Multi-stream service identification method and device and readable medium

Abstract

The invention provides a multi-stream service identification method, which comprises the following steps of: acquiring flow characteristics of data streams of a to-be-identified service type, preliminarily identifying the service type of each data stream according to the flow characteristics of the data streams of the to-be-identified service type by utilizing a first identification model, and identifying a target data stream of a first type or a second type in a preliminary identification result. The service type of each target data flow is determined to be the first type or the second type according to the flow quantity of the service to which each target data flow belongs, and the first type and the second type of services with similar flow characteristics can be accurately identified; service type identification is carried out based on the flow quantity of the data flow obtained through statistics, the service type identification algorithm is simple, the identification accuracy and the identification efficiency are high, and the method is suitable for identifying services with part of the same data flow types; the method can be realized without professional equipment and a mapping library, is low in cost, and can be applied to a WLAN scene with flexible layout. The invention also provides a multi-stream service identification device and a readable medium.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a multi-stream service identification method, device, and readable medium. Background Art

[0002] Service scenario recognition algorithms allow WLAN (Wireless Local Area Network) access devices and equipment to perceive the applications, services, and scenarios currently being accessed by users, such as gaming, voice, video, and live streaming. Based on the characteristics of these applications, services, and scenarios, access devices provide users with customized WLAN parameter configurations and services to maximize their network access experience. For example, prioritizing gaming packets provides users with a low-latency and low-stuttering gaming experience. The service type recognition algorithm, as the cornerstone of the entire process, is crucial for its performance.

[0003] Currently, service type identification primarily relies on deep packet inspection, message identification, and traffic signature recognition technologies. Deep packet inspection relies on proprietary equipment connected in series to the data path, making it unsuitable for the flexible layout of WLANs. Message identification, on the other hand, targets specific message formats, making its use and upgrade inflexible. Traffic signature recognition offers low computational complexity, ease of implementation, and wide coverage of service scenarios. However, it struggles to differentiate services with similar traffic characteristics. Summary of the Invention

[0004] The present disclosure provides a multi-stream service identification method, device, and readable medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a multi-flow service identification method, including:

[0006] Obtaining traffic characteristics of the data flow of the service type to be identified;

[0007] Using the first identification model, identifying the service type of each of the data flows according to the traffic characteristics, and obtaining a preliminary identification result;

[0008] For the target data stream whose preliminary identification result is a preset first type or a preset second type, the business type of each target data stream is determined to be the first type or the second type according to the number of streams of the business to which each target data stream belongs; wherein the number of streams of the business to which each target data stream belongs is determined based on the transmission information of each target data stream.

[0009] In another aspect, an embodiment of the present disclosure further provides a multi-flow service identification method, including:

[0010] Obtaining traffic characteristics and transmission information of the data flow of the service type to be identified;

[0011] Determining the number of flows of the service to which each of the data flows belongs according to the transmission information of each of the data flows;

[0012] The traffic characteristics and the number of flows of the business to which each of the data flows belongs are input into a second recognition model, so that the second recognition model can be used to identify the business type of each of the data flows as a preset first type or a preset second type based on the traffic characteristics, wherein the number of flows of the business to which the data flows belong is one of the training parameters of the second recognition model.

[0013] On the other hand, an embodiment of the present disclosure also provides a multi-stream service identification device, comprising: one or more processors; a memory on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement the multi-stream service identification method as described above; at least one I / O interface, connected between the processor and the memory, configured to implement information interaction between the processor and the memory.

[0014] On the other hand, an embodiment of the present disclosure further provides a computer-readable medium having a computer program stored thereon, wherein when the program is executed, the multi-flow service identification method as described above is implemented.

[0015] The multi-stream service identification method provided by the embodiment of the present disclosure obtains the traffic characteristics of the data flow of the service type to be identified, uses the first identification model to preliminarily identify the service type of each data flow according to the traffic characteristics of the data flow of the service type to be identified, and for the target data flow of the first type or the second type in the preliminary identification result, determines whether the service type of each target data flow is the first type or the second type according to the flow number of the service to which each target data flow belongs, and can accurately identify the first type and the second type of services with similar traffic characteristics; the embodiment of the present disclosure identifies the service type based on the flow number of the data flow obtained by statistics, and the service type identification algorithm is simple, with high identification accuracy and efficiency, and is suitable for identifying services with partially identical data flow types; it can be implemented without the aid of professional equipment and mapping libraries, with low cost, and can be applied to WLAN scenarios with flexible layouts.

