy by taking the LSTM model as an example:

specifically, in practical application, in order to quickly and accurately predict a TCP flow adjustment strategy suitable for an actual situation, a network model with an LSTM and a full connection layer may be pre-constructed, and then a TCP flow adjustment strategy prediction model may be obtained by performing iterative training on the constructed network model, so that each time new SMF data and UDP data are obtained, the two data may be directly merged and then input to the TCP flow adjustment strategy prediction model, and a result output by the TCP flow adjustment strategy prediction model may be used as the predicted TCP flow adjustment strategy.

The training of the TCP flow adjustment strategy prediction model is as follows:

first, sample data acquired in advance, that is, data obtained by combining SMF data and UDP data generated at a historical time is processed in accordance with the following equation 7: the scale of 3 is divided into a training set and a test set.

Then, the flow values in the training set in the time period of fixed time are normalized, so that the gradient item descending speed is increased.

Then, the flow values after normalization processing are sequentially input into the constructed network model for iterative training, for example: the input flow values are respectively the flow values at t-4, t-3, t-2 and t-1, the flow value at t is predicted based on the flow values at the times, then the predicted flow value at t is added into a training set, and the flow value at t +1 is predicted, and the overall process is as follows:

inputting: t-4, t-3, t-2, t-1; and (3) prediction: t is t

Inputting: t-3, t-2, t-1, t; and (3) prediction: t +1

Inputting: t-2, t-1, t-0, t + 1; and (3) prediction: t +2

The whole process is iterated continuously until the granularity needing to be predicted is achieved.

And then, the predicted value output after the time sequence is input into the network model is compared with the actual real value, the mean square error MSE loss function is adopted, the loss function is optimized by the ADAM optimization algorithm, gradient descent training is continuously carried out by optimizing the loss function, and the appropriate network model can be obtained along with the convergence of the loss function.

Finally, in the test set after normalization, the flow values corresponding to different times are input into the network model after iterative training, then the output predicted value is compared with the actual real value, when the performance index of the test result reaches the expectation, if the test value is basically the same as the real value or the error is in a certain range, the current network model can be determined as the TCP flow adjustment strategy prediction model, that is, the model can be used to predict the flow trend of the cell at a future time, and the subsequent TCP flow adjustment strategy suitable for the cell is formulated.

For example, once it is predicted that the traffic trend behind the cell shows a significant rising trend, if the increase rate of the rising trend exceeds a preset value, for example, the traffic rises by 30% in 30min, the private network IP corresponding to the IMSI causing the cell congestion and the destination IP accessed by the low-value APP, which are counted from the historical data, are determined as the destination IP requiring speed limitation, and the TCP flow adjustment policy is pushed to the TCP acceleration device, the TCP acceleration device limits the speed of the destination IP according to the needs, the TCP flow generated when the APP corresponding to the destination IP accesses the network is limited, otherwise, the TCP acceleration device accelerates the TCP flow generated when the APP corresponding to the destination IP accesses the network according to the destination IP requiring acceleration.

That is to say, in fig. 4, after receiving the TCP flow adjustment policy pushed by the NWDAF, the TCPO identifies the destination IP address in the UE internet access flow sent by the UPF according to the destination IP address to be accelerated recorded in the TCP flow adjustment policy, further accelerates the destination IP address to be accelerated, that is, the TCP flow corresponding to the high-value APP, and then sends the accelerated TCP flow to the external gateway; and identifying a target IP address in the UE internet access flow sent by the UPF according to the target IP address needing speed limitation and recorded in the TCP flow adjustment strategy, further limiting the speed of the target IP address needing speed limitation, namely the TCP flow corresponding to the low-value APP, and then sending the TCP flow after speed limitation to an external gateway.

