Method and communication device for data acquisition

Abstract

A method for data collection and a communication device, the method comprising: receiving, by a first device, a first signal carrying first data, the first data being collection data; and performing, by the first device, a first processing on second data to obtain a first tag, the first processing comprising at least encoding, the second data being obtained by performing a second processing on the first signal, the second processing comprising at least decoding, the second data being associated with the first data. After receiving the first signal carrying the first data, the first device can perform at least the decoding on the first signal, and perform a CRC check on the decoding result (e.g. the second data). The first data can be understood as the collection data, and the CRC check on the decoding result by the first device can be understood as that the first device judges the validity of the collection data through the CRC check.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a data acquisition method and a communication device. Background Technology

[0002] With the development of wireless communication technology, wireless networks can not only provide traditional wireless communication connectivity services, but also be deeply integrated with artificial intelligence (AI) to achieve network intelligence. AI or machine learning (ML) based on neural networks, when applied to wireless communication systems, can generally achieve accurate modeling and prediction of wireless data through data-driven training, enabling the completion of various wireless tasks, such as channel estimation, channel prediction, data traffic prediction, or beam prediction.

[0003] In future network architectures, base stations can be divided into primary and secondary devices. The primary device handles some higher-level physical layer functions, high-level protocol stacks, and control functions, while the secondary device handles radio frequency (RF) and some lower-level physical layer processing. In technical solutions that integrate wireless networks and AI, AI models can be deployed on base stations to achieve network intelligence. However, in a base station-separated architecture, assuming the AI ​​model is deployed on the secondary device (i.e., the input to the AI ​​model is located on the secondary device, while the corresponding performance information is located on the primary device), determining the validity of the collected data requires further discussion. Summary of the Invention

[0004] This application provides a data acquisition method and communication device, which enables the determination of the validity of acquired data in a scenario where the base station architecture is separated.

[0005] Firstly, a data acquisition method is provided. This method can be executed by a first device. Unless otherwise specified, "first device" can refer to the first device itself, a component of the first device (e.g., a processor, chip, or chip system), or a logic module or software that implements all or part of the first device. The component used in the first device can be a component within the first device, or it can be independent of the first device but coupled to it. For ease of understanding, the following description uses the execution of the first device in a network device as an example.

[0006] The method includes: a first device receiving a first signal from a second device, the first signal carrying first data; and, if the second data successfully undergoes a cyclic redundancy check (CRC), performing a first processing on the second data to obtain a first tag, the first processing including at least encoding, and the second data being obtained by performing a second processing on the first signal, the second processing including at least decoding.

[0007] For example, decoding, or what might be called decoding.

[0008] For example, the first processing may also include one or more of the following operations: scrambling, CRC check, or rate matching.

[0009] For example, the second process may also include one or more of the following operations: descrambling, rate matching dematching (or rate dematching), demodulation, equalization, or CRC check.

[0010] For example, the first data can be understood as the input data for the AI ​​functions included in the second device.

[0011] For example, the first signal carrying first data can be understood as the first data being transmitted through the first signal after some processing. As an example, this processing includes processing of AI functions, which at least include AI functions included in the second device.

[0012] According to the method provided in this application, after receiving a first signal carrying first data, the first device can first perform at least decoding on the first signal and perform CRC verification on the decoding result (or the decoded data). Here, the first data can be understood as collected data, and the CRC verification of the decoding result by the first device can be understood as judging the validity of the collected data. If the CRC verification is successful, the decoding result is then at least encoded. In other words, if the CRC verification fails, encoding is not required. This not only improves the accuracy of the encoded data but also reduces the resource waste caused by encoding invalid data. Furthermore, at least encoding the decoding result yields a tag. Since this tag is based on the decoding result, it can be used for subsequent operations. For example, when the input data for the AI ​​function in the second device is the first data, the first tag can be used as the output data of the AI ​​function, and the AI ​​function can be trained using the first data and the first tag.

[0013] In conjunction with the first aspect, in some possible implementations, the first label is associated with the first data.

[0014] For example, the first label is obtained by processing the second data through a first process. The second data is obtained by processing the first signal through a second process, whereby the first signal carries the first data, meaning the first label is associated with the first data. Associating the first label with the first data can be similar to the operation of an AI function, which may include training and / or fine-tuning the AI ​​function.

[0015] In conjunction with the first aspect, in some possible implementations, the method further includes: a first device sending a first identifier, the first identifier being used to indicate first data corresponding to a first tag.

[0016] In conjunction with the first aspect, in some possible implementations, the first device sends a first identifier, including: when T1 is greater than or equal to a first threshold, the first device sends an identifier set, the identifier set including the first identifier, the identifier set including the identifier corresponding to the data that has been successfully verified by CRC, wherein T1 represents the number of tags obtained after successful CRC verification, and the tags obtained after successful CRC verification include the first tag.

[0017] For example, T1 can represent the number of tags obtained after the first device has fully accumulated all tags that have successfully passed CRC verification, that is, T1 does not limit the time for the first device to accumulate tags that have successfully passed CRC verification; or, T1 can represent the number of tags obtained after the first device has accumulated all tags that have successfully passed CRC verification within a certain time period.

[0018] For example, a time period can be predefined, configured, or indicated.

[0019] For example, the identifier corresponding to the data, or it can also be called the data identifier.

[0020] Based on the above scheme, when T1 is greater than or equal to the first threshold, the first device sends an identifier set, thereby avoiding the first device from performing a CRC check every time. Furthermore, when the CRC check is successful, the first device sends the identifier corresponding to the successful CRC check to the second device, thus reducing the number of times the first device sends identifiers.

[0021] In conjunction with the first aspect, in some possible implementations, the method further includes: a first device storing a first sample, the first sample including a first tag and a first identifier, the first identifier being used to indicate first data corresponding to the first tag, and the first sample being used for operation of AI functions in a second device.

[0022] Based on the above example, if the second data passes the CRC check, the first device stores the first sample to avoid storing invalid data and the tag corresponding to the invalid data, thus wasting the storage resources of the first device.

[0023] In conjunction with the first aspect, in some possible implementations, the first identifier includes time information corresponding to the first data and the second data.

[0024] For example, the time information corresponding to the first data and the second data can refer to the information of the first data and the second data on the air interface time domain resources.

[0025] For example, this time information could be a time slot, a frame, or a timestamp, etc.

[0026] In conjunction with the first aspect, in some possible implementations, the method further includes: the first device sending a first tag.

[0027] In conjunction with the first aspect, in some possible implementations, the first device sends the first tag, including: when T2 is greater than or equal to the second threshold, the first device sends a tag set, the tag set including the first tag, the tag set including tags corresponding to data that has been successfully verified by CRC, wherein T2 represents the number of tags obtained after successful CRC verification.

[0028] For example, T2 can represent the number of tags obtained after the first device has fully accumulated all tags that have successfully passed CRC verification, that is, T2 does not limit the time for the first device to accumulate tags that have successfully passed CRC verification; or, T2 can represent the number of tags obtained after the first device has accumulated all tags that have successfully passed CRC verification within a certain time period.

[0029] For example, a time period can be predefined, configured, or indicated.

[0030] Based on the above scheme, when T2 is greater than or equal to the second threshold, the first device sends a set of tags, thereby avoiding the first device sending the tag corresponding to the successful CRC check every time it passes the CRC check, thus reducing the number of times the first device sends tags.

[0031] In conjunction with the first aspect, in some possible implementations, the method further includes: the first device sending the correspondence between the first data and the first tag.

[0032] In conjunction with the first aspect, in some possible implementations, the method further includes: a first device sending first request information, the first request information being used to request the acquisition of first data; and the first device receiving the first data.

[0033] In conjunction with the first aspect, in some possible implementations, the first request information includes a first identifier.

[0034] In conjunction with the first aspect, in some possible implementations, the first identifier is used to indicate the first data corresponding to the first label.

[0035] In conjunction with the first aspect, in some possible implementations, the method further includes: a first device storing a second sample, the second sample including a first label and first data, the second sample being used for the operation of AI functions in the second device.

[0036] In conjunction with the first aspect, among some possible implementation methods, the operation of the AI ​​function in the second device includes: training and / or fine-tuning.

[0037] In conjunction with the first aspect, in some possible implementations, the AI ​​functionality in the second device includes one or more of the following: AI channel estimation, AI equalization, or AI demodulation.

[0038] In conjunction with the first aspect, in some possible implementations, before the first device receives the first signal, the method further includes: the first device sending acquisition information corresponding to the first data, the acquisition information being used to indicate the acquisition time.

[0039] In conjunction with the first aspect, in some possible implementation methods, the information collected includes one or more of the following: collection start time, collection period information, offset value of collection start time, collection interval, or collection end time.

[0040] Secondly, a data acquisition method is provided. This method can be executed by a second device. Unless otherwise specified, "second device" can refer to the second device itself, a component of the second device (e.g., a processor, chip, or chip system), or a logic module or software that implements all or part of the second device. The component used in the second device can be a component within the second device, or it can be independent of the second device but coupled to it. For ease of understanding, the following description uses the execution of the method by a second device in a network device as an example.

[0041] The method includes: a second device sending a first signal carrying first data; the second device receiving a first identifier indicating the first data; and the second device storing a third sample including the first data and the first identifier, the third sample being used for the operation of AI functions in the second device.

[0042] It should be understood that the second aspect corresponds to the content in the first aspect above, and supplementary explanations and technical effects can be found in the description of the first aspect above.

[0043] In conjunction with the second aspect, in some possible implementations, the operation of the AI ​​function in the second device includes: training and / or fine-tuning.

[0044] In conjunction with the second aspect, in some possible implementations, the AI ​​functions in the second device include one or more of the following: AI channel estimation, AI equalization, or AI demodulation.

[0045] In conjunction with the second aspect, in some possible implementations, before the second device sends the first signal, the method further includes: the second device receiving acquisition information corresponding to the first data, the acquisition information being used to indicate the acquisition time; and the second device acquiring the first data based on the acquisition information.

[0046] In conjunction with the second aspect, in some possible implementation methods, the information collected includes one or more of the following: collection start time, collection period information, offset value of collection start time, collection interval, or collection end time.

[0047] Thirdly, a data acquisition method is provided. This method can be executed by a second device. Unless otherwise specified, "second device" can refer to the second device itself, a component of the second device (e.g., a processor, chip, or chip system), or a logic module or software that implements all or part of the second device. The component used in the second device can be a component within the second device, or it can be independent of the second device but coupled to it. For ease of understanding, the following description uses the execution of the method by a second device in a network device as an example.

[0048] The method includes: a second device sending a first signal carrying first data; the second device receiving a first tag associated with the first data; and the second device storing a second sample including the first data and the first tag, the second sample being used for the operation of AI functions in the second device.

[0049] It should be understood that the second aspect corresponds to the content in the first aspect above, and supplementary explanations and technical effects can be found in the description of the first aspect above.

[0050] In conjunction with the third aspect, in some possible implementations, the second device receives the first tag, which includes: the second device receiving a tag set, the tag set including the first tag, and the tag set including tags corresponding to data that have successfully passed cyclic redundancy check (CRC).

[0051] In conjunction with the third aspect, in some possible implementations, the method further includes: the second device receiving the correspondence between the first data and the first tag.

[0052] In conjunction with the third aspect, among some possible implementation methods, the operation of the AI ​​function in the second device includes: training and / or fine-tuning.

[0053] In conjunction with the third aspect, in some possible implementations, the AI ​​functions in the second device include one or more of the following: AI channel estimation, AI equalization, or AI demodulation.

[0054] In conjunction with the third aspect, in some possible implementations, before the second device sends the first signal, the method further includes: the second device receiving acquisition information corresponding to the first data, the acquisition information being used to indicate the acquisition time; and the second device acquiring the first data based on the acquisition information.

[0055] In conjunction with the third aspect, in some possible implementation methods, the information collected includes one or more of the following: collection start time, collection period information, collection start time offset, or collection end time.

[0056] Fourthly, a data acquisition method is provided. This method can be executed by a second device. Unless otherwise specified, "second device" can refer to the second device itself, a component of the second device (e.g., a processor, chip, or chip system), or a logic module or software that implements all or part of the second device. The component used in the second device can be a component within the second device, or it can be independent of the second device but coupled to it. For ease of understanding, the following description uses the execution of the method by a second device in a network device as an example.

[0057] The method includes: a second device sending a first signal carrying first data; the second device receiving first request information for requesting to obtain the first data; and sending the first data.

