Communication method and apparatus
By acquiring session traffic characteristic information and using AI to identify abnormal traffic, and dynamically updating network policies, the limitations of abnormal traffic identification in mobile networks are solved, thereby improving network stability and defense capabilities.
Patent Information
- Application Number
- PCT/CN2026/094261
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-08
- Filing Date
- 2026-04-30
- Publication Date
- 2026-11-12
AI Technical Summary
Existing methods for identifying abnormal traffic in mobile networks have limitations. Attackers can bypass firewalls by hiding traffic characteristics, resulting in insufficient network stability and defense capabilities.
By acquiring the traffic characteristics of a session, AI functions are used to identify abnormal traffic, and network policies are updated based on the identification results, including disabling traffic, implementing packet loss, and using specific QoS flows to dynamically handle abnormal traffic.
It improves the mobile network's ability to identify and defend against abnormal traffic and network attacks, reduces resource waste, and achieves more efficient network security performance.
Smart Images

Figure CN2026094261_12112026_PF_FP_ABST
Abstract
Description
A communication method and apparatus
[0001] Cross-references to related applications
[0002] This application claims priority to Chinese Patent Application No. 202510592129.4, filed on May 8, 2025, entitled "A Communication Method and Apparatus", the entire contents of which are incorporated herein by reference. Technical Field
[0003] This application relates to the field of mobile communications, and more particularly to a communication method and apparatus. Background Technology
[0004] Firewalls are typically used in mobile networks to deal with abnormal traffic attacks. The mechanism mainly involves configuring relatively static blacklists and other information. However, this method has obvious limitations: attackers can bypass the blacklist by changing the Internet Protocol (IP) and domain name characteristics of abnormal traffic through methods such as hiding traffic through legitimate services.
[0005] There is still a lack of clear solutions on how to identify abnormal traffic in order to improve the stability of mobile networks and their ability to defend against attacks. Summary of the Invention
[0006] This application provides a communication method and apparatus that can identify abnormal traffic, improve the stability of mobile networks, and enhance their ability to defend against attacks.
[0007] Firstly, a communication method is provided, which can be executed by a first network element or a chip or module within the first network element. The method includes: acquiring traffic characteristic information of a first session; and sending identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session.
[0008] This application provides a session-based (or session-granular) abnormal traffic identification scheme, namely: a first network element acquires traffic characteristic information of a first session and sends identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session. This scheme clarifies the specific method for mobile networks to identify abnormal traffic, and can be used to standardize the process of abnormal traffic analysis in mobile networks, thereby improving the ability of mobile networks to handle abnormal traffic and network attacks.
[0009] In one possible design, the first network element is used to provide network analysis functions, for example, the first network element is a network data analytics function (NWDAF) network element.
[0010] In this way, the first network element can provide the abnormal traffic identification function, and other network elements can use the abnormal traffic identification function provided by the first network element.
[0011] In one possible design, the method further includes: receiving a first request from a second network element, the first request being used to obtain identification information of abnormal traffic in a first session, the first request including identification information of the first session and an analysis identifier, the analysis identifier being used to indicate abnormal traffic analysis. Optionally, the first request may be a subscription message.
[0012] In this way, abnormal traffic analysis based on requests from the second network element can be performed on demand, thereby improving network security performance and avoiding the problem of resource waste caused by frequent triggering of abnormal traffic analysis.
[0013] In one possible design, the second network element is used to provide policy control functions, for example, the second network element is a policy control function (PCF) network element.
[0014] In this way, abnormal traffic analysis of second network element requests can be achieved.
[0015] In one possible design, the traffic characteristic information of the first session includes one or more of the following: protocol fingerprint information, service behavior information, service server information, packet protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and traffic quality of service (QoS) related information (such as latency, bit rate, packet loss rate, periodicity, burst information, traffic pattern, or congestion).
[0016] Of course, the above are just examples and are not limited to these.
[0017] This provides a variety of traffic characteristics information for the first session, improving the flexibility of the solution.
[0018] In one possible design, at least one of the traffic characteristics of the first session satisfies a preset condition.
[0019] Therefore, analyzing abnormal traffic based on the traffic characteristics of the first session that meets the preset conditions can improve accuracy.
[0020] In one possible design, the method may further include: acquiring historical traffic characteristic information. Correspondingly, based on the traffic characteristic information of the first session and the historical traffic characteristic information, identification information of abnormal traffic in the first session may also be sent.
[0021] Therefore, using historical traffic characteristic information to help determine the identification information of abnormal traffic in the first session can further improve the accuracy of abnormal traffic identification.
[0022] In one possible design, obtaining traffic characteristic information of the first session may include sending a second request to obtain the traffic characteristic information of the first session. Optionally, the second request may be a message subscription.
[0023] In this way, by sending a second request to obtain the traffic characteristic information of the first session, support is provided for abnormal traffic analysis based on the traffic characteristic information of the first session, thus improving the reliability of the solution.
[0024] In one possible design, the second request includes identification information from the first session;
[0025] This allows the recipient of the second request to clearly identify which session's traffic characteristics information to return, thus improving the reliability of the solution.
[0026] In one possible design, the identification information of the first session includes one or more of the following: the identifier of the first session, the identification information of the terminal device, and the attribute information of the first session; wherein, the first session is the session of the terminal device.
[0027] Of course, the above examples are just examples and not limitations.
[0028] This provides multiple identification information for the first session, improving the flexibility of the solution.
[0029] In one possible design, sending the second request includes sending the second request to at least one of a third network element, a fourth network element, or a fifth network element; wherein the third network element is used to provide user plane functions (e.g., the third network element is a user plane function (UPF) network element), the fourth network element is used to provide operation and maintenance management functions (e.g., the fourth network element is an operation administration and maintenance (OAM) network element), and the fifth network element is used to provide application functions (e.g., the fifth network element is an application function (AF) network element).
[0030] This provides multiple possible ways to obtain traffic characteristic information of the first session, improving the reliability and flexibility of the solution.
[0031] In one possible design, sending identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session may include: determining the identification information of abnormal traffic in the first session based on reasoning or analysis according to the traffic characteristic information of the first session.
[0032] In this way, a specific solution for improving network identification of abnormal traffic through artificial intelligence (AI) functions can be clearly defined. This solution can be used to standardize the process of NWDAF assisting UPF in abnormal traffic analysis, and further improve the network's ability to handle abnormal traffic and network attacks.
[0033] In one possible design, at least one of the following can also be sent based on the traffic characteristics of the first session:
[0034] First-time information includes one or more of the start time, end time, or duration of abnormal traffic in the first session;
[0035] The source of abnormal traffic in the first session;
[0036] Confidence level of the abnormal traffic analysis results in the first session.
[0037] In some embodiments, the first time information may also be an indication of the applicable time period.
[0038] This design enables the transmission of more information about abnormal traffic based on the traffic characteristics of the first session, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0039] In one possible design, the identification information of abnormal traffic in the first session is used to update the policy for the first session. Policies include, but are not limited to, policy and charging control (PCC) rules and N4 rules. N4 rules include, but are not limited to, packet detection rules (PDR), forwarding action rules (FAR), QoS enforcement rules (QER), and usage reporting rules (URR).
[0040] In this way, the identification information of abnormal traffic in the first session is used to update the policy of the first session, which can achieve the effect of handling abnormal traffic in advance and dynamically, and improve the network's ability to handle abnormal traffic and network attacks.
[0041] Secondly, a communication method is provided, which can be executed by a second network element or a chip or module within the second network element. The method includes: receiving identification information of abnormal traffic in a first session; and updating the policy of the first session based on the identification information of the abnormal traffic in the first session.
[0042] The embodiments of this application can realize the strategy of updating the identification information of abnormal traffic in the first session, thereby achieving the effect of handling abnormal traffic and improving the network's ability to handle abnormal traffic and network attacks.
[0043] In one possible design, the second network element is used to provide policy control functions; for example, the second network element is a PCF network element.
[0044] This clarifies the network element responsible for handling abnormal traffic.
[0045] In one possible design, a first request can also be sent. The first request is used to obtain the identification information of the abnormal traffic of the first session from the first network element. The first request includes the identification information of the first session and the analysis identifier, which is used to indicate abnormal traffic analysis.
[0046] In this way, the first network element can be triggered to provide abnormal traffic analysis function, thereby obtaining the identification information of abnormal traffic in the first session.
[0047] In one possible design, the first network element is used to provide network data analysis functions; for example, the first network element is an NWDAF network element.
[0048] Thus, the network element that provides abnormal traffic identification function is clearly defined.
[0049] In one possible design, sending the first request may include: sending the first request based on the attribute information of the first session and / or a third request; wherein the third request is used to indicate that the first session requests / needs identification information of abnormal traffic.
[0050] In this way, clarifying the conditions, timing, or scenario for triggering the first request helps to trigger abnormal traffic analysis on demand.
[0051] In one possible design, at least one of the following can also be accepted:
[0052] First-time information includes one or more of the start time, end time, or duration of abnormal traffic in the first session;
[0053] The source of abnormal traffic in the first session;
[0054] Confidence level of the abnormal traffic analysis results in the first session.
[0055] This design can obtain more information about abnormal traffic, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0056] In one possible design, updating the policy of the first session based on the identification information of the abnormal traffic of the first session includes: sending a first rule based on the identification information of the abnormal traffic of the first session; the first rule includes the identification information of the abnormal traffic of the first session and first information, the first information being used to indicate at least one of the following:
[0057] Disable abnormal traffic;
[0058] Packet loss is applied to abnormal traffic;
[0059] Use the first QoS parameter for abnormal traffic;
[0060] Use a separate QoS stream for abnormal traffic;
[0061] Implement caching for abnormal traffic.
[0062] The above design clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from the server. It can achieve the effect of handling abnormal traffic in advance and dynamically, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0063] In one possible design, the first rule may further include second time information, which indicates the time at which at least one action is taken on abnormal traffic. Optionally, the second time information may be associated with the first time information.
[0064] This clarifies the timing for handling abnormal traffic and improves the reliability of the solution.
[0065] In one possible design, the policy for the first session is updated based on the identification information of the abnormal traffic in the first session, including:
[0066] Release the first session; or,
[0067] Send a second message, which indicates the release of the first session;
[0068] Send a third message, which is used to instruct the terminal device corresponding to the first session to prohibit it from establishing a session within the first time period.
[0069] The above design clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from terminal devices. It can achieve the effect of handling abnormal traffic in advance and dynamically, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0070] Thirdly, a communication method is provided, which can be executed by a sixth network element or a chip or module within the sixth network element. The method includes:
[0071] Receive a first rule, which includes identification information of abnormal traffic in the first session and first information, the first information indicating at least one of the following: disable abnormal traffic; perform packet loss on abnormal traffic; use a first 5QI on abnormal traffic; use a separate QoS stream on abnormal traffic; send a second rule according to the first rule, the second rule including identification information of abnormal traffic in the first session and fourth information, the fourth information indicating at least one of the following: disable abnormal traffic; perform packet loss on abnormal traffic; use a first 5QI on abnormal traffic; use a separate QoS stream on abnormal traffic; or,
[0072] Receive the second message; trigger the release of the first session based on the second message.
[0073] In one possible design, the sixth network element is used to provide session control functions, such as a session management function (SMF) network element.
[0074] In one possible design, the first rule also includes second time information, which indicates the time at which at least one action is taken on abnormal traffic.
[0075] The beneficial effects of each design in the third aspect are the same as those of the corresponding designs in the first and second aspects, and will not be repeated here.
[0076] Fourthly, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes: receiving a second request for obtaining traffic characteristic information of a first session; and sending the traffic characteristic information of the first session according to the second request.
[0077] In one possible design, the third network element is used to provide user plane functions; for example, the third network element is a UPF network element.
[0078] In one possible design, the second request includes identification information of the first session; the identification information of the first session includes one or more of the following: the identifier of the first session, the identification information of the terminal device, and the attribute information of the first session; wherein, the first session is the session of the terminal device.
[0079] In one possible design, the traffic characteristic information of the first session includes one or more of the following: protocol fingerprint information, service behavior information, service server information, packet protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and QoS-related information of the traffic.
[0080] In one possible design, at least one of the traffic characteristics of the first session satisfies a preset condition.
[0081] In one possible design, the second request is also used to obtain historical traffic characteristic information; the method further includes: sending historical traffic characteristic information according to the second request.
[0082] One possible design also includes:
[0083] Receive a second rule, which includes identification information of abnormal traffic in the first session and a fourth piece of information. The fourth piece of information is used to indicate at least one of the following: disable abnormal traffic; perform packet loss on abnormal traffic; use a first QoS parameter on abnormal traffic; use a separate QoS stream on abnormal traffic; or perform caching on abnormal traffic.
[0084] Abnormal traffic from the first session will be handled according to the second rule.
[0085] The beneficial effects of each design in the fourth aspect are the same as those of the corresponding designs in the first and second aspects, and will not be repeated here.
[0086] Fifthly, a communication method is provided, which can be executed by a first network element or a chip or module within the first network element. The method includes: acquiring traffic characteristic information of a third network element; and sending identification information of abnormal traffic of the third network element based on the traffic characteristic information of the third network element.
[0087] This application provides an abnormal traffic identification scheme based on network elements (or network element granularity). Specifically, a first network element acquires traffic characteristic information from a third network element and sends identification information of the abnormal traffic of the third network element based on this information. This scheme clarifies the specific method for identifying abnormal traffic in mobile networks, and can be used to standardize the process of abnormal traffic analysis in mobile networks, thereby improving the ability of mobile networks to handle abnormal traffic and network attacks.
[0088] In one possible design, the first network element is used to provide network analysis functions; for example, the first network element is an NWDAF network element.
[0089] In this way, the first network element can provide the abnormal traffic identification function, and other network elements can use the abnormal traffic identification function provided by the first network element.
[0090] In one possible design, the third network element is used to provide user plane functions; for example, the third network element is a UPF network element.
[0091] In this way, abnormal traffic analysis of third-party network element requests can be achieved.
[0092] One possible design also includes: receiving a fourth request from a third network element, the fourth request being used to obtain identification information of abnormal traffic from the third network element, the fourth request including an analysis identifier, the analysis identifier being used to indicate abnormal traffic analysis.
[0093] In this way, abnormal traffic analysis based on requests from third-party network elements can be performed on demand, thereby improving network security performance and avoiding the problem of resource waste caused by frequent triggering of abnormal traffic analysis.
[0094] In one possible design, the traffic characteristic information of the third network element includes one or more of the following: protocol fingerprint information, service behavior information, service server information, message protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and QoS-related information of the traffic.
[0095] Of course, the above are just examples and are not limited to these.
[0096] This provides traffic characteristic information for various third-party network elements, improving the flexibility of the solution.
[0097] In one possible design, at least one of the traffic characteristic information of the third network element satisfies a preset condition.
[0098] Thus, by performing abnormal traffic analysis based on the traffic characteristic information of third network elements that meet preset conditions, the accuracy can be improved.
[0099] In one possible design, the method further includes: acquiring historical traffic characteristic information. Correspondingly, based on the traffic characteristic information of the third network element, identifying abnormal traffic of the third network element is sent, including: sending identifying abnormal traffic of the third network element based on both the traffic characteristic information and historical traffic characteristic information of the third network element.
[0100] Therefore, using historical traffic characteristic information to help determine the identification information of abnormal traffic of third network elements can further improve the accuracy of abnormal traffic identification.
[0101] In one possible design, obtaining traffic characteristic information of a third network element includes: sending a fifth request, which is used to obtain traffic characteristic information of the third network element.
[0102] In this way, by sending the fifth request to obtain the traffic characteristic information of the third network element, support is provided for abnormal traffic analysis based on the traffic characteristic information of the first session, thus improving the reliability of the solution.
[0103] In one possible design, the fifth request includes the identification information of the third network element.
[0104] In this way, the recipient of the fifth request can clearly identify which network element's traffic characteristic information to return, thus improving the reliability of the solution.
[0105] In one possible design, sending the fifth request includes sending the fifth request to at least one of the third network element, the fourth network element, or the fifth network element; wherein the third network element is used to provide user plane functions, the fourth network element is used to provide operation, maintenance, and management functions, and the fifth network element is used to provide application functions.
[0106] Of course, the above examples are just examples and not limitations.
