Communication methods, communication devices, communication system, storage medium and program product

Abstract

The present disclosure relates to communication methods, communication devices, a communication system, a storage medium and a program product, and belongs to the technical field of communications. A method comprises: a first device receives second data corresponding to first data transmitted by a second device, wherein the first data and the second data form training data pairs, and the training data pairs are used for training a first model. In the method provided in the present disclosure, while transmitted data of a transmitting end is unknown, the transmitted data of the transmitting end and received data of a receiving end are combined as training data pairs on the basis of the received data, so as to train a model, thereby improving the model training performance and accuracy.

Description

Communication methods, communication equipment, communication systems, storage media and software products Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to communication methods, communication devices, communication systems, storage media, and program products. Background Technology

[0002] Traditional demodulation methods are sensitive to phase and fading caused by channel factors, especially in low SNR scenarios, where high-order modulation performance is poor. Therefore, improving high-order modulation performance in low SNR scenarios is an issue that needs to be addressed.

[0003] Summary of the Invention

[0004] This disclosure provides a communication method, communication device, communication system, storage medium, and program product, which can be used in the field of communication technology to collect labels of training data for model training when the data sent by the unknown sender is unknown.

[0005] According to a first aspect of the present disclosure, a communication method is proposed, executed by a first device, comprising: receiving second data corresponding to first data sent by a second device, wherein the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0006] According to a second aspect of the present disclosure, a communication method is proposed, performed by a second device, comprising: sending first data, wherein the first data corresponds to second data received by the first device, the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0007] According to a third aspect of the present disclosure, a first device is provided, comprising: a transceiver module for receiving second data corresponding to first data sent by a second device, wherein the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0008] According to a fourth aspect of the present disclosure, a second device is provided, comprising: a transceiver module for transmitting first data, wherein the first data corresponds to second data received by the first device, the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0009] According to a fifth aspect of the present disclosure, a communication device is provided, which is capable of implementing the communication method described in any one of the first or second aspects of the present disclosure.

[0010] According to a sixth aspect of the present disclosure, a computer storage medium is provided, wherein the computer storage medium stores computer-executable instructions; after being executed by a processor, the computer-executable instructions are able to implement the communication method described in any one of the first and second aspects of the present disclosure.

[0011] According to a seventh aspect of the present disclosure, a communication system is provided, including a first device and a second device, wherein the first device is configured to perform a communication method described in any one of the first aspects of the present disclosure, and the second device is configured to perform a communication method described in any one of the second aspects of the present disclosure.

[0012] According to an eighth aspect of the present disclosure, a program product is provided, including at least one of a program and instructions, wherein when the program and instructions are executed by a communication device, they implement the communication method described in either the first or second aspect of the present disclosure.

[0013] According to the communication method proposed in this disclosure, under the premise that the receiving end does not know the data sent by the sending end, the actual data label of the sending end is obtained based on the real environment of the channel of the real wireless communication system for model training, thereby improving model performance and training accuracy. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings required for the description of the embodiments are introduced below. The following drawings are only some embodiments of this disclosure and do not impose specific limitations on the protection scope of this disclosure.

[0015] Figure 1A is a schematic diagram of the architecture of a communication system provided according to an embodiment of the present disclosure;

[0016] Figure 1B is a schematic diagram of the constellation for digital modulation;

[0017] Figure 2A is an interactive schematic diagram of a communication method provided according to an embodiment of the present disclosure;

[0018] Figure 2B is an interactive schematic diagram of a communication method provided according to an embodiment of the present disclosure;

[0019] Figure 2C is an interactive schematic diagram of a communication method provided according to an embodiment of the present disclosure;

[0020] Figure 2D is an interactive schematic diagram of a communication method provided according to an embodiment of the present disclosure;

[0021] Figure 3 is a schematic diagram of the communication method interaction according to an embodiment of the present disclosure;

[0022] Figure 4A is a schematic diagram of a preset transmission order provided according to an embodiment of the present disclosure;

[0023] Figure 4B is a schematic diagram of a resource configuration method provided according to an embodiment of the present disclosure;

[0024] Figure 5A is a schematic diagram of the structure of a first device provided according to an embodiment of the present disclosure;

[0025] Figure 5B is a schematic diagram of the structure of a second device provided according to an embodiment of the present disclosure;

[0026] Figure 6A is a schematic diagram of the structure of a communication device according to an embodiment of the present disclosure;

[0027] Figure 6B is a schematic diagram of the chip structure proposed in an embodiment of this disclosure. Detailed Implementation

[0028] This disclosure provides a communication method, communication device, communication system, storage medium, and program product.

[0029] In a first aspect, embodiments of this disclosure provide a communication method executed by a first device, comprising: receiving second data corresponding to first data sent by a second device, wherein the first data and the second data constitute a training data pair, and the training data pair is used to train a first model.

[0030] In the above embodiments, the receiving end, without knowing the first data sent by the sending end, combines the first data and the received second data into a training data pair, which is then used for training the AI ​​model of the communication system to improve model performance and training accuracy.

[0031] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: determining first information and second information based on protocol agreement or network device indication, wherein the first information is used for the second device to send first data and the first device to receive second data, and the second information is used for the second device to generate the first data and the first device to determine the first data.

[0032] In the above embodiments, the first device can determine the method of receiving data and the data label corresponding to the received data based on the protocol agreement or network device instruction, thereby improving the performance and accuracy of model training.

[0033] In conjunction with some embodiments of the first aspect, in some embodiments, receiving the second data includes: receiving the second data based on the first information; the method further includes: determining the first data based on the second information.

[0034] In conjunction with some embodiments of the first aspect, in some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, wherein the first symbol is jointly agreed upon by both the network device and the terminal, or the first symbol is indicated by the network device, and the first symbol is a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, wherein the first bit stream is jointly agreed upon by both the network device and the terminal, or the first bit stream is indicated by the network device, and the first bit stream is a bit stream that has not undergone channel coding; the first data is generated by performing channel coding on a second bit stream, rate matching and modulation on the coded bit stream, layer mapping, precoding, and resource mapping, wherein the second bit stream is jointly agreed upon by both the network device and the terminal, or the second bit stream is indicated by the network device.

[0035] In the above embodiments, the first device can determine the data label corresponding to the received data based on the instruction, thereby improving the performance and accuracy of model training.

