Communication anomaly handling method, apparatus and system

Abstract

The present application provides a communication anomaly handling method, an apparatus and a system. The method comprises: acquiring a first determination condition; determining, on the basis of the first determination condition and an artificial intelligence (AI) model, that a prediction result indicates an anomaly, and sending first indication information, the first indication information comprising a root cause analysis request; receiving second indication information, the second indication information comprising at least one of root cause analysis information or an adjustment solution which is generated according to the root cause analysis request; and adjusting parameters on the basis of the second indication information. The present application can determine causes of communication anomalies and perform adjustment, so as to reduce the probability of communication anomalies.

Description

Method, device and system for handling communication anomaly

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on February 23, 2024, with application number 202410204876.1 and application name “A method, device and system for handling communication anomalies”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communications, and in particular to a method, device, and system for handling communication anomalies. Background Art

[0003] Artificial intelligence (AI), a technology proposed in the 1950s, simulates the human brain to perform complex calculations. With the advancement of communications technology and the continuous improvement of data storage and computing capabilities, AI has become increasingly widely used. For example, 3GPP's Release 17 approved a study item (SI) proposing the application of AI to the new radio (NR) to improve network performance and user experience through intelligent data collection and analysis.

[0004] With the continuous development of network technology, communication networks are becoming increasingly large and complex. To address potential anomalies in communication networks, AI models are often used to predict future states, such as whether the network state is abnormal. However, when the AI ​​model (also known as a predictive model) is a complex model such as a neural network, if the predicted network state is abnormal or the network enters other unexpected states, it is impossible to determine the cause of the future network anomaly and thus unable to accurately optimize in advance to avoid the anomaly. Therefore, how to determine the cause of communication anomalies and make adjustments to reduce the possibility of communication anomalies has become a pressing issue.

[0005] Summary of the Invention

[0006] The present application provides a method, device, and system for handling communication anomalies, which can determine the cause of communication anomalies and make adjustments to reduce the probability of communication anomalies occurring.

[0007] In a first aspect, the present application provides a method for handling communication anomalies, which is applied to a base station. Specifically, the method is performed by the base station or a device in the base station. The method includes: obtaining a first judgment condition; and determining a prediction result based on the first judgment condition and in combination with an artificial intelligence (AI) model. If the prediction result is abnormal, sending a first indication message, the first indication message including a root cause analysis request; receiving a second indication message, the second indication message including at least one of root cause analysis information or an adjustment solution generated based on the root cause analysis request; and performing parameter adjustment based on the second indication message. If the prediction result is normal, no adjustment is made.

[0008] The method for handling communication anomalies in this application can determine whether an anomaly will occur in the future communication network by combining the first judgment condition obtained with the AI ​​model. At the same time, by interacting the predicted result of the anomaly with the XAI module when an anomaly may occur, the XAI module analyzes the cause of the possible anomaly, effectively solving the problem that the AI ​​model infers that an anomaly will occur under a more complex model, such as a neural network model, but cannot determine the cause of the anomaly. The base station can finally obtain an adjustment plan through one of the root cause analysis messages or adjustment plans provided by the AXI module, and then adjust the corresponding adjustable parameters according to the adjustment plan, effectively reducing the possibility of anomalies occurring in the future.

[0009] The base station can also verify the adjustment plan. If it is verified that the anomaly can be eliminated after adjustment according to the adjustment plan, the adjustment can be made according to the adjustment plan. If it is verified that the anomaly still occurs after adjustment according to the adjustment plan, the cause of the anomaly is obtained, and the XAI module is interacted with again. The XAI module provides an adjustment plan updated according to the cause of the anomaly, and the updated adjustment plan is verified again until the adjustment plan can eliminate the anomaly, or until the elimination of the anomaly is abandoned.

[0010] In one possible implementation, adjusting parameters based on the second indication information includes: determining, based on the second indication information and in combination with the AI ​​model, that the prediction result is normal, and adjusting parameters based on the second indication information. In other words, if the adjustment scheme can effectively eliminate the abnormality, the adjustment can be performed based on the adjustment scheme.

[0011] In one possible implementation, before the parameter adjustment is performed according to the second indication information, it also includes: based on the second indication information, combined with the AI ​​model, determining that the prediction result is abnormal, sending a third indication information, the third indication information includes a root cause analysis request and the cause of the abnormality; receiving the second indication information, the root cause analysis information or adjustment plan in the second indication information has been updated according to the root cause analysis request and the cause of the abnormality. In this case, according to the adjustment plan carried in the second indication information, and then calculated by the AI ​​model, it is determined that the prediction result is abnormal, then the cause of the abnormality is obtained by interacting with the XAI module again, and the adjustment plan is updated. The base station can verify again according to the updated adjustment plan until the abnormality can be effectively eliminated, and then make adjustments according to the updated adjustment plan.

[0012] In one possible implementation, the first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event. Exemplarily, the first judgment condition includes multiple second judgment conditions, each of which can be one of an abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event. This type of condition, which can be used to determine whether communication abnormalities will occur in the future, offers greater selectivity and a wider range of conditions, thereby improving the reliability and flexibility of prediction results.

[0013] In one possible implementation, the root cause analysis request includes at least one of the following: an input or output corresponding to the abnormal prediction result; or a second judgment condition for obtaining the abnormal prediction result; or an adjustable parameter for obtaining the second judgment condition; or a time node at which the second indication information is received based on the abnormal prediction result; or an explanation method of explainable AI (XAI). The XAI device can provide root cause analysis information or an adjustment plan based on the content in the root cause analysis request. For example, if the root cause analysis request carries the time node at which the second indication information is received, the XAI device can feedback the second indication information to the base station before the time node. The diverse content in the root cause analysis request, on the one hand, increases the accuracy and adaptability of the root cause analysis result obtained after the XAI device receives it, and on the other hand, it can also make the interaction between the base station and the XAI device more in line with the configuration and needs of the base station.

[0014] In one possible implementation, the adjustment scheme includes the type of parameter to be adjusted and the adjustment method for the parameter to be adjusted; or, the adjustment scheme includes the type of parameter to be adjusted, the adjustment method for the parameter to be adjusted, and a corresponding time for the parameter to be adjusted, where the corresponding time for the parameter to be adjusted includes at least one of the adjustment time for the parameter to be adjusted or the verification time after the adjustment of the parameter to be adjusted. The diversity of adjustment schemes can make base station adjustment more flexible.

[0015] In one possible implementation, obtaining the first judgment condition includes receiving the first judgment condition sent by an Operation Administration and Maintenance (OAM) network element or a core network (CN). The base station can obtain the first judgment condition in a variety of ways, such as by combining the first judgment condition based on conditions sent by one or more network elements in the OAM or CN, or by receiving the first judgment condition, thereby increasing the possibilities for obtaining the first judgment condition and enhancing system adaptability.

[0016] In a second aspect, the present application provides a method for handling communication anomalies, which is applied to an XAI device. Specifically, the method is performed by a separately configured XAI device, or an XAI device configured in a node other than a base station. The method includes: receiving first indication information, the first indication information including a root cause analysis request obtained according to a first judgment condition; generating at least one of root cause analysis information or an adjustment solution based on the root cause analysis request; and sending second indication information, the second indication information including at least one of the generated root cause analysis information or the adjustment solution.

[0017] In a possible implementation, the method further includes: receiving third indication information, the third indication information including a root cause analysis request and an abnormality cause; updating at least one of the root cause analysis information or an adjustment plan based on the root cause analysis request and the abnormality cause; and sending second indication information, the second indication information including at least one of the updated root cause analysis information or the adjustment plan.

[0018] In a possible implementation, the first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

[0019] In one possible implementation, the root cause analysis request includes at least one of the following: the input or output corresponding to the abnormal prediction result; or, obtaining the second judgment condition of the abnormal prediction result; or, obtaining the adjustable parameter of the second judgment condition; or, based on the abnormal prediction result, the time node of receiving the second indication information; or, an interpretation method of explainable artificial intelligence XAI.

[0020] In one possible implementation, the adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, the adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted, and the corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

[0021] It should be understood that the second aspect of this application corresponds to the technical solution of the first aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here.

[0022] In a third aspect, the present application provides a method for handling communication anomalies, which is applied to the user equipment (UE) side. Specifically, the method is performed by the UE or a device in the UE. The method includes: obtaining a third judgment condition; based on the third judgment condition, combined with an artificial intelligence (AI) model, determining that the prediction result is abnormal, sending a fourth indication message, the fourth indication message including a root cause analysis request; receiving a fifth indication message, the fifth indication message including at least one of the root cause analysis information or adjustment plan generated according to the root cause analysis request; and adjusting parameters according to the fifth indication message.

[0023] In one possible implementation, obtaining the third judgment condition includes pre-configuring the third judgment condition; or, according to the configuration indication of the first device, requesting the first device to obtain the third judgment condition and receiving the third judgment condition; or, the third judgment condition is sent by the OAM network element or CN, base station or network side (network, NW).

[0024] In one possible implementation, based on the fifth indication information and in combination with the AI ​​model, it is determined that the prediction result is normal, and parameters are adjusted according to the fifth indication information.

[0025] In one possible implementation, before adjusting the parameters according to the fifth indication information, it also includes: based on the fifth indication information, combined with the AI ​​model, determining that the prediction result is abnormal, sending a sixth indication information, the sixth indication information including a root cause analysis request and the cause of the abnormality; receiving the fifth indication information, the root cause analysis information or adjustment plan in the fifth indication information has been updated according to the root cause analysis request and the cause of the abnormality.

[0026] In one possible implementation, the root cause analysis request includes at least one of the following: the input or output corresponding to the abnormal prediction result; or, the fourth judgment condition for obtaining the abnormal prediction result; or, the adjustable parameter for obtaining the fourth judgment condition; or, the time node at which the second indication information is received based on the abnormal prediction result; or, the interpretation method of explainable artificial intelligence XAI.

[0027] In one possible implementation, the adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, the adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted, and the corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

[0028] In a possible implementation manner, the sending the fourth indication information includes: sending the fourth indication information to the XAI device, or sending the fourth indication information to the XAI device through a new radio (NR).

[0029] In a possible implementation manner, the sending the sixth indication information includes: sending the sixth indication information to the XAI device, or sending the sixth indication information to the XAI device through NR.

[0030] In a possible implementation manner, the receiving the fifth indication information includes: receiving the fifth indication information sent by an XAI device, or the fifth indication information sent through an NR, where the fifth indication information is generated by the XAI device.

[0031] The UE can interact with the XAI device through NR or directly with the XAI device, making the UE more adaptable in predicting and handling network anomalies and providing greater flexibility in network deployment.

[0032] It should be understood that the third aspect of this application corresponds to the technical solution of the first aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here.

[0033] In a fourth aspect, the present application provides a method for handling communication anomalies, which is applied to the XAI device side. Specifically, the method is performed by a separately set XAI device, or an XAI device set in a node outside the base station, and the XAI device includes an XAI module or a module capable of providing the explained machine learning. The method includes: receiving fourth indication information, the fourth indication information including a root cause analysis request obtained according to the third judgment condition; generating at least one of the root cause analysis information or the adjustment plan based on the root cause analysis request; and sending fifth indication information, the fifth indication information including at least one of the generated root cause analysis information or the adjustment plan.