[0016] Another embodiment of the present disclosure provides a multi-stream service identification method, which obtains the traffic characteristics and transmission information of the data flow of the service type to be identified, determines the number of flows of the service to which each data flow belongs based on the transmission information of the data flow of the service type to be identified, inputs the number of flows of the service to which each data flow belongs and the traffic characteristics of each data flow into a second identification model, and obtains that the service type of each data flow is a preset first type or a preset second type; the second identification model of the embodiment of the present disclosure is trained based on the number of flows of the service to which the data flow belongs, and the second model identifies the service type based on the traffic characteristics of the data flow obtained by statistics, and can accurately identify the first type and the second type of services with similar traffic characteristics. The service type identification algorithm is simple, and the identification accuracy and efficiency are high, and it is suitable for identifying services with partially identical data flow types; it can be implemented without the aid of professional equipment and mapping libraries, is low-cost, and can be applied to WLAN scenarios with flexible layouts. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the system architecture provided by an embodiment of the present disclosure;

[0018] Figure 2 Schematic diagram of the multi-stream service identification method provided in the embodiment of the present disclosure Figure 1 ;

[0019] Figure 3 Schematic diagram of the multi-stream service identification method provided in the embodiment of the present disclosure Figure 2 ;

[0020] Figure 4 Schematic diagram of the multi-stream service identification method provided in the embodiment of the present disclosure Figure 3 ;

[0021] Figure 5 A schematic diagram of a process for determining the number of flows of a service to which each target data flow belongs, provided in an embodiment of the present disclosure;

[0022] Figure 6 A schematic diagram of a process for determining whether a first data stream and a second data stream are different data streams of the same user and the same service, provided in an embodiment of the present disclosure;

[0023] Figure 7 Schematic diagram of the multi-stream service identification method provided in the embodiment of the present disclosure Figure 4 ;

[0024] Figure 8 This is a structural diagram of a multi-stream service identification device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] Example embodiments will be described more fully hereinafter with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the scope of this disclosure to those skilled in the art.

[0026] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0027] The terms used herein are used only to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a," "an," and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements, and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or groups thereof is not excluded.

[0028] The embodiments described herein may be described with reference to plan views and / or cross-sectional views, with the aid of idealized schematic diagrams of the present disclosure. Thus, the example illustrations may be modified based on manufacturing techniques and / or tolerances. Therefore, the embodiments are not limited to the embodiments shown in the accompanying drawings, but include modifications of the configurations formed based on the manufacturing process. Therefore, the regions illustrated in the accompanying drawings are schematic in nature, and the shapes of the regions shown in the drawings illustrate specific shapes of the regions of the elements, but are not intended to be limiting.

[0029] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined as such herein.

[0030] In related technologies, the commonly used business scenario identification solutions are as follows:

[0031] (1) Message-based business scenario identification. In a proprietary network, traffic packets containing business scenario type information are sent between devices. Business scenario identification can be achieved by interpreting the content of these traffic packets.

[0032] (2) Business scenario identification based on access characteristics. For specific scenarios, such as web browsing or gaming, specific ports and IP addresses are accessed. Business scenario identification is achieved by establishing a one-to-one mapping library between access characteristics and business scenarios.

[0033] (3) Business scenario identification based on deep packet inspection technology. The specific feature fields carried in the traffic of each scenario are extracted, and then the content of the traffic is compared with the pre-established feature library to determine the business scenario of the current traffic.

[0034] (4) Business scenario recognition based on machine learning. Based on pre-acquired data, algorithms are used to extract deep features of the data and establish a classification model. This type of algorithm generally has the advantages of simple calculation and good performance.

[0035] (5) Multi-business scenario recognition based on pre-trained decision forests. The message data is statistically analyzed to obtain statistical features such as data flow volume. The model will train the decision forest model based on these features to identify business scenarios in the data flow.