It is not difficult to find out through the above description that the adjustment method of TCP flow provided by this embodiment analyzes SMF data provided by SMF and UPF data provided by UPF by using NWDAF's intelligent analysis statistics plus trend prediction, and then estimates the adjustment policy of TCP flow generated when each APP accesses the network, if accelerating TCP flow of a certain type of APP, then limiting speed of a certain type, and finally pushing the estimated adjustment policy of TCP flow to a TCP acceleration device pre-accessed between UPF and an external gateway, so that the TCP acceleration device can reasonably adjust TCP flow from UPF, and further improve user network access experience of the mobile terminal as much as possible under limited network resources.

That is to say, based on the adjustment method of the TCP flow provided in this embodiment, it can be implemented that, under the condition that the mobile network is busy, the policy of the data flow packet is adjusted, for example, the TCP flow of the low-value APP is subjected to speed limitation, the TCP flow of the high-value APP is subjected to acceleration adjustment policy, and the TCP flow generated when each APP accesses the network is reasonably adjusted, instead of uniformly adjusting all the TCP flows, for example, speed limitation or acceleration is performed, so that limited mobile network resources can be reasonably utilized, thereby improving user experience, increasing user stickiness of the operator, and enhancing competitiveness of the operator.

The third embodiment of the present application relates to a TCP flow adjustment method, which is applied to a TCP acceleration apparatus.

For details of the TCP acceleration apparatus, other network element functional entities involved in implementing the TCP flow adjustment method, and the connection relationship between these network element functional entities, see fig. 1 and fig. 2 corresponding to the TCP flow adjustment system provided in the first embodiment of the present application.

As shown in fig. 5, a TCP flow adjustment method according to a second embodiment includes the following steps:

step 501, sending a TCP flow adjustment policy subscription instruction to the NWDAF.

Step 502, receiving a TCP traffic adjustment policy fed back by the NWDAF according to the TCP traffic adjustment policy subscription instruction, where the TCP traffic adjustment policy is determined by SMF data fed back by the NWDAF based on a session management function SMF and UPF data fed back by the UPF.

Step 503, according to the TCP flow adjustment policy, adjusting the TCP flow generated when each application APP accesses the network.

Specifically, the predicted TCP flow adjustment policy through the NWDAF records a destination IP address that needs to be accelerated and a destination IP address that needs to be speed-limited.

Therefore, when the TCP flow generated when each APP accesses the network is adjusted according to the TCP flow adjustment policy, the destination IP address that needs to be accelerated and the destination IP address that needs to be speed-limited are read from the TCP flow adjustment policy, and then the TCP flow generated when the APP corresponding to the destination IP address that needs to be accelerated accesses the network is accelerated based on the destination IP address that needs to be accelerated.

Correspondingly, based on the destination IP address needing speed limiting, the TCP stream generated when the APP corresponding to the destination IP address needing speed limiting accesses the network is subjected to speed limiting.

Further, in practical applications, there may be NWDAF analyzing based on SMF data and UPF data, and the estimated TCP flow adjustment policy only includes adjustment information for accelerating or limiting the TCP flow.

Therefore, in order to enable the TCP acceleration device to be better adapted to an actual usage scenario, before the TCP acceleration device reads a destination IP address to be accelerated and a destination IP address to be speed-limited from a TCP flow adjustment policy, it may be determined whether the TCP flow adjustment policy carries the destination IP address to be accelerated and the destination IP address to be speed-limited, and then a subsequent adjustment operation is triggered according to a determination result.

Specifically, if it is determined through the judgment that the TCP flow call-out policy carries the destination IP address to be accelerated and the destination IP address to be speed-limited, executing the step of accelerating the TCP flow generated when the APP corresponding to the destination IP address to be accelerated accesses the network according to the destination IP address to be accelerated, and limiting the speed of the TCP flow generated when the APP corresponding to the destination IP address to be speed-limited accesses the network according to the destination IP address to be speed-limited;

if the TCP stream call-out strategy only carries the target IP address needing to be accelerated through judgment, only executing the target IP address needing to be accelerated according to the requirement, and accelerating the TCP stream generated when the APP corresponding to the target IP address needing to be accelerated accesses the network;

and if the TCP stream calling strategy only carries the destination IP address needing speed limitation through judgment, only executing the destination IP address needing speed limitation according to the destination IP address needing speed limitation, and limiting the speed of the TCP stream generated when the APP corresponding to the destination IP address needing speed limitation accesses the network.