[0058] In conjunction with the fourth aspect, in some possible implementations, the first request information includes a first identifier.

[0059] In conjunction with the fourth aspect, in some possible implementations, the first identifier is used to indicate the first data.

[0060] In conjunction with the fourth aspect, among some possible implementation methods, the operation of the AI ​​function in the second device includes: training and / or fine-tuning.

[0061] In conjunction with the fourth aspect, in some possible implementations, the AI ​​functions in the second device include one or more of the following: AI channel estimation, AI equalization, or AI demodulation.

[0062] In conjunction with the fourth aspect, in some possible implementations, before the second device sends the first signal, the method further includes: the second device receiving acquisition information corresponding to the first data, the acquisition information being used to indicate the acquisition time; and the second device acquiring the first data based on the acquisition information.

[0063] In conjunction with the fourth aspect, in some possible implementation methods, the information collected includes one or more of the following: collection start time, collection period information, offset value of collection start time, collection interval, or collection end time.

[0064] Fifthly, a communication device is provided, which has the functions of the first device described above. The communication device has the functions of the first aspect described above. For example, the communication device includes modules, units or means corresponding to the operations involved in the first aspect described above. The modules, units or means can be implemented by software, or by hardware, or by a combination of software and hardware.

[0065] For example, the communication device may be the aforementioned terminal device, such as a module or unit (e.g., a chip, a chip system, or a circuit) corresponding to the method, operation, step, or action described in the first aspect above.

[0066] In one possible implementation, the communication device includes a transceiver unit (or communication module) and a processing unit (or processing module) connected to the transceiver unit.

[0067] For example, the transceiver unit is configured to: receive a first signal from a second device, the first signal carrying first data; and the processing unit is configured to: if the second data is successfully subjected to cyclic redundancy check (CRC) verification, perform a first processing on the second data to obtain a first tag, the first processing including at least encoding, the second data being obtained by performing a second processing on the first signal, the second processing including at least decoding.

[0068] In a sixth aspect, a communication device is provided, which has the functions of the second device described above. For example, the communication device includes modules, units, or means corresponding to the operations involved in any one of the second to fourth aspects described above. The modules, units, or means can be implemented by software, hardware, or a combination of software and hardware.

[0069] For example, the communication device may be the aforementioned network device, such as a module or unit (e.g., a chip, a chip system, or a circuit) corresponding to the method, operation, step, or action described in the second aspect above.

[0070] In one possible implementation, the communication device includes a transceiver unit (or communication module) and a processing unit (or processing module) connected to the transceiver unit.

[0071] For example, the transceiver unit is configured to: send a first signal, the first signal carrying first data; the transceiver unit is further configured to: receive a first identifier, the first identifier being used to indicate the first data; the processing unit is configured to: store a third sample, the third sample including the first data and the first identifier, the third sample being used for the operation of the AI ​​function in the second device.

[0072] For example, the communication device may be the aforementioned network device, such as a module or unit (e.g., a chip, a chip system, or a circuit) corresponding to the method, operation, step, or action described in the third aspect above.

[0073] In one possible implementation, the communication device includes a transceiver unit (or communication module) and a processing unit (or processing module) connected to the transceiver unit.

[0074] For example, the transceiver unit is configured to: send a first signal carrying first data; the transceiver unit is further configured to: receive a first tag associated with the first data; and the processing unit is configured to: store a second sample, the second sample including the first data and the first tag, the second sample being used for the operation of AI functions in the second device.

[0075] For example, the communication device may be the aforementioned network device, such as a module or unit (e.g., a chip, a chip system, or a circuit) corresponding to the method, operation, step, or action described in the fourth aspect above.

[0076] In one possible implementation, the communication device includes a transceiver unit (or communication module).

[0077] For example, the transceiver unit is configured to: send a first signal, the first signal carrying first data; the transceiver unit is further configured to: receive first request information, the first request information being used to request the acquisition of the first data; and the sending unit is configured to: send the first data.

[0078] Optionally, the communication device may also include a processing unit that can be used to perform processing operations other than receiving and sending.

[0079] In a seventh aspect, a communication device is provided. The communication device may be the aforementioned terminal device or network device. The communication device includes a processor configured to retrieve and execute the computer program from a memory, causing the communication device to perform the methods in any of the possible implementations of the first to fourth aspects described above.

[0080] Optionally, the communication device further includes a transceiver and a memory. The processor controls the transceiver to send and receive signals, and the memory stores computer programs. The memory can be integrated into the processor or set up independently.

[0081] Optionally, there may be one or more processors and one or more memories.

[0082] Alternatively, the memory can be integrated with the processor, or the memory can be set separately from the processor.

[0083] Optionally, the transceiver includes a transmitter and a receiver.

[0084] Eighthly, a communication device is provided, comprising one or more processors for executing computer programs or instructions, which, when executed, cause the communication device to implement the methods in any possible design or implementation of the first to fourth aspects described above. Optionally, the communication device further comprises a memory for storing part or all of the computer programs or instructions for implementing the functions involved in the first to fourth aspects described above.

[0085] In one possible design, the communication device may also include interface circuitry, wherein the processor is used to communicate with other devices or components via the interface circuitry.

[0086] The aforementioned communication device may be a terminal, or a communication module in a terminal, or a chip in a terminal that is responsible for communication functions, such as an MDEM chip, or a SoC chip or a system-in-a-package (SIP) chip that includes a modem module.

[0087] The aforementioned communication device may be a network device / terminal device, or a communication module in a network device / terminal device, or a circuit or chip in a network device / terminal device responsible for communication functions, or a functional module in a network device / terminal device capable of calling and executing programs.

[0088] Ninthly, a communication system is provided. The communication system includes a first device and / or a second device, wherein the first device is configured to perform the method in any possible implementation of the first aspect, and the second device is configured to perform the method in any possible implementation of the second to fourth aspects.

[0089] A tenth aspect provides a computer-readable storage medium. This computer-readable storage medium stores computer program code or instructions to cause the methods in any of the possible implementations of the first to fourth aspects to be implemented. For example, when the computer program code or instructions are executed, the methods in any of the possible implementations of the first to fourth aspects are implemented.

[0090] Eleventhly, a computer program product is provided. The computer program product includes computer program code or instructions to cause the methods in any of the possible implementations of the first to fourth aspects to be implemented. For example, when a computer reads and executes the computer program product, the methods in any of the possible implementations of the first to fourth aspects are implemented.

[0091] In a twelfth aspect, a computer program is provided. When the computer program is run, it causes the methods in any of the possible implementations of the first to fourth aspects to be implemented. Attached Figure Description

[0092] Figure 1 This is a schematic diagram of a communication system applicable to this application;

[0093] Figure 2 This is a schematic diagram of another communication system applicable to this application;

[0094] Figure 3 This is a schematic diagram of an artificial intelligence (AI) application framework applicable to this application;

[0095] Figure 4 This is a flowchart illustrating a communication method provided in an embodiment of this application;

[0096] Figure 5 This is a schematic diagram of a network architecture provided in an embodiment of this application;

[0097] Figure 6 This is a schematic block diagram of a communication device provided in an embodiment of this application;

[0098] Figure 7 This is a schematic block diagram of another communication device provided in the embodiments of this application;

[0099] Figure 8 This is a schematic diagram of the structure of an AI processor provided in an embodiment of this application. Detailed Implementation

[0100] To facilitate understanding of the above embodiments provided in this application, the following points are made:

[0101] 1) In this application, unless otherwise specified or in case of logical conflict, the terms and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0102] 2) In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can mean: a, or, b, or, c, or, a and b, or, a and c, or, b and c, or, a, b, and c. Here, a, b, and c can each be single or multiple.

[0103] 3) In this application, the terms "first," "second," and various numerical designations (e.g., #1, #2, etc.) indicate distinctions made for ease of description and are not intended to limit the scope of the embodiments of this application. For example, they may distinguish different messages, rather than describing a specific order or sequence. It should be understood that such descriptions can be interchanged where appropriate to describe solutions other than those in the embodiments of this application.

[0104] 4) In this application, descriptions such as “when…”, “under the circumstances of…” and “if” all refer to the fact that the device will make corresponding processing under certain objective circumstances. They are not time limits, nor do they require the device to make a judgment action when it is implemented, nor do they mean that there are other limitations.

[0105] 5) In this application, "instruction" or "for instruction" can include both direct and indirect instruction. When describing an instruction as being used to instruct A, it may include whether the instruction directly instructs A or indirectly instructs A, but does not necessarily mean that the instruction carries A.

[0106] The indication methods involved in the embodiments of this application should be understood to cover various methods that enable the party to be indicated to obtain the information to be indicated. The information to be indicated can be sent as a whole or divided into multiple sub-information and sent separately. Moreover, the sending period and / or sending time of these sub-information can be the same or different. This application does not limit the sending method, for example.

[0107] The "instruction information" in the embodiments of this application can be an explicit instruction, that is, a direct instruction through signaling, or an instruction obtained by combining other rules or parameters with the parameters indicated by the signaling, or by deduction. It can also be an implicit instruction, that is, an instruction obtained based on rules or relationships, or based on other parameters, or by deduction. This application does not specifically limit it in this regard.

[0108] 6) In this application, "communication" or "transmission" can also be described as "data transmission," "information transmission," "data processing," "signaling transmission," "broadcast channel transmission," etc. "Transmission" includes "sending" and / or "receiving." "Transmission" can be described as "output."

[0109] 7) In this application, "sending information to XX (device)" can be understood as the destination of the information being that device. This can include sending information directly or indirectly to that device. "Receiving information from XX (device), or receiving information from XX (device)" can be understood as the source of the information being that device, and can include receiving information directly or indirectly from that device. Information may undergo necessary processing between the source and destination, such as format changes, but the destination can understand the valid information from the source. Similar expressions in this application can be understood in a similar way, and will not be elaborated further here.

[0110] 8) In this application, the words “exemplary,” “example,” “for instance,” “e.g.,” are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as an “example” in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word “example” is intended to present a concept in a concrete manner. In the embodiments of this application, “of,” “corresponding, relevant,” and “corresponding” may sometimes be used interchangeably, and it should be noted that their intended meanings are consistent unless their distinction is emphasized.

[0111] 9) In this application, "predefined" can mean a standard protocol predefined, or it can mean something pre-agreed upon or negotiated between devices. Here, "protocol" can refer to a standard protocol in the field of communications, such as a fourth-generation (4G) protocol. th Generation 4G network, fifth generation (5G) network th This application does not limit the scope to network protocols such as 5G (generation, 5G), New Radio (NR), 5.5G, and related protocols applied in future communication networks. "Predefined" may include "preconfiguration," which can be achieved by pre-storing corresponding codes, tables, or other methods that can indicate relevant information in the device.

[0112] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0113] The technical solutions of this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, 5G systems, or New Radio (NR) and future communication systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems.

[0114] Furthermore, the embodiments of this application are applicable to both homogeneous and heterogeneous network scenarios, and there are no restrictions on the transmission points. They can be applied to systems such as multi-point collaborative transmission between macro base stations, micro base stations, and macro base stations. The embodiments of this application are applicable to both low-frequency and high-frequency scenarios, including terahertz and optical communications.

[0115] In a communication system, a device can send signals to or receive signals from another device. These signals may include reference signals, information, signaling, or data. In this application, "device" can be replaced by an entity, network entity, communication equipment, communication module, node, or communication node.

[0116] Figure 1 This is a schematic diagram of a communication system applicable to an embodiment of this application. For example... Figure 1 As shown, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. RAN 100 includes at least one RAN node (e.g., ...). Figure 1 110a and 110b, collectively referred to as 110) and at least one terminal (such as Figure 1 RAN100, denoted as RAN100, comprises RAN nodes 120a-120j, collectively referred to as RAN120. RAN100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment. Figure 1(Not shown in the image). Terminal 120 is connected to RAN node 110 wirelessly. RAN node 110 is connected to core network 200 wirelessly or via wired connection. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0117] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as a 4G mobile communication system, a 5G mobile communication system, or a future-oriented evolution system. RAN 100 can also be an open access network (O-RAN or ORAN), a cloud radioaccess network (CRAN), or a wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0118] RAN node 110, sometimes also referred to as network equipment, access network equipment, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in the communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative, for example... Figure 1 Network element 120i can be a helicopter or a drone, and it can be configured as a mobile base station. For terminals 120j that access RAN 100 through network element 120i, network element 120i is a base station; however, for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes referred to as communication devices, for example... Figure 1 Network elements 110a and 110b can be understood as communication devices with base station functions, while network elements 120a-120j can be understood as communication devices with terminal functions.