[0034] In a possible implementation, the method further includes: receiving sixth indication information, the sixth indication information including a root cause analysis request and an abnormality cause; updating at least one of the root cause analysis information or an adjustment plan based on the root cause analysis request and the abnormality cause; and sending fifth indication information, the fifth indication information including at least one of the updated root cause analysis information or the adjustment plan.

[0035] In a possible implementation, the third judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

[0036] In one possible implementation, the root cause analysis request includes at least one of the following: the input or output corresponding to the abnormal prediction result; or, obtaining the fourth judgment condition of the abnormal prediction result; or, obtaining the adjustable parameter of the fourth judgment condition; or, based on the abnormal prediction result, the time node of receiving the fifth indication information; or, an interpretation method of interpretable artificial intelligence XAI.

[0037] In a possible implementation, the adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or,

[0038] The adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted. The corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

[0039] In a possible implementation manner, the receiving the fourth indication information includes: receiving the fourth indication information sent by the UE, or receiving the fourth indication information sent by the UE through the new radio interface NR.

[0040] In a possible implementation manner, the receiving the sixth indication information includes: receiving the sixth indication information sent by the UE, or receiving the sixth indication information sent by the UE through the NR.

[0041] In a possible implementation manner, the sending the fifth indication information includes: sending the fifth indication information to the UE, or sending the fifth indication information to the UE through the NR.

[0042] It should be understood that the fourth aspect of this application corresponds to the technical solution of the third aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here.

[0043] In a fifth aspect, the present application provides a base station or a device in a base station, including: an acquisition module for acquiring a first judgment condition; a sending module for sending a first indication information after the processing module determines that the prediction result is abnormal based on the first judgment condition and combined with the AI ​​model, the first indication information includes a root cause analysis request; a receiving module for receiving a second indication information, the second indication information including at least one of the root cause analysis information or adjustment plan generated according to the root cause analysis request; the processing module is also used to adjust parameters according to the second indication information.

[0044] In one possible implementation, the processing module is specifically used to determine that the prediction result is normal based on the second indication information and in combination with the AI ​​model, and adjust parameters according to the second indication information.

[0045] In one possible implementation, the sending module is further used to send third indication information after the processing module determines that the prediction result is abnormal based on the second indication information and in combination with the AI ​​model, and the third indication information includes a root cause analysis request and the cause of the abnormality; the receiving module is used to receive the second indication information, and the root cause analysis information or adjustment plan in the second indication information has been updated according to the root cause analysis request and the cause of the abnormality.

[0046] In a possible implementation, the receiving module is further configured to receive the first judgment condition sent by the OAM or CN.

[0047] It should be understood that the fifth aspect of this application corresponds to the technical solution of the first aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here.

[0048] In a sixth aspect, the present application provides an XAI device, comprising: a receiving module for receiving first indication information, the first indication information including a root cause analysis request obtained according to a first judgment condition; an XAI module for generating at least one of root cause analysis information or an adjustment plan based on the root cause analysis request; and a sending module for sending second indication information, the second indication information including at least one of the generated root cause analysis information or the adjustment plan.

[0049] In one possible implementation, the receiving module is further used to receive third indication information, wherein the third indication information includes a root cause analysis request and an abnormality cause; the XAI module is further used to update at least one of the root cause analysis information or the adjustment plan based on the root cause analysis request and the abnormality cause; and the sending module is further used to send second indication information, wherein the second indication information includes at least one of the updated root cause analysis information or the adjustment plan.

[0050] It should be understood that the sixth aspect of this application corresponds to the technical solution of the second aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, so they will not be repeated here.

[0051] In the seventh aspect, the present application provides a UE or a device in the UE, including: an acquisition module for acquiring a third judgment condition; a sending module for sending a fourth indication information after the processing module determines that the prediction result is abnormal based on the third judgment condition and combined with the artificial intelligence AI model, and the fourth indication information includes a root cause analysis request; a receiving module for receiving fifth indication information, and the fifth indication information includes at least one of the root cause analysis information or adjustment plan generated according to the root cause analysis request; the processing module is also used to adjust parameters according to the fifth indication information.

[0052] In a possible implementation, the processing module is specifically used to determine that the prediction result is normal based on the fifth indication information and in combination with the AI ​​model, and adjust parameters according to the fifth indication information.

[0053] In one possible implementation, the sending module is further used to send sixth indication information after the processing module determines that the prediction result is abnormal based on the fifth indication information and in combination with the AI ​​model, and the sixth indication information includes a root cause analysis request and the cause of the abnormality; the receiving module is used to receive the fifth indication information, and the root cause analysis information or adjustment plan in the fifth indication information has been updated according to the root cause analysis request and the cause of the abnormality.

[0054] In one possible implementation, the processing module is specifically used to pre-configure the third judgment condition; the sending module is also used to request the first device to obtain the third judgment condition based on the configuration indication of the first device received by the receiving module, and the receiving module is also used to receive the third judgment condition; the receiving module is also used to obtain the third judgment condition sent by the OAM or CN, base station or NW.

[0055] In a possible implementation manner, the sending module is specifically configured to send the fourth indication information to the XAI device, or send the fourth indication information to the XAI device through NR.

[0056] In a possible implementation manner, the sending module is specifically configured to send the sixth indication information to the XAI device, or send the sixth indication information to the XAI device through NR.

[0057] In a possible implementation, the receiving module is specifically configured to receive the fifth indication information sent by the XAI device, or the fifth indication information sent through the NR, where the fifth indication information is generated by the XAI device.

[0058] It should be understood that the seventh aspect of this application corresponds to the technical solution of the third aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here.

[0059] In an eighth aspect, the present application provides an XAI device, comprising: a receiving module for receiving fourth indication information, the fourth indication information including a root cause analysis request obtained according to a third judgment condition; an XAI module for generating at least one of root cause analysis information or an adjustment plan based on the root cause analysis request; and a sending module for sending fifth indication information, the fifth indication information including at least one of the generated root cause analysis information or the adjustment plan.

[0060] In one possible implementation, the receiving module is further used for sixth indication information, and the sixth indication information includes a root cause analysis request and an abnormality cause; the XAI module is further used to update at least one of the root cause analysis information or the adjustment plan based on the root cause analysis request and the abnormality cause; the sending module is further used to send fifth indication information, and the fifth indication information includes at least one of the updated root cause analysis information or the adjustment plan.

[0061] In a possible implementation, the receiving module is further used to receive the fourth indication information sent by the UE, or to receive the fourth indication information sent by the UE through the new radio interface NR.

[0062] In a possible implementation, the receiving module is further used to receive the sixth indication information sent by the UE, or to receive the sixth indication information sent by the UE through the NR.

[0063] In a possible implementation manner, the sending module is further used to send the fifth indication information to the UE, or send the fifth indication information to the UE through the NR.

[0064] It should be understood that the eighth aspect of the present application corresponds to the technical solution of the fourth aspect of the present application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, so they will not be repeated here.

[0065] In a ninth aspect, the present application provides a communication device, which may be a base station or a device in a base station (e.g., a chip). The communication device includes a module for executing the method described in any one of the above aspects or any possible implementation of any one of the aspects, such as a processing module and a transceiver module. The processing module may be a processor, and the transceiver module may be a transceiver. When the communication device is a terminal device, the transceiver may be a radio frequency module. When the communication device is a device in a terminal device, the transceiver may be an input / output interface, a pin, or a circuit, etc.

[0066] In a tenth aspect, the present application provides a communication device, which may be a UE or a device in a UE (e.g., a chip). The communication device includes a module for executing the method described in any one of the above aspects or any possible implementation of any one of the aspects, such as a processing module and a transceiver module. The processing module may be a processor, and the transceiver module may be a transceiver. When the communication device is a network device, the transceiver may be a radio frequency module. When the communication device is a device in a network device, the transceiver may be an input / output interface, a pin, or a circuit, etc.

[0067] In an eleventh aspect, the present application provides a communication device, which may be an XAI device (e.g., a chip). The communication device includes a module for executing the method described in any one of the above aspects or any possible implementation of any one of the aspects, such as a processing module and a transceiver module. The processing module may be a processor, and the transceiver module may be a transceiver. When the communication device is a network device, the transceiver may be a radio frequency module. When the communication device is a device in a network device, the transceiver may be an input / output interface, a pin, or a circuit.

[0068] In a twelfth aspect, the present application provides a communication device, comprising at least one processor coupled to a storage medium, the storage medium storing instructions, wherein when the instructions are executed by the processor, the processor is configured to perform the method as described in any of the above aspects or any possible implementation of any of the aspects. The storage medium may be included in the device or may be located external to the device.

[0069] In a thirteenth aspect, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method described in any one of the above aspects or any possible implementation of any one of the aspects.

[0070] In a fourteenth aspect, the present application provides a computer program product comprising instructions that, when executed on a processor, implement the method as described in any one of the above aspects or any possible implementation of any one of the aspects.

[0071] In a fifteenth aspect, the present application provides a system, which includes the base station as described in the fifth aspect and the XAI device as described in the sixth aspect.

[0072] In a sixteenth aspect, the present application provides a system, which includes the UE described in the seventh aspect and the XAI device described in the eighth aspect.

[0073] It should be understood that the sixth to sixteenth aspects of the present application are consistent with or correspond to the technical solutions of the first, second, third, fourth or fifth aspects of the present application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0075] FIG1 is a schematic diagram of a network system 100 provided in an embodiment of the present application;

[0076] FIG2 is a schematic diagram of an architecture of AI applied in NR provided in an embodiment of the present application;

[0077] FIG3 is a flow chart of a method for handling communication anomalies provided in an embodiment of the present application;

[0078] FIG4 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application;

[0079] FIG5 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application;

[0080] FIG6 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application;

[0081] FIG7 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application;

[0082] FIG8 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application;

[0083] FIG9 is a schematic structural diagram of a base station or a base station device provided in an embodiment of the present application;

[0084] FIG10 is a schematic structural diagram of an XAI device provided in an embodiment of the present application;

[0085] FIG11 is a schematic structural diagram of a UE or a device in a UE provided in an embodiment of the present application;

[0086] FIG12 is a schematic structural diagram of an XAI device provided in an embodiment of the present application;

[0087] FIG13 is a schematic structural diagram of a device 80 according to an embodiment of the present application;

[0088] FIG14 is a schematic structural diagram of a device 70 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0089] In order to enable people in this technical field to better understand the solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in combination with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0090] The term "and / or" herein is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. A and B can be single or multiple. "At least one of the following" or similar expressions is used to indicate any combination of the listed items. For example, at least one of A, B, and / or C can mean: A exists alone, B exists alone, C exists alone, A and B exist at the same time, B and C exist at the same time, A and C exist at the same time, and A, B, and C exist at the same time. A, B, and C can be single or multiple.

[0091] In the description and claims of the embodiments of this application, the terms "first" and "second" are used to distinguish different objects, rather than to describe a specific order of objects. For example, the terms "first target object" and "second target object" are used to distinguish different objects, rather than to describe a specific order of objects.