It is obvious that steps 501 to 503 in this embodiment are related to steps 301 to 304 in the second embodiment, that is, when the adjustment of the TCP flow is implemented, the implementation of the whole scheme needs to involve all operations of steps 301 to 304 and steps 501 to 503. The specific implementation of step 501 to step 503 in this embodiment has already been described in the description of the second embodiment with reference to fig. 4, and is not described herein again.

Therefore, the method for adjusting TCP flows provided in this embodiment analyzes SMF data provided by SMF and UPF data provided by UPF by using NWDAF intelligent analysis statistics and trend prediction, and then estimates an adjustment policy for TCP flows generated when each APP accesses a network, and if accelerating TCP flows of certain types of APPs, speed limiting is performed on certain types, and finally the estimated adjustment policy for TCP flows is pushed to a TCP acceleration device pre-accessed between UPF and an external gateway, so that the TCP acceleration device can reasonably adjust TCP flows from UPF, and further, under limited network resources, user network access experience of a mobile terminal is improved as much as possible.

In addition, it should be understood that the above steps of the various methods are divided for clarity, and the implementation may be combined into one step or split into some steps, and the steps are divided into multiple steps, so long as the same logical relationship is included in the protection scope of the present patent; it is within the scope of the patent to add insignificant modifications to the algorithms or processes or to introduce insignificant design changes to the core design without changing the algorithms or processes.

A fourth embodiment of the present application relates to a TCP flow adjustment apparatus, and as shown in fig. 6, the TCP flow adjustment apparatus 600 includes: a log subscribing module 601, a data receiving module 602, a data analyzing module 603 and a policy pushing module 604.

The log subscription module 601 is configured to send a log subscription instruction to the session management function SMF and the user plane function UPF, respectively; a data receiving module 602, configured to receive SMF data fed back by the SMF according to the log subscription instruction, and UDF data fed back by the UPF according to the log subscription instruction; a data analysis module 603, configured to analyze the SMF data and the UDF data, and estimate a TCP flow adjustment policy; and a policy pushing module 604, configured to push the TCP flow adjustment policy to a TCP acceleration device pre-accessed between the UPF and an external gateway, so that the TCP acceleration device adjusts, according to the TCP flow adjustment policy, a TCP flow generated when each application APP accesses a network.

In addition, in another example, the data analysis module 603 is specifically configured to combine the SMF data and the UDF data according to preset field information to obtain data to be processed; and analyzing data in different time dimensions and different using areas in the data to be processed based on a preset trend prediction machine learning algorithm by taking a preset time granularity and a preset area as dimensions to obtain the TCP stream adjustment strategy.

In addition, in another example, the preset trend prediction based machine learning algorithm analyzes data in different time dimensions and different usage areas in the data to be processed with a preset time granularity and a preset area as dimensions to obtain the TCP flow adjustment policy, specifically:

counting source IP addresses of the user sides, destination IP addresses corresponding to the APPs in the user sides and historical flow use information of the APPs in different time dimensions and different use areas in the data to be processed by taking a preset time granularity and a preset area as dimensions;

traversing each APP, and establishing a mapping relation between the source IP address and a destination IP address and historical traffic use information corresponding to the traversed APP to obtain a historical traffic use statistical table of each APP;

and analyzing the data of each mapping relation recorded in the historical traffic usage statistical table based on a preset trend prediction machine learning algorithm to obtain the TCP flow adjustment strategy.