[0036] Solutions (1) and (2) are simple to implement, but they rely heavily on the accuracy and timeliness of the mapping library. As the number of applications on the market increases, the mapping library itself will become too large, reducing the speed of feature matching. The deep packet inspection in solution (3) generally requires professional equipment connected in series in the network link, which is expensive to use and is not suitable for WLANs with flexible layouts. At the same time, WLAN network equipment does not have the computing power to independently implement deep packet inspection technology. For solution (4), the machine learning method currently used can only identify specific services, such as video services, and does not have the ability to continuously improve performance in actual applications. For solution (5), the algorithm that uses decision forests for service identification is difficult to further distinguish services with similar traffic characteristics. Therefore, relying solely on existing technologies is not enough to achieve scene identification of services with some of the same data stream types while balancing cost and performance.

[0037] For example, online conferencing and online video services have at least two data streams: a video stream and a voice stream. Online video services only have a single video stream. Traffic-based service identification algorithms have very similar data stream characteristics for online video and online conferencing, making it difficult to distinguish them.

[0038] To solve the above problems, an embodiment of the present disclosure provides a multi-flow service identification method. Figure 1 This is a schematic diagram of the system architecture provided by the embodiment of the present disclosure. Figure 1As shown, the multi-stream service identification method is applied to a multi-stream service identification device, which includes but is not limited to an AP (Access Point) device. In the embodiments disclosed herein, the AP device is taken as an example for explanation. The AP device includes a radio frequency module (not shown in the figure), a physical link layer, and a media access control layer. The radio frequency module and the physical link layer are used to demodulate the received wireless signal and send it to the media access control layer. The media access control layer is used to extract message data. By analyzing the destination IP (Internet Protocol Address) and destination port information of the message, the AP device tracks the data flow of the message. The statistical features of the data flow are sent to the service identification module, which identifies the service type of the current data flow through calculation, so that the AP device can provide different wireless service parameters for the data flow according to the service type to improve the user experience.

[0039] like Figure 2 As shown, the multi-stream service identification method includes the following steps:

[0040] Step S11: Obtain the traffic characteristics of the data flow of the service type to be identified.

[0041] In this step, for each data flow of the service type to be identified, the traffic characteristics of the data flow are extracted. In the embodiment of the present disclosure, the traffic characteristics include but are not limited to the total number of uplink / downlink messages within a preset time period, the average exchange time of the messages, etc.

[0042] Step S12: using the first recognition model, identifying the service type of each data flow according to the traffic characteristics, and obtaining a preliminary recognition result.

[0043] This step is a preliminary identification step for the business type. In this step, the traffic characteristics of the data flow of the business type to be identified are input into the business identification module (i.e., the first identification model). The first identification model can preliminarily identify the business types as the first type and the second type of data flow based on the traffic characteristics. In the embodiment of the present disclosure, the first type of business can be an online conference business, and the second type of business can be an online video business, for example, a user watches a video online through a video website. The first identification model can be a decision forest model or a neural network model. When performing the preliminary identification of the business type, the first identification model does not need to carry the flow quantity information of the data flow.

[0044] Step S13, for the target data stream whose preliminary identification result is the preset first type or the preset second type, determine the business type of each target data stream as the first type or the second type according to the number of streams of the business to which each target data stream belongs; wherein the number of streams of the business to which each target data stream belongs is determined based on the transmission information of each target data stream.

[0045] Since the accuracy of the preliminary identification results is not high, based on the preliminary identification results, for the target data stream that needs to accurately distinguish the business type, the number of streams of the business to which the target data stream belongs is further used to accurately determine whether the business type of the target data stream is the first type (i.e., online meeting) or the second type (i.e., online video).

[0046] The number of target data flows can be determined based on the transmission information of the target data flow and other target data flows, and finally the number of flows of the services to which all target data flows belong is counted. The step of counting the number of flows of the services to which all target data flows belong can be performed directly after the step of preliminary identification of service types (i.e., step S12). The specific implementation process is subsequently combined with Figure 3 Further detailed description is given, in this way, the number of flows of the business to which the target data flow belongs can be directly counted. The step of counting the number of flows of the business to which the target data flow belongs can also be performed before performing the preliminary identification of the business type (i.e., step S12), obtaining the transmission information of all data flows, and after performing the preliminary identification of the business type (i.e., step S12) to determine the target data flow, screening out the transmission information of the target data flow, and obtaining the number of flows of the business to which the target data flow belongs based on the transmission information of the target data flow. The specific implementation process will be combined later Figure 4 Please explain in more detail.