[0036] In conjunction with some embodiments of the first aspect, in some embodiments, the first information and the second information include at least one of the following: time-domain resource allocation parameters; frequency-domain resource allocation parameters; frequency-domain resource allocation method; modulation method; modulation order; precoding matrix; antenna port; number of layers; relevant information of modulation symbols; waveform; physical resource block (PRB) binding size; resource mapping method; first bit stream corresponding to the first data; encoding method; code rate; redundancy version; modulation order.

[0037] In the above embodiments, the first device determines the method of receiving data and the data label corresponding to the received data by using the parameters indicated in the first information and the second information, thereby improving the performance and accuracy of model training.

[0038] In conjunction with some embodiments of the first aspect, in some embodiments, determining the first information and the second information based on protocol agreement or network device indication includes at least one of the following: receiving at least one of the first information and the second information indicated by the second device through a first signaling, wherein the first signaling includes either dynamic signaling or semi-static signaling; receiving a first list sent by the second device, wherein the first list includes at least one of the index of the first information, the index of the second information, the value of the first information, and the value of the second information; and receiving dynamic signaling sent by the second device, wherein the dynamic signaling is used to indicate all or part of the indexes of the first information and the second information, the value of the first information, and the value of the second information. The values ​​include at least one of all or part of the values; based on a first list preset by the protocol, the first list includes the index of the first information, the value of the first information, the index of the second information, and at least one of the values ​​of the second information; and receiving dynamic signaling sent by the second device, the dynamic signaling being used to indicate at least one of all or part of the index of the first information and the index of the second information, the value of the first information and the value of the second information, and determining at least one of the first information and the second information based on the protocol agreement; receiving third data sent by the second device, the third data including a first field, the first field being used to indicate at least one of the first information and the second information, or the first field being used to indicate whether the parameters for sending the third data can be reused as the parameters for sending the first data, the third data being communication data not used for model training.

[0039] In the above embodiments, the first device can determine the method of receiving data and the data label corresponding to the received data by at least one of the received signaling, list, and data, thereby improving the performance and accuracy of model training.

[0040] In conjunction with some embodiments of the first aspect, in some embodiments, the first device is a terminal and the second device is a network device. Determining the first information and the second information based on a protocol agreement or network device instruction includes: determining the first information and the second information based on a protocol agreement, or receiving the first information and the second information sent by the network device.

[0041] In the above embodiments, the terminal can determine the method of receiving data and the data label corresponding to the received data based on the protocol agreement or network device instructions, thereby improving the performance and accuracy of model training.

[0042] In conjunction with some embodiments of the first aspect, in some embodiments, the first device is a network device and the second device is a terminal. Determining the first information and the second information based on a protocol agreement or network device instruction includes: determining the first information and the second information based on a protocol agreement, or sending the first information and the second information to the terminal.

[0043] In the above embodiments, the network device can determine the method of receiving data and the data label corresponding to the received data based on the protocol agreement or network device instruction, thereby improving the performance and accuracy of model training.

[0044] In conjunction with some embodiments of the first aspect, in some embodiments, receiving the second data includes: receiving the second data in a first manner or a second manner, wherein the first manner is that the second device multiplexes the sending of the first data and the sending of the third data, and the second manner is that the second device does not multiplex the sending of the first data and the sending of the third data, and the third data is communication data not used for model training.

[0045] In the above embodiments, the first device can receive data in a reused or non-reuse manner, thereby determining the data label corresponding to the received data, which improves the performance and accuracy of model training.

[0046] In conjunction with some embodiments of the first aspect, in some embodiments, receiving the second data in a second manner includes: determining a first resource and a second resource based on first information; receiving the second data on the first resource; and receiving fourth data on the second resource, wherein the fourth data is data corresponding to the third data sent by the second device.

[0047] In the above embodiments, the first device can receive data in a non-reusable manner, thereby determining the data label corresponding to the received data, which improves the performance and accuracy of model training.

[0048] In conjunction with some embodiments of the first aspect, in some embodiments, the first information includes a first domain and a second domain, wherein the first domain is used to configure a first resource and the second domain is used to configure a second resource.

[0049] In conjunction with some embodiments of the first aspect, in some embodiments, determining the first resource and the second resource includes: receiving second signaling, or receiving first control information or a first control unit, wherein the second signaling includes a first parameter, and the first control information or the first control unit includes a first field; determining first information based on the first parameter or the first field, and determining the first resource and the second resource according to the first information; wherein the first parameter or the first field is used to indicate at least one of the following: whether to schedule the first resource or the second resource; the resource location of scheduling the first resource; the resource data of scheduling the first resource; the resource location of scheduling the second resource; and the resource data of scheduling the second resource.

[0050] In the above embodiments, the first device can determine the resource for receiving data by receiving signaling, so as to receive the data sent by the second device on the corresponding resource, thereby determining the data label corresponding to the received data, which improves the performance and accuracy of model training.

[0051] In conjunction with some embodiments of the first aspect, in some embodiments, receiving second data in a first manner includes: determining a first quantity of first data sent by the second device and a first location where the first data is sent; determining a second quantity of third data sent by the second device and a second location where the third data is sent; receiving second data based on the first quantity, the first location, the second quantity, and the second location, wherein the second device sends the first data and the third data in a multiplexed manner.

[0052] In the above embodiments, the first device can receive data in a reused manner, thereby determining the data label corresponding to the received data, which improves the performance and accuracy of model training.

[0053] In conjunction with some embodiments of the first aspect, in some embodiments, determining the first quantity of the first data sent by the second device includes any one of the following: determining the first quantity as 1 / M based on a protocol agreement, where M is the total number of available symbols, and M is determined by a protocol agreement or by an instruction from a network device; determining the first quantity as an integer multiple of all data corresponding to the current modulation order based on a protocol agreement or an instruction from a network device; or determining the first quantity based on at least one of the number of bits corresponding to the first data and the modulation order.

[0054] In the above embodiments, the first device can determine the method of receiving data based on the protocol agreement or network device instruction, and then determine the data label corresponding to the received data, thereby improving the performance and accuracy of model training.

[0055] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: demultiplexing the second data to obtain a bit stream corresponding to the first data and a bit stream corresponding to the third data, wherein the second device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data respectively before sending the bit stream corresponding to the first data and the bit stream corresponding to the third data.

[0056] In the above embodiments, the first device demultiplexes the data received in bit-level multiplexing mode to obtain training data for model training, thereby improving the performance and accuracy of model training.