[0092] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0093] In the description of the embodiments of this application, unless otherwise specified, "multiple" means two or more. For example, "multiple processing units" means two or more processing units; "multiple systems" means two or more systems.

[0094] For ease of understanding, the following first explains the relevant nouns or terms used in the embodiments of this application:

[0095] 1. Energy saving

[0096] The base station collects its own and neighboring cell's load, energy consumption, energy efficiency information, as well as UE's path information, measurement results, etc., to predict the direction of its own load, and combines the cell's purpose, key performance indicator (KPI) requirements, etc., to adopt energy-saving strategies in a timely and appropriate manner without affecting network coverage and user access. The simplest energy-saving strategy involves directly deactivating the cell. Other energy-saving strategies include carrier shutdown, channel shutdown, time slot shutdown, and transmission power reduction. More complex energy-saving strategies also include a combination of the above energy-saving strategies. When network coverage is affected or cannot meet UE access and service requirements, it is necessary to modify the current energy-saving strategy or directly restore to normal working state, and consider re-forecasting the load or changing the AI ​​model used for re-reasoning.

[0097] 2. Load balancing

[0098] The base station collects information on its own and neighboring cells' load, energy consumption, and energy efficiency, as well as UE path information and measurement results, to predict the direction of the load. Combined with the cell's purpose and KPI requirements, the base station reasonably selects some UEs to switch to neighboring cells, or receives UEs from neighboring cells, so that the load levels between base stations in the entire network are close, reducing the situation where some base stations are overloaded and affect normal services while some base station resources are idle. However, since the accuracy of the prediction is not 100%, it may lead to unreasonable UE selection or unreasonable switching target cells, resulting in handover failure or affected UE services, or inaccurate load prediction leading to poor load balancing effects, or temporary abnormal load changes causing the original load balancing strategy to no longer apply. In this case, it is necessary to exit or modify the current load balancing strategy, and consider re-forecasting the load or changing the AI ​​model used for re-reasoning.

[0099] 3. Mobility optimization

[0100] The base station collects historical UE path information and combines it with the UE's measurement information to predict the UE's future path. Based on the predicted path, it determines in advance whether the UE should switch, sends the handover configuration in advance, and notifies the target cell to prepare access resources. This reduces delays during the UE handover process and the probability of handover and access failures. However, because path prediction accuracy is not 100%, when the predicted path is incorrect, it will cause UE handover failure and service interruption. In this case, it is necessary to consider retraining the model and reasoning based on the abnormal situation, or consider replacing the model to avoid similar abnormal situations in subsequent UEs.

[0101] 4. Channel state information-reference signal feedback enhancement (CSI-RS Feedback Enhancement)

[0102] A process for implementing CSI-RS feedback enhancement between a base station and a user equipment terminal (UE) includes: first, the base station and the UE exchange a dictionary, typically a model pre-trained by the base station based on the UE's capabilities and its own requirements; then, an encoder and quantizer are sent to the UE; the UE compresses and quantizes the matrix to be fed back based on the measured channel matrix and the existing dictionary, and transmits the result B to the base station; the base station reversely recovers the original channel matrix based on the dictionary and the data reported by the UE to achieve CSI-RS feedback enhancement.

[0103] 5. Beam management enhancement

[0104] A process for implementing enhanced beam management between a base station and a user equipment (UE) includes: generating an initial model (e.g., having a certain number of UEs report the results of full-beam scanning of synchronization signals (SS) or physical broadcast channel (PBCH) blocks (SSBs) to train a sparse scanning matrix, which is typically unique to each cell); the base station sends the sparse model to the UE, which performs a first (P1) phase of beam scanning based on the matrix; based on the UE's sparse scanning results, the base station infers the optimal CSI-RS beam and begins a second (P2) phase of scanning of the UE, with the UE providing feedback on the optimal CSI-RS beam ID to implement enhanced beam management.

[0105] 6. Positioning management enhancement

[0106] A process for enhancing current positioning accuracy between a base station and a user equipment (UE) includes: using an operator-controlled reference UE to collect raw data; training models in the location management function (LMF) deployed in the location management node and the base station (such as a gNB). The AI ​​model in the LMF can infer the final location (latitude and longitude, etc.), and the AI ​​model in the gNB can infer the judgment result of the channel model. The channel model can be line of sight (LOS) or non-line of sight (NLOS) propagation to achieve enhanced current positioning accuracy.

[0107] 7. XAI

[0108] There are two options for model training: one is to select a model with a simple structure and easy to interpret, and then train it; the other is to train a complex optimal model and then develop interpretability techniques to explain it. Based on these two options, the interpretability of machine learning models (also known as AI models) can generally be divided into two categories: ex ante (ante-hoc) interpretability and ex post (post-hoc) interpretability. Ex ante (ante-hoc) interpretability is typically aimed at models with simple structures and good interpretability, such as naive Bayesian models, linear regression models, and decision tree models. The operational logic of such models can usually be understood from the model architecture itself, that is, the reasons for various inference results and the relationship between inputs and outputs can be understood. Ex post (post-hoc) interpretability is typically aimed at models with complex logical structures, such as neural network models and deep learning models. The complex logical structure of such models makes it difficult to intuitively determine the actual operating logic of the model and analyze the relationship between inputs and outputs. Explanation methods generally include global interpretability and local interpretability. A typical global interpretability method is model distillation. Typical local interpretability methods include the model-agnostic explanation method (LIME) and the Shapley Additive Explanation (SHAP) method. Global interpretability describes the algorithmic operation of the entire model and explains the model's global output. Local interpretability explains the contribution of a single input sample to the final inference result.

[0109] 8. A3 Incident

[0110] It refers to a measurement event configured by the base station for the UE. The event is usually triggered because the reference signal of the neighboring cell is better than that of the local cell, and the better exceeds a certain threshold (also called the trigger threshold). After the event is triggered, the UE will start measuring the reference signal quality of the neighboring cell and report it.

[0111] The communication anomaly handling method provided in the embodiments of the present application can be applied to different communication network systems. This embodiment provides a possible system 100, as shown in FIG1 , including multiple network devices. The network devices may be base stations (e.g., gNBs) in a radio access network (RAN). FIG1 takes gNB 10 as an example. In actual deployment, gNB 10 may be a combination of a centralized unit (CU) and a distributed unit (DU), such as gNB 10, or a separate CU and DU architecture, not limited to the example in FIG1 . System 100 also includes a core network (CN) 20. The RAN may be connected to CN 20, which may be a long term evolution (LTE) CN or a 5G CN. The CU and DU may be understood as components of the gNB. 10 Division from the perspective of logical functions. CU and DU can be physically separated or deployed together. Multiple DUs can share one CU (for example, two DUs share one CU in Figure 1), and one DU can also be connected to multiple CUs (not shown in Figure 1). The CU and DU can be connected through an interface, for example, an F1 interface (not shown in Figure 1). The CU and DU can be divided according to the protocol layer of the wireless network. One possible division method is: the CU is used to perform the functions of the radio resource control (RRC) layer, the service data adaptation protocol (SDAP) layer, and the packet data convergence protocol (PDCP) layer, while the DU is used to perform the functions of the radio link control (RLC) layer, the media access control (MAC) layer, the physical layer, etc. The division of the processing functions of the CU and DU in the embodiment of the present application is only an example and is not limited. The functions of the CU can be implemented by one entity or by different entities. For example, the functions of the CU can be further divided, for example, the control plane (CP) and the user plane (PDCP) can be divided into two parts. The CU-CP and CU-UP can be implemented by different functional entities and connected through an E1 interface, etc.The system 100 also includes a UE 30, which can be a wireless terminal or a wired terminal. A wireless terminal includes a device that provides at least one of voice or data connectivity to a user, a handheld device with wireless connection capabilities, or other processing devices connected to a wireless modem. Exemplarily, a UE can be an access terminal, a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent, or a user device. Such as mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, VR, AR, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, wearable devices, etc.

[0112] In one possible implementation, the system of the communication network can refer to the structural diagram of system 100 shown in Figure 1 and the architecture of Figure 1 to complete operations such as reasoning and prediction during the communication process. Figure 2 is a schematic diagram of the architecture of an AI application in NR provided in an embodiment of the present application. As shown in Figure 2, the framework of AI application in NR may include a data source, a training model, an inference model, and an actor entity. The data source is used to store data input from the next generation node base (gNB) (such as gNB 10 in Figure 1), a gNB-centralized unit (CU), a gNB-distributed unit (DU), a user equipment (UE), or other management entities as a database for AI model training and data analysis and reasoning; the training model (model training host) analyzes the training data provided by the data source to provide the optimal AI model. The model inference host uses AI models to generate reasonable AI-based predictions about network performance based on data source data, or to guide network policy adjustments. These policy adjustments are centrally planned by actor entities and sent to multiple network entities for execution. These network entities can be gNB 10, CN 20, or UE 30, as shown in Figure 2. After the relevant policies are applied, the specific performance of the communication network is stored in a database for use in updating the database.

[0113] Figure 3 is a flow chart of a method for handling communication anomalies provided in an embodiment of the present application. The method is executed by a base station or a base station device (or a device in a base station, such as a chip). This embodiment of the present application uses the base station as an example to illustrate the method. Other execution entities can refer to the base station's operation. The method includes: S101 to S104.

[0114] S101. A base station obtains a first judgment condition.

[0115] The first judgment condition includes at least one second judgment condition. Exemplarily, each second judgment condition may be an abnormal event, a threshold parameter corresponding to the abnormal event, or a judgment range corresponding to the abnormal event.

[0116] Exemplarily, abnormal events may be specific events in different scenarios, such as increased latency in the handover (HO) process, decreased throughput, service interruption, etc., or radio link failure (RLF) events for edge UEs, or service level agreement (SLA) violation events that fail to meet the perceived quality of experience (QoE) due to limited UE service transmission capabilities.

[0117] Threshold parameters corresponding to abnormal events, such as abnormal events during the HO process, may include: a latency increase of xx ms corresponding to a latency increase event; a throughput decrease of xx% corresponding to a throughput decrease event; or a service interruption duration exceeding xx ms corresponding to a service interruption event. Another example is a RLF occurrence rate exceeding xx% corresponding to an abnormal RLF event of edge UEs; or an occurrence rate exceeding xx% or a corresponding SLA violation UE percentage exceeding xx%. In this example, xx represents a numerical value and does not limit the specific value of xx in each example, nor does it limit the number of digits in the numerical value in each example.

[0118] Judgment range corresponding to abnormal events. For example, a specific judgment range can be preset for abnormal events. The specific judgment range can be a spatial range, such as a preset specific cell, a cell in a preset specific cell list, a preset specific UE, or a UE in a preset specific UE list; the specific judgment range can be a temporal range, such as an effective time range corresponding to a certain abnormal event, such as an effective time of xx s, or starting from aa:bb:cc and ending at dd:ee:ff, or starting from aa:bb:cc, or ending at aa:bb:cc, etc.; the specific judgment range can be a spatial and temporal range, such as one or more specific cells, the effective duration and effective time period in one or more abnormal events. For example, for a specific cell 1 and a specific edge UE 2, starting at 00:00:00, it is judged whether the RLF occurrence rate exceeds 10%; or, for a specific cell 1, it is judged whether the occurrence rate corresponding to the SLA violation event with an effective time of 60s exceeds 5%; or from 10:30:00 to 11:00:00, for a specific UE 1. Determine whether a 10ms service interruption event has occurred. In this example, xx represents a numerical value, and aa:bb:cc and dd:ee:ff represent time. The specific values ​​of xx, aa:bb:cc, and dd:ee:ff in each example are not limited, nor is the number of digits in xx in each example.