[0047] The multi-stream service identification method provided by the embodiment of the present disclosure obtains the traffic characteristics of the data flow of the service type to be identified, uses the first identification model to preliminarily identify the service type of each data flow according to the traffic characteristics of the data flow of the service type to be identified, and for the target data flow of the first type or the second type in the preliminary identification result, determines whether the service type of each target data flow is the first type or the second type according to the flow number of the service to which each target data flow belongs, and can accurately identify the first type and the second type of services with similar traffic characteristics; the embodiment of the present disclosure identifies the service type based on the flow number of the data flow obtained by statistics, and the service type identification algorithm is simple, with high identification accuracy and efficiency, and is suitable for identifying services with partially identical data flow types; it can be implemented without the aid of professional equipment and mapping libraries, with low cost, and can be applied to WLAN scenarios with flexible layouts.

[0048] In some embodiments, as Figure 3 As shown, after using the first identification model to identify the service type of each data flow according to the traffic characteristics and obtaining a preliminary identification result (i.e., step S12), the multi-flow service identification method further includes the following steps:

[0049] Step S21: Acquire transmission information of each target data flow, and determine the number of flows of the service to which each target data flow belongs according to the transmission information of each target data flow.

[0050] After executing step S12, a preliminary identification result is obtained, which includes target data flows whose business types are the first business type or the second business type. For these target data flows, their transmission information is obtained, and the number of flows of the business to which each target data flow belongs is determined based on the transmission information.

[0051] It should be noted that, in this embodiment, Figure 3 As shown, steps S11 and S13 still need to be executed, that is, after executing steps S11 and S12, steps S21 and S13 are executed in sequence, wherein the specific implementation methods of steps S11, S12, and S13 are as described above and will not be repeated here.

[0052] After performing preliminary identification of the service type, the number of flows of the service to which the target data flow belongs is directly determined. The range of the acquired transmission information is narrow, that is, only the transmission information of the target data flow is acquired, which can improve the efficiency of service type identification.

[0053] In some embodiments, as Figure 4 As shown, before using the first identification model to identify the service type of each data flow based on traffic characteristics and obtaining a preliminary identification result (i.e., step S12), the multi-flow service identification method may further include the following steps: Step S31, obtaining transmission information of the data flow of the service type to be identified. It should be noted that the execution order of step S31 and step S11 is not limited and can also be performed simultaneously.

[0054] After the first identification model is used to identify the service type of each data flow according to the traffic characteristics and a preliminary identification result is obtained (i.e., step S12), the multi-flow service identification method may further include the following steps:

[0055] Step S32: Filter the transmission information of each target data flow from the transmission information of the data flow of the service type to be identified.

[0056] In this step, the transmission information of data flows of various service types is screened according to the service type to obtain the transmission information of data flows of the first or second service type, ie, the transmission information of the target data flow.

[0057] Step S33: Determine the number of flows of the service to which each target data flow belongs based on the transmission information of each target data flow.

[0058] After executing step S33, step S13 needs to be executed to accurately distinguish whether the service type of each target data flow is the first type or the second type according to the number of flows of the service to which each target data flow belongs. The specific implementation of steps S11, S12, and S13 is as described above and will not be repeated here.

[0059] The following combination Figure 5 , the process of determining the number of flows of the business to which each target data flow belongs is described in detail. In some embodiments, Figure 5 As shown, the step of determining the number of flows of the service to which each target data flow belongs includes:

[0060] Step S41, for each first data stream in the target data stream, based on the transmission information of the first data stream and the transmission information of the second data stream, determine whether the first data stream and the second data stream are different data streams of the same user and the same service, so as to obtain the judgment result of each target data stream; wherein the second data stream is any other data stream in the target data stream except the first data stream.

[0061] Assume there are five target data streams, A, B, C, D, and E. Take any one of the target data streams (e.g., data stream A) as the current data stream and perform a judgment with any other target data stream (e.g., data stream B). Based on the transmission information of data streams A and B, determine whether they are different data streams for the same user and the same service, and obtain the judgment results for data stream A and data stream B. By traversing all target data streams and taking them as any other target data stream, the judgment results for the current target data stream and any other target data stream can be obtained. This completes a round of judgments and yields the judgment results corresponding to the current target data stream, namely, the judgment results for data streams A and B, A and C, A and D, and A and E. Taking data stream B as the current data stream, the judgment results for data stream B and A, B and C, B and D, and B and E can be obtained. Similarly, the five target data streams are respectively used as current data streams to obtain multiple rounds of judgment results corresponding to the five target data streams, which include the judgment results of any two target data streams.