[0107] This provides multiple possible ways to obtain traffic characteristic information of third network elements, improving the reliability and flexibility of the solution.
[0108] In one possible design, based on the traffic characteristic information of the third network element, the identification information of the abnormal traffic of the third network element is sent, including: determining the identification information of the abnormal traffic of the third network element based on reasoning or analysis according to the traffic characteristic information of the third network element.
[0109] In this way, a specific solution for improving network identification of abnormal traffic through AI functions can be clearly defined. This solution can be used to standardize the process of NWDAF assisting UPF in abnormal traffic analysis, thereby further improving the network's ability to handle abnormal traffic and network attacks.
[0110] One possible design also includes: sending at least one of the following based on the traffic characteristic information of the third network element:
[0111] Third time information includes one or more of the start time, end time, or duration of abnormal traffic of the third network element;
[0112] The source of abnormal traffic in third-party network elements;
[0113] Confidence level of abnormal traffic analysis results of third network elements.
[0114] This design can send more information about abnormal traffic based on the traffic characteristics of the third network element, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0115] In one possible design, the identification information of abnormal traffic from the third network element is used to update the policy of the third network element. Policies include, but are not limited to, PCC rules and N4 rules. N4 rules include, but are not limited to, PDR, FAR, QER, and URR.
[0116] In this way, the identification information of abnormal traffic of the third network element is used to update the policy of the third network element, which can achieve the effect of handling abnormal traffic in advance and dynamically, and improve the network's ability to handle abnormal traffic and network attacks.
[0117] In one possible design, the method further includes: determining the corresponding second network element based on the identification information of the abnormal traffic; sending the identification information of the abnormal traffic of the third network element, including: sending the identification information of the abnormal traffic of the third network element to the second network element; or,
[0118] Sending identification information of abnormal traffic from a third network element, including: sending identification information of abnormal traffic from a third network element to the third network element.
[0119] This provides multiple ways to send identification information for abnormal traffic, improving the flexibility of the solution.
[0120] Sixthly, a communication method is provided, which can be executed by a second network element or a chip or module within the second network element. The method includes: receiving identification information of abnormal traffic from a third network element; and updating the strategy of the third network element based on the identification information of the abnormal traffic from the third network element.
[0121] The embodiments of this application can implement a strategy of updating the identification information of abnormal traffic based on the third network element, thereby achieving the effect of handling abnormal traffic and improving the network's ability to handle abnormal traffic and network attacks.
[0122] In one possible design, the second network element is used to provide policy control functions; for example, the second network element is a PCF network element.
[0123] This clarifies the network element responsible for handling abnormal traffic.
[0124] In one possible design, the third network element is used to provide user plane functions, such as a UPF network element.
[0125] This clarifies the network elements that require abnormal traffic analysis and processing.
[0126] One possible design also accepts at least one of the following:
[0127] Third time information includes one or more of the start time, end time, or duration of abnormal traffic of the third network element;
[0128] The source of abnormal traffic in third-party network elements;
[0129] Confidence level of abnormal traffic analysis results of third network elements.
[0130] This design can obtain more information about abnormal traffic, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0131] In one possible design, the strategy of the third network element is updated based on the identification information of the abnormal traffic of the third network element, including:
[0132] Send a third rule based on the identification information of the abnormal traffic of the third network element;
[0133] The third rule includes identification information for abnormal traffic of the third network element and a fifth piece of information. The fifth piece of information is used to indicate at least one of the following:
[0134] Disable abnormal traffic;
[0135] Packet loss is applied to abnormal traffic;
[0136] Use the second QoS parameter for abnormal traffic;
[0137] Use a separate QoS stream for abnormal traffic;
[0138] Implement caching for abnormal traffic.
[0139] The above design clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from the server. It can achieve the effect of anticipating and dynamically handling abnormal traffic based on AI inference, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0140] In one possible design, the third rule also includes fourth time information, which indicates the time at which at least one of the abnormal traffic is executed.
[0141] This clarifies the timing for handling abnormal traffic and improves the reliability of the solution.
[0142] In one possible design, the strategy of the third network element is updated based on the identification information of the abnormal traffic of the third network element, including:
[0143] Release the session of the terminal device, which is the terminal device corresponding to the abnormal traffic; or...
[0144] Send the sixth message, which is used to indicate the release of the terminal device's session. The terminal device is the terminal device corresponding to the abnormal traffic.
[0145] Send the seventh message, which is used to instruct the terminal device corresponding to the abnormal traffic to prevent it from establishing a session in the second time period.
[0146] The above design clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from terminal devices. It can achieve the effect of handling abnormal traffic in advance and dynamically, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0147] A seventh aspect provides a communication method, which can be executed by a sixth network element or a chip or module within the sixth network element. The method includes: receiving a third rule, the third rule including identification information of abnormal traffic of the third network element and fifth information, the fifth information indicating at least one of the following: disabling abnormal traffic; performing packet loss on the abnormal traffic; using a second QoS parameter on the abnormal traffic; using a separate QoS stream on the abnormal traffic; sending a fourth rule according to the third rule, the fourth rule including identification information of abnormal traffic of the third network element and eighth information, the eighth information indicating at least one of the following: disabling abnormal traffic; performing packet loss on the abnormal traffic; using a second QoS parameter on the abnormal traffic; using a separate QoS stream on the abnormal traffic; performing buffering on the abnormal traffic; or...
[0148] Receive the sixth information and release the session of the terminal device according to the sixth information. The terminal device is the terminal device corresponding to the abnormal traffic of the third network element.
[0149] In one possible design, the sixth network element is used to provide session management functions, such as SMF.
[0150] In one possible design, the third network element is used to provide user plane functions, such as UPF.
[0151] In one possible design, the third rule also includes fourth time information, which indicates the time at which at least one of the abnormal traffic is executed.
[0152] The beneficial effects of each design in the seventh aspect are the same as those of the corresponding designs in the fifth and sixth aspects, and will not be repeated here.
[0153] Eighthly, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes: receiving a fifth request, the fifth request being used to obtain traffic characteristic information of the third network element; and sending the traffic characteristic information of the third network element according to the fifth request.
[0154] In one possible design, the third network element is used to provide user plane functions; for example, the third network element is a UPF network element.
[0155] In one possible design, the fifth request includes the identification information of the third network element.
[0156] In one possible design, the traffic characteristic information of the third network element includes one or more of the following: protocol fingerprint information, service behavior information, service server information, message protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and QoS-related information of the traffic.
[0157] In one possible design, at least one of the traffic characteristic information of the third network element satisfies a preset condition.
[0158] In one possible design, the fifth request is also used to obtain historical traffic characteristic information; the method also includes: sending historical traffic characteristic information according to the fifth request.
[0159] In one possible design, the method also includes receiving identification information of abnormal traffic from a third network element.
[0160] In one possible design, at least one of the following can also be accepted:
[0161] Third time information includes one or more of the start time, end time, or duration of abnormal traffic of the third network element;
[0162] The source of abnormal traffic in third-party network elements;
[0163] Confidence level of abnormal traffic analysis results of third network elements.
[0164] In one possible design, a fourth request can also be sent to obtain the identification information of abnormal traffic from the third network element. The fourth request includes an analysis identifier, which is used to indicate abnormal traffic analysis.
[0165] One possible design may also include:
[0166] The fourth rule is received, which includes the identification information of the abnormal traffic of the third network element and the eighth information. The eighth information is used to indicate at least one of the following: disabling the abnormal traffic; performing packet loss on the abnormal traffic; using the second QoS parameter on the abnormal traffic; using a separate QoS flow on the abnormal traffic; or performing buffering on the abnormal traffic.
[0167] Abnormal traffic from the third network element is handled according to the fourth rule.
[0168] One possible design may also include: based on the identification information of the abnormal traffic of the third network element, performing at least one of the following:
[0169] Disable abnormal traffic;
[0170] Packet loss is applied to abnormal traffic;
[0171] Use the second QoS parameter for abnormal traffic;
[0172] Use a separate QoS stream for abnormal traffic;
[0173] Implement caching for abnormal traffic.
[0174] One possible design also includes: releasing the session of the terminal device corresponding to the abnormal traffic.
[0175] One possible design may also include: sending identification information of abnormal traffic of the third network element, which is used to update the strategy of the third network element.
[0176] The beneficial effects of each design in the eighth aspect are the same as those of the corresponding designs in the fifth and sixth aspects, and will not be repeated here.
[0177] Ninthly, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes: receiving a model from a first network element, the model being used for abnormal traffic analysis; and determining identification information of the abnormal traffic based on the model.
[0178] This application provides a scheme for a third network element to acquire a model and perform local inference, namely: the third network element acquires the model from the first network element and determines the identification information of abnormal traffic based on the model. This scheme clarifies the specific method for mobile networks to utilize AI functions to identify abnormal traffic, achieving the effect of early and dynamic identification of abnormal traffic based on AI inference. This scheme can be used to standardize the process of abnormal traffic analysis in mobile networks, improving the ability of mobile networks to handle abnormal traffic and network attacks.
[0179] In one possible design, the first network element is used to provide network analysis functions; for example, the first network element is an NWDAF network element.
[0180] This clarifies the first network element providing the model and improves the reliability of the solution.
[0181] In one possible design, the third network element is used to provide user plane functions; for example, the third network element is a UPF network element.
[0182] In this way, the third network element can acquire the model and perform abnormal traffic analysis, thus improving the reliability of the solution.
[0183] One possible design may also include: sending a sixth request, which is used to obtain the model, and the sixth request includes an analysis identifier, which is used to indicate abnormal traffic analysis.
[0184] This enables on-demand request model deployment, improving the reliability of the solution.
[0185] In one possible design, the identification information of abnormal traffic is determined based on a model, including: determining the identification information of abnormal traffic based on traffic characteristic information and the model.
[0186] This clarifies that traffic characteristic information is used as the input parameter of the model, enabling the model to output identification information of abnormal traffic and improving the reliability of the solution.
[0187] In one possible design, traffic characteristic information includes one or more of the following: protocol fingerprint information, service behavior information, service server information, packet protocol information, domain name / address blacklist, suspicious / abnormal domain names / addresses, and QoS-related information of traffic.
[0188] Of course, the above are just examples and are not limited to these.
[0189] This provides a variety of traffic characteristic information, improving the flexibility of the solution.
[0190] In one possible design, at least one of the traffic characteristic information satisfies a preset condition.
[0191] Therefore, analyzing abnormal traffic based on traffic characteristic information that meets preset conditions can improve accuracy.
[0192] In one possible design, the identification information of abnormal traffic is determined based on a model, including: traffic characteristic information, historical traffic characteristic information, and the identification information of abnormal traffic determined by the model.
[0193] Therefore, using historical traffic characteristics to help determine the identification information of abnormal traffic can further improve the accuracy of abnormal traffic identification.
[0194] One possible design also includes determining at least one of the following based on the model:
[0195] The sixth time information includes one or more of the start time, end time, or duration of the abnormal traffic;
[0196] The source of abnormal traffic;
[0197] Confidence level of abnormal traffic analysis results.
[0198] This design can determine more information about abnormal traffic based on traffic characteristics, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0199] One possible design may also include: sending identification information for abnormal traffic, which is used to update the session's policy. Policies may include, but are not limited to, PCC rules and N4 rules. N4 rules include, but are not limited to, PDR, FAR, QER, and URR.
[0200] In this way, the strategy of notifying designated network elements (such as the second or sixth network element) of the identification information of abnormal traffic for updating the session clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from the server. This can achieve the effect of proactively and dynamically handling abnormal traffic based on AI inference, improving the network's ability to handle abnormal traffic and network attacks. At the same time, it is easily adaptable to the design logic of existing mobile networks with minimal modifications.
[0201] One possible design may also include: based on the identification information of the abnormal traffic, performing at least one of the following:
[0202] Disable abnormal traffic;
[0203] Packet loss is applied to abnormal traffic;
[0204] Use a third QoS parameter for abnormal traffic;
[0205] Use a separate QoS stream for abnormal traffic;
[0206] Implement caching for abnormal traffic.
[0207] In this way, the third network element can directly update the processing rules based on the identification information of abnormal traffic, clarifying the session-based abnormal traffic processing scheme in the scenario where abnormal traffic comes from the server, which can reduce signaling interaction and save resources.
[0208] One possible design could also include: triggering the release of the session of the terminal device corresponding to the abnormal traffic.
[0209] The above design clarifies the session-based abnormal traffic handling scheme in scenarios where abnormal traffic originates from terminal devices. It can achieve the effect of handling abnormal traffic in advance and dynamically, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0210] A tenth aspect provides a communication method, which can be executed by a first network element or a chip or module within the first network element. The method includes: receiving a sixth request from a third network element, the sixth request being used to obtain a model; wherein the sixth request includes an analysis identifier used to indicate abnormal traffic analysis; and sending the model, the model being used for abnormal traffic analysis.
[0211] For details on the specific design and beneficial effects of aspect 10, please refer to the details on the specific design and beneficial effects of aspect 9, which will not be repeated here.
[0212] Eleventhly, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes:
[0213] Receive a second rule, which includes identification information of the traffic of the first session and fourth information, wherein the fourth information is used to indicate at least one of the following: disable the traffic; perform packet loss on the traffic; perform caching on the traffic; use a first QoS parameter on the traffic; or use a separate QoS stream on the traffic.
[0214] The traffic of the first session is processed according to the second rule.
[0215] In one possible design, this traffic is considered abnormal traffic.
[0216] In one possible design, the traffic of the first session is processed according to the second rule during the third time period.
[0217] In one possible design, a second time information is also received. This second time information is used to indicate the effective or applicable time of the second rule and / or the fourth information. Alternatively, the second time information is used to indicate a third time period.
[0218] In one possible design, the second rule and / or fourth information includes second time information.
[0219] In one possible design, processing the traffic of the first session according to the second rule includes: processing the traffic / service flow corresponding to the identification information of the traffic of the first session according to the identification information of the traffic of the first session in the second rule.
[0220] In one possible design, processing the traffic of the first session according to the second rule includes: according to the fourth information in the second rule, performing at least one of the following on all traffic / service flows in the first session or traffic / service flows corresponding to the identification information of the traffic in the first session: disabling the traffic; performing packet loss on the traffic; performing caching on the traffic; applying the first QoS parameter to the traffic; or applying a separate QoS flow to the traffic.
[0221] In one possible design, the second rule is the N4 rule.
[0222] In one possible design, a second rule is received from the sixth network element.
[0223] For the specific design and beneficial effects of aspect ten, please refer to the corresponding design and beneficial effects of aspect three or four, and will not be repeated here.
[0224] In a twelfth aspect, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes:
[0225] Obtain traffic identification information;
[0226] Perform at least one of the following actions on the traffic: disable the traffic; perform packet loss on the traffic; perform caching on the traffic; apply a second QoS parameter to the traffic; or apply a separate QoS stream to the traffic.
[0227] In one possible design, this traffic is considered abnormal traffic.
[0228] In one possible design, at least one of the above is performed on the traffic during the fourth time period.
[0229] In one possible design, fifth time information is obtained, which is used to indicate the effective time or applicable time of the traffic identification information.
[0230] In one possible design, the fifth time information is used to indicate the fourth time period.
[0231] In a thirteenth aspect, a communication method is provided, which can be executed by a third network element or a chip or module within the third network element. The method includes:
[0232] Obtain traffic characteristic information of the first node or the first service; perform at least one of the following actions on the traffic of the first node or the first service based on the traffic characteristic information: disable traffic; perform packet loss on traffic; perform caching on traffic; apply a third QoS parameter to traffic; apply a separate QoS stream to traffic.
[0233] In one possible design, the traffic of the first node or the first service is abnormal traffic.
[0234] In one possible design, at least one of the above-mentioned actions is performed on the traffic of the first node or the first service during the fifth time period.
[0235] In one possible design, a sixth time information is obtained, which is used to indicate the effective time or applicable time of the traffic characteristic information.
[0236] In one possible design, the sixth time information is used to indicate the fifth time period.
[0237] In one possible design, the traffic of the first node or the first service is determined to be abnormal traffic based on the traffic characteristic information of the first node or the first service.
[0238] In one possible design, the load, congestion, or impact on the first node is determined based on the traffic characteristics of the first node or the first service.
[0239] Here, "first node load" can refer to the load on the first node exceeding a threshold. "First node congestion" can refer to the congestion level / grade of the first node exceeding a threshold. "First node impact" can refer to the first node being impacted by abnormal traffic, malicious traffic, or a large amount of traffic.