[0057] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: performing de-mapping on the second data to obtain a de-mapping bitstream; and demultiplexing the de-mapping bitstream to obtain a bitstream corresponding to the first data and a bitstream corresponding to the third data.

[0058] In the above embodiments, the first device processes the data received in a symbol-level multiplexing manner through de-mapping and de-multiplexing to obtain training data for model training, thereby improving the performance and accuracy of model training.

[0059] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: performing a second processing on the second data to obtain fifth data, wherein the fifth data is data that has not undergone demodulation and bitstream processing, or, the fifth data is data obtained by quantizing the sixth data obtained after the second processing according to a protocol agreement, wherein the sixth data is data that has not undergone demodulation; sending the fifth data to a second device, wherein the first data and the fifth data are used to train the first model.

[0060] In the above embodiments, the first device processes the received data using a quantization scheme to obtain the data label corresponding to the received data, thereby improving the performance and accuracy of model training.

[0061] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: receiving a trained first model sent by a second device, wherein the trained first model is trained by the second device, or the trained first model is provided by the second device to a third device as training data pairs and obtained from the third device.

[0062] In the above embodiments, the receiving end determines the labels required for model training under the AI-based signal demodulation method based on the received second data, so as to further improve the transmission performance of the wireless AI communication system and the accuracy and precision of model training.

[0063] Secondly, embodiments of this disclosure provide a communication method executed by a second device, comprising: sending first data, wherein the first data corresponds to second data received by the first device, the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0064] In the above embodiments, the sending device sends data to the receiving device, enabling the receiving device to determine the received data, thereby forming a training data pair with the data sent by the sending device for model training, which is then used for model training, thereby improving model performance and training accuracy.

[0065] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: determining first information and second information based on protocol agreement or network device indication, wherein the first information is used for the second device to send first data and the first device to receive second data, and the second information is used for the second device to generate the first data and the first device to determine the first data.

[0066] In conjunction with some embodiments of the second aspect, in some embodiments, sending the first data includes: sending the first data based on the first information; the method further includes: generating the first data based on the second information.

[0067] In conjunction with some embodiments of the second aspect, in some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, the first symbol being jointly agreed upon by both the network device and the terminal, or the first symbol being indicated by the network device, and the first symbol being a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, the first bit stream being jointly agreed upon by both the network device and the terminal, or the first bit stream being indicated by the network device, and the first bit stream being a bit stream that has not undergone channel coding; the first data is generated by performing channel coding on a second bit stream, rate matching and modulation on the coded bit stream, layer mapping, precoding, and resource mapping, the second bit stream being jointly agreed upon by both the network device and the terminal, or the second bit stream being indicated by the network device.

[0068] In conjunction with some embodiments of the second aspect, in some embodiments, the first information and the second information include at least one of the following: time-domain resource allocation parameters; frequency-domain resource allocation parameters; frequency-domain resource allocation method; modulation method; modulation order; precoding matrix; antenna port; number of layers; relevant information of modulation symbols; waveform; physical resource block (PRB) binding size; resource mapping method; first bit stream corresponding to the first data; encoding method; code rate; redundancy version; modulation order.

[0069] In conjunction with some embodiments of the second aspect, in some embodiments, based on protocol agreement or network device indication, determining the first information and the second information includes at least one of the following: sending a first signaling to the first device, wherein at least one of the first information and the second information is included in the first signaling, and the first signaling is either dynamic signaling or semi-static signaling; sending a first list to the first device, wherein the first list includes at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and sending dynamic signaling to the first device, wherein the dynamic signaling is used to indicate at least one of all or part of the index of the first information and the index of the second information, and all or part of the value of the first information and the value of the second information; determining the protocol. A preset first list includes at least one of the index of first information, the value of first information, the index of second information, and the value of second information. Dynamic signaling is used to indicate to the first device all or part of the index of first information and the index of second information, and at least one of all or part of the value of first information and the value of second information. Based on a protocol agreement, at least one of the first information and the second information is determined. Third data is sent to the first device, the third data including a first field, which indicates at least one of the first information and the second information, or the first field indicates whether the parameters for sending the third data can be reused as the parameters for sending the first data. The third data is communication data not used for model training.

[0070] In conjunction with some embodiments of the second aspect, in some embodiments, the first device is a terminal and the second device is a network device. Determining the first information and the second information includes: determining the first information and the second information based on a protocol agreement, or sending the first information and the second information to the terminal.

[0071] In conjunction with some embodiments of the second aspect, in some embodiments, the first device is a network device and the second device is a terminal. Determining the first information and the second information includes: determining the first information and the second information based on a protocol agreement, or receiving the first information and the second information sent by the network device.

[0072] In conjunction with some embodiments of the second aspect, in some embodiments, sending the first data includes: sending the first data in a first mode or a second mode, wherein the first mode is that the second device multiplexes the sending of the first data and the sending of the third data, and the second mode is that the second device does not multiplex the sending of the first data and the sending of the third data, and the third data is communication data not used for model training.

[0073] In conjunction with some embodiments of the second aspect, in some embodiments, sending the first data in a second manner includes: determining a first resource and a second resource based on first information; sending the first data on the first resource; and sending the third data on the second resource.

[0074] In conjunction with some embodiments of the second aspect, in some embodiments, the first information includes a first domain and a second domain, wherein the first domain is used to configure a first resource and the second domain is used to configure a second resource.

[0075] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: sending a second signaling to a first device, or sending first control information or a first control unit, wherein the second signaling includes a first parameter, the first control information or the first control unit includes a first field, the first parameter or the first field is used to indicate first information, and the first information is used to determine a first resource and a second resource; wherein the first parameter or the first field is used to indicate at least one of the following: whether to schedule the first resource or the second resource; the resource location of the scheduled first resource; the number of resources to schedule the first resource; the resource location of the scheduled second resource; and the number of resources to schedule the second resource.

[0076] In conjunction with some embodiments of the second aspect, in some embodiments, sending the first data in a first manner includes: determining a first quantity of the first data to be sent by the second device and a first location for sending the first data; determining a second quantity of the third data to be sent by the second device and a second location for sending the third data; and sending the first data and the third data in a multiplexed manner based on the first quantity, the first location, the second quantity, and the second location.