[0062] Step S42: Determine the number of flows of the service to which each target data flow belongs based on the judgment result of each target data flow.

[0063] In this step, statistics are collected on the judgment results of any two target data flows, and the number of target data flows belonging to the same service is accumulated to obtain the number of flows of each service.

[0064] In some embodiments, the transmission information may include IP address information and port number information. Figure 6 , describes the process of determining whether any two target data flows are different data flows of the same user and the same service.

[0065] like Figure 6As shown, the step of determining whether the first data stream and the second data stream are different data streams of the same user and the same service based on the transmission information of the first data stream and the transmission information of the second data stream (i.e., step S41) includes the following steps:

[0066] Step S51, determine whether the IP address information of the first data stream and the second data stream meets the preset first address judgment condition or the preset second address judgment condition. If the first address judgment condition or the second address judgment condition is met, execute step S52; if the first address judgment condition and the second address judgment condition are not met, execute step S54.

[0067] If the IP address information of the first data stream and the second data stream meets the first address judgment condition, or the IP address information of the first data stream and the second data stream meets the second address judgment condition, it means that the first data stream and the second data stream are data streams of the same service of the same user. At this time, it can be determined that the first data stream and the second data stream belong to the same service, but it is not certain whether the first data stream and the second data stream are the same data stream. Further judgment is required through the port number, so step S52 is executed.

[0068] If the IP address information of the first data stream and the second data stream does not meet the first address judgment condition, and the IP address information of the first data stream and the second data stream does not meet the second address judgment condition, it means that the first data stream and the second data stream are not data streams of the same service of the same user, that is, the first data stream and the second data stream belong to different services. In this case, there is no need to further judge the port number, and the next second data stream can be replaced to continue the next round of judgment with the first data stream (that is, the current data stream), that is, execute step S54, and this round of judgment ends.

[0069] In some embodiments, the preset first address judgment condition may include: the source IP address of the first data flow is the same as the source IP address of the second data flow, and the destination IP address of the first data flow is the same as the destination IP address of the second data flow.

[0070] In some embodiments, the preset second address determination condition may include: the source IP address of the first data stream is the same as the destination IP address of the second data stream, and the destination IP address of the first data stream is the same as the source IP address of the second data stream.

[0071] Step S52, when the IP address information of the first data stream and the second data stream meets the first address judgment condition, judge whether the port information of the first data stream and the second data stream does not meet the preset first port judgment condition; or, when the IP address information of the first data stream and the second data stream meets the second address judgment condition, judge whether the port information of the first data stream and the second data stream does not meet the preset second port judgment condition; if the first port judgment condition or the second port judgment condition is not met, execute step S53; if the first port judgment condition or the second port judgment condition is met, execute step S54.

[0072] If the IP address information of the first data stream and the second data stream meets the preset first address judgment condition and does not meet the preset first port judgment condition, or if the IP address information of the first data stream and the second data stream meets the preset second address judgment condition and does not meet the preset second port judgment condition, it means that the first data stream and the second data stream are data streams of the same service of the same user, that is, the first data stream and the second data stream belong to the same service, but the first data stream and the second data stream are not the same data stream, then step S53 is executed;

[0073] If the IP address information of the first data stream and the second data stream meets the preset first address judgment condition and the port number information of the first data stream and the second data stream meets the preset first port judgment condition, or the IP address information of the first data stream and the second data stream meets the preset second address judgment condition and the port number information of the first data stream and the second data stream meets the preset second port judgment condition, it means that the first data stream and the second data stream are data streams of the same service of the same user, that is, the first data stream and the second data stream belong to the same service, and the first data stream and the second data stream are the same data stream. In this case, it is impossible to determine whether there are other data streams for the service to which they belong based solely on the first data stream and the second data stream. Therefore, the next second data stream can be replaced with the first data stream (that is, the current data stream) for the next round of judgment, that is, step S54 is executed, and this round of judgment ends.