[0240] In a fourteenth aspect, a communication device is provided, comprising a module, unit, or technical means for performing the methods described in the first aspect or any possible design of the first aspect. Exemplarily, the communication device includes:
[0241] The transceiver module is used to obtain traffic characteristic information of the first session;
[0242] The processing module is used to control the transceiver module to send identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session.
[0243] In a fifteenth aspect, a communication device is provided, comprising a module, unit, or technical means for performing the method as described in the second aspect or any possible design of the second aspect. Exemplarily, the communication device includes:
[0244] The transceiver module is used to receive identification information of abnormal traffic in the first session;
[0245] The processing module is used to update the policy of the first session based on the identification information of the abnormal traffic in the first session.
[0246] A sixteenth aspect provides a communication device including modules, units, or technical means for performing the methods described in the third aspect or any possible design of the third aspect. Exemplarily, the communication device includes:
[0247] The transceiver module is used to receive a first rule, which includes identification information of abnormal traffic in the first session and first information. The first information is used to indicate at least one of the following: disabling abnormal traffic; performing packet loss on abnormal traffic; using a first QoS parameter on abnormal traffic; or using a separate QoS stream on abnormal traffic.
[0248] The processing module is used to control the transceiver module to send a second rule according to the first rule. The second rule includes identification information of abnormal traffic in the first session and fourth information. The fourth information is used to indicate at least one of the following: disabling abnormal traffic; performing packet loss on abnormal traffic; using a first QoS parameter on abnormal traffic; using a separate QoS stream on abnormal traffic; or performing caching on abnormal traffic.
[0249] or,
[0250] The processing module is used to receive the second information;
[0251] The processing module is used to release the first session based on the second information.
[0252] In a seventeenth aspect, a communication device is provided, comprising a module, unit, or technical means for performing the method as described in the fourth aspect or any possible design of the fourth aspect. Exemplarily, the communication device includes:
[0253] The transceiver module is used to receive the second request, which is used to obtain the traffic characteristic information of the first session;
[0254] The processing module is used to control the transceiver module to send the traffic characteristic information of the first session according to the second request.
[0255] Eighteenth aspect: A communication device is provided, including modules, units, or technical means for performing the methods described in the fifth aspect or any possible design of the fifth aspect. Exemplarily, the communication device includes:
[0256] The transceiver module is used to acquire traffic characteristic information of third-party network elements;
[0257] The processing module is used to control the transceiver module to send identification information of abnormal traffic of the third network element based on the traffic characteristic information of the third network element.
[0258] A nineteenth aspect provides a communication device including modules, units, or technical means for performing the methods described in the sixth aspect or any possible design of the sixth aspect. Exemplarily, the communication device includes:
[0259] The transceiver module is used to receive identification information of abnormal traffic from third-party network elements;
[0260] The processing module is used to update the policy of the third network element based on the identification information of the abnormal traffic of the third network element.
[0261] In a twentieth aspect, a communication device is provided, comprising a module, unit, or technical means for performing the method as described in the seventh aspect or any possible design of the seventh aspect. Exemplarily, the communication device includes:
[0262] The transceiver module is used to receive a third rule, which includes identification information of abnormal traffic from a third network element and a fifth information. The fifth information is used to indicate at least one of the following: disabling abnormal traffic; performing packet loss on abnormal traffic; using a second QoS parameter on abnormal traffic; or using a separate QoS flow for abnormal traffic.
[0263] The processing module is used to control the transceiver module to send a fourth rule according to the third rule. The fourth rule includes the identification information of the abnormal traffic of the third network element and the eighth information. The eighth information is used to indicate at least one of the following: disable the abnormal traffic; perform packet loss on the abnormal traffic; use the second QoS parameter on the abnormal traffic; use a separate QoS flow on the abnormal traffic; or perform caching on the abnormal traffic.
[0264] or,
[0265] The transceiver module is used to receive the sixth piece of information;
[0266] The processing module is used to release the session release of the terminal device based on the sixth information. The terminal device is the terminal device corresponding to the abnormal traffic of the third network element.
[0267] A twenty-first aspect provides a communication device including modules, units, or technical means for performing the methods described in the eighth aspect or any possible design of the eighth aspect. Exemplarily, the communication device includes:
[0268] The transceiver module is used to receive the fifth request, which is used to obtain the traffic characteristic information of the third network element;
[0269] The processing module is used to control the transceiver module to send the traffic characteristic information of the third network element according to the fifth request.
[0270] A twenty-second aspect provides a communication device including modules, units, or technical means for performing the methods described in the ninth aspect or any possible design of the ninth aspect. Exemplarily, the communication device includes:
[0271] The transceiver module is used to receive models from the first network element, and the models are used for abnormal traffic analysis.
[0272] The processing module is used to determine the identification information of abnormal traffic based on the model.
[0273] In a twenty-third aspect, a communication device is provided, comprising a module, unit, or technical means for performing the method as described in the tenth aspect or any possible design of the tenth aspect. Exemplarily, the communication device includes:
[0274] The transceiver module is used to receive a sixth request from a third network element. The sixth request is used to obtain the model. The sixth request includes an analysis identifier, which is used to indicate abnormal traffic analysis.
[0275] The processing module is used to control the sending and receiving module to send the model according to the sixth request. The model is used for abnormal traffic analysis.
[0276] A twentieth aspect provides a communication device comprising: at least one processor; and a communication interface communicatively connected to the at least one processor; wherein the at least one processor, by executing instructions stored in a memory, causes the device to perform, via the communication interface, the method described in any of the first to thirteenth aspects or any possible design of any of the aspects.
[0277] In a twenty-fifth aspect, a computer-readable storage medium is provided for storing instructions that, when executed, cause the method described in any of the first to thirteenth aspects or any possible design of any aspect to be implemented.
[0278] In a twenty-sixth aspect, a computer program product is provided, the computer program product storing instructions that, when run on a computer, cause the method described in any of the first to thirteenth aspects or any possible design of any aspect to be implemented.
[0279] A twenty-seventh aspect provides a communication system comprising a communication device as described in at least one of aspects fourteen to twenty-three.
[0280] For details on the specific designs and beneficial effects of aspects fourteen through twenty-seven, please refer to the corresponding designs and beneficial effects in aspects one through thirteen. Attached Figure Description
[0281] Figure 1 is a network architecture diagram of a communication system applicable to an embodiment of this application;
[0282] Figure 2 is a flowchart of a communication method provided in an embodiment of this application;
[0283] Figure 3 is an example diagram of the model reasoning process;
[0284] Figure 4 is an example diagram showing the identification information of abnormal traffic in the second network element requesting the first session;
[0285] Figure 5 is an example diagram of triggering a third request during the session establishment process;
[0286] Figure 6 is an example diagram of the first network element obtaining traffic characteristic information of the first session;
[0287] Figures 7A and 7B are example diagrams of strategies for updating the first session;
[0288] Figures 8A and 8B are example diagrams of abnormal traffic handling rules;
[0289] Figure 9 is a flowchart of another communication method provided in an embodiment of this application;
[0290] Figure 10 is a flowchart of another communication method provided in an embodiment of this application;
[0291] Figure 11 is an example diagram showing the identification information of abnormal traffic requested by a third network element from another third network element.
[0292] Figure 12 is an example diagram of the first network element obtaining the traffic characteristic information of the third network element;
[0293] Figure 13 is an example diagram of the strategy for updating the third network element;
[0294] Figure 14 is a flowchart of another communication method provided in an embodiment of this application;
[0295] Figure 15 is a flowchart of another communication method provided in an embodiment of this application;
[0296] Figure 16 is a flowchart of another communication method provided in an embodiment of this application;
[0297] Figure 17 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;
[0298] Figure 18 is a schematic diagram of another communication device provided in an embodiment of this application;
[0299] Figure 19 is a schematic diagram of the structure of a chip provided in an embodiment of this application. Detailed Implementation
[0300] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.
[0301] It is understood that the terms "system" and "network" in the embodiments of this application can be used interchangeably. "At least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, or b, or c, or a and b, or b and c, or a and c, or a and b and c.
[0302] Furthermore, unless otherwise stated, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects and are not used to limit the order, sequence, priority, or importance of multiple objects. For example, the first priority criterion and the second priority criterion are only used to distinguish different criteria, and do not indicate that the content, priority, or importance of these two criteria are different.
[0303] The technical solutions of the embodiments of this application can be applied to various communication systems, such as: 5th Generation (5G) communication systems, future communication systems, or other wireless communication systems that use wireless access technologies, etc., and the technical solutions of the embodiments of this application can be adopted.
[0304] Figure 1 shows a network architecture diagram of a communication system applicable to an embodiment of this application. For details, please refer to the 3rd Generation Partnership Project (3GPP) technical specification (TS) 23.501. The network functions and entities included in the system mainly include: user equipment (UE), radio access network (RAN) network elements, user plane function (UPF) network elements, data network (DN), access and mobility management function (AMF) network elements, session management function (SMF) network elements, policy control function (PCF) network elements, application function (AF) network elements, unified data management (UDM) network elements, network exposure function (NEF) network elements, network repository function (NRF) network elements, network data analytics function (NWDAF) network elements, binding support function (BSF), and operation administration and maintenance (OAM), etc.
[0305] In this context, user equipment (UE) can also be referred to as user equipment, terminal, mobile station, mobile terminal, etc. Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, etc. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, etc. The embodiments of this application do not limit the specific technologies or device forms used in the terminals.
[0306] Wireless access network elements can also be called access network elements, access network equipment, wireless access network equipment, access network, etc. Specific implementations can include base stations, evolved NodeBs (eNodeBs), transmission reception points (TRPs), next-generation NodeBs (gNBs) in fifth-generation mobile communication systems, next-generation base stations in sixth-generation mobile communication systems, base stations in future mobile communication systems, or access nodes in WiFi systems. They can also be modules or units that perform some of the functions of a base station; for example, they can be centralized units (CUs) or distributed units (DUs). Here, the CU performs the functions of the radio resource control protocol and packet data convergence protocol (PDCP) of the base station, and can also perform the functions of the service data adaptation protocol (SDAP). The DU performs the functions of the radio link control layer and medium access control (MAC) layer of the base station, and can also perform some or all of the physical layer functions. For specific descriptions of the above protocol layers, please refer to the relevant 3GPP technical specifications. Wireless access network equipment can be macro base stations, micro base stations, indoor stations, relay nodes, or donor nodes, etc. The embodiments of this application do not limit the specific technologies or equipment forms used in the wireless access network equipment.
[0307] Data network (DN): A data network that provides business services to users. Generally, the client is located on the terminal device, and the server is located on the data network. A data network can be a private network, such as a local area network (LAN), an external network not controlled by an operator, such as the Internet, or a dedicated network jointly deployed by operators, such as a network providing Internet Protocol (IP) multimedia core network subsystem (IMS) services.
[0308] The following is a brief introduction to the functions of some network elements in the core network:
[0309] AMF (Access Module) network elements are primarily responsible for access and mobility management in mobile networks, such as user registration management, connection management, and reachability management. Specific functions include non-access stratum signaling termination, registration area management, and access authentication.
[0310] SMF network elements are primarily responsible for session management in mobile networks, such as session establishment, modification, and release. Specific functions include assigning IP addresses to users and selecting UPFs that provide packet forwarding capabilities.
[0311] PCF network elements: support unified policy management of network behavior; provide policy rules to control plane functions for execution; obtain subscription-related information from user data repository (UDR) network elements for policy decision-making.
[0312] UDM network elements are mainly responsible for user contract management, access authorization, and authentication information generation.
[0313] AF network element: used by service providers (such as Edge Application Server (EAS) or edge application platform) to indicate their needs to the core network and subscribe to user plane events.
[0314] UPF network elements are primarily responsible for processing user packets, such as forwarding and billing statistics. In a session, the UPF directly connected to the DN via N6 is called the Protocol Data Unit (PDU) session anchor (PSA), and the UPF used for local traffic splitting is called the local PDU session anchor (L-PSA). The UPF serving as a splitting point is called a branching point (BP) or uplink classifier (ULCL).
[0315] NWDAF network elements: Used to perform model training and inference. They can collect input parameters for training and inference from other network elements and notify other network elements of the inference output parameters. Multiple NWDAF network elements can also transfer models to each other for further training or inference.
[0316] OAM network elements: primarily responsible for operation, management, and maintenance. They can be understood as network management systems, and are generally configured manually. NWDAF can also collect information from here (such as network element resource configuration, resource usage, etc., such as central processing unit (CPU) and memory resources).
[0317] The BSF network element is primarily responsible for dynamically managing the binding relationship between UEs and PCFs, ensuring accurate routing of policy requests and session continuity. It provides binding information to network elements such as NWDAF and SMF by storing, querying, and updating UE-PCF mapping relationships, supporting policy decisions and network function coordination.
[0318] UDR network element: Used for centralized storage and management of user data, including user subscription information, policy data, application data, etc.
[0319] It is understandable that different network elements can be deployed separately in separate physical entities, or they can be integrated into the same physical entity without restriction.
[0320] In practical applications, the above network architecture may also include other network elements, which are not limited in this application.
[0321] It should be noted that the names of the network elements in this application are merely examples, and this application does not preclude the possibility of other names for the network elements in the future, or the merging of functions between the network elements. With technological evolution, any device or network element capable of implementing the functions of the aforementioned network elements is within the scope of protection of this application. The interface names between the network elements are merely examples, and the interface names in specific implementations may be different; this application does not specifically limit them. Furthermore, the names of the messages (or signaling) transmitted between the aforementioned network elements are also merely examples and do not constitute any limitation on the function of the messages themselves.
[0322] For ease of description, each network element in this article can be represented by its corresponding English abbreviation, such as "AF" for application function network element, "SMF" for session management function network element, and so on.
[0323] In mobile networks, user plane functions (UPFs) serve as entry and exit points for user plane traffic and are typically protected by firewalls. However, their increasing prevalence in mobile networks has led to a surge in their numbers, making comprehensive coverage difficult with traditional static firewalls. Furthermore, firewalls have significant limitations: attackers can bypass blacklists by altering the IP addresses, domain names, and other characteristics of abnormal traffic through methods such as hiding traffic through legitimate services.
[0324] There is still a lack of clear solutions for identifying abnormal traffic and improving the stability and defense capabilities of mobile networks against attacks.
[0325] In view of this, the technical solution provided in the embodiments of this application can realize the identification of abnormal traffic based on traffic characteristic information and / or models (of the first session or the third network element), which can be used to standardize the process of abnormal traffic analysis in mobile networks, and can more timely cover all UPFs in the network, thereby improving the ability of mobile networks to handle abnormal traffic and network attacks.
[0326] To facilitate understanding, the network elements mentioned below will be introduced first:
[0327] First network element: Used to provide data analysis functions, such as performing model training and inference. It can collect training and inference input parameters from other network elements and notify other network elements of the inference output parameters. A specific example is NWDAF.
[0328] Second network element: used to provide policy control functions, such as PCF.
[0329] Third network element: Used to provide user plane functions (such as packet forwarding), specifically UPF.
[0330] The fourth network element is used to provide operation management and maintenance functions, such as OAM.
[0331] Fifth network element: used to provide application functions, such as AF.
[0332] The sixth network element is used to provide session management functions, such as SMF.
[0333] The seventh network element is used to provide security protection functions, such as a separately deployed firewall.
[0334] The eighth network element: used to provide unified data management functions, such as UDM.
[0335] Ninth network element: used to establish dynamic binding relationships between UE and PCF and to store the mapping relationship between UE and PCF, specifically such as the binding support function (BSF) network element.
[0336] The tenth network element: used to provide user data storage functions, such as UDR.
[0337] In the following text, the main description will focus on network elements as the executing entity. It should be understood that in actual applications, the executing entity can be the network element itself, or a chip or module within the network element.
[0338] Referring to Figure 2, a flowchart of a communication method provided in an embodiment of this application is shown. This method can realize session-based abnormal traffic identification. The method includes S301 to S302:
[0339] S301, the first network element (such as NWDAF) obtains the traffic characteristic information of the first session.
[0340] Optionally, the first session is a PDU session. Of course, this is just an example, and it is not limited to this in practice; it can be any connection used for terminals to access services in a mobile network.