[0077] In conjunction with some embodiments of the second aspect, in some embodiments, determining the first quantity of the first data sent by the second device includes at least one of the following: determining the first quantity as 1 / M based on a protocol agreement, where M is the total number of available symbols, and M is determined by the protocol agreement or indicated by the network device; determining the first quantity as Q based on a protocol agreement, where Q is an integer multiple of all data corresponding to the current modulation order; determining the first quantity based on at least one of the number of bits corresponding to the first data and the modulation order.

[0078] In some embodiments, in conjunction with the second aspect, the method further includes: performing a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data respectively, and multiplexing the bit stream corresponding to the first data and the bit stream corresponding to the third data after the first processing.

[0079] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: multiplexing and layer mapping the bit stream corresponding to the first data and the bit stream corresponding to the third data.

[0080] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: receiving fifth data sent by the first device, the fifth data being obtained by the first device performing a second processing on the second data, the fifth data being used to train the first model, the fifth data being data that has not undergone demodulation and bit-level processing, or the fifth data being data obtained by quantizing the sixth data according to a protocol-agreed method, the sixth data being data obtained by the first device performing a second processing on the second data, the sixth data being data that has not undergone demodulation.

[0081] In conjunction with some embodiments of the second aspect, in some embodiments, receiving the fourth data sent by the first device includes: using a third resource to receive the fifth data, wherein the third resource is a periodic resource or a dynamic resource.

[0082] In conjunction with some embodiments of the second aspect, in some embodiments, the method further includes: training a first model using a training data pair consisting of first data and fifth data, or sending a training data pair consisting of first data and fifth data to a third device and receiving a trained first model obtained by the third device based on the training data pair; and sending the trained first model to the first device.

[0083] In the above embodiments, the sending device sends first data to the receiving device so that the receiving device can determine the received second data without knowing the original data sent by the sending device. This allows the first data and the second data to form a training data pair for training the first model, thereby improving model performance and training accuracy.

[0084] Thirdly, embodiments of this disclosure provide a first device, including: a transceiver module, configured to receive second data corresponding to first data sent by a second device, wherein the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0085] Fourthly, embodiments of this disclosure provide a second device, including: a transceiver module for transmitting first data, wherein the first data corresponds to second data received by the first device, the first data and the second data form a training data pair, and the training data pair is used to train a first model.

[0086] Fifthly, embodiments of this disclosure provide a communication device for performing the methods described in any one of the first and second aspects of embodiments of this disclosure.

[0087] Sixthly, embodiments of this disclosure provide a storage medium storing instructions that, when executed on a communication device, cause the communication device to perform the method described in any one of the first or second aspects of embodiments of this disclosure.

[0088] In a seventh aspect, embodiments of the present disclosure provide a communication system, characterized in that it includes a first device and a second device, wherein the first device is used to perform the method described in any one of the first aspects of the present disclosure, and the second device is used to perform the method described in any one of the second aspects of the present disclosure.

[0089] Eighthly, embodiments of this disclosure provide a program product, including at least one of a program and instructions, wherein when the program or instructions are executed by a communication device, they implement the steps of the method described in any one of the first or second aspects of embodiments of this disclosure.

[0090] In a ninth aspect, embodiments of this disclosure provide a computer program that, when run on a computer, causes the computer to perform the methods described in the optional implementations of the first and second aspects.

[0091] In a tenth aspect, embodiments of this disclosure provide a chip or chip system. The chip or chip system includes processing circuitry configured to perform the methods described according to optional implementations of the first and second aspects above.

[0092] It is understood that the aforementioned first device, second device, communication system, communication equipment, storage medium, program product, computer program, chip, or chip system are all used to perform the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0093] This disclosure provides a communication method, a communication device, a communication system, a storage medium, and a program product. In some embodiments, terms such as communication method and information processing method can be used interchangeably, as can terms such as network device, information processing apparatus, and communication apparatus, and terms such as information processing system and communication system.

[0094] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments. In all embodiments of this disclosure, unless otherwise specified or logically conflicting, the terminology and / or descriptions between the embodiments are consistent and can be mutually referenced. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0095] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0096] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0097] In the embodiments disclosed herein, "multiple" refers to two or more.

[0098] In some embodiments, the terms "at least one of A or B, at least one of A and B", "one or more", "a plurality of", "multiple" and the like can be used interchangeably.

[0099] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of whether there is a branch B); in some embodiments, B (execute B regardless of whether there is a branch A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, both A and B are executed. The same applies when there are more branches such as A, B, C, etc.

[0100] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execute A regardless of whether a branch B exists); in some embodiments, B (execute B regardless of whether a branch A exists); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, and C.

[0101] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0102] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0103] In some embodiments, terms such as "time / frequency" and "time-frequency domain" refer to the time domain and / or frequency domain.

[0104] In some embodiments, terms such as “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “when…”, “if…”, etc. can be used interchangeably. These descriptions all refer to the device making a corresponding action under certain objective circumstances. They do not necessarily limit the time, nor do they require the device to make a judgment action when implementing it, nor do they mean that there must be other limitations.

[0105] In some embodiments, the terms “greater than,” “greater than or equal to,” “not less than,” “more than,” “more than or equal to,” “not less than,” “higher than,” “higher than or equal to,” “not lower than,” and “above” can be used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0106] In some embodiments, devices, etc., may be interpreted as physical or virtual, and their names are not limited to those described in the embodiments. Terms such as “device,” “equipment,” “circuit,” “network element,” “network function,” “network device,” “function,” “node,” “unit,” “section,” “system,” “network,” “chip,” “chip system,” “entity,” and “subject” are interchangeable.

[0107] In some embodiments, "network" can be interpreted as devices included in a network (e.g., access network devices, core network devices, etc.).

[0108] In some embodiments, the terms "access network device (AN device)," "radio access network device (RAN device)," "base station (BS)," "radio base station," "fixed station," "node," "access point," "transmission point (TP)," "reception point (RP)," "transmission / reception point (TRP)," "panel," "antenna panel," "antenna array," "cell," "macro cell," "small cell," "femto cell," "pico cell," "sector," "cell group," "serving cell," "carrier," "component carrier," and "bandwidth part (BWP)" can be used interchangeably.

[0109] In some embodiments, the terms "terminal", "terminal device", "user equipment (UE)", "user terminal", "mobile station (MS)", "mobile terminal (MT)", "subscriber station", "mobile unit", "subscriber unit", "wireless unit", "remote unit", "mobile device", "wireless device", "wireless communication device", "remote device", "mobile subscriber station", "access terminal", "mobile terminal", "wireless terminal", "remote terminal", "handset", "user agent", "mobile client", and "client" can be used interchangeably.