[0119] The values ​​of xx, aa:bb:cc and dd:ee:ff in the embodiments of the present application can be determined according to the requirements during use.

[0120] S102. The base station determines that the prediction result is abnormal based on the first judgment condition and the AI ​​model, and sends a first indication message, where the first indication message includes a root cause analysis request.

[0121] Based on the first judgment condition and in conjunction with the AI ​​model, a base station such as a gNB predicts whether anomalies will occur in the future of the communication network. If the prediction result is normal (i.e., no anomalies), it means that no adjustment to the communication network is required for the time being. If the prediction result is abnormal, it means that communication anomalies will occur, such as network anomalies. In this case, it is necessary to obtain the cause of the anomaly and make adjustments based on the cause to prevent future anomalies in the communication network after the adjustment.

[0122] Optionally, taking the example of a gNB combining an AI model to obtain prediction results, the AI ​​model can run in an AI module, which can be built into the gNB or not. The AI ​​module refers to a module with machine learning computing capabilities. For example, in a wireless communication system, the AI ​​module can be located in the OAM or the gNB. Alternatively, in a separate CU / DU architecture, the AI ​​module can be located in the CU. The AI ​​module can also be located in some UEs, or the AI ​​module can be a separate network element entity. In a wireless communication system, the AI ​​module can perform a series of AI calculations, such as model building, training approximation, and reinforcement learning, based on input data (e.g., data provided by the RAN or network operation data monitored by OAM, such as network load and channel quality). The trained model provided by the AI ​​module is referred to as an AI model in this embodiment of the present application. This AI model has the ability to predict network changes on the RAN side and can typically be used for load prediction, UE path prediction, etc., and based on the prediction, a prediction result is obtained to determine whether the network will experience anomalies in the future. In addition, the AI ​​module can also perform policy reasoning from the perspectives of network energy saving and mobility optimization based on the predicted results of the trained model on RAN network performance, in order to obtain reasonable and efficient energy-saving strategies and mobility optimization strategies.

[0123] The manner in which the gNB determines the prediction result in conjunction with the AI ​​model may vary depending on the location where the AI ​​model is deployed. For example, the gNB may use a built-in AI model (i.e., the AI ​​module is built-in to the gNB) and input the first judgment condition to obtain the prediction result. Alternatively, the gNB may send the first judgment condition to an AI model external to the gNB (i.e., the AI ​​module is not built-in to the gNB) and receive the prediction result fed back by the AI ​​model external to the gNB. Obtaining the prediction result by the gNB in ​​conjunction with the AI ​​model external to the gNB may include: sending at least one judgment condition to an AI model located in another node and receiving the prediction result fed back by the AI ​​model located in the other node (e.g., a node located on the CN side) to obtain the prediction result; or sending at least one judgment condition to a separately configured AI module running the AI ​​model and receiving the prediction result fed back by the AI ​​module to obtain the prediction result.

[0124] S103: The base station receives second indication information, where the second indication information includes at least one of root cause analysis information or an adjustment solution generated according to the root cause analysis request.

[0125] Exemplarily, the second indication information may be sent by an XAI module, and the second indication information includes root cause analysis information generated by the XAI module based on a root cause analysis request, or the second indication information includes an adjustment plan generated by the XAI module based on a root cause analysis request, or the second indication information includes both root cause analysis information and an adjustment plan generated by the XAI module based on a root cause analysis request. The root cause analysis information may be information related to an analysis of the cause of the anomaly calculated by the XAI module based on an anomaly prediction result sent by the base station, and the adjustment plan is a plan for adjusting certain adjustable parameters in the network based on the root cause analysis information so that the prediction result is no longer abnormal.

[0126] S104. The base station adjusts parameters according to the second indication information.

[0127] The second indication information received by the base station includes three situations: If the second indication information received by the base station includes an adjustment plan, the base station may perform parameter adjustment based on the adjustment plan. If the second indication information received by the base station includes root cause analysis information, the base station may determine an adjustment plan based on the root cause analysis information and perform corresponding parameter adjustment. If the second indication information received by the base station includes both the root cause analysis information and the adjustment plan, the base station may perform parameter adjustment based on the adjustment plan. Alternatively, the base station may obtain an adjustment plan based on the root cause analysis information and select one of the adjustment plans obtained by the base station and the adjustment plan received to perform parameter adjustment.

[0128] The method for handling communication anomalies provided in the embodiment of the present application can determine whether an anomaly will occur in the future communication network by obtaining the first judgment condition in combination with the AI ​​model. In the event that an anomaly may occur, the prediction result of the anomaly is interacted with the XAI module. The XAI module analyzes the cause of the possible anomaly, and the base station determines how to adjust the parameters based on the cause of the anomaly, which can effectively reduce the possibility of anomalies occurring in the future.

[0129] For the operation of the AI ​​model and XAI module provided in the embodiment of the present application to jointly provide root cause analysis information and adjustment solutions, reference can be made to the use case introduced on the current RAN side. That is, the base station can predict the future network state based on each use case calculated in the AI ​​model to obtain a prediction result, and then use the XAI module to analyze the root cause or adjustment solution for the abnormality, plan the corresponding optimization strategy in advance, and adjust the corresponding parameters to achieve the purpose of reducing the possibility of network abnormalities. Among them, the use case includes energy saving, load balancing, and mobility optimization. RAN1 includes CSI-RS feedback enhancement, beam scanning enhancement, and positioning enhancement.

[0130] For example, for load balancing, AI models typically predict the future load demand of adjacent cells, air interface load capacity, base station available resources, UE paths, and other factors, selecting appropriate UEs for handover to achieve future load balancing. However, when the AI ​​model uses complex models such as neural networks to generate predictions, even if an abnormal prediction result is obtained (including an abnormal predicted network state or other unexpected state), the AI ​​model cannot determine the cause of the future anomaly, making it unable to accurately make adjustments in advance to achieve optimization. For example, an AI model using a neural network may predict that a UE will experience a link failure or an abnormal decrease in throughput in the future, but the AI ​​model cannot determine the cause of these conditions, resulting in an inability to accurately adjust the corresponding parameters to avoid these anomalies. In this case, through interaction with the XAI module, the abnormal prediction result inferred by the AI ​​model is interpreted to determine the root cause of the anomaly, that is, to obtain root cause analysis information or further obtain an adjustment plan, and then make corresponding adjustments based on this information to reduce the possibility of future anomalies.

[0131] Figure 4 is a flowchart illustrating another method for handling communication anomalies provided in an embodiment of the present application. This method is executed by an XAI module or a module capable of providing root cause analysis information and adjustment solutions. The XAI module can be built into the gNB, a standalone node, another gNB, or a CN. This embodiment of the present application illustrates this method by an XAI module of a standalone node. Other execution entities and other configurations of the XAI module can be similar to the operation of the XAI module of the standalone node. As shown in Figure 4 , the method includes steps S201 to S203.

[0132] S201: An XAI module receives first indication information, where the first indication information includes a root cause analysis request obtained according to a first judgment condition.

[0133] The first indication information can be obtained by referring to S101 and S102, and will not be elaborated herein.

[0134] S202: The XAI module generates at least one of root cause analysis information and adjustment solutions according to the root cause analysis request.

[0135] Optionally, the XAI module generates the root cause analysis information according to the root cause analysis request, or the XAI module generates the root cause analysis information according to the root cause analysis request and then generates the adjustment plan according to the root cause analysis information.

[0136] S203: The XAI module sends second indication information, where the second indication information includes at least one of the generated root cause analysis information or the adjustment solution.

[0137] The second indication information only includes content generated by the XAI module. For example, if the XAI module generates root cause analysis information, the second indication information includes the root cause analysis information. If the XAI module generates root cause analysis information and obtains an adjustment plan based on the root cause analysis information, the second indication information may include the root cause analysis information, or may include the adjustment plan, or may include the root cause analysis information and the adjustment plan.

[0138] Optionally, if the second indication information includes the generated root cause analysis information, the base station may determine an adjustment plan based on the root cause analysis information and then make corresponding parameter adjustments. If the second indication information includes an adjustment plan, the base station may make corresponding parameter adjustments based on the adjustment plan.

[0139] The method for handling communication anomalies provided in the embodiment of the present application can analyze the causes of possible abnormalities in the communication network by receiving the first indication information sent by the base station, and indicate to the base station the root cause analysis information corresponding to the cause, or indicate an adjustment plan on how to adjust parameters according to the cause, so as to help the base station make corresponding parameter adjustments and reduce the possibility of future communication anomalies.

[0140] Figure 5 is a flowchart of another method for handling communication anomalies provided in an embodiment of the present application. As shown in Figure 5, the method is illustrated by taking the execution of the gNB, XAI module and OAM as an example, but is not limited thereto. For example, when the method is applied in the system 100 provided in Figure 1, the CN can refer to the operations performed by the OAM in this implementation to achieve corresponding effects, etc. The method includes: S301 to S313.

[0141] S301: OAM configures a first judgment condition.

[0142] Optionally, OAM can determine configuration information based on the actual deployment of the gNB. For example, based on the gNB's cell, gNB throughput, and other conditions, OAM can determine configuration information for the gNB and derive a first judgment condition based on the configuration information. The first judgment condition is a judgment condition for predicting whether a network anomaly will occur in the future. The first judgment condition includes multiple second judgment conditions, each of which can be one of an abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event. The first judgment condition can refer to the example provided in S101. The second judgment condition, if referring to this example, can be one of the following: an increase in HO process latency, a service interruption duration corresponding to a service interruption event exceeding a certain value, an occurrence rate corresponding to an SLA violation event exceeding a certain value, or a condition specific to a preset UE.

[0143] S302. OAM sends the first judgment condition to the gNB.

[0144] S303. The gNB receives the first judgment condition sent by OAM.

[0145] Optionally, the embodiment of the present application is illustrated by taking the example of the gNB receiving the first judgment condition sent by OAM to obtain the first judgment condition. In a possible implementation method, the gNB can also obtain the first judgment condition through pre-configuration. If the first judgment condition is obtained in this way, the method may include S303 to S312; or, the base station can obtain the first judgment condition by receiving the first judgment condition sent by other network elements, and the other network elements include CN, etc. If the first judgment condition is obtained in this way, the execution subject of S301 and S302 can be correspondingly changed to other network elements, such as CN.

[0146] In one possible implementation, if the gNB obtains a normal prediction result (i.e., no abnormality) based on the first judgment condition and the AI ​​model, no adjustment is required for the time being. If the prediction result is abnormal, the example operation of S304 is followed.

[0147] S304: The gNB determines, based on the first judgment condition and in combination with the AI ​​model, that the prediction result is abnormal, and then sends a first indication message to the XAI module. The first indication message includes a root cause analysis request.