[0341] Optionally, traffic refers to business data flow (simply put, business flow or data flow, such as the sequence of data packets when a terminal accesses a certain service), and abnormal traffic refers to abnormal business data flow (such as business data flow with abnormal behavior, business data flow that may be a network attack, business data flow that may cause negative impacts such as increased load on network devices / terminals / servers, etc.). Of course, this is just an example and is not limited to this in practice.
[0342] For example, the traffic characteristics information of the first session includes, but is not limited to, the following:
[0343] 1) The fingerprint information of the protocol;
[0344] For example: Secure Sockets Layer (SSL) / Transport Layer Security (TLS), etc.
[0345] 2) Business behavior information;
[0346] For example: heartbeat pattern, synchronization sequence numbers (SYN) - acknowledgment (ACK) characteristics (e.g., whether the number of TCP three-way handshake messages is obviously mismatched / different), transmission duration, transmission interval, traffic size, etc.
[0347] 3) Information about the business server;
[0348] A business server refers to a server that transmits business data, such as a server in a DN (Database Name).
[0349] Information about the business server, such as the business server's certificate.
[0350] 4) Protocol information of the message;
[0351] For example: port and protocol matching (e.g., match or no match), Hypertext Transfer Protocol (HTTP) header, Domain Name System (DNS) query mode (including recursive query, iterative query, or non-recursive query), message length, header length, plaintext header (such as time to live (TTL) or protocol fields), etc.
[0352] 5) Domain / address blacklist;
[0353] 6) Suspicious / abnormal domain names / addresses;
[0354] For example, if a service provider determines that one of its servers has been attacked, it will notify the operator of the server's domain name / address. Alternatively, if the operator's own security alert (AF) or a third-party security service provider discovers abnormal behavior on certain servers, it will notify the operator's core network of the server's domain name / address.
[0355] 7) Information related to the quality of service (QoS) of the traffic.
[0356] Examples include latency, bit rate, packet loss rate, periodicity, burst information, congestion, and traffic patterns. Another example is information related to Quality of Service (QoS) control of traffic, such as QoS-related parameters in 3GPP protocols TS23.501 and TS23.503, or information used to generate QoS parameters.
[0357] The fingerprint information, business behavior information, business server information, or domain / address blacklists mentioned above can be information from blacklists and / or whitelists. For example, if the information mentioned above is from a whitelist, then information that differs from the above information may be abnormal.
[0358] For ease of description, this application mainly uses a blacklist as an example (e.g., a TLS fingerprint blacklist), but does not limit the specific form of information obtained or the logic based on the information judgment. For example, when obtaining a blacklist, traffic that matches the blacklist is considered abnormal traffic. Alternatively, a whitelist can be obtained, and traffic that does not match the whitelist is considered abnormal traffic, etc., without limitation.
[0359] It is understood that 1)-7) above are merely examples, and there may be other characteristic information that can be used to detect network attacks or abnormal traffic. For example, the traffic characteristic information required for methods used to detect network attacks.
[0360] It should be understood that the traffic characteristic information of the first session includes the current traffic characteristic information of the first session, such as the traffic characteristic information generated within the current preset time range.
[0361] Obtaining the traffic characteristics information of the first session can also be referred to as the traffic characteristics information of the first session as input.
[0362] Optionally, the first network element can directly obtain the traffic characteristic information of the first session, such as 1)-7) above. Alternatively, the first network element can also obtain information used to determine the traffic characteristic information of the first session and determine the traffic characteristic information of the first session locally based on the obtained information (e.g., through hash calculation functions that can be configured locally by some first network elements). For example, NWDAF can directly obtain TLS fingerprint information (e.g., from UPF), or NWDAF can also obtain TLS protocol layer information (e.g., the content of client hellow and / or server hellow messages, including TLS protocol version, etc.), and calculate the hash value of the corresponding field locally based on the TLS protocol layer information to determine the TLS fingerprint information.
[0363] S302. The first network element sends the identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session.
[0364] Optionally, sending can be replaced by outputting or determining. S302 can be replaced by: the first network element determining the identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session; or, the first network element outputting the identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session.
[0365] Optionally, abnormal traffic can refer to abnormal or malicious traffic / business flows / data packets, or traffic / business flows / data packets used to access abnormal or malicious applications, or data packets / business flows / traffic used for network attacks, or traffic / business flows / data packets that do not conform to normal business logic, or traffic / business flows / data packets that negatively impact one or more devices on the data transmission path (e.g., increased load, congestion, performance or storage space consumption, affecting normal device operation, or harming information within the device). Any traffic / data packet that differs from or is significantly different from normal business data packets / business flows / traffic in any one or more of the following aspects (content, format, traffic characteristics, or traffic trend) can be called abnormal traffic; the name is not limited.
[0366] Optionally, the identification information for the abnormal traffic in the first session can be one or more of the following: application ID, IP 5-tuple, IP 3-tuple, domain name, etc., without limitation. Domain names include, for example, fully qualified domain names (FQDNs), uniform resource locators (URLs), or uniform resource identifiers (URIs). In this embodiment, the process by which the first network element determines / outputs / sends the identification information for the abnormal traffic in the first session based on the traffic characteristic information of the first session can be called abnormal traffic analysis (analytic / analyze), abnormal traffic identification, abnormal traffic inference, or abnormal traffic prediction, etc., or other names can be used without limitation.
[0367] The following are some methods used by the first network element to infer or analyze network attacks (i.e., abnormal traffic) based on traffic characteristic information:
[0368] 1. Network attack identification based on Secure Sockets Layer (SSL) / Transport Layer Security (TLS) fingerprinting.
[0369] A TLS fingerprint is a hash value generated from the plaintext portion of the client's Hello message or the server's Hello response message during the TLS handshake phase. Specific fields (such as supported TLS versions, cipher suites, and extension lists) are extracted and a specific algorithm (such as JA3's MD5 hash) is used to calculate this hash value. This hash value is the TLS fingerprint and can be used for client or server identification and classification.
[0370] Common TLS fingerprints include JA3 (based on Client Hello messages) / JA3S (based on Server Hello messages), which generate a hash value by concatenating specific extracted fields.
[0371] Fields extracted from Client Hello or Server Hello messages include, but are not limited to:
[0372] TLS version: The TLS protocol version supported by the client;
[0373] Cipher suite: A combination of encryption and key exchange algorithms negotiated between the client and server;
[0374] Extended list: Additional features or options allowed in the TLS protocol, such as supported elliptic curves, application layer protocol negotiation, etc.
[0375] Random number: A random value generated by each party in the TLS handshake, used for subsequent key generation.
[0376] TLS fingerprints can be used to detect network attacks or abnormal traffic. Malware and other unusual services often use specific TLS fingerprints for communication, and these services can be identified based on a blacklist of TLS fingerprints. For example, abnormal or malicious traffic may send a large number of abnormal / malicious packets after establishing a TLS connection to attack any one or more devices in the terminal device, server, routing device, or mobile network system (such as user plane network elements in a mobile network system).
[0377] 2. Synchronization (SYN) - Acknowledgment (ACK) attack.
[0378] The TCP connection establishment process requires a three-way handshake. The SYN-ACK packet appears in the second handshake and is used to confirm the first handshake. After receiving the SYN-ACK packet, one party first determines whether the packet falls within the scope of the three-way handshake. If the second handshake packet is received without completing the first handshake, a reset packet (RST) will be sent to the other party to inform them that the packet sent was erroneous and the connection cannot be established.
[0379] SYN-ACK attacks (also known as SYN-ACK flood attacks) exploit this vulnerability. Attackers use tools or manipulate botnets to send a large number of SYN-ACK packets to the target server. These packets are essentially second handshake packets that appear out of nowhere. The server is busy replying to RST packets, which leads to resource exhaustion and an inability to respond to normal requests.
[0380] When this type of attack occurs, there will be a significant mismatch in the number of three-way handshake messages (SYN, SYN-ACK, ACK) in the network (during normal connection establishment, the number of three-way handshake messages transmitted in the network is basically the same). Therefore, by detecting the number of three-way handshake messages using the SYN-ACK characteristic, it is possible to identify that the network has been attacked, and thus take measures against abnormal service flows.
[0381] 3. Other methods for detecting attacks.
[0382] Botnets typically have a fixed heartbeat frequency, and traffic originating from a botnet can be identified by recognizing the heartbeat frequency. When a botnet launches an attack, there will be a sudden surge in traffic. Network worms will rapidly initiate a large number of connections to probe for vulnerabilities in target devices. A server may suddenly receive a large number of accesses from uncommon addresses, or a large number of terminals may suddenly access uncommon addresses (for example, a sudden influx of foreign IPs accessing a server that is only used domestically).
[0383] Of course, the above are just examples, and there are actually more than these.
[0384] As can be seen, abnormal traffic can be identified by detecting special network behaviors through traffic feature information. For example, by comparing the traffic feature information of the current business data stream with historical traffic feature information (such as TLS fingerprint blacklist, mismatch / difference in the number of three-way handshake messages), it can be determined whether the current business data stream is abnormal traffic.
[0385] This application provides a session-based (or session-granular) abnormal traffic identification scheme, namely: a first network element acquires traffic characteristic information of a first session and sends identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session. This scheme clarifies the specific method for identifying abnormal traffic in mobile networks, can be used to standardize the process of abnormal traffic analysis in mobile networks, can more timely cover UPFs in the network, and improve the ability of mobile networks to handle abnormal traffic and network attacks.
[0386] In one possible design, the first network element can determine / output / send identification information of abnormal traffic in the first session based on inference or analysis, according to the traffic characteristic information of the first session. The result determined / output / sent by the first network element based on the traffic characteristic information of the first session can be called the analysis result or the inference result.
[0387] For example, the first network element takes the traffic characteristic information of the first session as the input parameter of the model, inputs the traffic characteristic information of the first session into the model, runs the model, and outputs the inference result, which includes the identification information of the abnormal traffic of the first session.
[0388] The 3GPP standard TS23.288 defines the process by which a consumer network element (NF) subscribes to model inference results from the NWDAF. The NF can indicate to the NWDAF, based on an analytics ID, which inference results it needs to subscribe to, such as congestion status, average traffic volume, and average end-to-end latency. The NWDAF can directly notify the NF of the inference results (output parameters) or notify the target network element indicated by the NF. To perform inference, the NWDAF needs to collect data (as input parameters). The standard also defines a data collection process, allowing the NWDAF to collect data from various network elements, UEs, OAM, etc. Figure 3 illustrates an example of the model inference process.
[0389] For example, the first network element uses the traffic characteristics of the first session as input parameters to perform analysis. The analysis results include the identification information of abnormal traffic in the first session (the analysis process can be the internal logic of the first network element).
[0390] Because the first network element is used to provide data analysis functions and has the ability to analyze or reason, it can realize the identification of abnormal traffic based on sessions by using analysis or AI functions, and can achieve the effect of early and dynamic identification of abnormal traffic.
[0391] This design clarifies specific schemes for improving network identification of abnormal traffic through analysis or reasoning. It can be used to standardize the process of NWDAF assisting UPF in abnormal traffic analysis (for example, the first network element is NWDAF, and in S302, NWDAF sends the identification information of abnormal traffic in the first session to UPF, so that all UPFs in the network can be covered more promptly), and further improve the network's ability to handle abnormal traffic and network attacks.
[0392] In one possible design, the first network element can also obtain historical traffic characteristic information of other sessions, which is used to help determine (such as analyze or infer) the identification information of abnormal traffic in the first session.
[0393] Historical traffic characteristics include, for example, traffic characteristics generated before a preset time range. Historical traffic characteristics may include those from the first session, as well as those from other sessions.
[0394] For example, the historical traffic characteristics of other sessions and the traffic characteristics of the first session can be input into the model, and the model can infer and output the identification information of the abnormal traffic of the first session.
[0395] Alternatively, for example, historical traffic characteristics of other sessions can be input into the model for training. After obtaining the trained model, the traffic characteristics of the first session can be input into the trained model, and the trained model can infer and output the identification information of abnormal traffic in the first session.
[0396] This design approach utilizes historical traffic characteristics to determine the identification information of abnormal traffic in the first session for analysis or inference, which can further improve the accuracy of abnormal traffic identification.
[0397] In one possible design, the first network element, based on the traffic characteristic information of the first session, in addition to determining / outputting / sending the identification information of the abnormal traffic of the first session, can also determine / output and send at least one of the following (or the analysis result or inference result can also include at least one of the following):
[0398] 1) First-time information, which includes one or more of the start time, end time or duration of abnormal traffic in the first session.
[0399] In some embodiments, the first time information can also indicate an applicable time period. For example, it may indicate the applicable time period for analysis results or inference results. For instance, the first time information may indicate a third time period.
[0400] 2) The source of abnormal traffic in the first session.
[0401] For example, abnormal traffic may originate from a terminal device, and / or from a server. It is understood that when there are multiple instances of abnormal traffic, all of it may originate from the terminal device, all from the server, or some may originate from the terminal device and some from the server, depending on the specific implementation.
[0402] In a specific example, the first network element (or the model used by the first network element) determines the source of the anomaly based on the uplink / downlink transmission behavior of the abnormal traffic, such as an abnormal terminal device for uplink and an abnormal server for downlink.
[0403] In a specific example, the first network element (or the model used by the first network element) determines the source of the anomaly based on the unique abnormal traffic characteristics of the terminal device or the server side. For example, TLS fingerprints can be identified through messages such as client hello or server hello, and an anomaly in the fingerprint of the client hello message may originate from the terminal device.
[0404] In one specific example, the first network element determines the source of the anomaly based on the number of handshake messages. For example, if there are far more handshake messages from the terminal device than response messages from the server, the anomaly may originate from the terminal device.
[0405] 3) Confidence level of the abnormal traffic analysis results of the first session.
[0406] For example, the model used by the first network element is based on the accuracy of the inference result output.
[0407] This design enables the first network element to send more information about abnormal traffic based on the traffic characteristics of the first session, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0408] In one possible design, the timing or triggering conditions for the first network element to send the identification information of abnormal traffic in the first session can be varied, such as a combination of one or more of the following:
[0409] 1) If the analysis results or inference results indicate that there is abnormal traffic in the first session, the first network element sends the identification information of the abnormal traffic in the first session;
[0410] This makes it easy to identify abnormal traffic and is simple to implement.
[0411] 2) The first network element sends the identification information of all traffic and indicates whether each traffic is normal or abnormal;
[0412] In this way, any abnormalities in all traffic during the first session can be clearly identified, making management easier.
[0413] 3) If there is abnormal traffic in the first session and the confidence level of the abnormal traffic exceeds the first threshold, the first network element sends the identification information of the abnormal traffic in the first session;
[0414] This avoids frequently triggering abnormal traffic processing, saving resource consumption.
[0415] 4) If the duration of abnormal traffic exceeds the second threshold or the current time is less than the start time of abnormal traffic than the third threshold, the first network element sends the identification information of abnormal traffic for the first session.
[0416] In this way, abnormal traffic can be handled in a timely manner while taking into account resource consumption.
[0417] Optionally, the way to indicate that traffic is abnormal traffic may be that the information carrying the abnormal traffic identifier or the message itself indicates that the traffic corresponding to the identifier is abnormal traffic, or other information carried in the same or different messages as the abnormal traffic identifier indicates that the traffic corresponding to the identifier is abnormal traffic (for example, a message indicating that the traffic corresponding to the identifier is abnormal traffic or normal traffic).
[0418] Of course, the above are just examples, and there are actually more than these.
[0419] The above design clarifies the timing or triggering conditions for the first network element to send the identification information of abnormal traffic in the first session, thus improving the reliability of the solution.
[0420] In one possible design, S301 to S302 may be executed in response to a request from a second network element.
[0421] For example, as shown in Figure 4, before S301, it also includes:
[0422] S300 and the second network element (such as PCF) send a first request, and the first network element receives the first request. The first request is used to obtain the identification information of abnormal traffic in the first session.
[0423] Accordingly, after receiving the first request, the first network element responds by executing S301 to S302. Further, S302 may specifically include: the first network element sending identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session, and the second network element receiving the identification information of abnormal traffic in the first session. Optionally, the first network element may also send the identification information of abnormal traffic in the first session to other network elements, such as a third network element, without restriction.
[0424] The first request may include identification information for the first session, used to indicate the first session. The identification information for the first session includes, but is not limited to, one or more of the following:
[0425] 1) Identifier of the first session. Such as PDU session ID.