[0110] In some embodiments, access network devices, core network devices, or network devices can be replaced by terminals. For example, embodiments of this disclosure can also be applied to structures where communication between access network devices, core network devices, or network devices and terminals is replaced by communication between multiple terminals (e.g., device-to-device (D2D), vehicle-to-everything (V2X), etc.). In this case, the structure can also be configured such that the terminal has all or part of the functions of the access network device. Furthermore, terms such as "uplink" and "downlink" can be replaced with terms corresponding to communication between terminals (e.g., "sidelink"). For example, uplink channel, downlink channel, etc., can be replaced with sidelink channel, and uplink link, downlink, etc., can be replaced with sidelink link.

[0111] In some embodiments, the terminal may be replaced by an access network device, a core network device, or a network device. In this case, the access network device, core network device, or network device may also be configured to have all or some of the functions of the terminal.

[0112] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0113] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0114] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0115] This disclosure proposes a communication method, communication device, communication system, storage medium, and program product. Based on the data received by the receiving end, the tag of the original data sent by the sending end is obtained, and data pairs are formed based on the data from the sending and receiving ends for model training.

[0116] The method proposed in this disclosure is applicable to various communication systems, including but not limited to 4G, 5G, 5G-advance and subsequent communication technologies (such as 6G).

[0117] The application fields of this disclosure are not limited to the field of AI, but can also include the fields of artificial intelligence, machine learning, etc. AI can handle nonlinear problems well. Therefore, especially in scenarios with low SNR, AI-based demodulation methods can be considered to obtain better demodulation performance.

[0118] Figure 1A is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure. As shown in Figure 1A, the communication system 100 may include a first device 101 and a second device 102.

[0119] In some embodiments, the first device may be a terminal, and the second device may be a network device.

[0120] In some embodiments, the first device may be a network device, and the second device may be a terminal.

[0121] In some embodiments, the network device may include at least one of an access network device and a core network device.

[0122] In some embodiments, the terminal includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication function, smart car, tablet computer, computer with wireless transceiver function, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0123] In some embodiments, the access network device is, for example, a node or device that connects a terminal to a wireless network. The access network device may include at least one of the following in a 5G communication system: evolved Node B (eNB), next-generation eNB (ng-eNB), next-generation Node B (gNB), node B (NB), home node B (HNB), home evolved node B (HeNB), radio backhaul device, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in a 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in a Wi-Fi system, but is not limited thereto.

[0124] In some embodiments, a core network device may be a single device comprising one or more network elements, or it may be multiple devices or a group of devices, each comprising all or part of the aforementioned one or more network elements. Network elements may be virtual or physical. The core network may include, for example, at least one of the following: Evolved Packet Core (EPC), 5G Core Network (5GCN), and Next Generation Core (NGC).

[0125] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within access network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0126] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the access network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0127] In some embodiments, the first device 101 may be a terminal.

[0128] In some embodiments, the first device 101 may be a network device, which includes at least one of a core network device and an access network device.

[0129] In some embodiments, the first device 101 may be a device for receiving second data.

[0130] In some embodiments, the first device 101 may be a device for determining first information.

[0131] In some embodiments, the first device 101 may be a device that receives the first information.

[0132] In some embodiments, the first device 101 may be a device for receiving third data.

[0133] In some embodiments, the first device 101 may be a device that receives second information.

[0134] In some embodiments, the first device 101 may be a device for determining the second information.

[0135] In some embodiments, the first device 101 may be a device that receives the first list.

[0136] In some embodiments, the first device 101 may be a device for determining a first resource and a second resource.

[0137] In some embodiments, the first device 101 may be a device for determining the first data.

[0138] In some embodiments, the first device 101 may be a device that transmits first data.

[0139] In some embodiments, the first device 101 may be a device that transmits fifth data.

[0140] In some embodiments, the first device 101 may be a device for performing a second process.

[0141] In some embodiments, the first device 101 may be a device for model training.

[0142] In some embodiments, the first device 101 may be a device that receives the first model.

[0143] In some embodiments, the name of the first device 101 is not limited, and may be, for example, "device for receiving second data", "device for determining first data", "device for sending first data", "device for model training", etc.

[0144] In some embodiments, the second device 102 may be a terminal.

[0145] In some embodiments, the second device 102 may be a network device, such as at least one of an access network device and a core network device.

[0146] In some embodiments, the second device 102 may be a device that transmits the first data.

[0147] In some embodiments, the second device 102 may be a device for determining the first data.

[0148] In some embodiments, the second device 102 may be a device for receiving second data.

[0149] In some embodiments, the second device 102 may be a device for determining the first information.

[0150] In some embodiments, the second device 102 may be a device that sends the first information.

[0151] In some embodiments, the second device 102 may be the device that sends the first list.

[0152] In some embodiments, the second device 102 may be a device that transmits third data.

[0153] In some embodiments, the second device 102 may be a device for determining the first resource and the second resource.

[0154] In some embodiments, the second device 102 may be a device that sends second information.

[0155] In some embodiments, the second device 102 may be a device for determining the second information.

[0156] In some embodiments, the second device 102 may be a device for receiving fifth data.

[0157] In some embodiments, the second device 102 may be a device for training the first model.

[0158] In some embodiments, the second device 102 may be a device that transmits the trained first model.

[0159] In some embodiments, the second device 102 may be a device that sends training data pairs.

[0160] In some embodiments, the second device 102 may be a device that receives the first model.

[0161] In some embodiments, the name of the second device 102 is not limited, and may be, for example, "device for sending first data", "device for receiving first data", "device for determining first data", "device for model training", "device for determining training data pairs", etc.

[0162] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions proposed in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions proposed in this disclosure are also applicable to similar technical problems.

[0163] The following embodiments of this disclosure can be applied to the communication system 100 shown in FIG1A, or to some of the main bodies, but are not limited thereto. The main bodies shown in FIG1A are illustrative. The communication system may include all or some of the main bodies in FIG1A, or it may include other main bodies outside of FIG1A. The number and form of each main body are arbitrary. Each main body may be physical or virtual. The connection relationship between the main bodies is illustrative. The main bodies may not be connected or may be connected. The connection can be in any way, it can be a direct connection or an indirect connection, it can be a wired connection or a wireless connection.