In addition, in another example, the analyzing, based on a preset trend prediction machine learning algorithm, data of each piece of mapping relation recorded in the historical traffic usage statistical table to obtain the TCP flow adjustment policy specifically includes:

acquiring the generation time of the history traffic use information in each piece of data with mapping relation recorded in the history traffic use statistical table;

and for each piece of data with mapping relation recorded in the historical flow use statistical table, analyzing by combining the generation time based on the trend prediction machine learning algorithm to obtain the TCP flow adjustment strategy.

analyzing each piece of data with a mapping relation recorded in the historical flow use statistical table based on a preset trend prediction machine learning algorithm, and identifying low-value APP and high-value APP with large occupied flow;

acquiring a destination IP address corresponding to the low-value APP and a destination IP address corresponding to the high-value APP;

and taking the target IP address corresponding to the low-value APP as a target IP address needing speed limiting, and taking the target IP address corresponding to the high-value APP as a target IP address needing acceleration, so as to obtain the TCP flow adjustment strategy.

In addition, in another example, the pushing the TCP flow adjustment policy to a TCP acceleration device pre-accessed between the UPF and an external gateway is used for the TCP acceleration device to adjust a TCP flow generated when each application APP accesses a network according to the TCP flow adjustment policy, and specifically includes:

and pushing the TCP flow adjustment strategy to a TCP acceleration device which is accessed between the UPF and an external gateway in advance, so that the TCP acceleration device accelerates the TCP flow generated when the APP corresponding to the target IP address which needs to be accelerated accesses the network according to the target IP address which needs to be accelerated recorded in the TCP flow adjustment strategy, and limits the speed of the TCP flow generated when the APP corresponding to the target IP address which needs to limit the speed accesses the network according to the target IP address which needs to be limited and is recorded in the TCP flow adjustment strategy.

It should be understood that this embodiment is a device embodiment corresponding to the first embodiment, and the embodiment can be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the first embodiment.

It should be noted that, all the modules involved in this embodiment are logic modules, and in practical application, one logic unit may be one physical unit, may also be a part of one physical unit, and may also be implemented by a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, a unit which is not so closely related to solve the technical problem proposed by the present invention is not introduced in the present embodiment, but this does not indicate that there is no other unit in the present embodiment.

A fifth embodiment of the present application relates to a TCP flow adjustment apparatus, and as shown in fig. 7, a TCP flow adjustment apparatus 700 includes: a policy subscription module 701, a policy receiving module 702, and a TCP flow adjustment module 703.

The policy subscription module 701 is configured to send a TCP flow adjustment policy subscription instruction to the NWDAF; a policy receiving module 702, configured to receive a TCP traffic adjustment policy fed back by the NWDAF according to the TCP traffic adjustment policy subscription instruction, where the TCP traffic adjustment policy is determined by the NWDAF based on SMF data fed back by a session management function SMF and UPF data fed back by the UPF; and a TCP flow adjusting module 703, configured to adjust, according to the TCP flow adjusting policy, a TCP flow generated when each application APP accesses the network.

In addition, in another example, the TCP flow adjusting module 703 is specifically configured to read a destination IP address that needs to be accelerated and a destination IP address that needs to be speed-limited from the TCP flow adjusting policy; according to the destination IP address needing to be accelerated, accelerating the TCP stream generated when the APP corresponding to the destination IP address needing to be accelerated accesses the network; and limiting the speed of the TCP stream generated when the APP corresponding to the destination IP address needing speed limitation accesses the network according to the destination IP address needing speed limitation.