[0074] In some embodiments, the preset first port determination condition includes: the source port number of the first data flow is the same as the source port number of the second data flow, and the destination port number of the first data flow is the same as the destination port number of the second data flow.

[0075] In some embodiments, the preset second port determination condition includes: the source port number of the first data flow is the same as the destination port number of the second data flow, and the destination port number of the first data flow is the same as the source port number of the second data flow.

[0076] It should be noted that in the embodiment of the present disclosure, if the source IP address and destination IP address, source port number and destination port number of the first data stream and the second data stream are exchanged at the same time, in this case, the first data stream and the second data stream are also considered to be the same data stream.

[0077] Step S53: Determine that the first data flow and the second data flow are different data flows of the same user and the same service.

[0078] Step S54, this round of judgment ends, and based on the transmission information of the first data stream and the transmission information of the other second data stream, it is judged whether the first data stream and the other second data stream are different data streams of the same user and the same service.

[0079] In this step, it is determined whether the IP address information of the first data stream and the other second data stream meets the preset first address determination condition or the preset second address determination condition, thereby performing the next round of determination.

[0080] Steps S51-S54 constitute a round of judgment between any two target data streams. Any two of the target data streams are combined and steps S51-S54 are executed to complete a round of judgment. After all target data streams are judged as the current data stream, the results of each round of judgment are obtained. Based on the services to which each target data stream belongs, the results of each round of judgment are summarized to obtain the number of flows for each service to which each target data stream belongs.

[0081] In some embodiments, determining whether the service type of each target data stream is the first type or the second type (i.e., step S13) is determined based on the number of streams of the service to which each target data stream belongs, including: when the number of streams of the same user of the same service to which the target data stream belongs is at least two, determining that the service type of the target data stream under the service is the first type, and the first type is online conferencing; when the number of streams of the same user of the same service to which the target data stream belongs is one, determining that the service type of the target data stream under the service is the second type, and the second type is online video. That is, after obtaining the number of streams of the service to which each target data stream belongs, the service type of each target data stream can be determined based on the number of streams of the service to which each target data stream belongs. If the number of streams of a data stream of a service is greater than 1, then the service is considered to be an online conferencing service, and the service type of the corresponding data stream is online conferencing; if the number of streams of a data stream of a service is 1, then the service is considered to be an online video service, and the service type of the corresponding data stream is online video.

[0082] The embodiment of the present disclosure also provides a multi-stream service identification method, such as Figure 7 As shown, the multi-stream service identification method includes the following steps:

[0083] Step S61: Acquire the traffic characteristics and transmission information of the data flow of the service type to be identified.

[0084] Traffic characteristics include, but are not limited to, the total number of uplink / downlink messages within a preset time period, the average message exchange time, etc. Transmission information includes IP address information and port number information.

[0085] Step S62: Determine the number of flows of the service to which each data flow belongs based on the transmission information of each data flow.

[0086] The process of determining the number of flows of the service to which each data flow belongs according to the transmission information of each data flow may follow steps S41 - S42 and steps S51 - S54 .

[0087] In step S63, the traffic characteristics and the number of flows of the business to which each data flow belongs are input into the second recognition model, so as to use the second recognition model to identify the business type of each data flow as the preset first type or the preset second type according to the traffic characteristics, wherein the number of flows of the business to which the data flow belongs is one of the training parameters of the second recognition model.

[0088] In the embodiment of the present disclosure, the second recognition model is a decision forest model, and the number of flows of the business to which each data flow belongs is input into the second recognition model as one of the characteristics of the data flow to identify the business type.

[0089] Before training a decision forest model, packet data flows must be statistically analyzed to obtain their statistical characteristics, namely, the number of flows. The model then uses these characteristics to identify the service scenarios for the data flows. Depending on the type of pre-trained decision forest model, any statistical characteristics of the data flow, such as labels and packet payloads, can be used as input data. However, each data flow does not contain information about the number of flows belonging to the same user and service at that time. When training the decision forest model, the number of flows for the service to which the data flow belongs must be used as one of the statistical characteristics of the data flow.

[0090] It should be noted that the second recognition model can also be a neural network model. However, the neural network model is more difficult to train and has higher computational complexity than the decision forest model, so it is not suitable for WLAN access point devices.