[0426] 2) Identification information of the terminal device. Among them, the first session is the session of the terminal device.
[0427] Terminal device identification information includes, for example: the user's permanent subscription identifier (SUPI), International Mobile Subscriber Identity (IMSI), permanent equipment identifier (PEI), 5G globally unique temporary ID (5G-GUTI), a encrypted version of SUPI (subscription concealed identifier (SUCI), address (such as IP address), or a generic public subscription identifier (GPSI), etc.
[0428] 3) Attribute information of the first session.
[0429] Examples include (data network name, DNN) and single network slice selection assistance information (S-NSSAI).
[0430] Of course, the above three are just examples, and there are actually more than three examples.
[0431] The first request may include an analytics ID, which can be used to indicate to the first network element the inference result, inference type, or analytics type subscribed to by the second network element. For example, the analytics ID can be used to indicate abnormal traffic analysis.
[0432] In practice, the second network element sending the first request can be triggered by one or more of the following:
[0433] 1) The third request is used to indicate that the first session requires / needs identification information of abnormal traffic, or in other words, the third request is used to indicate that the first session requires / needs abnormal traffic analysis.
[0434] For example, a third request can be triggered by the session establishment process of the first session. As shown in Figure 5, this includes:
[0435] S201. The terminal device initiates a session establishment request. The session establishment request contains the identification information of the first session. The sixth network element (such as SMF) receives the session establishment request. The session establishment request is used to request the establishment of the first session.
[0436] Optionally, session establishment requests can also be used to indicate abnormal traffic analysis.
[0437] The specific implementation of the identification information for the first session can be found above, and will not be repeated here.
[0438] S202. The sixth network element obtains the terminal device's subscription information from the eighth network element (such as UDM) based on the session establishment request.
[0439] S203, the sixth network element sends a third request to the second network element based on the subscription information of the terminal device, and the second network element receives the third request.
[0440] For example, if the terminal device's subscription information indicates that the terminal device is allowed / authorized for abnormal traffic analysis, then the sixth network element and the second network element send a third request. The third request may include the identifier of the first session, the attribute information of the first session, or the identifier of the terminal device, etc.
[0441] Alternatively, for example, if the second network element determines, based on local logic or locally configured policy information, or policy information obtained from the eighth network element, that the terminal device or the first session is allowed / authorized for abnormal traffic analysis, then the sixth network element sends a third request to the second network element. The third request may include the identifier of the first session, the attribute information of the first session, or the identifier of the terminal device, etc.
[0442] S204, the second network element, the sixth network element, etc., execute the remaining session establishment process.
[0443] It is understandable that the above is triggered by the third request during the session establishment process of the first session. In fact, the third request can also be triggered after the session establishment process of the first session is completed (i.e., S204), that is, steps S203, S300-302 are executed after S204.
[0444] Furthermore, the third request can also be triggered by other network elements. For example, the above S2010 to S204 can be replaced with:
[0445] The fifth network element sends the "Y" request to the second network element, and the second network element receives the "Y" request. The second network element then sends the third request based on the "Y" request.
[0446] For example, the AF may instruct the PCF that a certain terminal device or service requires abnormal traffic analysis. The "Y" request may include identification information of the service (such as application identifier, IP 5-tuple, IP 3-tuple, etc.), identification of the terminal device, or indication information indicating that abnormal traffic analysis is required.
[0447] For example, the "Y" request can be an AF request. The AF can send the "Y" request directly to the PCF, or send the "Y" request to the PCF through the forwarding of network elements such as NEF and UDR. During the sending process, the content and form of the message can be changed or remain the same (for example, when the AF sends it to the NEF, it is message 1, and when the NEF forwards it to the PCF, it is message 2, and the information included inside may be added or reduced, or the name of some information may also be changed. The key is that the function of the information remains unchanged).
[0448] After receiving the third request, the second network element responds by sending the first request.
[0449] 2) Attribute information of the first session.
[0450] For example, the second network element may not consider whether the terminal device or the first session is allowed / authorized for abnormal traffic analysis, but instead send a first request to the second network element based on the attribute information of the first session (such as DNN or S-NSSAI). The first request may include the identifier of the first session or the attribute information of the first session.
[0451] For example, in scenarios with high security requirements for DNN and / or S-NSSAI, abnormal traffic analysis of the first session is required regardless of whether the terminal device or the first session is allowed / authorized, which can improve network security.
[0452] Optionally, the second network element can obtain the attribute information of the first session from the tenth network element (such as UDR).
[0453] 3) Local configuration.
[0454] For example, the first network element periodically sends the first request according to the locally configured time interval, thereby periodically triggering the abnormal traffic analysis of the first session.
[0455] 4) Based on the terminal device's subscription / policy information
[0456] For example, the second network element can determine the terminal device's request for / authorization of abnormal traffic analysis based on the terminal's subscription / policy information obtained from the UDR network element or its local configuration, and then send the first request.
[0457] Of course, the above four are just examples, and there are actually more than these.
[0458] In the above design, the first network element executes S301 to S302 based on the request from the second network element, which can realize the on-demand triggering of abnormal traffic analysis, improve network security performance, and simultaneously avoid the problem of resource waste caused by frequent triggering of abnormal traffic analysis.
[0459] In one possible design, the first network element obtains the traffic characteristic information of the first session from other network elements.
[0460] For example, as shown in Figure 6, S301 specifically includes:
[0461] S301A, the first network element sends a second request, the second request being used to obtain traffic characteristic information (including traffic characteristic information of the first session and / or historical traffic characteristic information, etc.), and correspondingly, at least one network element receives the second request;
[0462] S301B: At least one network element sends traffic characteristic information, and the first network element receives the traffic characteristic information.
[0463] Optionally, the second request is to subscribe to messages.
[0464] Optionally, the second request is used to obtain traffic characteristic information of the first session. The second request includes the identification information of the first session, so that the recipient of the second request can clearly return the traffic characteristic information of which session.
[0465] The identification information for the first session includes, but is not limited to, one or more of the following:
[0466] 1) Identifier of the first session. Such as PDU session ID.
[0467] 2) Identification information of the terminal device. The first session refers to the terminal device's session. Identification information of the terminal device includes, for example: SUPI, PEI, 5G-GUTI, SUCI, IP address, IMSI, or GPSI, etc.
[0468] 3) Attribute information of the first session. For example, DNN, S-NSSAI, etc.
[0469] Of course, the above three are just examples, and there are actually more than three examples.
[0470] Optionally, the second request may also include an event ID, such as an event ID used to indicate session traffic characteristics. The event ID can indicate which information is being subscribed to.
[0471] For details on how the first network element acquires historical traffic characteristic information, please refer to the specific implementation method for acquiring traffic characteristic information of the first session by the first network element; it will not be elaborated here. The acquired historical traffic characteristic information can also be used by the first network element to perform abnormal traffic analysis, such as training and / or inference.
[0472] In practice, the traffic characteristic information and historical traffic characteristic information of the first session can originate from one or more network elements. For example, the second request sent by the first network element may include at least one of the following:
[0473] 1) The first network element sends a second request to the third network element (such as UPF). Correspondingly, the third network element receives the second request and sends the traffic characteristic information of the first session according to the second request.
[0474] For example, a first network element can subscribe to protocol fingerprint information, service behavior information, service server information, packet protocol information, traffic QoS-related information, domain name / address blacklists, etc., from a third network element (such as a UPF). The information subscribed to by the first network element from the third network element can be current information, such as the information on traffic currently being forwarded by the third network element, or, for example, the information on traffic in the first session.
[0475] 2) The first network element sends a second request to the fourth network element (such as OAM). Accordingly, the fourth network element receives the second request and sends the traffic characteristic information of the first session and / or other sessions according to the second request.
[0476] For example, a first network element can subscribe to protocol fingerprint information, service behavior information, service server information, packet protocol information, traffic QoS-related information, domain name / address blacklists, etc., from a fourth network element (such as OAM). The information subscribed to by the first network element from the fourth network element can be historical information. Historical information can refer to information before the first session was established or before receiving the subscription message from the first network element, such as the protocol fingerprint information regarding the normal / abnormal status of a certain service (also known as a whitelist / blacklist) before receiving the subscription message from the first network element. In one possible implementation, the information subscribed by the first network element from the fourth and fifth network elements corresponds to the information subscribed by the first network element from the third network element, such as having the same information type. The information types subscribed by the first network element from the third network element can be fewer than the information types subscribed by the first network element from the fourth and fifth network elements.
[0477] 3) The first network element sends a second request to the fifth network element (such as AF). Accordingly, the fifth network element receives the second request and sends the traffic characteristic information of the first session based on the second request.
[0478] For example, the first network element can subscribe to the fifth network element (such as AF) for suspicious / abnormal domain names / addresses, protocol fingerprint information, service behavior information, service server information, message protocol information, traffic QoS-related information, domain name / address blacklists, etc. The information subscribed by the first network element to the fifth network element can be historical information or current information.
[0479] Optionally, at least one of the traffic characteristic information in the first session satisfies a preset condition. Different pieces of information can satisfy different preset conditions. Correspondingly, the third, fourth, or fifth network element determines the traffic characteristic information that satisfies the preset conditions before sending the traffic characteristic information.
[0480] For example, the TLS fingerprint information of the business data flow may match an abnormal fingerprint or not match a normal fingerprint; for example, the port and protocol may not match; for example, a third network element may determine from the SYN-ACK characteristics that the number of TCP three-way handshake signaling messages is unequal, and optionally, if the difference in number is significant (e.g., exceeding a threshold), then the first network element will be notified of the traffic characteristic information of the first session. If the threshold is not exceeded, no notification will be made; for example, the plaintext header may contain abnormal TTL or protocol fields, etc.
[0481] In this design, the first network element requests (e.g., subscribes to) the traffic characteristic information of the first session from other network elements, providing support for the first network element to perform abnormal traffic analysis based on the traffic characteristic information of the first session, thus improving the reliability of the solution.
[0482] In one possible design, the identification information of abnormal traffic in the first session is used to update the policy (or session policy) of the first session. Updating the policy of the first session includes, but is not limited to, updating one or more of the following: policy and charging control (PCC) rules and N4 rules. N4 rules include, but are not limited to, packet detection rules (PDR), forwarding action rules (FAR), quality of service (QoS) enforcement rules (QER), and usage reporting rules (URR).
[0483] For example, S302 above includes identification information of abnormal traffic received by the second network element in the first session. As shown in Figure 7A, after S302, the following is also executed:
[0484] S303 and the second network element (such as PCF) update the policy of the first session based on the identification information of the abnormal traffic of the first session.
[0485] In one possible example, the abnormal traffic in the first session originates from the server. In this case, as shown in Figure 7B, the second network element updates the policy of the first session based on the identification information of the abnormal traffic, which may include the following process:
[0486] S303A and the second network element send the first rule based on the identification information of the abnormal traffic in the first session; correspondingly, the sixth network element (such as SMF) receives the first rule. The first rule is, for example, a PCC rule.
[0487] The first rule includes identification information for the abnormal traffic of the first session and first information, the first information being used to indicate at least one of the following:
[0488] 1) Disable abnormal traffic;
[0489] Disabling abnormal traffic can involve actions such as packet loss or adjusting QoS parameters.
[0490] 2) Perform packet loss on abnormal traffic;
[0491] 3) Apply the first QoS parameter to abnormal traffic;
[0492] For example, the first QoS parameter may include the 5G system Quality of Service Flow Identifier (5QI).
[0493] Understandably, 5QI can be used specifically for QoS parameters that indicate abnormal traffic, such as limiting the maximum bit rate.
[0494] It is understandable that when using special (or new) QoS parameters for abnormal traffic, the sixth network element also needs to configure QoS rules on terminal devices and access network devices (such as QoS rules on terminal devices and QoS profiles on the RAN).
[0495] 4) Use a separate QoS stream for abnormal traffic;
[0496] 5) Cache abnormal traffic.
[0497] S303B and the sixth network element send the second rule according to the first rule, and the third network element (such as UPF) receives the second rule. The second rule is, for example, the N4 rule.
[0498] Optionally, the sixth network element determines the correspondence between the first rule and the QoS flow according to the first rule, for example, following the rules of QoS flow binding and QoS flow mapping in 3GPP protocols TS23.501 and TS23.503. For example, if the first 5QI in the first rule is the first 5QI of abnormal traffic, the service data flow of abnormal traffic is bound to the QoS flow with the first 5QI, for example, this QoS flow contains only abnormal service data flow. As another example, if the first rule indicates the use of a separate QoS flow, then the service data flow included in the first rule (e.g., the abnormal service data flow) is bound to the separate QoS flow.
[0499] The second rule includes identification information for abnormal traffic in the first session and a fourth piece of information. The fourth piece of information is used to indicate at least one of the following: disabling abnormal traffic; performing packet loss on abnormal traffic; using a first QoS parameter on abnormal traffic; using a special QoS parameter on abnormal traffic; using a separate QoS stream on abnormal traffic; or performing caching on abnormal traffic.
[0500] Figure 8A shows an example of packet loss for abnormal traffic, with the second rule including PDR and FAR, and the fourth information including the content of FAR. Figure 8B shows an example of using special QoS parameters for abnormal traffic (optional, using a separate QoS flow for abnormal traffic), with the second rule including PDR, FAR, and QER, and the fourth information including the content of FAR and QER. An example of caching for abnormal traffic is similar to Figure 8A, except that the FAR action is caching.
[0501] It is understandable that PDR is used to filter incoming data packets, for example, based on packet header information such as the IP 5-tuple and 3-tuple.
[0502] FAR and QER are associated with PDR. PDR is used to filter out a portion of data packets, and then the corresponding FAR and QER are used to perform forwarding and control actions on that type of data packets.
[0503] FAR is used to indicate the forwarding action. For example, it indicates whether the packet should be forwarded, buffered, or dropped. If the action is forwarding, it also indicates information such as which tunneling protocol user plane tunnel (GPRS tunneling protocol user plane tunnel, GTP-U) the packet should be forwarded to.
[0504] QER is used to indicate actions related to QoS flows and QoS control. For example, a QER includes a QoS Flow Identifier (QFI), which is written into the GTP-U header of the data packet to indicate to the RAN that the data packet belongs to the corresponding QoS Flow and to perform the corresponding QoS control. Another example is that a QER includes the QoS parameters (maximum flow bit rate, guaranteed flow bit rate, average window, etc.) corresponding to the QoS Flow or the data packet, which are used by the UPF to perform rate measurement and rate limiting of traffic.
[0505] Optionally, if the first network element in S302 also outputs first time information, the first rule and / or the second rule may also include second time information. The second time information is used to indicate the time when at least one of the above is performed on abnormal traffic. For example, the second time information is used to indicate a third time period (e.g., indicating the operation time in FAR or QER). The first time information is associated with the second time information.
[0506] Alternatively, a local timer can be configured on the third, second, or sixth network element. This timer is started when at least one of the above-mentioned operations (such as packet loss) is performed on abnormal traffic. When the timer expires, the relevant traffic is allowed again. Subsequent observation reveals that if abnormalities reappear, at least one of the above-mentioned operations (such as packet loss) is performed on the abnormal traffic again.
[0507] For example, if the analysis or inference results of the first network element in S302 indicate that the abnormal traffic lasts only 1 minute, or if the timer configured locally by the third, second, or sixth network element is 1 minute, then after 1 minute, operations such as packet loss on the abnormal traffic will be stopped to restore it to normal. For example, the PCF will reissue the PCC rule, indicating the previous QoS parameters. Another example is that the SMF will store the previous QoS parameters locally and update the N4 rule. Yet another example is that the N4 rule includes QoS parameters for both abnormal and normal traffic processing. After the duration is reached, the UPF will perform QoS control according to the normal QoS parameters.
[0508] The S303C and the third network element handle abnormal traffic in the first session according to the second rule.
[0509] For example, the third network element performs packet loss operation on abnormal traffic according to the FAR shown in Figure 8A; or for example, the third network element performs QoS control on abnormal traffic according to the QER shown in Figure 8B.
[0510] In one possible example, the abnormal traffic in the first session originates from the terminal device. In this case, as shown in Figure 7B, the second network element updates the policy of the first session based on the identification information of the abnormal traffic, which may include:
[0511] The S303D and the second network element trigger the release process of the first session.