[0119] In one possible scenario, a RAN node can be a base station (BS), an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a base station in a future mobile communication system, or an access node in a WiFi system, etc. Figure 1110a), micro base stations or indoor stations (such as Figure 1 In CRAN scenarios, RAN nodes can be 110b), relay nodes or donor nodes, or wireless controllers. Optionally, RAN nodes can also be servers, wearable devices, vehicles, or in-vehicle equipment. For example, in vehicle-to-everything (V2X) technology, the access network equipment can be a roadside unit (RSU).

[0120] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be central units (CU), distributed units (DU), CU-control plane (CU-CP), CU-user plane (CU-UP), radio units (RU), or CU-radio units (CU-RU), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radioheads (RRHs).

[0121] In different systems, CU (including open CU-CP (O-CU-CP) and open CU-UP (O-CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called an open central unit (O-CU), DU can also be called an open distributed unit (O-DU), CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0122] Terminal 120 can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. A terminal can also be referred to as user equipment (UE), terminal, user device, access terminal, user unit, user station, mobile station, mobile station (MS), remote station, remote terminal, mobile device, user terminal, terminal unit, terminal station, terminal device, wireless communication equipment, user agent, or user device. A terminal typically contains a communication module, circuit, or chip that performs the corresponding communication functions. The terminal may also be configured with program instructions for performing these communication functions.

[0123] For example, the terminal in this application embodiment can be a mobile phone, a personal digital assistant (PDA) computer, a laptop computer, a tablet computer, a drone, a computer with wireless transceiver capabilities, a machine type communication (MTC) terminal, a virtual reality (VR) terminal, an augmented reality (AR) terminal, an Internet of Things (IoT) terminal, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical care, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home (e.g., game consoles, smart TVs, smart speakers, smart refrigerators, and fitness equipment), a transport vehicle with wireless communication capabilities, a communication module, or a roadside unit (RSU) with terminal capabilities.

[0124] RAN 100 and terminal 120 can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the scenarios in which RAN 100 and terminal 120 are located.

[0125] Communication between access network devices and terminal devices follows a specific protocol layer structure. This protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: radio resource control (RRC) layer, packet data convergence protocol (PDCP) layer, radio link control (RLC) layer, media / medium access control (MAC) layer, or physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: service data adaptation protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or physical layer, etc.

[0126] The correspondence between network elements and their achievable protocol layer functions in the ORAN system can be found in Table 1 below.

[0127] Table 1

[0128] ORAN network elements 3GPP protocol layer functions O-CU-CP RRC+PDCP-Control Plane (PDCP-C) O-CU-UP SDAP+PDCP - User Plane (PDCP-U) O-DU RLC+MAC+PHY-high O-RU PHY-low

[0129] CN 200 can be a 5G core network or an evolved 5G core network. Taking a 5G core network as an example, CN 200 includes access and mobility management (AMF) network elements responsible for mobility management and access management services; session management (SMF) network elements responsible for session management; user plane (UPF) network elements responsible for user plane packet routing and forwarding and quality of service (QoS) control; and policy control (PCF) network elements. These core network elements can operate independently or be combined to implement certain control functions; for example, AMF, SMF, and PCF can be combined into a single core network device.

[0130] The communication system 10 provided in this application may further include AI network elements for implementing some or all AI-related operations. AI network elements may also be referred to as AI nodes, AI devices, AI entities, AI modules, AI models, or AI units, etc. The AI ​​network elements may be built into the network elements of the communication system. For example, an AI network element may be an AI module built into access network equipment, core network equipment, cloud servers, or operation, administration and maintenance (OAM) management systems to implement AI-related functions. The OAM may be the management system for core network equipment and / or the management system for access network equipment. Alternatively, the AI ​​network element may be an independently configured network element in the communication system. Optionally, the terminal or its built-in chip may also include an AI entity for implementing AI-related functions.

[0131] Figure 2 This is a schematic diagram of a possible application framework in a communication system applicable to embodiments of this application. For example... Figure 2 As shown, network elements in a communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes (RAN nodes), terminals, or one or more devices in the OAM, are equipped with one or more AI modules (for clarity, ...). Figure 2 (Only one is shown in the image). The access network node can be a single RAN node or can include multiple RAN nodes, such as RU, CU, and DU. One or more of the RU, CU, or DU can also be equipped with one or more AI modules. Optionally, the CU can also be split into CU-CP and CU-UP. One or more AI modules are provided in CU-CP and / or CU-UP.

[0132] The AI ​​module is used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can implement different functions. The AI ​​module model can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or bias in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The bias in the activation function can also be referred to as the neural network bias.

[0133] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0134] It should be understood that the above naming is defined solely for the purpose of distinguishing different functions and should not constitute any limitation on this application. This application does not preclude the possibility of using other naming conventions in 5G networks and other future networks. For example, in future networks, some or all of the above-mentioned network elements may use the terminology from 5G, or they may use other names, etc.

[0135] Understandable. Figure 1 or Figure 2 This is merely an example and does not constitute a limitation on the scope of protection of this application. The communication method provided in the embodiments of this application may also involve... Figure 1 The network elements not shown in the diagram may also include, of course, the communication method provided in this application embodiment, only network elements not shown in the diagram. Figure 1 Some of the network elements are shown.

[0136] To facilitate understanding of the embodiments of this application, the basic concepts involved in this application will be explained first.

[0137] 1. AI: AI enables machines to possess human-like intelligence, for example, allowing machines to use computer hardware and software to simulate certain intelligent human behaviors. To achieve artificial intelligence, machine learning methods can be employed. In machine learning, machines learn (or train) models using training data. This model represents the mapping between inputs and outputs. The learned model can be used for reasoning (or prediction), that is, it can be used to predict the output corresponding to a given input. This output can also be called the reasoning result (or prediction result).

[0138] This document explains some basic concepts in the field of AI, which does not limit the scope of protection of the embodiments of this application.

[0139] (1) Machine learning (ML):

[0140] Machine learning is a crucial technological approach to achieving AI. AI endows machines with human-like intelligence, using computer hardware and software to simulate certain intelligent human behaviors, including machine learning and other methods. Machine learning refers to learning models or rules from raw data, such as neural networks, decision trees, and support vector machines. Machine learning can be categorized into supervised learning, unsupervised learning, and reinforcement learning.

[0141] Supervised learning, based on collected sample values ​​and labels, uses machine learning algorithms to learn the mapping relationship between sample values ​​and labels, and expresses this learned mapping relationship using a machine learning model. The process of training the machine learning model is the process of learning this mapping relationship. For example, in signal detection, the noisy received signal is the sample, and the corresponding real constellation point is the label. Machine learning aims to learn the mapping relationship between samples and labels through training, that is, to enable the machine learning model to learn a signal detector. During training, the model parameters are optimized by calculating the error between the model's predicted values ​​and the real labels. Once the mapping relationship is learned, it can be used to predict the sample label of each new sample. The mapping relationship learned in supervised learning can include linear mappings and nonlinear mappings. Based on the type of label, the learning task can be divided into classification tasks and regression tasks.

[0142] Unsupervised learning relies solely on collected sample values, using algorithms to discover inherent patterns within the samples. One type of unsupervised learning algorithm uses the samples themselves as supervisory signals; that is, the model learns the mapping relationship from sample to sample, which is called self-supervised learning. During training, model parameters are optimized by calculating the error between the model's predictions and the samples themselves. Self-supervised learning can be used for signal compression and decompression recovery applications; common algorithms include autoencoders and generative adversarial networks.

[0143] Reinforcement learning, unlike supervised learning, is a type of algorithm that learns problem-solving strategies through interaction with the environment. Unlike supervised and unsupervised learning, reinforcement learning problems do not have explicit "correct" action labels. The algorithm needs to interact with the environment to obtain reward signals from the environment, and then adjust its decision actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each terminal device based on the total system throughput feedback from the wireless network, aiming to achieve a higher system throughput. The goal of reinforcement learning is also to learn the mapping relationship between the environment state and the optimal / near-optimal decision actions. However, because the label of the "correct action" cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action." Reinforcement learning training is achieved through iterative interaction with the environment.

[0144] Deep neural networks (DNNs) are a specific implementation of machine learning. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Traditional communication systems rely on extensive expert knowledge to design communication modules, while DNN-based deep learning communication systems can automatically discover hidden pattern structures from large datasets, establish mapping relationships between data, and achieve performance superior to traditional modeling methods.

[0145] Based on their construction method, DNNs can be divided into feedforward neural networks (FNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs). FNNs can be neural networks where neurons in adjacent layers are completely connected pairwise, which makes FNNs typically require a large amount of storage space and have high computational complexity.

[0146] CNNs are neural networks specifically designed to process data with a grid-like structure. For example, time-series data (discrete sampling along the time axis) and image data (two-dimensional discrete sampling) can both be considered grid-like data. CNNs do not use all the input information at once for computation; instead, they use a fixed-size window to extract a portion of the information for convolution operations, which significantly reduces the computational cost of model parameters. Furthermore, depending on the type of information extracted by the window (such as people and objects in an image representing different types of information), each window can use different convolution kernels, allowing CNNs to better extract features from the input data.

[0147] Recurrent Neural Networks (RNNs) are a type of distributed neural network (DNN) that utilizes feedback time-series information. Their input includes the current input value and their own output value from the previous time step. RNNs are well-suited for acquiring temporally correlated sequence features, and are particularly applicable to applications such as speech recognition and channel coding / decoding.

[0148] AI models refer to function models that map a certain-dimensional input to a certain-dimensional output, and their parameters can be obtained through machine learning training. For example, f(X) = aX² + b is a quadratic function model, which can be viewed as an AI model. a and b correspond to the model's parameters and can be obtained through machine learning training. Data used for model training, validation, and / or testing in machine learning can form datasets or training datasets. The quantity and / or quality of data in these datasets or training datasets will affect the effectiveness of machine learning. Model training involves selecting an appropriate loss function (which measures the difference between the model's predictions and the true values) and using optimization algorithms to train the model parameters to minimize the loss function value. Model testing involves evaluating the model's performance using test data after training. Model application involves using the trained model to solve real-world problems.

[0149] A neural network, or artificial neural network, is a mathematical model that mimics the behavioral characteristics of animal neural networks to perform distributed parallel information processing. It is a special form of AI model.

[0150] (2) Model training:

[0151] Model training involves selecting an appropriate function (such as a loss function) and using optimization algorithms to train the model parameters so that the difference between the model's predicted values ​​and the ground truth (or target values, labels) tends to be minimized.

[0152] For example, model training methods include, but are not limited to, supervised learning, self-supervised learning, and knowledge distillation.

[0153] (3) Model file and model parameters:

[0154] Model files and / or model parameters can be used to determine the model. Optionally, the model in this application may refer to the model itself, or it may refer to the model files and / or model parameters used to determine the model.

[0155] The model file can be used to indicate the model structure, which may include, but is not limited to, FNN, CNN, or RNN. The model file can have a fixed format, such as a standard predefined format, or a format pre-negotiated by both ends of the interface. Model parameters can refer to parameters in the neural network model, such as, but not limited to, the number of layers in the neural network, the type and weights of neurons in each layer, etc. This application does not limit the method of distributing model parameters.

[0156] Take DNN as an example. The idea behind DNN comes from the neuronal structure of the brain. Each neuron can perform a weighted summation operation on its inputs and then use the result of the weighted summation operation to generate the output through a non-linear function. For example, the input of a neuron is x = [x0, x1, ..., x...]. N-1 The weights corresponding to the inputs are w = [w0, w1, ..., w] N-1 The bias of the weighted summation is b. The nonlinear function f() can take many forms; for example, the nonlinear function f() can be the maximum value function max{0, x}. Then the effect of a neuron's execution is... Where N is a positive integer, and n is a positive integer greater than or equal to 0 and less than or equal to (N-1). The weights of the weighted summation operation of neurons in a neural network and the nonlinear function are called the parameters of the neural network. The parameters of all neurons in a neural network constitute the parameters of the neural network.