[0148] For example, the gNB needs to obtain a corresponding prediction result. This prediction result is obtained based on each second condition in the first judgment condition. That is, the input corresponding to each second judgment condition is jointly calculated to obtain an overall prediction result. The overall prediction result may include the corresponding prediction for each second judgment condition. For example, if a second judgment condition is a HO process for a specific UE 1, the prediction result obtained by the gNB must include the prediction for the HO process for the specific UE 1.

[0149] In one possible implementation, if the AI ​​model is not built into the gNB (e.g., the AI ​​model runs in a separate AI module), the gNB may send the first judgment condition to the AI ​​module and instruct the AI ​​module, when providing a prediction result, to include the prediction results corresponding to each of the second judgment conditions in the first judgment condition. Referring to the above example, the prediction result must include the prediction result for the HO process of the specific UE 1. Optionally, the gNB may also request the AI ​​module to provide input information corresponding to the prediction result obtained by the AI ​​model for the first judgment condition. For example, referring to the above example, the prediction result must include the HO process of the specific UE 1. The input information may include the path of the specific UE.

[0150] If the AI ​​model is built into the gNB, when the AI ​​model feeds back a prediction result, the prediction result corresponding to the first judgment condition and the input information corresponding to the prediction result for the first judgment condition can be obtained in the prediction result.

[0151] Exemplarily, the root cause analysis request sent by the gNB to the XAI module includes at least one of the following: an input or output corresponding to the abnormal prediction result; or, a second judgment condition for obtaining the abnormal prediction result; or, an adjustable parameter for obtaining the second judgment condition; or, a time node at which the second indication information is received based on the abnormal prediction result; or, an XAI interpretation method.

[0152] For example, an abnormal prediction result may include an abnormal HO process for a specific UE 1. The input or output corresponding to the abnormal prediction result may include, for example, the path for the specific UE and the service interruption duration for the specific UE 1 exceeding 30 ms. The second judgment condition for obtaining the abnormal prediction result (i.e., the condition indicating resolving the abnormality) may include the HO process for the specific UE 1. For example, the adjustable parameter for obtaining the second judgment condition may include adjustable configuration information corresponding to the service interruption duration (configurations on the gNB may be adjustable or non-adjustable), such as the trigger threshold for the A3 event. Based on the abnormal prediction result, the time point at which the second indication information is received may include the time point at which feedback from the XAI module is required, such as within xx seconds or before aa:bb:cc. The XAI interpretation method may include a specific interpretation method recommended or required for the XAI module to perform root cause analysis.

[0153] S305: The XAI module receives first indication information.

[0154] S306 : The XAI module generates an adjustment plan based on the root cause analysis request.

[0155] Exemplarily, a specific adjustment plan needs to include the type of parameter to be adjusted and the specific adjustment method. For example, the adjustment plan includes the type of parameter to be adjusted and the adjustment method for the parameter to be adjusted; or, the adjustment plan includes the type of parameter to be adjusted, the adjustment method for the parameter to be adjusted, and the corresponding time for the parameter to be adjusted, where the corresponding time for the parameter to be adjusted includes at least one of the adjustment time for the parameter to be adjusted or the verification time after the adjustment of the parameter to be adjusted.

[0156] For example, for anomalies that may occur during the HO process, an adjustment plan may include: indicating the need to adjust the trigger threshold for the A3 event and increase or decrease the power parameter, such as increasing the power parameter by xx decibels (dB). Furthermore, the adjustment plan may include requiring the adjustment to be completed within xx seconds, before aa:bb:cc, or between aa:bb:cc and dd:ee:ff. The adjustment plan may also include the XAI method used by the XAI module, such as LIME or SHAP. The adjustment plan may also include requiring the gNB to verify the adjustment plan and feedback the verification result within xx seconds, before aa:bb:cc, or between aa:bb:cc and dd:ee:ff. In this example, xx represents a numerical value, and aa:bb:cc and dd:ee:ff represent time. The specific values ​​of xx, aa:bb:cc, and dd:ee:ff in each example are not limited, nor is the number of digits in xx.

[0157] Optionally, due to the different specific locations of the XAI module, the impact of the interface between the XAI module and the gNB on the adjusted parameters should also be considered when generating the adjustment plan. For example, if the XAI module is built into the gNB and its interface is the Application Programming Interface (API), if the XAI module is installed separately, its interface with the gNB may be the Nx interface, if the XAI module is installed in another gNB, its interface may be the XN interface between base stations, if the XAI module is installed on the CN side, its interface may be the NG interface, and so on. In this example, if the XAI module is installed separately and its interface with the gNB is the Nx interface, assuming that the Nx interface may cause a 1ms delay, the adjustment plan for the XAI module should take this delay into account when adjusting the timing parameters.

[0158] S307. The XAI module sends second indication information to the gNB, indicating the adjustment plan.

[0159] Optionally, the second indication information sent by the XAI module to the gNB may indicate an adjustment plan by carrying content such as in the example of S202, such as carrying the adjustment plan in the second indication information, or allowing the gNB to obtain the adjustment plan based on the carried root cause analysis information.

[0160] S308. The gNB determines whether the prediction result is abnormal based on the received second indication information and the AI ​​model. If so, execute S309; ​​otherwise, execute S313.

[0161] After receiving the adjustment plan, the gNB can first verify whether the adjustment plan can solve the abnormal problem that may occur in the future, and then determine whether to use the adjustment plan. If the adjustment plan and the AI ​​model determine that the prediction result is abnormal, the XAI module will again obtain the cause of the abnormality and update the adjustment plan. If the adjustment plan and the AI ​​model determine that the prediction result is normal, the adjustment plan can be used.

[0162] S309. The gNB sends a third indication message, where the third indication message includes a root cause analysis request and the cause of the abnormality.

[0163] The root cause analysis request may refer to the root cause analysis request carried in the first indication information of S304. The cause of the abnormality may be the cause of the abnormality determined when the gNB obtains an abnormal prediction result after calculating the adjustment plan through the AI ​​model.

[0164] S310: The XAI module receives third indication information.

[0165] S311. The XAI module updates the adjustment plan based on the root cause analysis request and the cause of the abnormality.

[0166] The XAI module may refer to S306 to obtain an adjustment plan, and consider the cause of the abnormality when obtaining the adjustment plan, and finally obtain an updated adjustment plan.

[0167] S312. The XAI module again sends second indication information to the gNB. The second indication information includes an updated adjustment plan.

[0168] After S312, S308 is executed, where the gNB determines whether the prediction result is abnormal based on the received second indication information, i.e., the updated adjustment plan, and the AI ​​model.

[0169] S313. The gNB adjusts parameters according to the second indication information.

[0170] In the embodiment of the present application, S308 to S313 are the process of the gNB verifying whether the anomaly that will occur in the communication network can be eliminated after the parameter adjustment is performed according to the adjustment plan based on the feedback of the XAI module. If the result of the judgment is that the anomaly has not been eliminated, that is, the prediction result is still an anomaly, the reason for the non-elimination is obtained, and a third indication message is sent to the XAI again, indicating that the anomaly has not been eliminated (in this example, the root cause indication request is carried to indicate that the anomaly has not been eliminated). The reason for the non-elimination is also carried in the third indication message, so that the XAI module updates an adjustment plan and then determines whether the anomaly can be eliminated. If the result of the judgment is that the anomaly has been eliminated, that is, the prediction result is that there is no anomaly, the parameter adjustment can be performed according to the second indication message.

[0171] Optionally, the adjustment scheme indicated by the second execution information may include parameters that need to be adjusted by the gNB, and may also include parameters that need to be adjusted by other nodes in the network. When making corresponding adjustments, the gNB may also indicate to other nodes the parameters that need to be adjusted. Referring to Figure 1, if the adjustment scheme corresponds to the HO process, the triggering threshold of the A3 event needs to be modified, and there are parameters that need to be adjusted by the UE, the gNB can instruct the UE to modify the relevant parameters of the triggering threshold of the A3 event, and the UE will modify the parameters.

[0172] In this embodiment of the present application, the gNB uses a first judgment condition and an AI model to determine whether a future network anomaly will occur. The gNB then interacts with the XAI module to further analyze the cause of the anomaly and develops an adjustment plan to address the cause of the anomaly. The adjustment plan is then verified until it is determined that the adjustment plan can resolve the anomaly, and then the corresponding adjustment is made. This method embodiment can proactively avoid potential network anomalies, achieve automatic network optimization, and enhance network robustness.

[0173] The method for handling communication anomalies provided in an embodiment of the present application can be applied in a system 100 including the above-described architecture. The method is executed by a UE (or a terminal device) or a device in the UE (or a device in the terminal device, such as a chip). The embodiment of the present application uses the method executed by the UE as an example for explanation. Other execution entities can refer to the operation of the UE for execution. Figure 6 is a flow diagram of another method for handling communication anomalies provided in an embodiment of the present application, which includes: S401 to S404.

[0174] S401. The UE obtains a third judgment condition.

[0175] The third judgment condition includes at least one fourth judgment condition. Exemplarily, each fourth judgment condition may be one of an abnormal event, a threshold parameter corresponding to an abnormal event, or a judgment range corresponding to an abnormal event:

[0176] Exemplarily, abnormal events may be specific events in different scenarios, such as increased latency in the HO process, decreased throughput, service interruption, etc., as well as RLF events for edge UEs, or SLA violation events that fail to meet QoE due to limited UE service transmission capabilities.

[0177] Threshold parameters corresponding to abnormal events, such as abnormal events during the HO process, may include: a latency increase of xx ms corresponding to a latency increase event; a throughput decrease of xx% corresponding to a throughput decrease event; or a service interruption duration exceeding xx ms corresponding to a service interruption event. Another example is a RLF occurrence rate exceeding xx% corresponding to an abnormal RLF event of edge UEs; or an occurrence rate exceeding xx% or a corresponding SLA violation UE percentage exceeding xx%. In this example, xx represents a numerical value and does not limit the specific value of xx in each example, nor does it limit the number of digits in the numerical value in each example.

[0178] Judgment range corresponding to abnormal events. For example, a specific judgment range can be preset for abnormal events. The specific judgment range can be a spatial range, such as a preset specific cell, a cell in a preset specific cell list, a preset specific UE, or a UE in a preset specific UE list; the specific judgment range can be a temporal range, such as an effective time range corresponding to a certain abnormal event, such as an effective time of xx s, or starting from aa:bb:cc and ending at dd:ee:ff, or starting from aa:bb:cc, or ending at aa:bb:cc, etc.; the specific judgment range can be a spatial and temporal range, such as one or more specific cells, the effective duration and effective time period in one or more abnormal events. For example, for a specific cell 1 and a specific edge UE 2, starting at 00:00:00, it is judged whether the RLF occurrence rate exceeds 10%; or, for a specific cell 1, it is judged whether the occurrence rate corresponding to the SLA violation event with an effective time of 60s exceeds 5%; or from 10:30:00 to 11:00:00, for a specific UE 1. Determine whether a 10ms service interruption event has occurred. In this example, xx represents a numerical value, and aa:bb:cc and dd:ee:ff represent time. The specific values ​​of xx, aa:bb:cc, and dd:ee:ff in each example are not limited, nor is the number of digits in xx in each example.