[0512] For example, the second network element releases the first session.
[0513] For example, the second network element sends a second message, which is used to indicate the release of the first session; the sixth network element receives the second message and releases the first session according to the second message.
[0514] Optionally, the second network element may also prohibit the terminal device corresponding to the first session from establishing a session within the first time period; or, the second network element may send third information, which is used to instruct the terminal device corresponding to the first session to establish a session within the first time period, and the sixth network element may receive the third information and prohibit the terminal device corresponding to the first session from establishing a session within the first time period according to the third information.
[0515] Optionally, the third information is specifically used to instruct the terminal device corresponding to the first session to prohibit it from establishing a session with the same attributes (such as DNN or S-NSSAI) as the first session within a first time period. The first time period can be indicated by the third information, for example, the first time period can be associated with the aforementioned second time information or the first time information; or, the first time period can also be configured locally by the third network element, and so on.
[0516] It should be noted that when there are multiple abnormal traffic flows, these multiple abnormal traffic flows may all originate from the terminal device, all from the server, or some from the terminal device and some from the server, depending on the specific implementation. Accordingly, S303 to S303C and S303D mentioned above may only be partially executed, or they may all be executed, depending on the specific implementation.
[0517] Alternatively, in S302 above, the first network element can also send analysis results or inference results (such as the identification information of abnormal traffic in the first session) to the third network element. The third network element can then directly update the policy of the first session based on the analysis results or inference results received from the first network element (such as performing packet loss or using special QoS parameters to handle abnormal traffic). That is, S303A to S303B can be skipped, and S303C or S303D can be executed directly. The second rule in S303C can be generated by the third network element based on the received analysis results or inference results.
[0518] It is understandable that the third network element may not be aware of whether the traffic indicated by the received identification information is abnormal or normal traffic. For example, in the scenario where the third network element receives the second rule, the third network element processes the relevant traffic according to the second rule without being aware of whether the traffic indicated by the second rule is abnormal.
[0519] The above design clarifies the session-based abnormal traffic handling scheme, which can achieve the effect of handling abnormal traffic in advance and dynamically, thereby improving the network's ability to handle abnormal traffic and network attacks.
[0520] To better understand the identification and handling of session-based abnormal traffic described above, here is another specific example, as shown in Figure 9, which includes the following process:
[0521] S401. The terminal device initiates a session establishment request. The session establishment request contains the identification information of the first session. The sixth network element (such as SMF) receives the session establishment request. The session establishment request is used to request the establishment of the first session.
[0522] S402, the sixth network element obtains the terminal device's subscription information from the eighth network element (such as UDM) based on the session establishment request.
[0523] S403, the sixth network element sends a third request to the second network element (such as PCF) based on the subscription information of the terminal equipment, and the second network element receives the third request.
[0524] The above S401 to S403 can be referred to the specific implementation of S201 to S203 above, and will not be repeated here.
[0525] S404. The second network element sends a first request based on the third request. The first network element (such as NWDAF) receives the first request. The first request is used to obtain the identification information of abnormal traffic in the first session.
[0526] It is understandable that S404 can also be replaced by: the second network element sends a first request to the first network element based on the attribute information of the first session (such as DNN or S-NSSAI, etc.), and the first network element receives the first request.
[0527] The above S404 can be referred to the specific implementation of S300 above, and will not be repeated here.
[0528] S405, the second network element, the sixth network element, etc., execute the remaining session establishment process.
[0529] It should be understood that S404 can be executed before or after S405. That is, during the session establishment process, if the second network element receives the third request in S403, S404 can be triggered according to the third request. Alternatively, the second network element can receive the third request but not trigger S404 immediately, but send it at any time after S403 (e.g., after the session is established).
[0530] S401 to S403 and S405 mentioned above are optional.
[0531] After receiving the first request, the first network element (S406-S408) subscribes to relevant input parameters (i.e., traffic characteristic information) from the third network element (such as UPF), the fourth network element (such as OAM), and the fifth network element (such as AF), for example, by sending the second request.
[0532] It is understandable that the first network element's subscription to AF and OAM is an optional step. Specifically, the information the first network element subscribes to from AF is optional; analysis or inference can be completed without obtaining this information, but obtaining it can improve the effectiveness of analysis or inference (e.g., improve efficiency and confidence). The information the first network element subscribes to from OAM is necessary for analysis or inference, but the relevant information can also be configured locally on the NWDAF; therefore, the step of the first network element subscribing to information from OAM is optional.
[0533] For the specific implementation of S406 to S408, please refer to the specific implementation of S301A to S301B mentioned above.
[0534] S409. The first network element notifies the second network element of the analysis or inference results.
[0535] The analysis results or inference results (i.e., output parameters) include: identification information of abnormal traffic in the first session.
[0536] Optionally, the analysis results or inference results may also include: any one or more of the start time, end time, and duration of the abnormal traffic; the source of the abnormal traffic (such as from a terminal device or server); the confidence level of the analysis results or inference results (e.g., how accurate the AI model is in its inference results), etc.
[0537] For a detailed implementation of S409, please refer to the relevant description of S302 above.
[0538] If the abnormal traffic originates from the server, then execute steps S410 through S412:
[0539] S410, the second network element sends a first rule based on the identification information of the abnormal traffic in the first session; correspondingly, the sixth network element (such as SMF) receives the first rule. The first rule, for example, is a PCC rule.
[0540] S411. The sixth network element sends the second rule according to the first rule, and the third network element (such as UPF) receives the second rule. The second rule is, for example, the N4 rule.
[0541] S412, the third network element handles the abnormal traffic of the first session according to the second rule.
[0542] The specific implementation methods of S410 to S412 can be referred to the specific implementation methods of S303A to S303C above, which will not be repeated here.
[0543] If the abnormal traffic originates from the terminal device, then execute S413:
[0544] S413, the second network element triggers the release process of the first session.
[0545] The specific implementation of S413 can be found in the implementation of S303D above, and will not be repeated here.
[0546] The steps marked with dashed lines in Figure 9 are optional steps.
[0547] In the above process, the second network element subscribes to the abnormal traffic analysis of the first session from the first network element and handles the abnormal traffic through policy instructions, which can achieve the effect of handling abnormal traffic in advance and dynamically.
[0548] Referring to Figure 10, which is a flowchart of another communication method provided in an embodiment of this application, this method can realize abnormal traffic identification based on network elements. The method includes S501 to S502:
[0549] S501, the first network element (such as NWDAF) obtains the traffic characteristic information of the third network element (such as UPF).
[0550] The traffic characteristic information of the third network element here can be the characteristic information of all traffic passing through the third network element (such as the traffic characteristic information of all sessions on the third network element), or it can be the characteristic information of some traffic passing through the third network element (such as the traffic characteristic information of some sessions or some services on the third network element), without restriction. For ease of description, the following text mainly takes the example of the traffic characteristic information of the third network element being the traffic characteristic information of all sessions passing through the third network element.
[0551] It should be understood that the traffic characteristic information of the third network element includes the current traffic characteristic information of the third network element, such as the traffic characteristic information generated within the current preset time range.
[0552] For details regarding the specific content of the traffic characteristic information, please refer to the detailed description of the traffic characteristic information in the embodiment shown in Figure 2 above, which will not be elaborated here. The difference is that Figure 2 shows the traffic characteristic information of the first session, while this embodiment shows the traffic characteristic information of the third network element.
[0553] S502. The first network element sends the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element.
[0554] Optionally, sending can be replaced by outputting or determining. S502 can be replaced by: the first network element determining the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element; or, the first network element outputting the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element.
[0555] Optionally, the identification information for abnormal traffic of the third network element can be one or more of the following: application identifier, IP 5-tuple, IP triplet, domain name (FQDN / URL / URI), etc., without restriction.
[0556] It is understood that, in the embodiments of this application, the process by which the first network element determines / outputs / sends the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element can be called abnormal traffic analysis (analytic / analyze), abnormal traffic identification, abnormal traffic inference, or abnormal traffic prediction, etc., and other names can also be used without limitation.
[0557] The specific implementation method of the first network element determining / outputting / sending the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element can refer to the specific implementation method of the first network element determining / outputting / sending the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the first session in S302 above, and will not be elaborated here. Unlike the embodiment shown in Figure 2 above, which is based on session (or session granularity) traffic characteristic analysis, this embodiment is based on third network element (or network element granularity) traffic characteristic analysis.
[0558] This application provides an abnormal traffic identification scheme based on third-party network elements (or third-party network element granularity). Specifically, a first network element acquires traffic characteristic information from a third network element and sends an identification information for the abnormal traffic of that third network element based on this information. This scheme clarifies the specific method for identifying abnormal traffic in mobile networks, can be used to standardize the process of abnormal traffic analysis in mobile networks, can more promptly cover all UPFs in the network, and improve the ability of mobile networks to handle abnormal traffic and network attacks.
[0559] In one possible design, the first network element can determine / output / send the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element, using inference or analysis. The result determined / output / sent by the first network element based on the traffic characteristic information of the third network element can be called the analysis result or the inference result.
[0560] For example, the first network element uses the traffic characteristic information of the third network element as the input parameter of the model, inputs the traffic characteristic information of the third network element into the model, runs the model, and outputs the inference result, which includes the identification information of the abnormal traffic of the third network element.
[0561] For example, the first network element uses the traffic characteristic information of the third network element as input parameters to perform analysis. The analysis results include the identification information of abnormal traffic of the third network element (the analysis process can be the internal logic of the first network element).
[0562] Because the first network element is used to provide data analysis functions and has reasoning capabilities, it can realize the identification of abnormal traffic based on sessions using AI functions, and can achieve the effect of early and dynamic identification of abnormal traffic based on AI reasoning.
[0563] This design clarifies specific schemes for improving network identification of abnormal traffic through analysis or reasoning. It can be used to standardize the process of NWDAF assisting UPF in abnormal traffic analysis (for example, the first network element is NWDAF, and in S502, NWDAF sends the identification information of abnormal traffic of UPF to UPF, so that all UPFs in the network can be covered more timely), and further improve the network's ability to handle abnormal traffic and network attacks.
[0564] In one possible design, the first network element can also acquire historical traffic characteristic information, which is used to assist in determining (such as analyzing or inferring) the identification information of abnormal traffic in the first session.
[0565] Historical traffic characteristic information includes, for example, traffic characteristic information generated before a preset time range. Historical traffic characteristic information may include historical traffic characteristic information of third network elements, as well as historical traffic characteristic information of other network elements, such as historical traffic characteristic information of one or more UPFs.
[0566] For example, the historical traffic characteristics of other network elements and the traffic characteristics of the third network element are input into the model, and the model infers and outputs the identification information of the abnormal traffic of the third network element.
[0567] Alternatively, for example, the historical traffic characteristics of other network elements can be input into the model for training. After obtaining the trained model, the traffic characteristics of the third network element can be input into the trained model, and the trained model can infer and output the identification information of the abnormal traffic of the third network element.
[0568] This design utilizes historical traffic characteristics to help determine the identification information of abnormal traffic in third network elements, which can further improve the accuracy of abnormal traffic identification.
[0569] In one possible design, the first network element, based on the traffic characteristic information of the third network element, in addition to determining / outputting / sending the identification information of the abnormal traffic of the third network element, can also determine / output / send at least one of the following (or the analysis result or reasoning result can also include at least one of the following):
[0570] 1) Third-time information: Third-time information includes one or more of the start time, end time, or duration of abnormal traffic of a third network element. For a detailed description, please refer to the first-time information above.
[0571] In some embodiments, the third time information can also indicate an applicable time period. For example, it could indicate the applicable time period for analysis results or inference results. Alternatively, the third time information could indicate a fourth time period.
[0572] 2) Sources of abnormal traffic from third-party network elements. For a detailed description, please refer to the section on sources of abnormal traffic in the first session above.
[0573] 3) Confidence level of abnormal traffic analysis results for the third network element. For a detailed description, please refer to the confidence level of abnormal traffic analysis results in the first session above.
[0574] This design enables the first network element to send more information about abnormal traffic based on the traffic characteristics of the third network element, which helps to further improve the network's ability to handle abnormal traffic and network attacks.
[0575] In one possible design, the timing or triggering conditions for the first network element to send the identification information of abnormal traffic from the third network element can be varied, such as a combination of one or more of the following:
[0576] 1) If the analysis or reasoning results indicate that there is abnormal traffic in the third network element, the first network element sends the abnormal traffic identification information of the third network element.
[0577] This makes it easy to identify abnormal traffic and is simple to implement.
[0578] 2) The first network element sends the identification information of all traffic and indicates whether each traffic is normal or abnormal;
[0579] This clearly identifies all traffic anomalies within the third network element, facilitating centralized management.
[0580] 3) If there is abnormal traffic in the third network element and the confidence level of the abnormal traffic exceeds the first threshold, the first network element sends the identification information of the abnormal traffic of the third network element.
[0581] This avoids frequently triggering abnormal traffic processing, saving resource consumption.
[0582] 4) If the duration of abnormal traffic exceeds the second threshold or the current time is less than the start time of abnormal traffic than the third threshold, the first network element sends the abnormal traffic identification information of the third network element.
[0583] In this way, abnormal traffic can be handled in a timely manner while taking into account resource consumption.
[0584] Of course, the above are just examples, and there are actually more than these.
[0585] The methods for indicating abnormal traffic can be found in the previous descriptions, and will not be repeated here.
[0586] The above design clarifies the timing or triggering conditions for the first network element to send abnormal traffic identification information to the third network element, thus improving the reliability of the solution.
[0587] In one possible design, S501 to S502 may be executed in response to a request from a third network element.
[0588] For example, as shown in Figure 11, before S501, it also includes:
[0589] S500, the third network element sends the fourth request, and the first network element receives the fourth request.
[0590] The fourth request is used to obtain the identification information of abnormal traffic of the third network element. The fourth request includes an analysis ID, which is used to indicate abnormal traffic analysis.
[0591] Accordingly, after receiving the fourth request, the first network element responds to the fourth request by executing S501 to S502. Further, S502 may specifically include: the first network element sending the identification information of the abnormal traffic of the third network element according to the traffic characteristic information of the third network element, and the third network element receiving the identification information of the abnormal traffic of the third network element.
[0592] Optionally, the fourth request is to subscribe to messages.
[0593] Alternatively, the first network element can also send the identification information of the abnormal traffic of the third network element to other network elements, such as the second network element, without restriction.
[0594] Alternatively, the fourth request can also come from other network elements, such as the second network element.
[0595] Furthermore, the specific implementation of the fourth request can also refer to the first request mentioned above.
[0596] In the above design, the first network element executes S501 to S5302 based on the request from the third network element, which can realize the on-demand triggering of abnormal traffic analysis, improve network security performance, and simultaneously avoid the problem of resource waste caused by frequent triggering of abnormal traffic analysis.
[0597] In one possible design, the first network element obtains the traffic characteristic information of the first session from other network elements.
[0598] For example, as shown in FIG12, S501 may include:
[0599] S501A, the first network element sends a fifth request, which is used to obtain traffic characteristic information (including traffic characteristic information of the third network element and / or historical traffic characteristic information, etc.). Correspondingly, at least one network element (such as at least one of the third network element, the fourth network element, or the fifth network element) receives the fifth request.
[0600] Optionally, the fifth request is to subscribe to messages.
[0601] S501B: At least one network element sends traffic characteristic information, and the first network element receives traffic characteristic information from the third network element.
[0602] Optionally, the fifth request is used to obtain the traffic characteristic information of the third network element. The fifth request includes the identification information of the third network element (such as the address of the third network element), so that the receiver can clearly return the traffic characteristic information of which session.
[0603] For the specific implementation of the first network element obtaining traffic characteristic information from at least one of the third, fourth, or fifth network elements, please refer to the specific implementation of the first network element obtaining traffic characteristic information from at least one of the third, fourth, or fifth network elements in the relevant embodiment of Figure 2 above, which will not be described in detail here.
[0604] In this design, the first network element requests (e.g., subscribes to) the traffic characteristic information of the third network element from other network elements, providing support for the first network element to perform abnormal traffic analysis based on the traffic characteristic information of the third network element, thus improving the reliability of the solution.
[0605] In one possible design, the identification information of abnormal traffic from a third network element is used to update the third network element's policy (or session policy). Updating the third network element's policy includes, but is not limited to, updating one or more of the following: PCC rules and N4 rules. N4 rules include, but are not limited to, PDR, FAR, QER, and URR.