[0157] A DNN typically has multiple neural network layers, including an input layer, one or more hidden layers, and an output layer. Generally, the first layer is the input layer, the last layer is the output layer, and the layers in between are hidden layers. Each layer contains multiple neurons. Layers are fully connected; that is, any neuron in the i-th layer is connected to any neuron in the (i+1)-th layer. The input layer processes the received values ​​(i.e., the DNN's input) through neurons and then passes them to the hidden layers. Similarly, the hidden layers pass the computation results to the final output layer, producing the DNN's output. This application does not limit the structure and parameters used in the AI ​​model.

[0158] One of the model structure or model parameters can be predefined, while the other can be sent by the sender (e.g., the network side). Alternatively, both the model structure and model parameters can be sent by the sender (e.g., the network side). This application does not impose any restrictions on this.

[0159] Sending a model can refer to sending a model file and / or model parameters, while receiving a model can refer to receiving a model file and / or model parameters.

[0160] (4) Training dataset and inference data:

[0161] In the field of machine learning, ground truth usually refers to data that is considered accurate or real.

[0162] Training datasets are used to train AI models. A training dataset can include the input to the AI ​​model, or it can include both the input and the target output of the AI ​​model. Specifically, a training dataset includes one or more training data points, which can include training samples input to the AI ​​model or the target output of the AI ​​model. The target output can also be referred to as a label, sample label, or labeled sample. The label is the ground truth value.

[0163] In the field of communications, training datasets can include simulation data collected through simulation platforms, experimental data collected from experimental scenarios, or measured data collected in actual communication networks. Because the geographical environment and channel conditions where the data is generated vary—for example, indoor / outdoor conditions, movement speed, frequency bands, or antenna configurations—the collected data can be categorized during acquisition. For instance, data with the same channel propagation environment and antenna configuration can be grouped together.

[0164] Model training essentially involves learning certain features from training data. In training AI models (such as neural network models), the goal is to make the model's output as close as possible to the desired predicted value. This is achieved by comparing the network's current predictions with the target value and updating the weight vector of each layer based on the difference. (Of course, there's usually an initialization process before the first update, where parameters are pre-configured for each layer.) For example, if the network's prediction is too high, the weight vector is adjusted to predict a lower value. This adjustment continues until the AI ​​model can predict the target value or a value very close to it. Therefore, it's necessary to predefine "how to compare the difference between the predicted and target values," which is the loss function or objective function. These are important equations used to measure the difference between the predicted and target values. Taking the loss function as an example, a higher output value (loss) indicates a greater difference. Therefore, training the AI ​​model becomes a process of minimizing this loss, making the loss function value less than a threshold, or making the loss function value meet the target requirements. For example, if the AI ​​model is a neural network, adjusting the model parameters of the neural network includes adjusting at least one of the following parameters: the number of layers of the neural network, the width, the weights of the neurons, or the parameters in the activation function of the neurons.

[0165] Inference data can be used as input to a trained AI model for inference. During the model's inference process, the inference data is input into the AI ​​model, and the corresponding output, which is the inference result, is obtained.

[0166] Figure 3 This is a schematic diagram of an AI application framework applicable to this application. This AI application framework can be used to implement model lifecycle management (LCM), and the core functions of LCM may include, for example... Figure 3 The diagram shows a data acquisition module, a model training module, a management module, an inference module, and a model storage module. The management module interacts with and manages the other modules, the model storage module stores and manages the AI ​​models, and the model training and inference modules are examples of AI modules.

[0167] In the data collection phase, the data source provides both training and inference datasets. In the model training phase, the AI ​​model is obtained by analyzing or training the data provided by the data source. The AI ​​model represents the mapping relationship between the model's inputs and outputs.

[0168] For example, such as Figure 1 and Figure 2 The network devices, terminal devices, gNB CUs, DUs, or other management entities shown can input data into the data acquisition module, which serves as a database for AI model training and data analysis inference. The model training module analyzes and trains the network based on the training data output by the data acquisition module, providing a usable AI model, which is then deployed to the model management module. The model inference module, based on the inference data output by the data acquisition module, provides reasonable predictions about network operation based on the AI ​​model (obtained through training), and feeds back the AI ​​model's performance data to the model training module. The model training module then continues training the model based on the feedback performance data and feeds the updated AI model back to the data acquisition module for storage.

[0169] It is understood that a communication system may include network elements (or devices, functions) with artificial intelligence capabilities. The aforementioned AI model design-related steps can be executed by one or more network elements with artificial intelligence capabilities. In one possible design, AI functions (such as AI modules or AI entities) can be configured within existing network elements in the communication system to implement AI-related operations, such as AI model training and / or inference. For example, the existing network element could be a network device or a terminal device. Alternatively, in another possible design, an independent network element can be introduced into the communication system to perform AI-related operations, such as training an AI model. This independent network element can be called an AI network element or an AI node, etc., and this application embodiment does not limit this name. Exemplarily, the AI ​​network element can be directly connected to network devices in the communication system, or it can be indirectly connected to network devices through third-party network elements. The third-party network element can be a core network element such as an authentication management function (AMF) network element, a user plane function (UPF) network element, an operation administration and maintenance (OAM) network element, a cloud server, or other network elements, without limitation. For example, the independent network element can be deployed on one or more of the following: the network device side, the terminal device side, or the core network side. Optionally, it can be deployed on a cloud server.

[0170] The training processes of different models can be deployed on different devices or nodes, or on the same device or node. Similarly, the inference processes of different models can be deployed on different devices or nodes, or on the same device or node. For example, the model parameters of an AI model may include one or more of the following: model structure parameters (e.g., the number of layers, and / or weights), model input parameters (e.g., input dimension, number of input ports), or model output parameters (e.g., output dimension, number of output ports). It can be understood that the input dimension refers to the size of an input data set; for example, when the input data is a sequence, the corresponding input dimension can indicate the length of the sequence. The number of input ports can refer to the quantity of input data. Similarly, the output dimension can refer to the size of an output data set; for example, when the output data is a sequence, the corresponding output dimension can indicate the length of the sequence. The number of output ports can refer to the quantity of output data.

[0171] With the development of wireless communication technology, intelligence will further evolve at the wireless network architecture level. AI and wireless networks will be further deeply integrated to achieve network intelligence. In network intelligence scenarios, in addition to providing traditional communication connectivity services, the network can also combine with AI to provide AI services, realizing the inclusiveness, real-time nature, and high security of network intelligence.

[0172] To achieve network intelligence, AI models can be deployed in base stations. These AI models can be used to implement physical layer functions in network devices. In future network architectures, the separation of base station architecture may be a common design. For example, in a scenario with a separated base station architecture, the base station can be divided (or segmented) into multiple parts. In other words, a base station can include multiple parts, or it can be composed of multiple parts, or multiple parts can constitute a single base station. For instance, a base station can be divided into a first device and a second device. The first device can be responsible for some higher-layer physical layer functions in the base station, and may further include one or more of the higher-layer (also called higher-level) protocol stack and some control functions, etc. The second device can be responsible for radio frequency and some lower-layer physical layer processing functions in the base station. In technical solutions that integrate wireless networks and AI, AI models can be deployed in base stations to achieve network intelligence. In uplink transmission scenarios, the second device sends data to the first device. Assuming the second device has deployed an AI model, meaning the input to the AI ​​model is located in the second device, and the corresponding performance information is located in the first device, how to determine the validity of the collected data in the uplink transmission scenario requires further discussion.

[0173] Based on the above description, this application provides a data acquisition method, aiming to provide a method for judging the validity of acquired data in a scenario where the base station architecture is separated.

[0174] In this application embodiment, in a network device (e.g., base station) architecture separation scenario, the method provided by this application is described using the example of a network device comprising two parts (e.g., a first part and a second part). Of course, the method provided by this application also applies to cases where the network device includes multiple parts (e.g., a third part, or a fourth part, etc.). For example, a network device divided into two parts may include a first device and a second device. The second device may be responsible for radio frequency and some lower-level physical layer processing functions in the network device, while the first device may be responsible for some higher-level physical layer functions in the base station, and may further include one or more of the following: a higher-level (also called a higher-layer) protocol stack and some control functions. Considering an architecture separation scenario, the first device carries a portion of the functions in the network device, and the second device carries another portion of the functions in the network device. Alternatively, it can be understood that the functions included in the network device can be divided (or segmented) into two parts, with one part carried by the first device and the other by the second device. It is evident that the functions carried by the first device and the second device may differ depending on the division location (or segmentation location).

[0175] It should be understood that in some scenarios, the first device may be referred to as the first part, the first unit, the first module, or the baseband high unit (BBH). The second device may be referred to as the second part, the second unit, the second module, or the baseband low unit (BBL). In the O-RAN scenario, the first device may also be a DU in the O-RAN, and the second device may also be a RU in the O-RAN. The RU and DU in the O-RAN are also applicable to the method provided in this application.

[0176] The methods provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings, and can be applied to the above-mentioned methods. Figure 1 or Figure 2 The communication system shown. It should be understood that the embodiments of this application can be applied to scenarios where the sending end and the receiving end communicate.

[0177] It should also be understood that the embodiments shown below do not specifically limit the structure of the execution subject of the method provided in the embodiments of this application, as long as it is possible to communicate according to the method provided in the embodiments of this application by running the code or program that records the method provided in the embodiments of this application. For example, the method provided in the embodiments of this application can be executed by a first device or a second device in a network device. Unless otherwise specified, the "first device" in this application can refer to the first device itself, or a component in the first device (e.g., a communication module, processor, circuit, chip (such as a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, or a chip system, etc.), or it can be a logic module or software that can implement all or part of the functions of the first device. The "second device" in this application can refer to the second device itself, or a component in the second device (e.g., a communication module, processor, circuit, chip, or chip system, etc.), or it can be a logic module or software that can implement all or part of the functions of the second device.

[0178] For ease of description, the following embodiments use a first device (e.g., BBH) and a second device (e.g., BBL) as examples.

[0179] Figure 4 This is a schematic flowchart illustrating a data acquisition method provided in an embodiment of this application. Figure 4 As shown, the method may include the following steps.

[0180] It should be understood that Figure 4 The method shown is applicable to uplink transmission scenarios, where BBL includes AI functionality and BBH includes a functional unit for obtaining decoding results.

[0181] 401, BBL sends the first signal to BBH. Correspondingly, BBH receives the first signal from BBL.

[0182] The first signal carries the first data.

[0183] For example, the first signal carrying first data can be understood as the first data being transmitted through the first signal after some processing. As an example, this processing includes processing of AI functions, which at least include AI functions included in the second device.

[0184] For example, after the first data is processed by one or more functional units, it is carried in a first signal and sent to the BBH by the BBL. The one or more functional units include all or some of the functional units in the BBL; specifically, it may include the AI ​​function in the BBL.

[0185] The functional units in the BBL are associated with the splitting positions of the BBL and BBH in the network device.

[0186] For example, Figure 5 This illustrates various possible splitting points in a network device separation architecture scenario. A splitting point refers to the location where the functions within a network device are divided into BBL and BBH. For example, if the functions within a network device are split at a certain location, resulting in two parts, one part handled by BBL and the other by BBH, this location is referred to as the splitting point for ease of description. Furthermore, the term "splitting" is mentioned multiple times in the embodiments of this application; this "splitting" can refer to a logical division of functions within the network device.

[0187] It should be understood that the network device separation architecture may include various segmentation locations, which are not listed here. This application only uses the following four segmentation locations as examples for illustration, and is not a limitation. For different segmentation locations, this application mainly uses the location of the channel estimation unit, equalization unit, and demodulation unit in the BBH or BBL as examples for introduction. Of course, in addition to the above three functional units, the network device may also include other functional units (such as one or more of the following: radio frequency unit, layer mapping unit, descrambling unit, or derate matching unit, etc.). For different segmentation locations, the specific locations of the above other functional units may also be different.

[0188] like Figure 5 As shown, the cutting positions may include cutting position 1, cutting position 2, cutting position 3 and cutting position 4.

[0189] For the segmentation position 1, BBH includes a channel estimation unit, an equalization unit, and a demodulation unit;

[0190] For the corresponding segmentation position 2: BBL includes a channel estimation unit, and BBH includes an equalization unit and a demodulation unit;

[0191] For the corresponding segmentation position 3: BBL includes a channel estimation unit and an equalization unit, and BBH includes a demodulation unit;

[0192] For the segmentation position 4, the BBL includes a channel estimation unit, an equalization unit, and a demodulation unit.

[0193] For example, for segmentation position 2, the AI ​​functions included in the BBL may include AI channel estimation; for segmentation position 3, the AI ​​functions included in the BBL may include AI channel estimation and / or AI equalization; for segmentation position 4, the AI ​​functions included in the BBL may include one or more of AI channel estimation, AI equalization, or AI demodulation.