[0179] The values ​​of xx, aa:bb:cc and dd:ee:ff in the embodiments of the present application can be determined according to the requirements during use.

[0180] S402. The UE determines that the prediction result is abnormal based on the third judgment condition and the AI ​​model, and sends a fourth indication message, where the fourth indication message includes a root cause analysis request.

[0181] The UE-integrated AI model can run in an AI module, which can be built into the UE or not. For example, the AI ​​module can be set in the OAM, in the UE, or as a separate network element entity. The functions of the AI ​​module can refer to the example in S102. The UE combines the AI ​​model to obtain prediction results, and the gNB combines the AI ​​model to obtain prediction results in S102. This is not repeated here.

[0182] S403. The UE receives fifth indication information, where the fifth indication information includes at least one of root cause analysis information or an adjustment solution generated according to the root cause analysis request.

[0183] Exemplarily, the fifth indication information may be sent by the XAI module, and the fifth indication information includes the root cause analysis information generated by the XAI module according to the root cause analysis request, or the fifth indication information includes the adjustment plan generated by the XAI module according to the root cause analysis request, or the fifth indication information includes the root cause analysis information and the adjustment plan generated by the XAI module according to the root cause analysis request.

[0184] S404. The UE adjusts parameters according to the fifth indication information.

[0185] If the fifth indication information received by the UE includes an adjustment scheme, the UE may adjust the parameters based on the adjustment scheme. If the fifth indication information received by the UE includes only root cause analysis information, the UE may determine the adjustment scheme based on the root cause analysis information and adjust the parameters.

[0186] The method for handling communication anomalies provided in the embodiment of the present application can determine whether an anomaly will occur in the future communication network by obtaining the first judgment condition in combination with the AI ​​model. In the event that an anomaly may occur, the prediction result of the anomaly is interacted with the XAI module. The XAI module analyzes the cause of the possible anomaly, and the UE determines how to adjust the parameters based on the cause of the anomaly, which can effectively reduce the possibility of anomalies occurring in the future.

[0187] FIG7 is a flow chart illustrating another method for handling communication anomalies provided in an embodiment of the present application. This method is executed by an XAI module or a module capable of providing root cause analysis information and adjustment solutions. The XAI module may be built into the UE, or an independent node, or configured in the gNB, or configured on the CN side. This embodiment of the present application illustrates this method using an independent node's XAI module as an example. Other execution entities and other configurations of the XAI module may refer to the operation of the independent node's XAI module. As shown in FIG7 , the method includes: S501 to S503.

[0188] S501: The XAI module receives fourth indication information, where the fourth indication information includes a root cause analysis request obtained according to a third judgment condition.

[0189] The first indication information can be obtained by referring to S401 and S402, and will not be described in detail.

[0190] S502: The XAI module generates at least one of root cause analysis information and adjustment solutions according to the root cause analysis request.

[0191] Optionally, the root cause analysis information may be information related to an analysis of the cause of the anomaly, calculated by the XAI module based on the anomaly prediction result sent by the UE. Optionally, the XAI module may further obtain an adjustment plan based on the root cause analysis information, and then send second indication information including the adjustment plan to the UE, for instructing the UE on how to make the adjustment.

[0192] S503: The XAI module sends fifth indication information, where the fifth indication information includes at least one of the generated root cause analysis information or the adjustment solution.

[0193] The fifth indication information only includes the content generated by the XAI module. For example, if the XAI module generates root cause analysis information, the fifth indication information includes the root cause analysis information; if the XAI module generates root cause analysis information and obtains an adjustment plan based on the root cause analysis information, the fifth indication information includes at least one of the root cause analysis information or the adjustment plan.

[0194] Optionally, the fifth indication information includes the generated root cause analysis information, and the UE determines an adjustment scheme based on the root cause analysis information and performs parameter adjustment. The fifth indication information includes the adjustment scheme, and the UE can perform parameter adjustment based on the adjustment scheme.

[0195] FIG8 is a flow chart of another method for handling communication anomalies provided in an embodiment of the present application. As shown in FIG8 , the method is described by taking the method executed by the UE, the XAI module and the NW as an example. The method includes: S601 to S612.

[0196] S601. The NW indicates the third judgment condition to the UE.

[0197] The third judgment condition includes multiple fourth judgment conditions, each of which can be an abnormal event, a parameter corresponding to each abnormal event, or one of the judgment ranges corresponding to each abnormal event. The third judgment condition and the fourth judgment condition can refer to the example provided in S401.

[0198] S602: The UE obtains a third judgment condition.

[0199] Optionally, the UE may obtain the third judgment condition by pre-configuring the third judgment condition on its own; or, according to the configuration instruction of the first device, request the first device to obtain the third judgment condition. The first device may be a device deployed on the NW side, such as the UE actively requests the relevant abnormality judgment condition from the NW, and the NW indicates to the UE that it has the root cause optimization capability and the third judgment condition; or, the UE receives the third judgment condition after being sent by the OAM, CN, base station (such as gNB) or NW. Alternatively, the UE obtains it from a third-party node through the application layer, for example, the UE obtains the third judgment condition from a network element outside 3GPP through user plane data. Among them, the third judgment condition refers to the example of S401 and will not be repeated here.

[0200] S603. The UE determines that the prediction result is abnormal based on the third judgment condition and the AI ​​model, and then sends fourth indication information to the XAI module, where the fourth indication information includes a root cause analysis request.

[0201] The method for the UE to determine the prediction result based on the third judgment condition and the AI ​​model can refer to the method for the gNB to determine the prediction result based on the first judgment condition and the AI ​​model in S304, and is not further described here. The AI ​​model can be built-in to the UE or not built-in to the UE.

[0202] Optionally, the UE may send the fourth indication information to the XAI module through the NW, that is, the UE sends the fourth indication information to the NW, and the NW forwards it to the XAI module; or, the UE directly sends the fourth indication information to the XAI module.

[0203] Exemplarily, the XAI module can be deployed in different locations. For example, if the XAI module is located on the 3GPP network side (including gNB, CN, and OAM), the UE can receive or forward the fourth indication message to the XAI module through the air interface by the gNB; if the XAI module is located inside the UE, the UE can allow the XAI module to obtain the fourth indication message through the internal API interface; if the XAI module is located on a third-party node on the non-3GPP network side, the UE can establish a link with the third-party node and send the fourth indication message, or the UE requests the NW to forward the fourth indication message to the third-party node. For example, the UE can request the NW to forward the message by indicating the IP address of the third-party node.

[0204] Exemplarily, the root cause analysis request sent by the UE to the XAI module includes at least one of the following: an input or output corresponding to the abnormal prediction result; or a fourth judgment condition for obtaining the abnormal prediction result; or an adjustable parameter for obtaining the fourth judgment condition; or a time point at which the second indication information is received based on the abnormal prediction result; or an XAI interpretation method. For an example, see S304.

[0205] S604: The XAI module receives fourth indication information.

[0206] Optionally, the XAI module may receive fourth indication information sent by the UE, or fourth indication information forwarded from the NW.

[0207] S605 : The XAI module generates an adjustment plan based on the root cause analysis request.

[0208] For this step, refer to the example of S306.

[0209] S606. The XAI module sends fifth indication information to the UE, indicating the adjustment solution.

[0210] Optionally, the XAI module may send the fifth indication information to the UE through the NW, that is, the XAI module sends the fifth indication information to the NW, and the NW forwards it to the UE; or, the XAI module directly sends the fifth indication information to the UE.

[0211] S607. The UE determines whether the prediction result is abnormal based on the received fifth indication information and the AI ​​model. If so, execute S608; if not, execute S612.

[0212] Exemplarily, in this example, if the information between the UE and the XAI module is forwarded based on the NW, the NW may obtain the fifth indication information and indicate it to the UE through the air interface, and the UE may verify the adjustment solution.

[0213] For example, if the fifth indication message is obtained directly by the UE from the XAI module, the UE may report to the NW that it has obtained the adjustment plan of the XAI module. Optionally, the UE may also instruct the NW to refer to step S607 to verify whether the adjustment plan can eliminate the anomaly. This embodiment of the present application uses the UE as an example to illustrate the execution entity of this verification, but is not limited to this.

[0214] The embodiment of the present application is illustrated by taking the fifth indication information including the adjustment plan as an example. The case where the fifth indication information includes the root cause analysis information can refer to this example, and the UE generates the adjustment plan by itself after obtaining the analysis information.

[0215] S608: Send sixth indication information, where the sixth indication information includes a root cause analysis request and an abnormality cause.

[0216] Optionally, the UE may send the sixth indication information to the XAI module via the NW; or the UE may directly send the sixth indication information to the XAI module.

[0217] S609: The XAI module receives sixth indication information.

[0218] Optionally, the XAI module may receive sixth indication information sent by the UE, or sixth indication information forwarded from the NW.

[0219] S610 , the XAI module updates the adjustment plan based on the root cause analysis request and the cause of the abnormality.

[0220] The XAI module may refer to S605 to obtain an adjustment plan, and consider the cause of the abnormality when obtaining the adjustment plan, and finally obtain an updated adjustment plan.

[0221] S611. The XAI module sends fifth indication information to the UE again, where the fifth indication information includes an updated adjustment solution.

[0222] After S611, S607 is executed, and the UE determines whether the prediction result is abnormal based on the received fifth indication information, that is, the updated adjustment plan, combined with the AI ​​model again.

[0223] S612. The UE adjusts parameters according to the fifth indication information.

[0224] The XAI module completes the root cause analysis based on the UE request, obtains an adjustment plan, or an updated adjustment plan, and indicates the obtained adjustment plan to the UE through the fifth indication information. The UE interacts with the NW based on the adjustment plan. If there are parameters that the NW wants to adjust in the adjustment plan, the fifth indication information is sent to the NW or the parameters that need to be adjusted by the NW are sent.

[0225] Exemplarily, the UE executes S607 and determines that the prediction result obtained according to the adjustment scheme carried by the fifth indication message is normal, then the adjustment of the relevant parameters is performed. If it is still abnormal, S608 is performed until the prediction result is normal. For parameter adjustment, if it is a parameter inside the UE, it is performed by the UE. If it is a parameter outside the UE, such as a parameter of the NW, an indication can be sent to the NW to indicate the parameters that the UE needs the NW to adjust. The UE can also be instructed to adjust the relevant parameters for the parameters that need to be adjusted by the NW, and the NW can be informed that the parameter adjustment is to reduce the possibility of network abnormalities.

[0226] If the NW completes the verification of whether the prediction result is abnormal, and the NW determines that the prediction result is normal, the relevant parameters are adjusted. If the prediction result is still abnormal, the NW can indicate the current abnormal prediction result to the UE, which cannot be circumvented. The NW or the UE determines whether to continue to execute S608 and try to update the adjustment plan to eliminate the abnormality. If the NW determines that the prediction result is normal, the parameters that can be adjusted are NW internal parameters. If they are UE parameters, an indication can be sent to the UE to indicate the parameters that the UE needs to adjust. The NW can also be instructed to adjust the relevant parameters for the parameters that the UE needs to adjust, and the UE can be informed that the parameter adjustment is to reduce the possibility of network abnormalities.