[0091] The multi-stream service identification method provided by the embodiment of the present disclosure obtains the traffic characteristics and transmission information of the data flow of the service type to be identified, determines the flow number of the service to which each data flow belongs based on the transmission information of the data flow of the service type to be identified, inputs the flow number of the service to which each data flow belongs and the traffic characteristics of each data flow into the second identification model, and obtains that the service type of each data flow is a preset first type or a preset second type; the second identification model of the embodiment of the present disclosure is trained based on the flow number of the service to which the data flow belongs, and the second model identifies the service type based on the traffic characteristics of the data flow obtained by statistics, and can accurately identify the first type and the second type of services with similar traffic characteristics. The service type identification algorithm is simple, and the identification accuracy and efficiency are high, and it is suitable for identifying services with partially identical data flow types; it can be implemented without the aid of professional equipment and mapping libraries, is low-cost, and can be applied to WLAN scenarios with flexible layouts.

[0092] To address the problem that the traffic characteristics of individual data streams for online video and online conferencing services are highly similar, making it difficult to distinguish between service types, the disclosed embodiments propose a multi-stream recognition method for identifying video and conferencing service scenarios. Compared to existing solutions that simply identify service types based on data traffic characteristics, this method can more accurately distinguish between online conferencing and video services with almost no increase in computational complexity, thereby effectively improving the user experience.

[0093] The embodiments of the present disclosure are applicable to business scenarios in the WLAN field and are applied to any WLAN access point device.

[0094] Online video and online conferencing are two different service types, each with its own characteristics and requirements. For example, online conferencing, as a real-time service, is more sensitive to latency jitter; online video, with its buffering and other mechanisms to combat latency, requires greater downlink bandwidth for high-definition video. The disclosed embodiments can accurately distinguish between online conferencing and online video services, allowing access point devices to employ different parameter configuration strategies to optimize access performance for each service.

[0095] like Figure 8 As shown, the embodiment of the present disclosure further provides a multi-stream service identification device, the multi-stream service identification device comprising:

[0096] at least one processor 801;

[0097] a memory 802 storing at least one program, which, when executed by the at least one processor, enables the at least one processor to implement the multi-flow service identification method described above;

[0098] At least one I / O interface 803 is connected between the processor and the memory and is configured to implement information exchange between the processor and the memory.

[0099] Among them, the processor 801 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 802 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically such as SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) 803 is connected between the processor 801 and the memory 802, and can realize information exchange between the processor 801 and the memory 802, including but not limited to a data bus (Bus), etc.

[0100] In some embodiments, the processor 801 , the memory 802 , and the I / O interface 803 are connected to each other via a bus, and further connected to other components of the computing device.

[0101] The embodiments of the present disclosure further provide a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed, the multi-flow service identification method provided in the aforementioned embodiments is implemented.

[0102] It will be appreciated by those skilled in the art that all or some of the steps in the method disclosed above, and the functional modules / units in the device can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0103] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for purposes of limitation. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly indicated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A multi-stream service identification method, characterized in that: include: Obtaining traffic characteristics of the data flow of the service type to be identified; Using the first identification model, identifying the service type of each of the data flows according to the traffic characteristics, and obtaining a preliminary identification result; For the target data stream whose preliminary identification result is a preset first type or a preset second type, the business type of each target data stream is determined to be the first type or the second type according to the number of streams of the business to which each target data stream belongs; wherein the number of streams of the business to which each target data stream belongs is determined based on the transmission information of each target data stream.

2. The method according to claim 1, wherein After using the first identification model to identify the business type of each data flow according to the traffic characteristics and obtaining a preliminary identification result, the transmission information of each target data flow is obtained, and the number of flows of the business to which each target data flow belongs is determined based on the transmission information of each target data flow.

3. The method according to claim 1, wherein Before using the first identification model to identify the service type of each of the data flows according to the traffic characteristics and obtaining a preliminary identification result, the method further includes: obtaining transmission information of the data flow of the service type to be identified; After using the first identification model to identify the service type of each of the data flows according to the traffic characteristics and obtaining a preliminary identification result, the method further includes: Filtering the transmission information of each target data flow from the transmission information of the data flow of the service type to be identified; The number of flows of the service to which each target data flow belongs is determined according to the transmission information of each target data flow.