[0194] For example, the segmentation position in the network device is segmentation position 2. The BBL includes AI signal estimation. The first data can be the input data of AI channel estimation. The first signal carries the first data, which can be the output data after the first data has been processed by AI channel estimation.

[0195] For example, the segmentation position in the network device is segmentation position 3. This BBL includes AI signal estimation. The first data can be the input data for AI channel estimation. The first signal carries the first data, which can be the output data of the first data after processing by the AI ​​channel estimation and equalization unit included in the BBL. Specifically, the first data is input to the AI ​​channel estimation unit, which processes the first data and outputs the processed first data. The processed first data is then used as the input to the equalization unit, which processes the processed first data and outputs the processed data.

[0196] For another example, the segmentation position in the network device is segmentation position 3. The BBL includes AI equalization. The first data can be the input data of AI equalization. The first signal carries the first data, which can be the output data after the first data has been processed by AI equalization.

[0197] For example, the segmentation position in the network device is segmentation position 3. This BBL includes AI channel estimation and AI equalization. The first data can be the input data for AI channel estimation. The first signal carries the first data, which can be the data output after the first data has been processed by the AI ​​channel estimation and AI equalization units included in the BBL. Specifically, the first data is input to AI channel estimation, which processes the first data and outputs the processed first data. The processed first data is then used as the input to AI equalization, which processes the processed first data and outputs the processed data.

[0198] For example, consider segmentation position 4 in the network device. This BBL includes AI signal estimation. The first data can be the input data for AI channel estimation. The first signal carries the first data, which can be the output data after processing by the AI ​​channel estimation, equalization unit, and demodulation unit included in the BBL. Specifically, the first data is input to the AI ​​channel estimation unit, which processes the first data and outputs the processed first data. This processed first data is then used as input to the equalization unit, which processes the processed first data and outputs the processed data. Finally, the processed first data is used as input to the demodulation unit, which processes the processed data and outputs the processed data.

[0199] It should be understood that when the segmentation position is segmentation position 4, BBL includes AI channel estimation and AI equalization, BBL includes AI equalization and AI demodulation, or BBL includes AI channel estimation and AI demodulation, which are similar to the examples above, and will not be elaborated here.

[0200] The following will provide an exemplary description of AI channel estimation, AI equalization, and AI demodulation. It is understood that other methods may also be used for AI channel estimation, AI equalization, and AI demodulation, and this application does not limit them.

[0201] For example, the input data for AI channel estimation can include signals on multiple resource units (e.g., resource elements, REs), some of which carry known pilot signals. The output data for AI channel estimation can include channel estimates (or estimated channel values) for all REs. The training data for AI channel estimation can include input features and label data. Input features can be the input data for AI channel estimation (e.g., channel observations corresponding to multiple pilot REs) and the location information corresponding to the pilots. Label data can be the actual channel values ​​(or channel responses, ideal channel estimates, or channel values ​​obtained from actual measurements) at all RE locations. AI channel estimation can be further optimized using the error between the predicted channel and the actual channel (e.g., the parameters in AI channel estimation can be iteratively optimized based on the backpropagation algorithm). The error between the predicted channel parameters and the actual channel parameters can be represented by one or more of the following: mean squared error (MSE), root mean square error, normalized mean square error, or similarity-guided channel selection (SGCS). The smaller the error or the higher the similarity, the higher the accuracy and reliability of the AI ​​channel estimation, and the better the performance.

[0202] For example, the input data for AI equalization can include the received signals from Nr antennas, and the output data can include equalized symbols from Nt data streams, where Nr and Nt are both positive integers. The training data for AI equalization can include input features and label data. The input features can be the AI ​​equalization input data (distorted data containing noise), and the label data can be the original transmitted data (noise-free / distortion-free data). The training data can include input data affected by channel distortion and noise, and the corresponding original transmitted data. The training objective can be to minimize the error between the equalized data output by the model and the original transmitted symbols (e.g., MSE, cross-entropy loss) and optimize the parameters (e.g., neural network weights, biases, etc.) in AI equalization through backpropagation (e.g., gradient descent). The evaluation metric for AI equalization can be determined by the error between the equalized data and the original transmitted data. A smaller error indicates better equalization performance, and the recovered symbols are closer to the true values.

[0203] For example, the input data for AI demodulation can include equalized data, and the output data can include the bit depth or log-likelihood ratio (LLR) corresponding to the equalized data. The training data for AI demodulation can include the equalized output data, and the training objective can include bit decision or LLR output. The performance of AI demodulation can be evaluated using bit error rate or LLR accuracy. A lower bit error rate indicates better AI demodulation performance, and a closer LLR distribution to the Shannon limit results in higher decoding gain and better AI demodulation performance.

[0204] It should be understood that the AI ​​channel estimation, AI equalization, and AI demodulation described above can also be obtained using the training methods illustrated in the following example. Taking a neural network as an example, this neural network can be any one of the AI ​​channel estimation, AI equalization, and AI demodulation methods:

[0205] For example, a neural network can be understood as a mapping from an input data set to an output data set. Generally, the initialization of a neural network can be random; the process of obtaining a mapping from existing data using random weights w and biases b can be called neural network training. Specifically, training can be implemented by using a loss function to evaluate the output of the neural network and backpropagating the error, iteratively optimizing w and b using gradient descent until the loss function is less than or equal to a certain threshold. The loss function can be represented by the cross-entropy or MSE between the output data and the real data of the neural network. For example, a neural network includes AI channel estimation; the loss function corresponding to this AI channel estimation can be represented by the cross-entropy or MSE between the output data of the AI ​​channel estimation and the real data (e.g., real channel data). The loss functions for AI equalization and AI demodulation are similar to those for AI channel estimation, and will not be listed here. The gradient descent process can be represented as... θ represents the parameters to be optimized (e.g., w and b), L is the loss function, and η is the learning rate used to control the step size of gradient descent. The backpropagation process can utilize the chain rule for partial derivatives; for example, the gradient of the parameters in the previous layer can be recursively calculated from the gradient of the parameters in the next layer. The specific expression can be given as: Where w ij Let s be the weight of the connection between node j and node i. i The weighted sum of the inputs at node i.

[0206] It should be understood that the AI ​​functions included in BBH in this application can be deployed within BBH, on other devices, or in the cloud, etc., one or more of these methods. Similarly, the AI ​​functions included in BBL can be deployed within BBL, on other devices, or in the cloud, etc., one or more of these methods.

[0207] It should also be understood that, prior to step 401, the method may further include step 402:

[0208] 402, BBH sends a second message to BBL, and BBL receives the second message from BBH accordingly.

[0209] The second piece of information includes the collection information corresponding to the first data. This collection information is used to indicate the collection time.

[0210] For example, the collected information may include one or more of the following: collection start time, collection period information, offset value of collection start time, collection interval or collection end time, etc.

[0211] Among them, the acquisition start time can be used to indicate the start time or initial time of acquiring the first data; the acquisition period information can be used to indicate the time span between the start time and the end time of acquiring the first data, or it can be understood as the time range of acquiring the first data; the acquisition start time offset value is used to indicate the offset value based on a certain reference time, which can be compared with the time of acquiring the first data obtained from the reference time; the acquisition interval can be used to indicate the time interval between two acquisitions of the first data; and the acquisition end time can be used to indicate the end time of acquiring the first data.

[0212] The following section will introduce the relevant schemes for collecting the first data.

[0213] In one possible implementation, the BBL acquires the first data based on the acquisition information corresponding to the first data. After acquiring the first data, the BBL processes the first data according to the functional units included in the BBL and sends a first signal carrying the first data to the BBH. A detailed description of the BBL sending the first signal to the BBH can be found in step 401, and will not be repeated here.

[0214] For example, the segmentation location in the network device is segmentation location 2. The BBL includes AI channel estimation, which obtains the input data for AI channel estimation based on the acquisition information corresponding to the first data. The first data includes the input data for AI channel estimation.

[0215] For another example, the segmentation position in the network device is segmentation position 3. The BBL includes AI channel estimation, or AI channel estimation and AI equalization. This BBL obtains the input data for AI channel estimation based on the collected information corresponding to the first data. The first data includes the input data for AI channel estimation. Assuming the segmentation position in the network device is segmentation position 3, the BBL includes AI equalization, and this BBL obtains the input data for AI equalization based on the collected information corresponding to the first data. The first data includes the input data for AI equalization.

[0216] For another example, if the segmentation position in the network device is segmentation position 4, the BBL includes AI channel estimation, or AI channel estimation and AI equalization, or AI channel estimation and AI demodulation, or AI channel estimation, AI equalization, and AI demodulation. This BBL obtains the input data for AI channel estimation based on the acquisition information corresponding to the first data. The first data includes the input data for AI channel estimation. Assuming the segmentation position in the network device is segmentation position 4, the BBL includes AI equalization, or AI equalization and AI demodulation. This BBL obtains the input data for AI equalization based on the acquisition information corresponding to the first data. The first data includes the input data for AI equalization. Assuming the segmentation position in the network device is segmentation position 4, the BBL includes AI demodulation. This BBL obtains the input data for AI demodulation based on the acquisition information corresponding to the first data. The first data includes the input data for AI demodulation.

[0217] For example, the second information may also include communication configuration information, which includes antenna configuration and / or pilot configuration. This communication configuration information is used for data communication between the BBL and BBH. The antenna configuration involves the design of the antenna's physical or logical parameters, and can be used to optimize the performance of data transmission and reception between the BBL and BBH, and improve channel capacity, etc. The pilot configuration can be used to ensure channel visibility, and can also provide a basis for synchronization, equalization, and adaptive control.

[0218] It should be understood that the communication configuration information and the collection information corresponding to the first data described above can be transmitted in the same message (e.g., the second information), or they can be transmitted separately in different messages. This application does not limit this.

[0219] 403, BBH performs a second processing on the first signal to obtain the second data.

[0220] For example, after BBH receives a first signal from BBL, it performs a second process on the first signal to obtain second data. This second process includes at least decoding, or it can be understood that the second data is obtained by at least decoding the first signal.

[0221] For example, the second process may also include one or more of equalization, demodulation, descrambling, rate dematching (or rate dematching), or CRC check.

[0222] For example, in the network device architecture, BBL and BBH correspond to splitting position 2. After receiving the first signal from BBL, BBH uses the first signal as input to the equalization unit and sequentially executes the functional units included in BBH. The functional units included in BBH can be found in [link to relevant documentation]. Figure 5Examples, such as BBH, may include an equalization unit, a demodulation unit, a descrambling unit, a rate matching unit, a decoding unit, and a CRC check unit.

[0223] Assuming an AI model is deployed on the equalization unit of a BBH (Balanced Shield Hierarchy), the BBH includes AI equalization. The BBH can use the received first signal as input to the AI ​​equalization unit, sequentially executing the functional units within the BBH. For example, the AI ​​equalization unit processes the first signal, and the processed first signal is used as input to the demodulation unit. The demodulation unit demodulates the processed first signal, using the demodulated result as input to the descrambling unit, and so on. The first signal is processed and transmitted sequentially within the functional units of the BBH. Similarly, assuming an AI model is deployed on the demodulation unit of the BBH, the BBH includes AI demodulation. The BBH can use the received first signal as input to the equalization unit, sequentially executing the functional units within the BBH. For example, the equalization unit processes the first signal, and the processed first signal is used as input to the AI ​​demodulation unit. The AI ​​demodulation unit demodulates the processed first signal, using the demodulated result as input to the descrambling unit, and so on. The first signal is processed and transmitted sequentially within the functional units of the BBH. Furthermore, assuming that AI models are deployed on both the equalization and demodulation units in the BBH, after the BBH receives the first signal, it will sequentially process the first signal through AI equalization and AI demodulation, and so on. The first signal will be processed and transmitted sequentially in the functional units included in the BBH.

[0224] For example, suppose the segmentation position of BBL and BBH in the network device architecture is segmentation position 3. After receiving the first signal from BBL, BBH uses the first signal as the input to the demodulation unit and executes the functional units included in BBH sequentially. The functional units included in BBH can be found in [link to relevant documentation]. Figure 5 Examples, such as BBH, may include a demodulation unit, a descrambling unit, a rate matching unit, a decoding unit, and a CRC check unit.