[0227] In this embodiment of the application, the UE uses the third judgment condition, combined with the AI ​​model, to determine whether a future network anomaly will occur. Furthermore, through interaction with the XAI module, the UE further analyzes the cause of the anomaly and obtains an adjustment plan corresponding to the anomaly that will occur. The adjustment plan is then verified until it is determined that the adjustment plan can resolve the anomaly, and then the corresponding adjustment is made. This method embodiment can avoid possible network anomalies in advance, achieve automatic network optimization, and enhance network robustness.

[0228] Figure 9 is a structural diagram of a base station or a base station device provided in an embodiment of the present application. As shown in Figure 9, the base station 10 or the device 10 in the base station includes: an acquisition module 101, a sending module 102, a receiving module 103 and a processing module 104.

[0229] The acquisition module 101 is configured to acquire a first judgment condition.

[0230] The sending module 102 is used to send a first indication message after the processing module 104 determines that the prediction result is abnormal based on the first judgment condition and the AI ​​model, and the first indication message includes a root cause analysis request.

[0231] The receiving module 103 is configured to receive second indication information, where the second indication information includes at least one of root cause analysis information or an adjustment solution generated according to the root cause analysis request.

[0232] The processing module 104 is further configured to adjust parameters according to the second indication information.

[0233] In one possible implementation, the processing module 104 is specifically configured to determine, based on the second indication information and in combination with the AI ​​model, that the prediction result is normal, and to adjust parameters according to the second indication information.

[0234] In one possible implementation, sending module 102 is further configured to send third indication information after the processing module determines that the prediction result is abnormal based on the second indication information and the AI ​​model, the third indication information including a root cause analysis request and the cause of the abnormality. Receiving module 103 is configured to receive the second indication information, wherein the root cause analysis information or adjustment plan in the second indication information has been updated based on the root cause analysis request and the cause of the abnormality.

[0235] In a possible implementation, the receiving module 102 is further configured to receive a first judgment condition sent by the OAM or CN.

[0236] In a possible implementation, the first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

[0237] In one possible implementation, the root cause analysis request includes at least one of the following: an input or output corresponding to an abnormal prediction result; or a second judgment condition for obtaining an abnormal prediction result; or an adjustable parameter for the second judgment condition; or

[0238] The time point at which the second indication information is received according to the abnormal prediction result; or, the interpretation method of XAI.

[0239] In one possible implementation, the adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, the adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted, and the corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

[0240] It should be understood that the modules shown in FIG9 are merely examples, and each module may perform its operations with reference to the method portion in the embodiments of the present application, or perform variations of its operations, such as the operations performed by a base station (e.g., a gNB) in FIG1 , FIG3 , and FIG5 . In the examples provided in the embodiments of the present application, each module of the base station may also perform other operations, and the examples in the embodiments of the present application are not limiting.

[0241] FIG10 is a schematic structural diagram of an XAI device according to an embodiment of the present application. As shown in FIG10 , the XAI device 40 includes a receiving module 401 , an XAI module 402 , and a sending module 403 .

[0242] The receiving module 401 is configured to receive first indication information, where the first indication information includes a root cause analysis request obtained according to a first judgment condition.

[0243] The XAI module 402 is configured to generate at least one of root cause analysis information and adjustment solutions according to the root cause analysis request.

[0244] The sending module 403 is configured to send second indication information, where the second indication information includes at least one of the generated root cause analysis information and the adjustment solution.

[0245] In one possible implementation, the receiving module 401 is further used to receive third indication information, which includes a root cause analysis request and an abnormality cause; the XAI module 402 is further used to update at least one of the root cause analysis information or adjustment solutions based on the root cause analysis request and the abnormality cause; and the sending module 403 is further used to send second indication information, which includes at least one of the updated root cause analysis information or adjustment solutions.

[0246] In a possible implementation, the first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

[0247] In one possible implementation, the root cause analysis request includes at least one of the following: an input or output corresponding to an abnormal prediction result; or a second judgment condition for obtaining an abnormal prediction result; or an adjustable parameter for the second judgment condition; or

[0248] The time point at which the second indication information is received according to the abnormal prediction result; or, the interpretation method of XAI.

[0249] In one possible implementation, the adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, the adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted, and the corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

[0250] It should be understood that the modules shown in FIG10 are merely examples, and each module may perform its operations with reference to the method portion in the embodiments of the present application, or perform variations of its operations, such as the operations performed by the XAI device in FIG4 and FIG5. In the examples provided in the embodiments of the present application, each module of the XAI device may also perform other operations, and is not limited to the examples in the embodiments of the present application.

[0251] Figure 11 is a structural diagram of a UE or a device in a UE provided in an embodiment of the present application. As shown in Figure 11, the UE 30 or the device 30 in the UE includes: an acquisition module 301, a sending module 302, a receiving module 303 and a processing module 304.

[0252] The acquisition module 301 is used to acquire the third judgment condition.

[0253] The sending module 302 is used to send a fourth indication message after the processing module 304 determines that the prediction result is abnormal based on the third judgment condition and the artificial intelligence AI model, and the fourth indication message includes a root cause analysis request.

[0254] The receiving module 303 is configured to receive fifth indication information, where the fifth indication information includes at least one of root cause analysis information or an adjustment solution generated according to the root cause analysis request.

[0255] The processing module 304 is further configured to adjust parameters according to the fifth indication information.

[0256] In a possible implementation, the processing module 304 is specifically configured to determine, based on the fifth indication information and in combination with the AI ​​model, that the prediction result is normal, and to adjust parameters according to the fifth indication information.

[0257] In one possible implementation, the sending module 302 is also used to send the sixth indication information after the processing module 304 determines that the prediction result is abnormal based on the fifth indication information and combined with the AI ​​model, and the sixth indication information includes the root cause analysis request and the cause of the abnormality; the receiving module 303 is used to receive the fifth indication information, and the root cause analysis information or adjustment plan in the fifth indication information has been updated according to the root cause analysis request and the cause of the abnormality.

[0258] In a possible implementation, the processing module 304 is specifically configured to preconfigure a third judgment condition.

[0259] The sending module 302 is further configured to request the first device to obtain the third judgment condition according to the configuration instruction of the first device received by the receiving module 303. The receiving module 303 is further configured to receive the third judgment condition.

[0260] The receiving module 303 is further configured to obtain a third judgment condition sent by the OAM or CN, the base station or the NW.

[0261] In a possible implementation, the sending module 302 is specifically configured to send the fourth indication information to the XAI device, or send the fourth indication information to the XAI device through the NR.

[0262] In a possible implementation, the sending module 302 is specifically configured to send the sixth indication information to the XAI device, or send the sixth indication information to the XAI device through the NR.

[0263] In a possible implementation, the receiving module 303 is specifically configured to receive fifth indication information sent by the XAI apparatus, or fifth indication information sent through the NR, where the fifth indication information is generated by the XAI apparatus.

[0264] It should be understood that the modules shown in FIG11 are merely examples, and each module may perform its operations with reference to the method portion in the embodiments of the present application, or perform variations of its operations, such as the operations performed by the UE in FIG1 , FIG6 , and FIG8 . In the examples provided in the embodiments of the present application, each module of the UE may also perform other operations, and is not limited to the examples in the embodiments of the present application.

[0265] FIG12 is a schematic structural diagram of an XAI device according to an embodiment of the present application. As shown in FIG12 , the XAI device 50 includes a receiving module 501 , an XAI module 502 , and a sending module 503 .

[0266] The receiving module 501 is configured to receive fourth indication information, where the fourth indication information includes a root cause analysis request obtained according to the third judgment condition.

[0267] The XAI module 502 is configured to generate at least one of root cause analysis information and adjustment solutions according to the root cause analysis request.

[0268] The sending module 503 is configured to send fifth indication information, where the fifth indication information includes at least one of the generated root cause analysis information or the adjustment solution.

[0269] In one possible implementation, receiving module 501 is further configured to receive sixth indication information, where the sixth indication information includes a root cause analysis request and an exception cause. XAI module 502 is further configured to update at least one of the root cause analysis information and an adjustment plan based on the root cause analysis request and the exception cause. Sending module 503 is further configured to send fifth indication information, where the fifth indication information includes at least one of the updated root cause analysis information and an adjustment plan.

[0270] In a possible implementation, the receiving module 501 is further configured to receive fourth indication information sent by the UE, or to receive the fourth indication information sent by the UE through the new radio interface NR.

[0271] In a possible implementation, the receiving module 501 is further configured to receive sixth indication information sent by the UE, or to receive the sixth indication information sent by the UE through the NR.

[0272] In a possible implementation, the sending module 503 is further configured to send fifth indication information to the UE, or to send the fifth indication information to the UE through the NR.

[0273] It should be understood that the modules shown in FIG12 are merely examples, and each module may perform its operations with reference to the method portion in the embodiments of the present application, or perform variations of its operations, such as the operations performed by the XAI device in FIG7 and FIG8. In the examples provided in the embodiments of the present application, each module of the XAI device may also perform other operations, and is not limited to the examples in the embodiments of the present application.

[0274] In the embodiment of the present application, the base station or the device in the base station provided in Figure 9 can be applied in the scenario provided in Figure 1, as a component of the base station 10 or UE 10 to implement the communication method provided in the embodiment of the present application. The UE or the device in the UE provided in Figure 11 can be applied in the scenario provided in Figure 1, as a component of the UE 30 or UE 30 to implement the communication method provided in the embodiment of the present application. Exemplarily, the system 100 provided in Figure 1 can be a 5G NR system, and the 5G NR system includes CN 20 and RAN, and the interface between RAN and CN 20, such as 5GC, is NG (usually denoted as NG-RAN). 5GC includes a network element access and mobility management function (AMF) and a user plane function (UPF), and NG-RAN can include at least one network device, such as a 5G network device (usually denoted as gNB) and a 4G network device connected to the 5GC (usually denoted as ng-eNB). When the network device provides services to the UE, the gNB is responsible for providing the 5G NR user plane and control plane protocol functions for the UE, and the ng-eNB is responsible for providing the 4G E-UTRA user plane and control plane protocol functions for the UE. Figure 1 is an example of a system architecture for implementing an embodiment of the present application, which is used to illustrate a possible implementation scenario of the embodiment of the present application. The communication method of the embodiment of the present application can also be used in other systems and is not limited thereto. In some actual use scenarios, the gNB can send information to the UE, such as paging information, information related to cell residence determination, etc. The UE can send information to the gNB, such as an RRC connection establishment request message or a recovery request message, etc. The communication anomaly handling method provided in the embodiment of the present application can be combined and applied in these scenarios.

[0275] In addition, as shown in Figure 13, Figure 13 is a schematic diagram of the structure of a device 80 according to an embodiment of the present application. The device 60 shown in Figure 13 includes a transceiver 601 and a processor 602. The device 60 can be used to perform methods S101 to S104, or S201 to S203, or S301 to S313, or S401 to S404, or S501 to S503, or S601 to S612 in the above embodiments. The device 60 is equivalent to the base station (e.g., gNB), XAI device, or UE mentioned in the method.