[0606] As shown in Figure 13, depending on the receiving end of the abnormal traffic identification information of the third network element, the specific implementation method of the strategy for updating the third network element can be different. Several possible methods are listed below:
[0607] Method 1: The first network element sends the identification information of the abnormal traffic of the third network element, the third network element receives the identification information of the abnormal traffic of the third network element (S502A), the third network element forwards the identification information of the abnormal traffic of the third network element to the second network element (S503A), and the second network element updates the policy of the third network element according to the identification information of the abnormal traffic of the third network element (S504A).
[0608] It is understandable that if the first network element sends other information (such as third-time information, the source of abnormal traffic of the third network element, the confidence level of the abnormal traffic analysis results of the third network element, etc.), then the third network element can also send the other information to the second network.
[0609] In one possible example, the abnormal traffic from the third network element originates from the server. In this case, the second network element updates the policy of the first session based on the identification information of the abnormal traffic from the first session, which may include the following process:
[0610] A. The second network element sends the third rule based on the abnormal traffic identification information of the third network element, and the sixth network element receives the third rule;
[0611] The third rule includes identification information for abnormal traffic of the third network element and a fifth piece of information. The fifth piece of information is used to indicate at least one of the following:
[0612] Disable abnormal traffic (specific actions to disable abnormal traffic may include packet loss, adjusting QoS parameters, etc.);
[0613] Packet loss is applied to abnormal traffic;
[0614] Use a second QoS parameter (which may include 5QI) for abnormal traffic;
[0615] Use a separate Quality of Service (QoS) stream for abnormal traffic;
[0616] Implement caching for abnormal traffic.
[0617] The third network element handles abnormal traffic according to the fourth rule.
[0618] Optionally, the third rule may also include fourth time information, which indicates the time at which at least one action is taken against abnormal traffic, and is associated with the third time information. For example, the fourth time information indicates a fourth time period.
[0619] The third rule can be referenced from the first rule mentioned above.
[0620] B. The sixth network element sends the fourth rule according to the third rule, and the third network element receives the fourth rule.
[0621] The fourth rule includes identification information for abnormal traffic of the third network element and the eighth information. The eighth information is used to indicate at least one of the following: disabling abnormal traffic; performing packet loss on abnormal traffic; using the second QoS parameter on abnormal traffic; using a separate QoS flow on abnormal traffic; or performing caching on abnormal traffic.
[0622] Optionally, the fourth rule may also include fifth time information, which indicates the time at which at least one of the abnormal traffic measures is performed, and is associated with the fourth time information. For example, the fifth time information indicates the fourth time period.
[0623] C. The third network element handles abnormal traffic according to the fourth rule.
[0624] The fourth rule can be found in the second rule mentioned above.
[0625] In another possible example, the abnormal traffic from the third network element originates from the terminal device. In this case, the second network element updates the policy of the first session based on the identification information of the abnormal traffic in the first session, which may include one or more of the following:
[0626] 1) The second network element releases the session of the terminal device, which is the terminal device corresponding to the abnormal traffic.
[0627] 2) The second network element sends the sixth information, which is used to indicate the release of the terminal device's session. The terminal device is the terminal device corresponding to the abnormal traffic. The sixth network element receives the sixth information and releases the session of the terminal device corresponding to the abnormal traffic of the third network element according to the sixth information.
[0628] 3) The second network element prohibits the terminal device corresponding to the abnormal traffic from establishing a session in the second time period; or, the second network element sends a seventh message, which is used to instruct the terminal device corresponding to the abnormal traffic to be prohibited from establishing a session in the second time period. The sixth network element receives the seventh message and prohibits the terminal device corresponding to the abnormal traffic from establishing a session in the second time period according to the seventh message.
[0629] Method 2: The first network element sends the identification information of the abnormal traffic of the third network element, the second network element receives the identification information of the abnormal traffic of the third network element (S502B), and the second network element updates the policy of the third network element according to the identification information of the abnormal traffic of the third network element (S503B).
[0630] The first network element can determine the second network element based on the identification information of the abnormal traffic. For example, if the identification information of the abnormal traffic includes an IP 5-tuple, the first network element can query the second network element from the ninth network element (such as BSF) based on the UE IP in the IP 5-tuple. After that, the first network element can send the identification information of the abnormal traffic of the third network element to the second network element.
[0631] It is understandable that if the first network element sends other information (such as third-time information, the source of abnormal traffic of the third network element, the confidence level of the abnormal traffic analysis results of the third network element, etc.), the second network element can also update the strategy of the third network element in combination with other information.
[0632] The specific implementation of the strategy for the second network element to update the third network element based on the identification information of the abnormal traffic of the third network element can be found in the relevant description in Method 2, and will not be repeated here.
[0633] Method 3: The first network element sends the identification information of the abnormal traffic of the third network element, the third network element receives the identification information of the abnormal traffic of the third network element (S502C), and the third network element updates the processing rules according to the identification information of the abnormal traffic of the third network element (S503C).
[0634] In one possible example, the abnormal traffic from the third network element originates from the server. In this case, the third network element updates its processing rules based on the identification information of the abnormal traffic, which may include at least one of the following:
[0635] Disable abnormal traffic;
[0636] Packet loss is applied to abnormal traffic;
[0637] Use the second QoS parameter for abnormal traffic;
[0638] Use a separate Quality of Service (QoS) stream for abnormal traffic;
[0639] Implement caching for abnormal traffic.
[0640] The details of the above information can be found in the text above and will not be repeated here.
[0641] In one possible example, the abnormal traffic from the third network element originates from a terminal device. In this case, the third network element updates its policy based on the identification information of the abnormal traffic, which may include: the third network element triggering the sixth network element or the second network element to release the session of the terminal device corresponding to the abnormal traffic of the third network element.
[0642] The third network element can determine the N4 context (session granularity context between the third and sixth network elements) based on the UE address in the abnormal traffic identification information (e.g., the IP 5-tuple). It can then indicate to the sixth network element, through the N4 context corresponding to the N4 session, that the UE has abnormal traffic, causing the sixth network element to release the UE's session. Alternatively, the first network element can send the UE address corresponding to the abnormal traffic, and the third network element can notify the sixth network element that the UE at that address has abnormal traffic, causing the sixth network element to release the UE's session.
[0643] Optionally, the third network element can also instruct the sixth network element to prohibit the terminal device corresponding to the abnormal traffic from establishing a session in the second time period.
[0644] It is understandable that the third network element may not be aware of whether the traffic indicated by the received identification information is abnormal or normal traffic. For example, in the scenario where the third network element receives the fourth rule, the third network element processes the relevant traffic according to the fourth rule and is unaware of whether the traffic indicated by the second rule is abnormal.
[0645] The above design clarifies the abnormal traffic handling scheme based on third network elements, which can achieve the effect of handling abnormal traffic in advance and dynamically, and improve the network's ability to handle abnormal traffic and network attacks.
[0646] To better understand the identification and handling of session-based abnormal traffic described above, here is another specific example, as shown in Figure 14, which includes the following process:
[0647] S601, the third network element (such as UPF) sends the fourth request, and the first network element (such as NWDAF) receives the fourth request;
[0648] The fourth request is used to obtain the identification information of abnormal traffic of the third network element.
[0649] For a detailed implementation of S601, please refer to the previous section on S500.
[0650] S602-S604, the first network element obtains traffic characteristic information (including traffic characteristic information of the third network element, and optionally historical traffic characteristic information) from the third network element, the fourth network element (such as OAM), or the fifth network element (such as AF), for example:
[0651] S602, The first network element obtains the traffic characteristic information of the third network element from the third network element;
[0652] S603, The first network element obtains historical traffic characteristic information from the fourth network element;
[0653] S604. The first network element obtains the traffic characteristic information of the third network element from the fifth network element;
[0654] It is understandable that the traffic characteristics of the third network element in S602 and S604 may be different.
[0655] For the specific implementation of S602-S604, please refer to the previous sections S501A to S501B.
[0656] S605. The first network element determines the analysis result or inference result (including the identification information of abnormal traffic of the third network element) based on the obtained traffic characteristic information.
[0657] When abnormal traffic from a third network element originates from a server, there are three possible scenarios:
[0658] Method 1, S606-S607:
[0659] S606. The first network element notifies the third network element of the analysis results or inference results (including the identification information of abnormal traffic of the third network element);
[0660] After receiving the notification, S607 and the third network element discard all packets corresponding to the abnormal traffic.
[0661] Method 2, S606, S608~S611:
[0662] S606. The first network element notifies the third network element of the analysis results or inference results (including the identification information of abnormal traffic of the third network element);
[0663] S608 After receiving the notification, the third network element will send the analysis results or inference results (including the identification information of the abnormal traffic of the third network element) to the second network element (such as PCF) through the sixth network element (such as SMF);
[0664] S609, The second network element updates the PCC rules to the sixth network element;
[0665] S610, the sixth network element updates the N4 rules to the third network element;
[0666] S611, the third network element processes abnormal traffic according to the updated N4 rules.
[0667] Method 3, S612~S613:
[0668] S612, the first network element queries the second network element corresponding to the UE IP from the ninth network element (such as BSF);
[0669] S613. The first network element sends the analysis results or inference results (including the identification information of abnormal traffic of the third network element) to the second network element;
[0670] S614, Strategy for updating the third network element from the second network element.
[0671] When abnormal traffic from a third network element originates from a terminal device, there are two possibilities:
[0672] When NWDAF determines that abnormal traffic originates from the UE, there are two methods:
[0673] Method 1, S615-S617:
[0674] S615. The first network element can query the sixth network element corresponding to the terminal device identifier from the UDM.
[0675] S616, The third network element triggers the sixth network element to release the session of the terminal device corresponding to the abnormal traffic of the third network element;
[0676] S617. The sixth network element releases the session of the terminal device corresponding to the abnormal traffic of the third network element.
[0677] Method 2, S618-S620:
[0678] S618. The first network element queries the second network element corresponding to the UE IP from the ninth network element (such as BSF);
[0679] S619. The first network element sends the analysis results or inference results (including the identification information of the abnormal traffic of the third network element) to the second network element, and the second network element receives the analysis results or inference results.
[0680] S620, the second network element releases or triggers the sixth network element to release the abnormal traffic of the third network element, corresponding to the terminal device's session.
[0681] For specific implementation methods of various cases in S606 to S620, please refer to the specific implementation methods of the corresponding cases in the embodiment of Figure 13 above.
[0682] The steps marked with dashed lines in Figure 14 are optional steps.
[0683] In the above process, the third network element can subscribe to the abnormal traffic analysis of the third network element from the first network element and process the abnormal traffic, thereby achieving the effect of handling abnormal traffic in advance and dynamically.
[0684] Referring to Figure 15, which is a flowchart of another communication method provided in an embodiment of this application, this method enables a third network element to perform abnormal traffic identification locally. The method includes S701 to S702:
[0685] S701, the first network element (such as NWDAF) sends the model, and the third network element (such as UPF) receives the model from the first network element. The model is used for abnormal traffic analysis.
[0686] Optionally, the third network element can send a sixth request to obtain a model. This sixth request includes an analysis identifier, which indicates abnormal traffic analysis. Optionally, the first network element sends a model corresponding to the analysis identifier to the third network element. Optionally, the sixth request is a subscription message.
[0687] S702, the third network element determines the identification information of abnormal traffic based on the model.
[0688] Specifically, the third network element outputs inference results through model inference, which include identification information of abnormal traffic.
[0689] Optionally, the inference results of the model may also include (i.e., the third network element may also be determined based on the model) at least one of the following:
[0690] 1) Sixth time information, which includes one or more of the start time, end time, or duration of abnormal traffic;
[0691] The implementation of the sixth time information can be referenced from the previous sections on the first time information or the fourth time information.
[0692] 2) Sources of abnormal traffic;
[0693] The source of abnormal traffic can be found in the previous section on the implementation of abnormal traffic sources.
[0694] 3) Confidence level of abnormal traffic analysis results.
[0695] The confidence level of the abnormal traffic analysis results can be referenced in the previous section on the implementation of the confidence level of abnormal traffic analysis results.
[0696] In specific implementations, the third network element can identify abnormal traffic at different granularities as needed. For example, if the identification granularity is at the session level, the abnormal traffic identification information determined by the third network element based on the model can be the abnormal traffic identification information of a certain session (such as the first session); if the identification granularity is at the network element level, the abnormal traffic identification information determined by the third network element based on the model can be the abnormal traffic identification information of a certain network element (such as the third network element); if the identification granularity is at the node level, the abnormal traffic identification information determined by the third network element based on the model can be the abnormal traffic identification information of a certain node (such as the first node); if the identification granularity is at the service level, the abnormal traffic identification information determined by the third network element based on the model can be the abnormal traffic identification information of a certain service (such as the first service), and so on. This application does not impose any restrictions.
[0697] For details on how the third network element determines the identification information of abnormal traffic based on the model, please refer to the details on how the first network element determines the identification information of abnormal traffic based on the model in S302 or S502 above.
[0698] For example, the third network element can determine the inference result based on traffic feature information and the model (the inference result includes identification information of abnormal traffic). The traffic feature information includes, but is not limited to, one or more of the following: protocol fingerprint information, service behavior information, service server information, packet protocol information, domain name / address blacklists, suspicious / abnormal domain names / addresses, and QoS-related information of the traffic. Optionally, at least one of the traffic feature information satisfies a preset condition.
[0699] For example, the third network element can also acquire historical traffic characteristic information. Based on the traffic characteristic information, historical traffic characteristic information, and the model, the third network element can determine the inference result (the inference result includes the identification information of abnormal traffic). Using historical traffic characteristic information to assist in the identification of abnormal traffic can further improve the accuracy of abnormal traffic identification.
[0700] This application provides a scheme for a third network element to acquire a model and perform local inference. Specifically, the third network element acquires the model from a first network element and determines the identification information of abnormal traffic based on the model. This scheme clarifies the specific method for mobile networks to utilize AI functions to identify abnormal traffic, achieving the effect of early and dynamic identification of abnormal traffic based on AI inference. This scheme can be used to standardize the process of abnormal traffic analysis in mobile networks, enabling more timely coverage of all UPFs in the network and improving the mobile network's ability to handle abnormal traffic and network attacks.
[0701] In one possible design, the third network element can send inference results (which contain identification information of abnormal traffic) to update the session's policy.
[0702] For example, a third network element can send inference results (which include identification information of abnormal traffic) to a second or sixth network element, etc., so that the second or sixth network element can update the session policy based on the inference results (which include identification information of abnormal traffic). For the specific implementation of the second or sixth network element updating the session policy, please refer to the relevant descriptions in the embodiments above, which will not be detailed here.
[0703] In this design, the third network element notifies the designated network element (such as the second or sixth network element) of the inference result, and the designated network element updates the session's strategy. This approach is easily adaptable to the design logic of existing mobile networks and requires minimal modification.
[0704] In another possible design, the third network element can update the processing rules based on the inference results (which include identification information of abnormal traffic).
[0705] For example, when abnormal traffic originates from a server, the third network element may perform at least one of the following:
[0706] Disable abnormal traffic;
[0707] Packet loss is applied to abnormal traffic;
[0708] Use a third QoS parameter for abnormal traffic;
[0709] Use a separate QoS stream for abnormal traffic;
[0710] Implement caching for abnormal traffic.
[0711] The implementation of the above information can be found in the relevant content above, and will not be repeated here.
[0712] Optionally, the third network element may perform at least one of the above actions during the fifth time period.
[0713] For example, when abnormal traffic originates from a terminal device, the third network element triggers the sixth network element or the second network element to release the session of the terminal device corresponding to the abnormal traffic. For instance, the third network element sends a ninth message, which instructs the release of the session of the terminal device corresponding to the abnormal traffic, and the sixth network element receives the ninth message; the sixth network element releases the session of the terminal device corresponding to the abnormal traffic according to the ninth message, or the sixth network element sends the ninth message to the second network element, and the second network element releases the session of the terminal device corresponding to the abnormal traffic according to the ninth message.
[0714] This design allows the third network element to directly update the processing rules based on the inference results, which can reduce signaling interactions and save resources.
[0715] Optionally, the third network element can also acquire sixth time information, which is used to indicate the effective time (or applicable time) of traffic characteristic information, models, or abnormal traffic identification information, etc. Optionally, the sixth time information is used to indicate the fifth time period.