[0164] The embodiments disclosed herein can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), Super 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0165] When training AI models, supervised machine learning is typically employed. For supervised machine learning, it's essential to consider how to obtain labels for the training dataset. Currently, offline AI training methods are primarily considered, such as acquiring datasets through simulation or obtaining datasets transmitted in real-world air interface environments through testing. Both simulated and real-world datasets usually consider channel information of signals passing through the wireless air interface environment. However, the channel in simulation is often a mathematical model of the real channel, which may deviate from the actual channel environment. When collecting AI datasets, it's crucial to collect data that has experienced real-world channel environments so that the trained model can be more accurately applied to real communication systems. Furthermore, considering the generalization ability of AI models, or to improve the transmission capabilities of wireless AI, online learning methods are used to fine-tune the AI ​​model; alternatively, the original AI model may be obtained through online training. For wireless AI model training and inference, there are two common approaches: Case 1, model training and inference are performed on the same node; Case 2, model training is performed on a remote node, and model inference is performed on the local node. Therefore, for AI-based demodulation methods, when conducting online model training or offline training by acquiring real datasets, it is necessary to consider how to obtain the training dataset and its corresponding labels.

[0166] Digital modulation uses a "constellation diagram" as shown in Figure 1B. The constellation diagram defines the basic parameters of the modulation technique: signal distribution and the mapping relationship between it and the modulated digital bits. Constellation diagrams can be divided into QPSK constellation diagrams, 8-PSK constellation diagrams, 16QAM constellation diagrams, 32QAM constellation diagrams, and 64QAM constellation diagrams. As shown in Figure 1B, the constellation diagram shows the available symbols in different formats. For example, in the 16QAM constellation diagram, each symbol represents the possible combinations of four binary bits, resulting in 16 combinations for four binary bits.

[0167] The following explains the technical terms used in this disclosure:

[0168] 1. DCI: Downlink Control Information, carried by the downlink physical control channel PDCCH, is the downlink control information sent from the eNB to the UE, including uplink and downlink resource allocation, HARQ information, power control, etc.

[0169] 2. PDCCH: Physical Downlink Control Channel.

[0170] 3. PHICH: Physical Hybrid ARQ Indicator Channel.

[0171] 4. PCFICH: Physical Control Format Indicator Channel.

[0172] 5. PBCH: Physical Broadcast Channel.

[0173] 6. URLLC: Ultra-Reliable Low-Latency Communications (Ultra-reliable and ultra-low-latency communications).

[0174] 7. PDSCH: Physical Downlink Shared Channel.

[0175] 8. SPS: Semi-Persistent Scheduling.

[0176] 9. MAC CE: MAC - Control Element, Media Access Control - Control Unit.

[0177] 10. DMRS: Demodulation Reference Signal, used in LTE for correlation demodulation of PUSCH and PUCCH channels.

[0178] 11. QAM: Quadrature Amplitude Modulation (QAM) is a modulation method that modulates amplitude on two orthogonal carriers. These two carriers are typically sine waves with a phase difference of 90 degrees (π / 2), hence the name "quadrature carriers." This modulation method is named accordingly.

[0179] 12. RNTI: Radio Network Temporary Identity.

[0180] The following is an interactive schematic diagram of a communication method provided in this disclosure. Embodiments of this disclosure relate to a communication method that can be executed by a communication system, such as the communication system 100 shown in FIG1A. The communication system includes a first device and a second device. The communication method may include the following specific interaction methods:

[0181] Figure 2A is an interactive schematic diagram of one of the communication methods provided in this disclosure. As shown in Figure 2A, the first device is a terminal and the second device is a network device; or, the first device is a network device and the second device is a terminal. The method includes the following steps:

[0182] Step 2101: Determine the first information and the second information.

[0183] In some embodiments, the first information is used for the second device to send first data and the first device to receive second data, and the second information is used for the second device to generate the first data and the first device to determine the first data.

[0184] In some embodiments, the first information and the second information can be carried in the same information, and can simultaneously enable the second device to send the first data and the first device to receive the second data, the second device to generate the first data, and the first device to determine the first data. For example, the first device is a terminal, the second device is a network device, the first information is used by the network device to send the first data to the terminal and by the terminal to receive the second data, and the second information is used by the network device to generate the first data and by the terminal to determine the first data based on the received second data.

[0185] For example, the first device is a network device, the second device is a terminal, the first information is used for the terminal to send first data to the network device and the resources used by the network device to receive second data, and the second information is used for the terminal to generate first data and the network device to determine first data based on the received second data.

[0186] In some embodiments, the first information may be second information, and the first information is used by the second device to generate the first data and by the first device to determine the first data.

[0187] In some embodiments, the second information may be the first information, and the second information is used for the second device to send the first data and the first device to receive the second data.

[0188] In some embodiments, the first information includes a first domain and a second domain, wherein the first domain is used to configure a first resource and the second domain is used to configure a second resource.

[0189] In some embodiments, the second device determines the first information and the second information based on a protocol agreement, and the first device determines the first information and the second information based on a protocol agreement; or, the network device indicates the first information and the second information to the terminal so that the terminal can determine the first information and the second information based on the network device's indication.

[0190] In some embodiments, the second device and the first device may agree on first information and second information based on a protocol.

[0191] In some embodiments, the first information is used to indicate parameters related to the transmission of first data or a first bit stream by the second device. The first data corresponds to a first bit stream, a second bit stream, or a first symbol. The first information and the second information include at least one of the following: time-domain resource allocation parameters; frequency-domain resource allocation parameters; frequency-domain resource allocation method; modulation method; modulation order; precoding matrix; antenna port; number of layers; relevant information of modulation symbols; waveform; physical resource block (PRB) binding size; resource mapping method; the first bit stream corresponding to the first data; encoding method; code rate; redundancy version; modulation order.

[0192] In some embodiments, the relevant information of the modulation symbols may be the transmission order of the modulation symbols, and the first bit stream may be the value of the first bit stream or the transmission order.

[0193] For example, for method one of determining the first data, the parameters include at least one of the following: time-domain resource allocation (location and number), frequency-domain resource allocation parameters (based on these parameters and parameters such as resource allocation method to determine the location and number of frequency-domain resources), modulation method, modulation order, precoding matrix (uplink), antenna port, number of layers, modulation symbol transmission order, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency-domain resource allocation method (continuous or discrete), resource mapping method, etc.