4. The method according to claim 1, wherein The step of determining the number of flows of the service to which each target data flow belongs includes: For each first data stream in the target data stream, determining, based on transmission information of the first data stream and transmission information of the second data stream, whether the first data stream and the second data stream are different data streams of the same user and the same service, to obtain a determination result for each target data stream; wherein the second data stream is any data stream other than the first data stream in the target data stream; According to the judgment result of each target data flow, the number of flows of the service to which each target data flow belongs is determined.

5. The method according to claim 4, wherein The transmission information includes IP address information and port number information, and the determining, based on the transmission information of the first data stream and the transmission information of the second data stream, whether the first data stream and the second data stream are different data streams of the same user and the same service includes: When the IP address information of the first data stream and the second data stream meets the preset first address judgment condition and does not meet the preset first port judgment condition, or when the IP address information of the first data stream and the second data stream meets the preset second address judgment condition and does not meet the preset second port judgment condition, it is determined that the first data stream and the second data stream are different data streams of the same user and the same service.

6. The method according to claim 4, wherein The transmission information includes IP address information and port number information, and the determining, based on the transmission information of the first data stream and the transmission information of the second data stream, whether the first data stream and the second data stream are different data streams of the same user and the same service includes: If the IP address information of the first data stream and the second data stream does not satisfy a preset first address judgment condition and does not satisfy a preset second address judgment condition, determining, based on the transmission information of the first data stream and the transmission information of another second data stream, whether the first data stream and the another second data stream are different data streams of the same user and the same service; or When the IP address information of the first data stream and the second data stream meets the preset first address judgment condition and the port number information of the first data stream and the second data stream meets the preset first port judgment condition, or when the IP address information of the first data stream and the second data stream meets the preset second address judgment condition and the port number information of the first data stream and the second data stream meets the preset second port judgment condition, determine whether the first data stream and the other second data stream are different data streams of the same user and the same service based on the transmission information of the first data stream and the transmission information of the other second data stream.

7. The method according to claim 5 or 6, characterized in that The preset first address judgment condition includes: the source IP address of the first data flow is the same as the source IP address of the second data flow, and the destination IP address of the first data flow is the same as the destination IP address of the second data flow; The preset first port judgment condition includes: the source port number of the first data flow is the same as the source port number of the second data flow, and the destination port number of the first data flow is the same as the destination port number of the second data flow.

8. The method according to claim 5 or 6, wherein: The preset second address judgment condition includes: the source IP address of the first data flow is the same as the destination IP address of the second data flow, and the destination IP address of the first data flow is the same as the source IP address of the second data flow; The preset second port judgment condition includes: the source port number of the first data flow is the same as the destination port number of the second data flow, and the destination port number of the first data flow is the same as the source port number of the second data flow.

9. The method according to claim 1, wherein The determining, according to the number of flows of the services to which each of the target data flows belongs, whether the service type of each of the target data flows is the first type or the second type, includes: When the number of flows of the same user belonging to the same service of the target data flow is at least two, determining that the service type of the target data flow under the service is the first type, and the first type is an online conference; When the number of flows of the same user of the same service to which the target data flow belongs is one, it is determined that the service type of the target data flow under the service is the second type, and the second type is online video.

10. A multi-stream service identification method, characterized in that: include: Obtaining traffic characteristics and transmission information of the data flow of the service type to be identified; Determining the number of flows of the service to which each of the data flows belongs according to the transmission information of each of the data flows; The traffic characteristics and the number of flows of the business to which each of the data flows belongs are input into a second recognition model, so that the second recognition model can be used to identify the business type of each of the data flows as a preset first type or a preset second type based on the traffic characteristics, wherein the number of flows of the business to which the data flows belong is one of the training parameters of the second recognition model.

11. A multi-stream service identification device, wherein: include: one or more processors; a memory having one or more programs stored therein; When the one or more programs are executed by the one or more processors, the one or more processors implement the multi-flow service identification method according to any one of claims 1 to 9, or the multi-flow service identification method according to claim 10; At least one I / O interface is connected between the processor and the memory and is configured to implement information interaction between the processor and the memory.

12. A computer-readable medium having a computer program stored thereon, wherein: When the program is executed, the multi-stream service identification method according to any one of claims 1 to 9 or the multi-stream service identification method according to claim 10 is implemented.

Images