[0225] Assuming an AI model is deployed on the demodulation unit in the BBH, then the BBH includes AI demodulation. The BBH can use the received first signal as the input of the AI ​​demodulation and execute the functional units included in the BBH in sequence. For example, the AI ​​demodulation processes the first signal, and the first signal after AI demodulation processing is used as the input of the demodulation unit. The demodulation unit descrambles the first signal after AI demodulation processing and uses the descrambling result as the input of the derate matching unit. And so on, the first signal is processed and transmitted sequentially in the functional units included in the BBH.

[0226] For example, the corresponding splitting position for BBL and BBH in the network device architecture is splitting position 4. After receiving the first signal from BBL, BBH uses the first signal as input to the descrambling unit and sequentially executes the functional units included in BBH. The functional units included in BBH can be found in [link to relevant documentation]. Figure 5 Examples, such as BBH, may include a descrambling unit, a rate matching unit, a decoding unit, and a CRC check unit.

[0227] For example, BBH can use the received first signal as the input of the descrambling unit and execute the functional units included in BBH in sequence. For example, the descrambling unit processes the first signal, and the first signal after processing by the descrambling unit is used as the input of the derate matching unit. The derate matching unit performs derate matching on the first signal after processing by the descrambling unit and uses the result of derate matching as the input of the decoding unit. The decoding unit decodes the first signal after processing by the derate matching unit and uses the decoded result (e.g., the second data) as the input of the CRC check unit. The CRC check unit performs CRC check on the second data to determine whether the CRC check of the second data is successful.

[0228] Assume the data transmitted via the wireless channel can be represented in binary as 011001101101, with a length of 12 bits. The first 9 bits, "011001101", represent the original data, and the last 3 bits, "101", represent the CRC checksum. If the data output by the decoding unit (including the CRC checksum) is "101001101101", the CRC check unit performs a CRC check on the first 9 bits, "101001101", to obtain the CRC checksum. If the CRC checksum obtained by the CRC check unit based on "101001101" matches the CRC checksum in the data output by the decoding unit, the CRC checksum of the data from the decoding unit is successful. If the CRC checksum obtained by the CRC check unit based on "101001101" differs from the CRC checksum in the data output by the decoding unit, the CRC checksum of the data from the decoding unit fails. For example, if the CRC check unit performs a CRC check on "101001101" and obtains a CRC check code of "101", it means that the data "101001101101" output by the decoding unit has been successfully checked for CRC.

[0229] In one possible implementation, if the BBH fails to perform a CRC check on the second data, the BBH will not be executed further. Figure 4 The subsequent operations.

[0230] In another possible implementation, BBH successfully performs a CRC check on the second data, such as... Figure 4 As shown, the method may also include the following steps:

[0231] 404, BBH performs the first processing on the second data to obtain the first label.

[0232] For example, BBH performs a CRC check on the second data, and if the CRC check is successful, BBH performs a first process on the second data to obtain a first tag. This first process includes at least encoding, or it can be understood that the first tag is obtained by at least encoding the second data.

[0233] For example, the first label is obtained by processing the second data first, and the second data is obtained by processing the first data second. That is, the first label is associated with the first data, or in other words, the first label corresponds to the first data.

[0234] For example, the first process may also include one or more of the following operations: CRC check, rate matching, or scrambling.

[0235] For example, BBL and BBH correspond to one of the following segmentation positions in the network device architecture: segmentation position 2, segmentation position 3, and segmentation position 4. BBH performs a second processing on the received first signal to obtain second data. If the CRC check of the second data is successful, BBH can use the functional unit corresponding to the downlink transmission to perform a first processing on the second data to obtain the first tag.

[0236] For example, a BBH including an encoding unit can use second data as input to the encoding unit, and the output of the encoding unit can be used as the first tag. As another example, when a BBH including an encoding unit also includes CRC checksum, scrambling code, and rate matching, the BBH can use second data as input to the CRC checksum. This second data is then processed sequentially by the CRC checksum, encoding, rate matching, and scrambling units, and the output of the scrambling unit can be used as the first tag. Again, as another example, when a BBH including an encoding unit also includes CRC checksum and rate matching, the BBH can use second data as input to the CRC checksum. This second data is then processed sequentially by the CRC checksum, encoding, and rate matching units, and the output of the rate matching unit can be used as the first tag.

[0237] It should be understood that steps 403 and 404 are internal implementation operations of BBH, and are optional steps in the signaling interaction process between BBL and BBH.

[0238] For example, if the CRC check of the second data is successful, BBH determines the first label based on the second data. BBH and / or BBL can store samples (e.g., first sample, second sample, third sample), which can be used for subsequent operations of AI functions in BBL. For example, operations of AI functions in BBL may include: training and / or fine-tuning.

[0239] Training refers to the process of building a model from scratch, which typically requires a large amount of data and resources to adjust all parameters of the model (e.g., AI functions) so that it learns common patterns and features from the data. Fine-tuning, on the other hand, refers to further adjusting some parameters of a trained model using small-scale data to enable it to perform certain specific tasks.

[0240] The following examples will illustrate how to store samples in BBH and / or BBL, and how to perform corresponding operations on the AI ​​functions in BBL using these stored samples.

[0241] Example 1

[0242] like Figure 4 The method, further comprising:

[0243] 405, BBH sends the first identifier. Correspondingly, BBL receives the first identifier.

[0244] The first identifier is used to indicate the first data corresponding to the first label.

[0245] For example, the first identifier may include time information corresponding to the first data and the second data.

[0246] Here, the first data and the second data refer to data from the same air interface time-frequency resource information. The time information corresponding to the first data and the second data can be the air interface time-frequency resource information corresponding to the first data and the second data. Alternatively, it can be understood that: the time attributes of the first data and the second data are consistent; or, the first data and the second data correspond to the same bit information; or, in a certain description method, the first data and the second data have the same index, tag, or indication, etc. For example, the relationship between the first data and the second data can be represented by time information. For instance, during data transmission, the time information corresponding to the first data is the same as the time information corresponding to the second data.

[0247] For example, the time representation of the first and second data can be: time slot, timestamp, or frame, etc.

[0248] For example, the time information corresponding to the first data and the second data is a frame, and the first identifier can be a frame identifier (frame ID); or the time information corresponding to the first data and the second data is a slot, and the first identifier can be a slot identifier (slot ID); or the time information corresponding to the first data and the second data is a timestamp, and the first identifier can be a timestamp identifier.

[0249] In one possible implementation, after BBH performs the first processing on the second data to obtain the first tag in step 404, BBH can send the first identifier to BBL. Accordingly, after receiving the first identifier, BBL stores a third sample, which includes the first identifier and the first data. A detailed description of BBL storing the third sample can be found in the description of step 406 below.

[0250] In another possible implementation, BBH performs a first processing on the second data in step 404 to obtain a first tag. If T1 is greater than or equal to a first threshold, BBH sends an identifier set to BBL, which includes the first identifier. This identifier set includes identifiers corresponding to data that have successfully undergone CRC verification. T1 represents the number of tags obtained after successful CRC verification, and T1 is an integer greater than or equal to 1.

[0251] It should be understood that T1 can be obtained by counting the number of tags obtained after a successful CRC check without considering time limits, or by counting the number of tags obtained after a successful CRC check within a certain time period. Specifically, T1 can be obtained by whichever method is used; it can be predefined, configured, or indicated.

[0252] For example, the count of tags T1 obtained after a successful CRC check can be independent of time constraints; that is, BBH can use a fully cumulative approach to count the tags after a successful CRC check. If T1 is greater than or equal to a first threshold, BBH sends the identifiers corresponding to the successfully checked CRC data to BBL in the form of an identifier set.

[0253] For example, the count of the number of tags T1 that successfully completed CRC checks can be based on a specific time period. That is, BBH can count the number of tags that successfully completed CRC checks within a certain time period. If the number of tags T1 that successfully completed CRC checks within that time period is greater than or equal to a first threshold, BBH will send the identifiers corresponding to the successfully checked data to BBL in the form of an identifier set.

[0254] It should be understood that the size of the first threshold can be predefined, configured, or indicated.

[0255] 406, BBL stores the third sample.

[0256] For example, after receiving the first identifier, the BBL uses the first identifier to indicate the first data corresponding to the first tag. The BBL determines the third sample based on the first identifier and the first data, and stores the third sample. The third sample includes the first data and the first identifier.

[0257] For example, when BBL receives a set of identifiers, BBL determines the data corresponding to each identifier in the identifier set and stores the identifier and its corresponding data. Similar to the third sample described above, this application uses the first identifier and first data as an example for illustrative description. In the specific implementation, BBL may store multiple samples, each similar to the third sample described above, which will not be elaborated upon here.

[0258] 407, BBH stores the first sample.

[0259] The first sample includes a first label and a first identifier.

[0260] For example, step 407 can be executed after step 404, such as after BBH performs the first processing on the second data to obtain the first tag, BBH can execute step 407 to store the first sample; or, step 407 can be executed after step 405, such as after BBH sends the first identifier to BBL, BBH stores the first sample. The storage of the first sample by BBH and the storage of the third sample by BBL can be performed simultaneously or sequentially, and this application does not limit this.

[0261] For example, when BBH sends a set of identifiers to BBL, BBH can store each identifier in the identifier set along with its corresponding identifier. Similar to the first sample described above, this application uses the first identifier and the first tag as an example for illustrative description. In the actual implementation, BBH may store multiple samples, each similar to the first sample described above, which will not be elaborated upon here.

[0262] For example, if BBL stores a third sample and BBH stores a first sample, BBL can use the third sample and BBH can use the first sample to perform the following operations on the AI ​​functions included in BBL: training and / or fine-tuning.

[0263] As an example, the AI ​​receiver includes channel estimation, equalization, and demodulation functions. At least one of these functions—channel estimation, equalization, and demodulation—is AI-enabled; that is, the channel estimation included in the AI ​​receiver can be AI channel estimation, the equalization can be AI equalization, and the demodulation can be AI demodulation. Where the BBL includes at least one of AI channel estimation, AI equalization, or AI demodulation, the input to the AI ​​receiver is located on the BBL side. The BBL uses a stored third sample, and the BBH uses a stored first sample to train and / or fine-tune the AI ​​functions included in the BBL. Alternatively, it can be understood that the BBL using the third sample and the BBH using the first sample can be used to train and / or fine-tune the AI ​​receiver.

[0264] For example, the corresponding segmentation position of BBL and BBH in the network device architecture is segmentation position 2. BBL includes AI channel estimation, meaning the AI ​​receiver includes the AI ​​channel estimation in BBL. The input data for AI channel estimation can be understood as the input data for the AI ​​receiver, such as the first data being the input data for the AI ​​receiver. The first data in the first sample stored in BBL is used as the input to the AI ​​receiver, and the first label in the third sample stored in BBH is used as the output label for the AI ​​receiver. This is used to train the AI ​​receiver, or to fine-tune an already trained AI receiver, or to first train the AI ​​receiver and then fine-tune it to improve the performance of the AI ​​receiver.

[0265] It should be understood that the AI ​​receiver may also include AI functions in the BBH. For example, for segmentation position 2, the BBL includes AI channel estimation functionality, and the BBH may include AI equalization and / or AI demodulation functionality.

[0266] Example 2

[0267] like Figure 4 The method, further comprising:

[0268] 408, BBH sends the first tag. Correspondingly, BBL receives the first tag.

[0269] In one possible implementation, after the BBH performs the first processing on the second data to obtain the first tag in step 404, the BBH can send the first tag to the BBL. Accordingly, after receiving the first tag, the BBL can store a second sample, which includes the first tag and the first data. A detailed description of how the BBL stores the second sample can be found in step 409 below.

[0270] In another possible implementation, BBH performs a first processing on the second data in step 404 to obtain a first tag. If T2 is greater than or equal to a second threshold, BBH sends a tag set to BBL, which includes the first tag. This tag set includes tags corresponding to data that have successfully passed CRC verification. T2 represents the number of tags obtained after successful CRC verification, and T2 is an integer greater than or equal to 1.

[0271] It should be understood that T2 can be obtained by counting the number of tags obtained after a successful CRC check without considering time limits, or by counting the number of tags obtained after a successful CRC check within a certain time period. Specifically, T2 can be obtained based on which method it is defined, configured, or indicated.

[0272] For example, the count of tags T2 obtained after a successful CRC check can be independent of time constraints; that is, BBH can use a fully cumulative approach to count the tags after a successful CRC check. If T2 is greater than or equal to the second threshold, BBH sends the tags corresponding to the successfully checked CRC data to BBL in the form of a tag set.