[0194] For example, in method two for determining the first data, the parameters include at least one of the following: time / frequency domain resource allocation parameters, modulation scheme, modulation order, precoding matrix (uplink), antenna port, number of layers, "0" and "1" information bit streams, coding scheme, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency domain resource allocation scheme (continuous or discrete), resource mapping scheme (downlink, whether interleaving is allowed under continuous frequency domain resource allocation), code rate, etc.

[0195] For example, for method three of determining the first data, the parameters include at least one of the following: time / frequency domain resource allocation parameters, modulation scheme, modulation order, precoding matrix (uplink), antenna port, number of layers, "0" and "1" bit streams, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency domain resource allocation method, resource mapping method (downlink, whether interleaving is enabled), redundancy version, etc.

[0196] In some embodiments, the names of information, etc., are not limited to the names described in the embodiments. Terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", and "chip" can be used interchangeably.

[0197] In some embodiments, the terms "uplink", "uplink", and "physical uplink" can be used interchangeably, as can the terms "downlink", "downlink", and "physical downlink", as well as the terms "sidelink", "sidelink", "sidelink communication", "sidelink communication", "direct connection", "direct link", "direct communication", and "direct link communication".

[0198] Option 1: The first device is a terminal, and the second device is a network device.

[0199] In some embodiments, the second device determines the first information and the second information based on a protocol agreement.

[0200] In some embodiments, the second device may send and receive data in a manner agreed upon by a protocol.

[0201] In some embodiments, the second device and the first device may agree on at least one of the first information and the second information based on a protocol.

[0202] For example, the base station and the terminal can determine the values ​​of each parameter and the method of data transmission based on the agreed-upon protocol.

[0203] In some embodiments, the second device may determine the value of the parameter included in any one of the methods from method one to method three in step 2102 based on the protocol agreement.

[0204] In some embodiments, the first device determines the first information and the second information based on a protocol agreement or network device instruction.

[0205] In some embodiments, the first device determines at least one of the first information and the second information based on a protocol agreement. In some embodiments, the second device and the first device determine the values ​​of the parameters involved in different methods in step 2102 and the method of sending data through a protocol agreement.

[0206] In some embodiments, the first device determines the first information and the second information based on network device instructions.

[0207] In some embodiments, the first device determines the first information and the second information based on network device indication, including at least one of the following: receiving at least one of the first information and the second information indicated by the second device via first signaling, wherein the first signaling includes either dynamic signaling or semi-static signaling; receiving a first list sent by the second device, wherein the first list includes at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and receiving dynamic signaling sent by the second device, wherein the dynamic signaling is used to indicate at least one of all or part of the index of the first information and the index of the second information, the value of the first information, and the value of the second information; determining the protocol pre- The system includes a first list comprising at least one of an index of first information, a value of first information, an index of second information, and a value of second information; and receives dynamic signaling sent by a second device, the dynamic signaling indicating at least one of all or part of the index of the first information and the index of the second information, the value of the first information, and the value of the second information; and receives third data sent by the second device, the third data comprising a first field indicating at least one of the first information and the second information, or the first field indicating whether the parameters for sending the third data can be reused as the parameters for sending the first data, the third data being communication data not used for model training.

[0208] In some embodiments, the first device determines at least one of first information and second information based on a first signaling received from a second device. The first signaling includes either dynamic signaling or semi-static signaling.

[0209] In some embodiments, the first device determines at least one of the first information and the second information based on a first list received from the second device, and at least one of all or part of the indices of the first information and the second information indicated by the second device through dynamic signaling, and at least one of all or part of the values ​​of the first information and the second information.

[0210] In some embodiments, the first device determines at least one of the first information and the second information based on the third data received from the second device. The third data includes a first field, which is used to indicate at least one of the first information and the second information, or the first field is used to indicate whether the parameters for sending the third data can be reused as the parameters for sending the first data. The third data is communication data that is not used for model training.

[0211] In some embodiments, the third data may be ordinary business data or high-level signaling data.

[0212] In some embodiments, the second device sends at least one of a first signaling, a first list, and third data to enable the first device to determine first information and second information based on network device instructions.

[0213] In some embodiments, the second device sends a first signaling to the first device, wherein at least one of the first information and the second information is included in the first signaling, and the first signaling is dynamic signaling or semi-static signaling.

[0214] For example, the base station configures the terminal device via semi-static signaling. One possible approach is that the semi-static signaling can be RRC signaling, such as RRCRecofiguration, RRCsetup, RRCResume, RRCestablishment, RRCelease, etc.

[0215] For example, the base station instructs the terminal device via dynamic signaling, which could be either DCI or MAC CE.

[0216] In some embodiments, the second device sends a first list to the first device, the first list including the index and / or value of the first information and / or the index and / or value of the second information; and sends dynamic signaling to the first device, the dynamic signaling being used to indicate all or part of the indexes of the first information and the indexes of the second information and / or all or part of the values ​​of the first information and the values ​​of the second information. For example, the first list includes multiple rows of data, where each row corresponds to an index or value, and the second device indicates one or more rows through dynamic signaling to determine the corresponding index or value.

[0217] For example, the base station first configures a list of values ​​for relevant parameters, and then dynamic signaling indicates a row in the table to indicate the specific value of the relevant parameter to the terminal device.

[0218] In some embodiments, the second device sends third data to the first device. The third data includes a first field, which is used to indicate at least one of the first information and the second information, or the first field is used to indicate whether the parameters for sending the third data can be reused as the parameters for sending the first data. The third data is communication data that is not used for model training.

[0219] In some embodiments, the third data may be ordinary business data or high-level signaling data.

[0220] For example, the base station reuses some parameters of ordinary data transmission, such as PDSCH (Time Domain Allocation List) and datascramblingIdentityPDSCH. For instance, known modulation symbols are placed in preset resource locations and multiplexed with ordinary data for transmission. Method 1: Reusing the DMRS location for multiplexing with ordinary data; Method 2: Placing known modulation symbols in preset locations and multiplexing with ordinary data for transmission. The preset locations are jointly agreed upon by the terminal and gNB through a certain negotiation method, including: the protocol presets the location of known symbols; the gNB configures / indicates the location of known symbols. The above two methods are combined to determine the location of known modulation symbols. Based on these two methods, the transmission of the uplink or downlink physical channel does not need to carry the DMRS (Demodulation Reference Signal) again; the receiving end can perform channel estimation and signal detection of the uplink or downlink physical channel based on the aforementioned known modulation symbols.