In another example, before the reading of the destination IP address that needs to be accelerated and the destination IP address that needs to be speed-limited from the TCP flow adjustment policy, the method further includes:

judging whether the TCP stream adjustment strategy carries the destination IP address needing acceleration and the destination IP address needing speed limitation;

if the destination IP address needing to be accelerated and the destination IP address needing to be speed-limited are carried, executing the destination IP address needing to be accelerated according to the requirement, accelerating the TCP stream generated when the APP corresponding to the destination IP address needing to be accelerated accesses the network, and executing the step of limiting the speed of the TCP stream generated when the APP corresponding to the destination IP address needing to be speed-limited accesses the network according to the destination IP address needing to be speed-limited;

if the target IP address which needs to be accelerated is carried, only executing the target IP address which needs to be accelerated according to the requirement, and accelerating the TCP stream generated when the APP corresponding to the target IP address which needs to be accelerated accesses the network;

and if the destination IP address which needs to limit the speed is carried, only executing the step of limiting the speed of the TCP stream generated when the APP corresponding to the destination IP address which needs to limit the speed accesses the network according to the destination IP address which needs to limit the speed.

It should be understood that the present embodiment is a device embodiment corresponding to the second embodiment, and the present embodiment and the second embodiment can be implemented in cooperation. The related technical details mentioned in the second embodiment are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the second embodiment.

A sixth embodiment of the present application relates to a network data analysis function, as shown in fig. 8, including: includes at least one processor 801; and a memory 802 communicatively coupled to the at least one processor; the memory 802 stores instructions executable by the at least one processor 801, and the instructions are executed by the at least one processor 801, so that the at least one processor 801 can execute the method for adjusting the TCP flow applied to the network data analysis function.

The memory 802 and the processor 801 are coupled by a bus, which may include any number of interconnecting buses and bridges that couple one or more of the various circuits of the processor 801 and the memory 802 together. The bus may also connect various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 801 is transmitted over a wireless medium through an antenna, which receives the data and transmits the data to the processor 801.

The processor 801 is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. While memory 802 may be used to store data used by processor 801 in performing operations.

A seventh embodiment of the present application relates to a TCP accelerating device, as shown in fig. 9, including: includes at least one processor 901; and, a memory 902 communicatively coupled to the at least one processor; the memory 902 stores instructions executable by the at least one processor 901, and the instructions are executed by the at least one processor 901, so that the at least one processor 901 can execute an adjustment method applied to a TCP stream of the TCP acceleration apparatus.

The memory 902 and the processor 901 are coupled by a bus, which may comprise any number of interconnected buses and bridges that couple one or more of the various circuits of the processor 901 and the memory 902. The bus may also connect various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 901 is transmitted over a wireless medium via an antenna, which further receives the data and transmits the data to the processor 901.

The processor 901 is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And memory 902 may be used for storing data used by processor 901 in performing operations.

An eighth embodiment of the present application relates to a computer-readable storage medium storing a computer program. The computer program realizes the adjustment method of the TCP stream applied to the network data analysis function when being executed by the processor; or a method for adjusting a TCP stream applied to a TCP acceleration device.

That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the present application, and that various changes in form and details may be made therein without departing from the spirit and scope of the present application in practice.

Claims

1. A method for adjusting a TCP stream, applied to a network data analysis function (NWDAF), the method comprising:

respectively sending a log subscription instruction to a Session Management Function (SMF) and a User Plane Function (UPF);

and pushing the TCP stream adjustment strategy to a TCP acceleration device which is accessed between the UPF and an external gateway in advance, so that the TCP acceleration device can adjust the TCP streams generated when the APP of each application program accesses the network according to the TCP stream adjustment strategy.

2. The method for TCP flow adjustment according to claim 1, wherein the analyzing the SMF data and the UDF data and predicting a TCP flow adjustment policy includes:

combining the SMF data and the UDF data according to preset field information to obtain data to be processed;

and analyzing data in different time dimensions and different using areas in the data to be processed based on a preset trend prediction machine learning algorithm by taking a preset time granularity and a preset area as dimensions to obtain the TCP stream adjustment strategy.