[0273] For example, the count of the number of tags T2 that successfully completed CRC checks can be based on a specific time period. That is, BBH can count the number of tags that successfully completed CRC checks within a certain time period. If the number of tags T2 that successfully completed CRC checks within that time period is greater than or equal to a first threshold, BBH will send the tags corresponding to the successfully checked data to BBL in the form of a tag set.

[0274] It should be understood that the size of the second threshold can be predefined, configured, or indicated.

[0275] It should also be understood that the first threshold in Example 1 and the second threshold in Example 2 may be the same or different.

[0276] 409, BBL stores the second sample.

[0277] The second sample includes the first label and the first data.

[0278] For example, after receiving the first tag, BBL associates the first tag with the first data. BBL then determines the second sample based on the first tag and the first data, and stores the second sample. The second sample includes the first data and the first tag.

[0279] For example, when BBL receives a set of identifiers, BBL determines the data corresponding to each identifier in the identifier set and stores the tag and the data corresponding to that tag. Similar to the second sample described above, this application uses the first tag and the first data as an example for illustrative description. In the specific implementation, BBL may store multiple samples, each similar to the second sample described above, which will not be elaborated upon here.

[0280] It should also be understood that, in the case of BBL storing a second sample, BBL can use the second sample to perform the following operations on the AI ​​functions included in BBL: training and / or fine-tuning.

[0281] As an example, the AI ​​receiver includes channel estimation, equalization, and demodulation functions. At least one of these functions—channel estimation, equalization, and demodulation—is AI-enabled; that is, the channel estimation, equalization, and demodulation included in the AI ​​receiver can be AI channel estimation, AI equalization, or AI demodulation. When the BBL includes at least one of AI channel estimation, AI equalization, or AI demodulation, the input to the AI ​​receiver is located on the BBL side. The BBL uses stored second samples to train and / or fine-tune the AI ​​functions included in the BBL; or, as can be understood, the second samples can be used to train and / or fine-tune the AI ​​receiver.

[0282] For example, BBL and BBH correspond to segmentation position 3 in the network device architecture. BBL includes AI channel estimation and AI equalization, meaning the AI ​​receiver includes AI channel estimation and AI equalization from BBL, and the input data of the AI ​​receiver is the input data of AI channel estimation. The first data in the second sample stored in BBL is used as the input of the AI ​​receiver, and the first label in the second sample stored in BBL is used as the label of the AI ​​receiver output. This is used to train the AI ​​receiver, or to fine-tune an already trained AI receiver, or to first train the AI ​​receiver and then fine-tune it to improve the performance of the AI ​​receiver.

[0283] For example, the corresponding segmentation position of BBL and BBH in the network device architecture is segmentation position 3. BBL includes AI equalization, meaning the AI ​​receiver includes the AI ​​equalization in BBL, and the input data of the AI ​​receiver is the input data of the AI ​​equalization. The first data in the second sample stored in BBL is used as the input of the AI ​​receiver, and the first label in the second sample stored in BBL is used as the label of the AI ​​receiver output. This allows for fine-tuning of the AI ​​receiver, either by fine-tuning an already trained AI receiver, or by first training the AI ​​receiver and then fine-tuning the trained AI receiver, thereby improving the performance of the AI ​​receiver.

[0284] It should be understood that the AI ​​receiver may also include AI functions in the BBH. For example, for segmentation position 3, the BBL may include AI channel estimation and AI demodulation, and the BBH may include AI demodulation. Therefore, the AI ​​receiver may include AI channel estimation and AI equalization in the BBL, as well as AI demodulation in the BBH.

[0285] Example 3

[0286] like Figure 4 The method, further comprising:

[0287] 410, BBH sends the first request message. Correspondingly, BBL receives the first request message.

[0288] The first request information is used to request the acquisition of the first data.

[0289] For example, the first request information may include a first identifier, which can be used by BBL to determine the first data corresponding to the first identifier based on the first request information. A detailed description of the first identifier can be found in step 405 above.

[0290] In one possible implementation, if the BBH successfully performs a CRC check on the second data, the BBH can send a first request message to the BBL.

[0291] In another possible implementation, after the BBH performs the first processing on the second data in step 404 to obtain the first tag, the BBH can send a first request message to the BBL.

[0292] In another possible implementation, BBH performs a first processing on the second data in step 404 to obtain a first tag. If T1 is greater than or equal to a first threshold, BBH sends a first request message to BBL. This first request message can be used to request data corresponding to an identifier set, which includes a first identifier, and the data corresponding to the identifier set includes first data. Alternatively, if T2 is greater than or equal to a second threshold, BBH sends a first request message to BBL. This first request message can be used to request data corresponding to a tag set, which includes a first tag, and the data corresponding to the identifier set includes first data. For a detailed explanation of T1, T2, the first threshold, and the second threshold, please refer to the descriptions in Examples 1 and 2 above.

[0293] It should be understood that in Example 3, step 410 may be performed before or after step 404, and this application does not limit this.

[0294] 411, BBL sends the first data to BBH. Correspondingly, BBH receives the first data from BBL.

[0295] For example, after receiving the first request information, BBL sends the first data to BBH based on the first request information; or, after receiving the first request information, BBL determines the first data based on the first identifier in the first request information and sends the first data to BBH.

[0296] For example, when BBL receives a set of identifiers, BBL determines the data corresponding to each identifier in the identifier set and sends the data corresponding to that identifier to BBH. The identifier set includes the first identifier corresponding to the first data.

[0297] 412, BBH stores the second sample.

[0298] For example, after BBH receives the first data from BBL, BBH stores the second sample. The second sample includes the first data and the first label.

[0299] For example, when BBH sends an identifier set to BBL, BBH receives the data corresponding to each identifier in the identifier set. BBH then stores the data corresponding to each identifier in the identifier set, along with the corresponding tag. Similar to the second sample described above, this application uses the first tag and first data as an example for illustrative description. In the actual implementation, BBH may store multiple samples, each similar to the second sample described above, which will not be elaborated upon here.

[0300] For example, when BBH sends a tag set to BBL, BBH receives the data corresponding to each tag in the tag set and stores the data corresponding to each tag in the tag set. Similar to the second sample described above, this application uses the first tag and the first data as an example for illustrative description. In the specific implementation, BBH may store multiple samples, each similar to the second sample described above, which will not be elaborated upon here.

[0301] For example, if BBH stores a second sample, BBH can use the second sample to perform the following operations on the AI ​​functions included in BBL: training and / or fine-tuning.

[0302] As an example, the AI ​​receiver includes channel estimation, equalization, and demodulation functions. Specifically, at least one of the channel estimation, equalization, and demodulation functions in the AI ​​receiver has AI functionality; that is, the channel estimation included in the AI ​​receiver can be AI channel estimation, the equalization included in the AI ​​receiver can be AI equalization, and the demodulation included in the AI ​​receiver can be AI demodulation. Where the BBL includes at least one of AI channel estimation, AI equalization, or AI demodulation, the input of the AI ​​receiver is located on the BBL side. The BBH uses stored second samples to perform training and / or fine-tuning on the AI ​​functions included in the BBL; or, in other words, the second samples can be used to train and / or fine-tune the AI ​​receiver.

[0303] For example, suppose that BBL and BBH correspond to segmentation position 4 in the network device architecture. BBL includes AI channel estimation and AI equalization, meaning the AI ​​receiver includes AI channel estimation and AI equalization in BBL. The input data for AI channel estimation can be understood as the input data for the AI ​​receiver, such as the first data being the input data for the AI ​​receiver. The first data in the second sample stored in BBH is used as the input to the AI ​​receiver, and the first label in the second sample stored in BBH is used as the output label for the AI ​​receiver. This is used to train the AI ​​receiver, or to fine-tune an already trained AI receiver, or to first train the AI ​​receiver and then fine-tune it to improve the performance of the AI ​​receiver.

[0304] For example, suppose that BBL and BBH correspond to segmentation position 2 in the network device architecture. BBL includes AI channel estimation, meaning the AI ​​receiver includes the AI ​​channel estimation in BBL. The input data for AI channel estimation can be understood as the input data for the AI ​​receiver. The first data in the second sample stored in BBH is used as the input to the AI ​​receiver, and the first label in the second sample stored in BBH is used as the output label for the AI ​​receiver. This allows for training the AI ​​receiver, fine-tuning an already trained AI receiver, or training the AI ​​receiver first and then fine-tuning it to improve its performance.

[0305] It should be understood that the AI ​​receiver may also include AI functions in the BBH. For example, for segmentation position 2, the BBL includes AI channel estimation functionality, and the BBH may include AI equalization and / or AI demodulation functionality.

[0306] In combination with the above Figure 4The method shown assumes that all AI models are deployed on the BBH side, meaning the BBH includes AI functionality and a functional unit for acquiring decoding results. The BBH acquires collected data (e.g., first data), which is then processed and transmitted sequentially within the functional units of the BBH. The BBH performs CRC verification on the transmitted decoding results. If the CRC verification is successful, the BBH generates a first tag corresponding to the first data and stores a second sample.

[0307] As an example, assume that BBL and BBH correspond to segmentation position 1 in the network device architecture. BBH includes AI channel estimation, equalization unit, demodulation unit, descrambling unit, rate dematching unit, decoding unit, and CRC check unit. The first data can be the input data of the AI ​​channel estimation unit. The first data sequentially passes through the AI ​​channel estimation, equalization, demodulation, descrambling, rate dematching, and decoding units. The output of the decoding unit can be considered as the second data. BBH inputs the second data into the CRC check unit and performs CRC check on the second data. If the CRC check on the second data is successful, BBH sequentially processes the second data through encoding, rate matching, scrambling, modulation, and equalization to obtain the first tag. BBH can store a second sample, which includes the first tag and the first data.

[0308] As another example, suppose BBL and BBH correspond to segmentation position 2 in the network device architecture. BBH includes an AI equalizer unit, demodulation unit, descrambling unit, rate matching unit, decoding unit, and CRC check unit. The first data can be the input data of the AI ​​equalizer unit. The first data passes through the AI ​​equalizer unit, demodulation unit, descrambling unit, rate matching unit, and decoding unit in sequence. The output of the decoding unit can be considered as the second data. BBH inputs the second data into the CRC check unit and performs CRC check on the second data. If the CRC check on the second data is successful, BBH sequentially encodes, rate-matches, scrambles, and modulates the second data to obtain the first tag. BBH can store a second sample, which includes the first tag and the first data.

[0309] For example, if BBH stores a second sample, BBH can use the second sample to perform the following operations on the AI ​​functions included in BBH: training and / or fine-tuning.

[0310] As an example, the AI ​​receiver includes channel estimation, equalization, and demodulation functions. Specifically, at least one of the channel estimation, equalization, and demodulation functions in the AI ​​receiver has AI functionality; that is, the channel estimation included in the AI ​​receiver can be AI channel estimation, the equalization included in the AI ​​receiver can be AI equalization, and the demodulation included in the AI ​​receiver can be AI demodulation. Where the BBH includes at least one of AI channel estimation, AI equalization, or AI demodulation, the input of the AI ​​receiver is located on the BBH side. The BBH uses stored second samples to perform training and / or fine-tuning on the AI ​​functions included in the BBH; or, it can be understood that the BBH uses the second samples to train and / or fine-tune the AI ​​receiver.

[0311] For example, BBL and BBH correspond to segmentation position 1 in the network device architecture. BBH includes AI channel estimation, meaning the AI ​​receiver includes the AI ​​channel estimation in BBH. The input data for AI channel estimation can be understood as the input data for the AI ​​receiver, such as the first data being the input data for the AI ​​receiver. The first data in the second sample stored in BBH serves as the input to the AI ​​receiver, and the first label in the second sample stored in BBH serves as the output label for the AI ​​receiver. This is used to train the AI ​​receiver, or to fine-tune an already trained AI receiver, or to first train the AI ​​receiver and then fine-tune it to improve its performance.

[0312] For example, BBL and BBH correspond to segmentation position 1 in the network device architecture. BBH includes AI channel estimation and AI equalization, meaning the AI ​​receiver includes AI channel estimation and AI equalization within BBH. The input data for AI channel estimation can be understood as the input data for the AI ​​receiver, such as the first data being the input data for the AI ​​receiver. The first data in the second sample stored in BBH serves as the input to the AI ​​receiver, and the first label in the second sample stored in BBH serves as the output label for the AI ​​receiver. This is used to train the AI ​​receiver, or to fine-tune an already trained AI receiver, or to first train the AI ​​receiver and then fine-tune it to improve its performance.