[0276] It should be noted that the division of the various parts in the embodiments of the present application is schematic and is merely a logical functional division. In actual implementation, other division methods may be used. The various functions in the embodiments of the present application may be integrated into a single processor, or the transceiver and processor may exist separately. The aforementioned integrated devices may be implemented in the form of hardware, such as a chip, or in the form of software functional units.

[0277] In addition, an embodiment of the present application further provides a device 70, as shown in FIG14 , which is a schematic diagram of the structure of a device 70 provided in an embodiment of the present application. As shown in FIG14 , the device 70 may include a processor 701, a memory 702 coupled to the processor 701, and a transceiver 703. The transceiver 703 may include an MR, an LR, a communication interface, an optical module, etc., for receiving messages or data information, etc. The processor 701 may include a central processing unit (CPU), a network processor (NP), or a combination of a CPU and an NP, for executing the relevant steps of the wake-up signal processing in the device exemplified in the above embodiment. The processor may also be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above-mentioned PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. The processor 701 may refer to a single processor or may include multiple processors. The memory 702 may include a volatile memory, such as a random-access memory (RAM); the memory may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); the memory 702 may also include a combination of the above types of memory. The memory 702 may refer to a single memory or may include multiple memories for storing program instructions. In one embodiment, the memory 702 stores computer-readable instructions, which include multiple software modules, such as a sending module, a processing module, and a receiving module. After executing each software module, the processor 701 may perform corresponding operations according to the instructions of each software module. In this embodiment, the operation performed by a software module actually refers to the operation performed by the processor 701 according to the instructions of the software module.Optionally, the processor 701 may also store program codes or instructions for executing the embodiments of the present application. In this case, the processor 701 does not need to read the program codes or instructions from the memory 702.

[0278] The device 70 may be configured to execute the methods in the above embodiments. Specifically, the device 70 may execute methods S101 to S104 in the above embodiments, or execute the operations in S301 to S313 performed by a base station (e.g., a gNB). Alternatively, the device 70 may execute methods S401 to S404 in the above embodiments, or execute the operations in S601 to S612 performed by a UE. Alternatively, the device 70 may execute methods S201 to S203, S301 to S313, S501 to S503, or S601 to S612 performed by an XAI device in the above embodiments.

[0279] In addition, embodiments of the present application further provide a communication device. The communication device includes a storage medium and a processor connected to the storage medium. The storage medium stores instructions, and when the instructions are executed by the processor, the processor is configured to implement some or all of the operations of any of the methods in any of the aforementioned embodiments.

[0280] An embodiment of the present application also provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is executed on a processor, it implements part or all of the operations in any of the methods in any of the aforementioned embodiments.

[0281] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed on a processor, implements part or all of the operations in any of the methods in any of the aforementioned embodiments.

[0282] The present application also provides a chip including an interface circuit and a processor connected to each other, wherein the processor is configured to cause the chip to execute part or all of the operations in any of the methods in any of the aforementioned embodiments.

[0283] An embodiment of the present application also provides a chip system, including: a processor, the processor is coupled to a memory, the memory is used to store programs or instructions, when the program or instructions are executed by the processor, the chip system implements part or all of the operations of any one of the methods of any one of the embodiments described above.

[0284] Optionally, there may be one or more processors in the chip system. The processor may be implemented in hardware or software. When implemented in hardware, the processor may be a logic circuit, an integrated circuit, etc. When implemented in software, the processor may be a general-purpose processor implemented by reading software code stored in a memory.

[0285] Optionally, the memory in the chip system may be one or more. The memory may be integrated with the processor or may be provided separately from the processor, which is not limited in the embodiments of the present application. For example, the memory may be a non-transient processor, such as a read-only memory (ROM), which may be integrated with the processor on the same chip or provided on different chips. The embodiments of the present application do not specifically limit the type of memory or the configuration of the memory and the processor.

[0286] Exemplarily, the chip system can be an FPGA, an ASIC, a system on chip (SoC), a CPU, an NP, a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0287] The present application also provides a system including one or more of the above-mentioned devices, apparatuses, computer-readable storage media, computer program products, chips, or chip systems, which can be applied to the scenario shown in FIG1 , but is not limited thereto.

[0288] The terms "first," "second," "third," "fourth," and the like (if any) in the specification and claims of this application and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or sequential sequence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0289] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0290] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical business division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0291] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0292] In addition, each business unit in each embodiment of the present application can be integrated into a processing unit, each unit can exist physically separately, or two or more units can be integrated into a single unit. The above-mentioned integrated units can be implemented in the form of hardware or software business units.

[0293] If the integrated unit is implemented in the form of a software business unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the technical solution of the present application can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, ROM, RAM, Random Access Memory, disk or optical disk, etc. Various media that can store program code.

[0294] Those skilled in the art will appreciate that, in one or more of the examples above, the services described herein may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these services may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one location to another. Storage media may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0295] The above specific implementation methods further describe in detail the purpose, technical solutions and beneficial effects of this application. It should be understood that the above are only specific implementation methods of this application.

[0296] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for handling communication anomalies, characterized in that: include: Obtaining a first judgment condition; Based on the first judgment condition and in combination with the artificial intelligence (AI) model, determining that the prediction result is abnormal, and sending first indication information, where the first indication information includes a root cause analysis request; receiving second indication information, where the second indication information includes at least one of root cause analysis information or an adjustment plan generated according to the root cause analysis request; Perform parameter adjustment according to the second indication information.

2. The method according to claim 1, characterized in that The adjusting parameters according to the second indication information includes: Based on the second indication information and in combination with the AI ​​model, it is determined that the prediction result is normal, and parameters are adjusted according to the second indication information.

3. The method according to claim 1 or 2, characterized in that Before adjusting the parameters according to the second indication information, the method further includes: Based on the second indication information and in combination with the AI ​​model, determining that the prediction result is abnormal, and sending third indication information, wherein the third indication information includes a root cause analysis request and a cause of the abnormality; The second indication information is received, where the root cause analysis information or the adjustment plan in the second indication information has been updated according to the root cause analysis request and the abnormality cause.

4. The method according to any one of claims 1 to 3, characterized in that The first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

5. The method according to any one of claims 1, 3 or 4, characterized in that The root cause analysis request includes at least one of the following: The input or output corresponding to the abnormal prediction result; or A second judgment condition for obtaining an abnormal prediction result; or Obtaining an adjustable parameter of the second judgment condition; or, The time point at which the second indication information is received according to the abnormal prediction result; or Explanation methods for explainable artificial intelligence (XAI).

6. The method according to any one of claims 1 to 5, characterized in that The adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, The adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted. The corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

7. The method according to any one of claims 1 to 6, characterized in that The obtaining of the first judgment condition includes: Receive the first judgment condition sent by the operation, maintenance and management OAM network element or the core network CN.

8. A method for handling communication anomalies, characterized in that: include: receiving first indication information, where the first indication information includes a root cause analysis request obtained according to a first judgment condition; generating at least one of root cause analysis information or an adjustment plan according to the root cause analysis request; Second indication information is sent, where the second indication information includes at least one of the generated root cause analysis information or the adjustment solution.

9. The method according to claim 8, characterized in that Also includes: receiving third indication information, wherein the third indication information includes a root cause analysis request and an abnormality cause; updating at least one of root cause analysis information or an adjustment plan according to the root cause analysis request and the cause of the abnormality; Second indication information is sent, where the second indication information includes at least one of the updated root cause analysis information or the adjustment solution.

10. The method according to claim 8 or 9, characterized in that The first judgment condition includes at least one abnormal event, a parameter corresponding to each abnormal event, or a judgment range corresponding to each abnormal event.

11. The method according to any one of claims 8 to 10, characterized in that The root cause analysis request includes at least one of the following: The input or output corresponding to the abnormal prediction result; or A second judgment condition for obtaining an abnormal prediction result; or Obtaining an adjustable parameter of the second judgment condition; or, The time point at which the second indication information is received according to the abnormal prediction result; or Explanation methods for explainable artificial intelligence (XAI).

12. The method according to any one of claims 8 to 10, characterized in that The adjustment scheme includes the type of the parameter to be adjusted and the adjustment method of the parameter to be adjusted; or, The adjustment scheme includes the type of the parameter to be adjusted, the adjustment method of the parameter to be adjusted and the corresponding time of the parameter to be adjusted. The corresponding time of the parameter to be adjusted includes the adjustment time of the parameter to be adjusted, or at least one of the verification time after the adjustment of the parameter to be adjusted.

13. A base station, characterized in that: include: An acquisition module, configured to acquire a first judgment condition; a sending module, configured to send first indication information including a root cause analysis request after the processing module determines that the prediction result is abnormal based on the first judgment condition and in combination with the artificial intelligence (AI) model; a receiving module, configured to receive second indication information, where the second indication information includes at least one of root cause analysis information or an adjustment plan generated according to the root cause analysis request; The processing module is further configured to adjust parameters according to the second indication information.

14. The base station according to claim 13, characterized in that The processing module is specifically used to determine that the prediction result is normal based on the second indication information and in combination with the AI ​​model, and adjust parameters according to the second indication information.

15. The base station according to claim 13 or 14, characterized in that The sending module is further configured to send third indication information after the processing module determines that the prediction result is abnormal based on the second indication information and in combination with the AI ​​model, where the third indication information includes a root cause analysis request and a cause of the abnormality; The receiving module is configured to receive the second indication information, wherein the root cause analysis information or the adjustment plan in the second indication information has been updated according to the root cause analysis request and the abnormality cause.

16. The base station according to any one of claims 13 to 15, characterized in that: The receiving module is further configured to receive an operation, maintenance and management OAM network element or a core network CN.

17. An explainable artificial intelligence (XAI) device, characterized in that: include: a receiving module, configured to receive first indication information, where the first indication information includes a root cause analysis request obtained according to a first judgment condition; An XAI module is configured to generate at least one of root cause analysis information or an adjustment plan according to the root cause analysis request; The sending module is used to send second indication information, where the second indication information includes at least one of the generated root cause analysis information or the adjustment plan.

18. The device according to claim 17, characterized in that The receiving module is further configured to receive third indication information, wherein the third indication information includes a root cause analysis request and an abnormality cause; The XAI module is further configured to update at least one of root cause analysis information or an adjustment plan based on the root cause analysis request and the cause of the abnormality; The sending module is further configured to send second indication information, where the second indication information includes at least one of the updated root cause analysis information or the adjustment solution.

19. A communication device, characterized in that: The communication device comprises a processor, wherein the processor is configured to execute the method according to any one of claims 1 to 7, or configured to execute the method according to any one of claims 8 to 12.

20. A computer-readable storage medium, characterized in that The computer-readable storage medium comprises instructions, which, when executed, enable the method according to any one of claims 1 to 7 to be implemented, or enable the method according to any one of claims 8 to 12 to be implemented.

21. A computer program product, characterized in that The computer program product comprises instructions, which, when executed, enable the method according to any one of claims 1 to 7 to be implemented, or enable the method according to any one of claims 8 to 12 to be implemented.

Images