3. The method for adjusting a TCP flow according to claim 2, wherein the analyzing data in different time dimensions and different usage areas in the data to be processed by using a preset time granularity and a preset area as dimensions based on a preset trend prediction machine learning algorithm to obtain the TCP flow adjustment policy includes:

4. The method according to claim 3, wherein the analyzing data, which is recorded in the statistical table of historical traffic usage and has a mapping relationship, of each piece of data based on a preset trend prediction machine learning algorithm to obtain the TCP flow adjustment policy includes:

obtaining the generation time of the history flow use information in each piece of data with mapping relation recorded in the history flow use statistical table;

and analyzing each piece of data with a mapping relation recorded in the historical traffic use statistical table based on the trend prediction machine learning algorithm and the generation time to obtain the TCP flow adjustment strategy.

5. The method according to claim 3 or 4, wherein the analyzing, based on a preset trend prediction machine learning algorithm, data of each mapping relation recorded in the historical traffic usage statistics table to obtain the TCP flow adjustment policy includes:

6. The method for adjusting a TCP flow according to claim 5, wherein the pushing the TCP flow adjustment policy to a TCP acceleration device pre-accessed between the UPF and an external gateway, so that the TCP acceleration device adjusts a TCP flow generated when each application APP accesses a network according to the TCP flow adjustment policy, includes:

and pushing the TCP flow adjustment strategy to a TCP acceleration device which is pre-accessed between the UPF and an external gateway, supplying the TCP acceleration device to accelerate TCP flows generated when APP corresponding to the target IP addresses needing acceleration accesses the network according to the target IP addresses needing acceleration recorded in the TCP flow adjustment strategy, and limiting the TCP flows generated when APP corresponding to the target IP addresses needing speed limiting accesses the network according to the target IP addresses needing speed limiting recorded in the TCP flow adjustment strategy.

7. A method for adjusting a TCP stream, applied to a TCP acceleration device, wherein the TCP acceleration device is respectively in communication connection with a network data analysis function NWDAF, a user plane function UPF and an external gateway, and the method comprises the following steps:

sending a TCP stream adjustment policy subscription instruction to the NWDAF;

8. The method for adjusting a TCP flow according to claim 7, wherein the adjusting, according to the TCP flow adjustment policy, a TCP flow generated when each application APP accesses a network includes:

reading a destination IP address needing acceleration and a destination IP address needing speed limitation from the TCP stream adjustment strategy;

according to the destination IP address needing to be accelerated, accelerating the TCP stream generated when the APP corresponding to the destination IP address needing to be accelerated accesses the network;

and limiting the speed of the TCP stream generated when the APP corresponding to the destination IP address needing speed limitation accesses the network according to the destination IP address needing speed limitation.

9. The TCP flow adjustment method according to claim 8, wherein before reading the destination IP address that needs to be accelerated and the destination IP address that needs to be rate-limited from the TCP flow adjustment policy, the method further comprises:

10. An apparatus for adjusting a TCP flow, comprising:

11. An apparatus for adjusting a TCP flow, comprising:

12. A network data analysis function, comprising:

at least one processor; and the number of the first and second groups,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of adjusting a TCP flow according to any one of claims 1 to 6.

13. A TCP acceleration apparatus, comprising:

at least one processor; and the number of the first and second groups,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of adjusting a TCP flow according to any one of claims 7 to 9.

14. A computer-readable storage medium storing a computer program, wherein the computer program is configured to implement the TCP flow adjustment method according to any one of claims 1 to 6 or the TCP flow adjustment method according to any one of claims 7 to 9 when executed by a processor.

15. A TCP flow adjustment system, comprising:

the TCP acceleration device of claim 13; and the number of the first and second groups,

the network data analysis function NWDAF, user plane function UPF, and external gateway of claim 12 communicatively coupled to the TCP acceleration device; and the number of the first and second groups,

the NWDAF is also communicatively coupled with the UPF;

the NWDAF is configured to perform the method of adjusting a TCP flow of any one of claims 1 to 6;

the TCP acceleration apparatus is configured to execute the TCP flow adjustment method according to any one of claims 7 to 9.