[0313] It should be understood that the method provided in this application can also be used in O-RAN scenarios, wherein the above-mentioned Figure 4 In this context, BBL can be considered as RU in O-RAN, and BBH can be considered as DU in O-RAN.

[0314] For example, in an O-RAN architecture, the RU includes AI functionality. The RU can send data to the OAM, the DU can send a first label to the OAM, and the OAM stores a second sample. The OAM can use the second sample to train and / or fine-tune the AI ​​functionality in the RU.

[0315] The above text combined Figures 1 to 5 The communication method embodiments of this application are described in detail below, and will be combined with... Figures 6 to 8 This application describes in detail the communication device-side embodiments. It should be understood that the descriptions of the device embodiments correspond to the descriptions of the method embodiments; therefore, any parts not described in detail can be found in the preceding method embodiments.

[0316] Figure 6 This is a schematic block diagram of the communication device 1000 provided in an embodiment of this application. Figure 6 As shown, the communication device 1000 includes a processing module 1010 and a communication module 1020. The communication device 1000 can be a transmitting device (e.g., a second device, i.e., BBL), or a communication device applied to or used in conjunction with a transmitting device to implement a method executed by the transmitting device, such as a chip, chip system, or circuit; or, the communication device 1000 can be a receiving device (e.g., a first device, i.e., BBH), or a communication device applied to or used in conjunction with a receiving device to implement a method executed by the receiving device, such as a chip, chip system, or circuit.

[0317] The communication module can also be called a transceiver module, transceiver, transceiver unit, or transceiver device. The processing module can also be called a processor, processing board, processing unit, or processing device. Optionally, the communication module is used to execute the sending and receiving operations of the sending and receiving devices in the above method. The device in the communication module that implements the receiving function can be considered a receiving unit, and the device in the communication module that implements the sending function can be considered a sending unit; that is, the communication module includes a receiving unit and a sending unit.

[0318] Optionally, the communication device 1000 may further include a storage module 1030 for storing device program code and / or data.

[0319] In one example, when the communication device 1000 is applied to the first device, the processing module 1010 can be used to implement the first device in the above embodiments (e.g., Figure 4 The communication module 1020 can be used to implement the transmit and receive functions of the first device in the above embodiments, and the BBH processing function in the communication module 1020 can be used to implement the transmit and receive functions of the first device in the above embodiments.

[0320] In another example, when the communication device 1000 is applied to the second device, the processing module 1010 can be used to implement the second device in the above embodiments (e.g., Figure 4 The BBL processing function in the above embodiment is used to implement the transmit and receive function of the second device.

[0321] Furthermore, it should be noted that the aforementioned communication module and / or processing module can be implemented through virtual modules. For example, the processing module can be implemented through software functional units or virtual devices, and the communication module can be implemented through software functions or virtual devices. Alternatively, the processing module or communication module can also be implemented through physical devices, such as chips / circuits (e.g., integrated circuits or logic circuits). The communication module can be an input / output circuit and / or a communication interface, performing input operations (corresponding to the aforementioned receiving operation) and output operations (corresponding to the aforementioned sending operation); the processing module is an integrated processor, microprocessor, or circuit (e.g., integrated circuits or logic circuits).

[0322] The module division in this application is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. Furthermore, the functional modules in the various examples of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0323] In one example, the functional unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as: one or more application-specific integrated circuits (ASICs), or one or more central processing units (CPUs), one or more microcontroller units (MCUs), one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0324] In one example, storage module 1030 may include random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory and / or registers, etc.

[0325] Figure 7This is a schematic block diagram of a communication device 2000 provided in an embodiment of this application. Optionally, the communication device 2000 may be a chip or a chip system. Optionally, in this application, the chip system may be composed of chips or may include chips and other discrete devices.

[0326] like Figure 7 As shown, the communication device 2000 can be used to implement the functions of any device (e.g., terminal device, network device) in the communication system described in the foregoing examples. The communication device 2000 may include at least one processor 2010. Optionally, the processor 2010 is coupled to a memory, which may be located within the device, integrated with the processor, or located outside the device. For example, the communication device 2000 may also include at least one memory 2020. The memory 2020 stores the computer programs, computer programs or instructions, and / or data necessary for implementing any of the above examples; the processor 2010 may execute the computer programs stored in the memory 2020 to complete the methods in any of the above examples.

[0327] The communication device 2000 may also include a communication interface 2030, through which the communication device 2000 can interact with other devices. Exemplarily, the communication interface 2030 may be a transceiver, circuit, bus, module, pin, or other type of communication interface. When the communication device 2000 is a chip-based device or circuit, the communication interface 2030 in the device 2000 may also be an input / output circuit, capable of inputting information (or receiving information) and outputting information (or sending information). The processor 2010 may be an integrated processor, microprocessor, integrated circuit, or logic circuit, etc., and the processor can determine the output information based on the input information.

[0328] In one example, when the communication device 2000 is applied to the first device (e.g., BBL), the processor 2010 can be used to implement the processing function of the first device in the above embodiments, and the communication interface 2030 can be used to implement the sending and receiving function of the first device in the above embodiments.

[0329] In another example, when the communication device 2000 is applied to the second device (e.g., BBH), the processor 2010 can be used to implement the processing function of the second device in the above embodiments, and the communication interface 2030 can be used to implement the sending and receiving function of the second device in the above embodiments.

[0330] In another example, when the communication device 2000 is applied to a third device (e.g., BBH or OAM), the processor 2010 can be used to implement the processing function of the third device in the above embodiments, and the communication interface 2030 can be used to implement the sending and receiving function of the third device in the above embodiments.

[0331] The coupling in this application refers to indirect coupling or communication connection between devices, units, or modules, which can be electrical, mechanical, or other forms, used for information exchange between devices, units, or modules. The processor 2010 may operate in conjunction with the memory 2020 and the communication interface 2030. This application does not limit the specific connection medium between the processor 2010, the memory 2020, and the communication interface 2030.

[0332] Optionally, such as Figure 7 As shown, the processor 2010, the memory 2020, and the communication interface 2030 are interconnected via a bus 2040. Optionally, the bus may include buses of the types such as address bus, data bus, and control bus. Furthermore, for ease of illustration, Figure 7 The diagram shows a bus 2040, but does not imply that there is only one bus or only one type of bus.

[0333] It should be understood that the processor mentioned in the embodiments of this application can be one of the following devices or a portion of the circuitry used for processing functions: a central processing unit (CPU), a processor for AI, or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0334] For example, the processor used for AI can be one or more of the following: graphics processing unit (GPU), neural processing unit (NPU), tensor processing unit (TPU), and data processing unit (DPU).

[0335] For example, one possible implementation of a processor for AI could be... Figure 8 The AI ​​processor 2100 shown can be understood to be used in many other possible ways, which are not limited in this application. Figure 8 This is a schematic diagram of the structure of the AI ​​processor provided in the embodiments of this application.

[0336] like Figure 8 As shown, the AI ​​processor 2100 may include one or more of the following: an AI core, a digital vision pre-processing (DVPP) module, a task scheduler (TS), an L3 cache, an AI CPU, a control CPU, an L2 cache, a universal serial bus (USB) interface, a network interface card (NIC), a peripheral component interconnect express (PCIe) interface (PCIe is a high-speed serial computer expansion bus standard), a double data rate (DDR) / high bandwidth memory (HBM) interface, a general purpose input / output (GPIO) / inter-integrated circuit (I2C) bus, etc. It is understood that the specific meanings of these terms are well known to those skilled in the art and will not be elaborated upon here.

[0337] It should also be understood that the memory mentioned in the embodiments of this application can be volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM). For example, RAM can be used as an external cache. By way of example and not limitation, RAM includes the following forms: static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0338] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) can be integrated into the processor.

[0339] It should also be noted that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0340] This application also provides a computer-readable storage medium storing computer instructions for implementing the methods executed by a communication device (e.g., a first device and / or a second device) in the above-described method embodiments.

[0341] This application also provides a computer program product comprising instructions which, when executed by a computer, implement the methods performed by a communication device (e.g., a first device and / or a second device) in the above-described method embodiments.

[0342] This application also provides a communication system, which includes the first device and / or the second device described in the above embodiments.

[0343] Optionally, the communication system may further include the first device and / or the second device described in the above embodiments.

[0344] The explanations and beneficial effects of the relevant contents in any of the devices provided above can be found in the corresponding method embodiments provided above, and will not be repeated here.

[0345] In the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0346] This application will present various aspects, embodiments, or features relating to systems that may include multiple devices, components, modules, etc. It should be understood and appreciated that individual systems may include additional devices, components, modules, etc., and / or may not include all the devices, components, modules, etc. discussed in conjunction with the accompanying drawings. Furthermore, combinations of these approaches are also possible.

[0347] In this application, examples may reference each other without logical contradiction. For example, methods and / or terms between method embodiments may reference each other, functions and / or terms between device embodiments may reference each other, and functions and / or terms between device examples and method examples may reference each other.

[0348] It should be understood that the above embodiments are mainly illustrated using devices in existing network architectures as examples, and the specific form of the devices is not limited in the embodiments of this application. For example, any device that can achieve the same function in the future is applicable to the embodiments of this application.

[0349] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0350] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be described again here.

[0351] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0352] The units described as separate components may or may not be physically separate. The 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0353] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0354] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to existing solutions, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, external hard drives, ROM, RAM, magnetic disks, or optical disks.

[0355] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for data acquisition, characterized in that, The network device includes a first device and a second device, the method is applied to the first device or to a chip in the first device, the method includes: Receive a first signal from a second device, the first signal carrying first data; If the second data is successfully checked by cyclic redundancy check (CRC), the second data is processed to obtain a first tag. The first processing includes at least encoding. The second data is obtained by processing the first signal. The second processing includes at least decoding.

2. The method according to claim 1, characterized in that, The first label is associated with the first data.

3. The method according to claim 1 or 2, characterized in that, The method further includes: Send a first identifier, which is used to indicate the first data corresponding to the first tag.

4. The method according to claim 3, characterized in that, The sending of the first identifier includes: If T1 is greater than or equal to the first threshold, a set of identifiers is sent, the set of identifiers including the first identifier and the identifiers corresponding to data with successful CRC verification. Wherein, T1 represents the number of tags obtained after successful CRC verification, and the tags obtained after successful CRC verification include the first tag.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: The first sample is stored, which includes the first tag and the first identifier. The first identifier is used to indicate the first data corresponding to the first tag. The first sample is used for the operation of the AI ​​function in the second device.

6. The method according to any one of claims 3 to 5, characterized in that, The first identifier includes the time information corresponding to the first data and the second data.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Send the first tag.

8. The method according to claim 7, characterized in that, Sending the first tag includes: If T2 is greater than or equal to the second threshold, a tag set is sent, the tag set including the first tag, and the tag set including tags corresponding to data with successful CRC verification. Wherein, T2 represents the number of tags obtained after successful CRC verification.

9. The method according to claim 7 or 8, characterized in that, The method further includes: Send the correspondence between the first data and the first tag.

10. The method according to any one of claims 1 to 9, characterized in that, The method further includes: Send a first request message, which is used to request the acquisition of the first data; Receive the first data.

11. The method according to claim 10, characterized in that, The first request information includes a first identifier.

12. The method according to claim 11, characterized in that, The first identifier is used to indicate the first data corresponding to the first tag.

13. The method according to any one of claims 10 to 12, characterized in that, The method further includes: The second sample is stored, which includes the first tag and the first data, and is used for the operation of the AI ​​function in the second device.

14. The method according to claim 5 or 13, characterized in that, The operation of the AI ​​function in the second device includes training and / or fine-tuning.

15. The method according to claim 14, characterized in that, The AI ​​functions in the second device include one or more of the following: AI channel estimation, AI equalization, or AI demodulation.

16. The method according to any one of claims 1 to 15, characterized in that, Before receiving the first signal, the method further includes: Send the collection information corresponding to the first data, the collection information being used to indicate the collection time.

17. The method according to claim 16, characterized in that, The collected information includes one or more of the following: Data collection start time, data collection cycle information, offset of data collection start time, data collection interval or data collection end time.

18. A communication device, characterized in that, Includes modules for implementing the method as described in any one of claims 1 to 17.

19. A communication device, characterized in that, Includes a processor for executing a computer program or instructions in memory to cause the method as described in any one of claims 1 to 17 to be performed.

Images