[0716] It should be understood that the specific implementation of the update strategy or rule can be referred to the relevant description in the above embodiments, and will not be described in detail here.
[0717] Alternatively, the methods described above, which are performed by the third network element, can also be performed by the firewall of the third network element. The firewall of the third network element can be integrated into the same physical entity as the third network element, or it can be deployed in different physical entities (e.g., as a separate network element, such as the seventh network element), without any restrictions.
[0718] To better understand the identification and handling of session-based abnormal traffic described above, here is another specific example, as shown in Figure 16, which includes the following process:
[0719] S801: The third network element subscribes to the model used for abnormal traffic analysis from the first network element;
[0720] S802~S803: The third network element subscribes to historical traffic characteristic information from the fourth network element and subscribes to traffic characteristic information from the fifth network element;
[0721] It is understandable that the third network element can also obtain traffic characteristic information from the local area; S802 and S803 are optional steps.
[0722] S804: The third network element inputs the acquired information into the model, determines the inference result based on the model, and the inference result includes at least the identification information of abnormal traffic.
[0723] Furthermore, the third network element can also handle abnormal traffic:
[0724] When abnormal traffic originates from the server, there are three possibilities:
[0725] Method 1, S805: The third network element directly discards all packets corresponding to the abnormal traffic;
[0726] Method 2, S806-S809:
[0727] S806, the third network element notifies the second network element corresponding to the session to which the abnormal traffic belongs of the inference result;
[0728] S807, The second network element updates the PCC rules to the sixth network element;
[0729] S808, the sixth network element updates the N4 rules to the third network element;
[0730] S809 and the third network element handle abnormal traffic according to the updated N4 rules.
[0731] Method 3, S810-S811:
[0732] S810 and the third network element will notify the firewall (such as the seventh network element) of the inference results;
[0733] S811: The firewall directly drops all packets corresponding to abnormal traffic.
[0734] When abnormal traffic originates from the UE, S812 can be executed:
[0735] S812: The third network element triggers the sixth network element to release the session corresponding to the abnormal traffic.
[0736] The steps marked with dashed lines in Figure 16 are optional steps.
[0737] In the above process, the third network element subscribes to the model from the first network element and performs inference locally to analyze and process abnormal traffic, achieving the effect of anticipating and dynamically processing abnormal traffic based on AI inference.
[0738] It is understood that the above embodiments are introduced starting from the first network element side. In actual applications, the methods and steps executed by each network element can be implemented separately.
[0739] For example, embodiments of this application also provide a communication method, including:
[0740] The third network element receives the second rule, which includes identification information of the traffic of the first session and fourth information. The fourth information is used to indicate at least one of the following: disable the traffic; perform packet loss on the traffic; perform caching on the traffic; use the first QoS parameter on the traffic; or use a separate QoS flow on the traffic.
[0741] The third network element processes the traffic of the first session according to the second rule.
[0742] In one possible design, this traffic is considered abnormal traffic.
[0743] In one possible design, the traffic of the first session is processed according to the second rule during the third time period.
[0744] In one possible design, a second time information is also received. This second time information is used to indicate the effective or applicable time of the second rule and / or the fourth information. Alternatively, the second time information is used to indicate a third time period.
[0745] In one possible design, the second rule and / or fourth information includes second time information.
[0746] In one possible design, processing the traffic of the first session according to the second rule includes: processing the traffic / service flow corresponding to the identification information of the traffic of the first session according to the identification information of the traffic of the first session in the second rule.
[0747] In one possible design, processing the traffic of the first session according to the second rule includes: according to the fourth information in the second rule, performing at least one of the following on all traffic / service flows in the first session or traffic / service flows corresponding to the identification information of the traffic in the first session: disabling the traffic; performing packet loss on the traffic; performing caching on the traffic; applying the first QoS parameter to the traffic; or applying a separate QoS flow to the traffic.
[0748] In one possible design, the second rule is the N4 rule.
[0749] In one possible design, a second rule is received from the sixth network element.
[0750] For details on the implementation of the above methods, please refer to the corresponding content above; they will not be elaborated here.
[0751] For example, embodiments of this application also provide a communication method, including:
[0752] The third-party network element obtains the identification information of the traffic;
[0753] Perform at least one of the following on the traffic: disable the traffic; perform packet loss on the traffic; perform caching on the traffic; apply a second QoS parameter to the traffic; or apply a separate QoS stream to the traffic.
[0754] In one possible design, the traffic is abnormal traffic.
[0755] In one possible design, at least one of the above is performed on the traffic during the fourth time period.
[0756] In one possible design, fifth time information is obtained, which is used to indicate the effective time or applicable time of the traffic identification information.
[0757] In one possible design, the fifth time information is used to indicate the fourth time period.
[0758] For details on the implementation of the above methods, please refer to the corresponding content above; they will not be elaborated here.
[0759] For example, embodiments of this application also provide a communication method, including:
[0760] The third network element obtains traffic characteristic information of the first node or the first service;
[0761] Based on the traffic characteristic information, the third network element performs at least one of the following actions on the traffic of the first node or the first service: disable the traffic; perform packet loss on the traffic; perform caching on the traffic; apply a third QoS parameter to the traffic; or apply a separate QoS stream to the traffic.
[0762] In one possible design, the traffic of the first node or the first service is abnormal traffic.
[0763] In one possible design, at least one of the above-mentioned actions is performed on the traffic of the first node or the first service during the fifth time period.
[0764] In one possible design, a sixth time information is obtained, which is used to indicate the effective time or applicable time of the traffic characteristic information.
[0765] In one possible design, the sixth time information is used to indicate the fifth time period.
[0766] In one possible design, the traffic of the first node or the first service is determined to be abnormal traffic based on the traffic characteristic information of the first node or the first service.
[0767] In one possible design, the load, congestion, or impact on the first node is determined based on the traffic characteristics of the first node or the first service.
[0768] Here, "first node load" can refer to the load on the first node exceeding a threshold. "First node congestion" can refer to the congestion level / grade of the first node exceeding a threshold. "First node impact" can refer to the first node being impacted by abnormal traffic, malicious traffic, or a large amount of traffic.
[0769] For details on the implementation of the above methods, please refer to the corresponding content above; they will not be elaborated here.
[0770] Of course, the above are just some examples, and the actual situation is not limited to these.
[0771] The above embodiments can be implemented individually or in combination.
[0772] The methods provided by the embodiments of this application have been described above with reference to the accompanying drawings. The apparatus provided by the embodiments of this application will be described below with reference to the accompanying drawings.
[0773] Based on the same technical concept, embodiments of this application provide a communication device, which includes a module / unit / means for executing the method performed by any network element in the above method embodiments. This module / unit / means can be implemented in software, or in hardware, or implemented in hardware executing corresponding software.
[0774] For example, referring to FIG17, the device may include a transceiver module 1101 and a processing module 1102.
[0775] For example, when the device is used to implement the method described in S301 to S302 above:
[0776] The transceiver module 1101 is used to obtain traffic characteristic information of the first session;
[0777] The processing module 1102 is used to control the transceiver module 1101 to send the identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session.
[0778] For example, when the device is used to implement the method described in S501 to S502 above:
[0779] The transceiver module 1101 is used to acquire traffic characteristic information of the third network element;
[0780] The processing module 1102 is used to control the transceiver module 1101 to send the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element.
[0781] For example, when the device is used to implement the method described in S701 to S702 above:
[0782] Processing module 1102 is used to receive a model from the first network element, and the model is used for abnormal traffic analysis;
[0783] The transceiver module 1101 is used to determine the identification information of abnormal traffic based on the model.
[0784] Of course, the above are just examples of devices corresponding to some of the methods and steps. Devices corresponding to other methods and steps can be found in the above text, and will not be elaborated here.
[0785] It should be understood that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.
[0786] In practical implementation, the above-mentioned device can take many product forms. Several possible product forms are introduced below.
[0787] As shown in Figure 18, this application embodiment also provides a communication device, including:
[0788] At least one processor 1201; and a communication interface 1203 communicatively connected to the at least one processor 1201; the at least one processor 1201 causes the device to perform the method steps in the above method embodiment through the communication interface 1203 by executing instructions stored in the memory 1202.
[0789] Optionally, the memory 1202 is located outside the device.
[0790] Optionally, the device includes the memory 1202, which is connected to the at least one processor 1201, and stores instructions executable by the at least one processor 1201. Figure 18 shows, with dashed lines, that the memory 1202 is optional for the device.
[0791] The processor 1201 and the memory 1202 can be coupled through an interface circuit or integrated together; no restriction is imposed here.
[0792] This embodiment does not limit the specific connection medium between the processor 1201, memory 1202, and communication interface 1203. In Figure 18, the processor 1201, memory 1202, and communication interface 1203 are connected via a bus 1204, which is represented by a thick line. The connection methods between other components are for illustrative purposes only and are not intended to be limiting. The bus can be classified as an address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used in Figure 18, but this does not indicate that there is only one bus or one type of bus.
[0793] Based on the same technical concept, this application also provides a chip, as shown in Figure 19. This chip may include logic circuitry and input / output interfaces. Optionally, it may also include a memory. The input / output interfaces can be used to receive code instructions (stored in the memory, which can be read directly from the memory or through other devices) and transmit them to the logic circuitry; the logic circuitry can be used to execute the code instructions to perform the methods described in the above method embodiments.
[0794] It should be understood that the processor mentioned in the embodiments of this application can be implemented in hardware or software. When implemented in hardware, the processor can be a logic circuit, integrated circuit, etc. When implemented in software, the processor can be a general-purpose processor, implemented by reading software code stored in memory.
[0795] For example, the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.
[0796] It should be understood that the memory mentioned in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).
[0797] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) can be integrated into the processor.
[0798] It should be noted that the memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.
[0799] Based on the same technical concept, this application also provides a computer-readable storage medium storing a computer program or instructions, which, when executed by a communication device, implements the method steps described in the above method embodiments.
[0800] Based on the same technical concept, this application also provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are run by a communication device, the method steps in the above method embodiments are executed.
[0801] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0802] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.
[0803] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.
[0804] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.
Claims
1. A communication method, characterized in that, include: Obtain traffic characteristic information for the first session; Based on the traffic characteristic information of the first session, send the identification information of the abnormal traffic of the first session.
2. The method as described in claim 1, characterized in that, Also includes: A first request is received from a second network element. The first request is used to obtain the identification information of abnormal traffic of the first session. The first request includes the identification information of the first session and an analysis identifier, and the analysis identifier is used to indicate abnormal traffic analysis.
3. The method as described in claim 2, characterized in that, The second network element is used to provide policy control functions.
4. The method according to any one of claims 1-3, characterized in that, The traffic characteristic information of the first session includes one or more of the following: protocol fingerprint information, service behavior information, service server information, message protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and traffic quality of service related information.
5. The method as described in claim 4, characterized in that, At least one of the traffic characteristic information of the first session satisfies a preset condition.
6. The method according to any one of claims 1-5, characterized in that, Also includes: Obtain historical traffic characteristic information; The step of sending identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session includes: Based on the traffic characteristic information of the first session and the historical traffic characteristic information, send the identification information of the abnormal traffic of the first session.
7. The method according to any one of claims 1-6, characterized in that, The acquisition of traffic characteristic information for the first session includes: Send a second request, which is used to obtain traffic characteristic information of the first session.
8. The method as described in claim 7, characterized in that, The second request includes the identification information of the first session; The identification information of the first session includes one or more of the following: the identifier of the first session, the identification information of the terminal device, and the attribute information of the first session; wherein, the first session is the session of the terminal device.
9. The method as described in claim 7 or 8, characterized in that, Sending the second request includes: Send the second request to at least one of the third, fourth, or fifth network elements; The third network element is used to provide user plane functions, the fourth network element is used to provide operation, maintenance and management functions, and the fifth network element is used to provide application functions.
10. The method according to any one of claims 1-9, characterized in that, The step of sending identification information of abnormal traffic in the first session based on the traffic characteristic information of the first session includes: Based on the traffic characteristic information of the first session, the identification information of abnormal traffic in the first session is determined by reasoning.
11. The method according to any one of claims 1-10, characterized in that, Also includes: Based on the traffic characteristics information of the first session, send at least one of the following: First-time information, which includes one or more of the start time, end time, or duration of abnormal traffic in the first session; The source of the abnormal traffic in the first session; The confidence level of the abnormal traffic analysis results of the first session.
12. The method according to any one of claims 1-11, characterized in that, The identification information of abnormal traffic in the first session is used to update the policy of the first session.
13. A communication method, characterized in that, include: Receive identification information for abnormal traffic in the first session; Update the policy of the first session based on the identification information of the abnormal traffic of the first session.
14. The method as described in claim 13, characterized in that, Also includes: Send a first request, which is used to obtain the identification information of abnormal traffic of the first session from the first network element. The first request includes the identification information of the first session and the analysis identifier, which is used to indicate abnormal traffic analysis.
15. The method as described in claim 14, characterized in that, The first network element is used to provide network data analysis functions.
16. The method according to any one of claims 13-15, characterized in that, Sending the first request includes: Based on the attribute information of the first session and / or the third request, the first request is sent; wherein the third request is used to indicate the identification information of abnormal traffic requested / needed by the first session.
17. The method as described in claim 13, characterized in that, Also includes: Receive at least one of the following: First-time information, which includes one or more of the start time, end time, or duration of abnormal traffic in the first session; The source of the abnormal traffic in the first session; The confidence level of the abnormal traffic analysis results of the first session.
18. The method according to any one of claims 13-17, characterized in that, The step of updating the policy for the first session based on the identification information of the abnormal traffic of the first session includes: Send the first rule based on the identification information of the abnormal traffic in the first session; The first rule includes identification information of abnormal traffic in the first session and first information, the first information being used to indicate at least one of the following: Disable the abnormal traffic; Packet dropping is performed on the abnormal traffic; The first quality of service parameter is used for the abnormal traffic; A separate Quality of Service (QoS) flow is used for the abnormal traffic; Implement caching for abnormal traffic.
19. The method as described in claim 18, characterized in that, The first rule also includes second time information, which indicates the time at which the at least one item is performed on the abnormal traffic.
20. A communication method, characterized in that, include: Obtain traffic characteristic information of the third network element; Based on the traffic characteristic information of the third network element, send the identification information of the abnormal traffic of the third network element.
21. The method as described in claim 20, characterized in that, Also includes: A fourth request is received from the third network element. The fourth request is used to obtain the identification information of the abnormal traffic of the third network element. The fourth request includes an analysis identifier, which is used to indicate abnormal traffic analysis.
22. The method as described in claim 20 or 21, characterized in that, The third network element is used to provide user plane functions.
23. The method according to any one of claims 20-22, characterized in that, The traffic characteristic information of the third network element includes one or more of the following: protocol fingerprint information, service behavior information, service server information, message protocol information, domain name / address blacklist, suspicious / abnormal domain name / address, and traffic quality of service related information.
24. The method as described in claim 23, characterized in that, At least one of the traffic characteristic information of the third network element satisfies a preset condition.
25. The method according to any one of claims 20-24, characterized in that, The method further includes: acquiring historical traffic characteristic information; The step of sending the identification information of the abnormal traffic of the third network element based on the traffic characteristic information of the third network element includes: Based on the traffic characteristic information of the third network element and the historical traffic characteristic information, the identification information of the abnormal traffic of the third network element is sent.
26. A communication device, characterized in that, It includes a module for performing the method as described in any one of claims 1-12, or includes a module for performing the method as described in any one of claims 13-19, or includes a module for performing the method as described in any one of claims 20-25.
27. A communication device, characterized in that, include: At least one processor; as well as, A communication interface that is communicatively connected to the at least one processor; The at least one processor, by executing instructions stored in the memory, causes the device to perform the method as described in any one of claims 1-12, or the method as described in any one of claims 13-19, or the method as described in any one of claims 20-25 via the communication interface.
28. A computer-readable storage medium, characterized in that, The readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1-12, or the method as described in any one of claims 13-19, or the method as described in any one of claims 20-25 to be implemented.
29. A computer program product, characterized in that, The computer program product stores instructions that, when executed on a computer, cause the method as described in any one of claims 1-12, or the method as described in any one of claims 13-19, or the method as described in any one of claims 20-25 to be implemented.