[0221] In some embodiments, the terms “downlink control information (DCI),” “downlink (DL) assignment,” “DL DCI,” “uplink (UL) grant,” and “UL DCI” can be used interchangeably.

[0222] In some embodiments, terms such as "physical downlink shared channel (PDSCH)" and "DL data" can be used interchangeably, as can terms such as "physical uplink shared channel (PUSCH)" and "UL data".

[0223] In some embodiments, the second device may send first information and second information to the first device by sending different combinations of first signaling, first list and third data, so that the first device can determine the first information and the second information.

[0224] Option 2: The first device is a network device, and the second device is a terminal.

[0225] In some embodiments, the second device and the first device may agree on at least one of the first information and the second information based on a protocol.

[0226] In some embodiments, the second device determines the first information and the second information based on network device instructions.

[0227] In some embodiments, the first device sends first information and second information to the second device so that the second device can determine the first information and the second information.

[0228] In some embodiments, the second device determines that the first information and the second information include at least one of the following: sending a first signaling to the first device, wherein at least one of the first information and the second information is included in the first signaling, and the first signaling is either dynamic signaling or semi-static signaling; determining a first list preset by the protocol, wherein the first list includes at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and sending dynamic signaling to the first device, wherein the dynamic signaling is used to indicate at least one of all or part of the index of the first information and the index of the second information, and at least one of all or part of the value of the first information and the value of the second information; and sending third data to the first device, wherein the third data includes a first field, wherein the first field is used to indicate at least one of the first information and the second information, or whether the parameter of the first field used to send the third data can be reused as the parameter for sending the first data, and the third data is communication data not used for model training.

[0229] For example, when using DCI for parameter indication, existing fields or newly added fields can be used. When using existing fields, such as in method one or method three in step 2102, since LDPC encoding and rate matching are no longer performed, fields such as RV / MCS / NDI / DAI / PDSCH-to-HARQ_feedback timing indicator can be used to indicate the aforementioned parameters. For example, this can be used to indicate the index in the bitstream configuration table in Example 1 below. The MCS field can be used in conjunction with other parameters to indicate modulation order, such as indicating bitstream order / modulation symbol stream order, or indicating modulation order with some bits and bitstream order with others.

[0230] Example 1: Based on traditional uplink or downlink physical channel transmission, dynamic scheduling is implemented. At least some parameters are indicated by the gNB via dynamic signaling. Optionally, other parameters can be configured by the gNB or preset by the protocol. The protocol presets the modulation scheme, for example, the set of preset modulation schemes is {QPSK, 16QAM, 64QAM, 256QAM, 1024QAM}, etc., and the QAM modulation can be rectangular QAM modulation. The base station sends a modulation symbol / information bit stream / bit stream transmission order configuration table to the terminal via semi-static signaling; or the protocol presets a modulation symbol / information bit stream / bit stream transmission order configuration table, and different configuration tables can be used for different modulation schemes. The transmitting and receiving ends determine the transmission order of the modulation symbol / information bit stream / bit stream based on a row in the configuration table, and generate all modulation symbols / information bit streams / bit streams transmitted over the air interface based on the order. Furthermore, during each dynamic scheduling / semi-static configuration, the gNB can indicate the modulation order and table index, thereby indicating the transmission order or bit stream corresponding to the corresponding modulation order to the terminal device.

[0231] For example, in step 2102, in method one, the modulation method is QPSK, and the four modulation symbols are numbered sequentially. For instance, they can be numbered according to the order of quadrants 1 / 2 / 3 / 4 (modulation symbols #0 / 1 / 2 / 3 in sequence). Alternatively, they can be numbered according to the bitstream value corresponding to each modulation symbol; then, the modulation symbols corresponding to "00", "01", "10", and "11" can be numbered as symbols #0 / 1 / 2 / 3 respectively. Furthermore, the protocol presets or the gNB configures a transmission order table for the modulation symbols. For example, the preset transmission order table is shown in Figure 4A. When performing dynamic scheduling / semi-static configuration, the gNB can indicate one row index in the table below, and the terminal determines the transmission order of the modulation symbols based on this indication. Modulation symbols are then generated based on this order and transmitted over the air interface. Furthermore, if it is a dynamic indicator table index, the table index can be indicated based on existing fields in DCI (RRC / DCI additionally displays an indication that AI+demodulation dataset collection is enabled) or newly added fields (based on the newly added fields, it is determined whether AI+demodulation dataset collection is enabled, and the sending order is determined based on the indication information of the newly added fields).

[0232] For example, the protocol presets or the gNB configures a transmission order table for the modulation symbols. For instance, the preset transmission order table is shown in Figure 4A. When performing dynamic scheduling / semi-static configuration, the gNB can indicate an index in one row of the table below, and the terminal determines the transmission order of the modulation symbols based on this indication. Modulation symbols are then generated based on this order and transmitted over the air interface. Furthermore, if the table index is dynamically indicated, it can be based on existing fields in the DCI (RRC / DCI additionally displays an indication that AI+demodulation dataset collection is enabled), or newly added fields (based on the newly added fields, it determines whether AI+demodulation dataset collection is enabled, and determines the transmission order based on the indication information of the newly added fields). Determining the transmission order of modulation symbols based on the index indication includes: the indicated index is the transmission order of the modulation symbols. For example, if the index is 0, then the number of modulation symbols is determined according to the resources allocated by the gNB (time-frequency spatial domain resources), and then modulation symbols are generated according to the order indication. For example, if the number of modulation symbols is 10, the transmitted QSPK symbols will be: 0, 1, 2, 3, 0, 1, 2, 3, 0, 1; or the starting transmission order will be determined based on the index indication. If the number of symbols determined based on resources is greater than the number of symbols under a certain modulation order, then the transmission will be cyclically performed based on the index corresponding to the starting transmission order. For example, assuming the number of symbols is 40 and the starting transmission order indication is 2, then the transmission will be cyclically performed with indexes 2, 3, 0, 1, meaning the order of the modulation symbols to be transmitted will be: 0, 2, 1, 3, 1, 3, 0, 2, 0, 1, 2, 3, 3, 2, 1, 0, 0, 2, 1, 3, 1, 3, 0, 2…

[0233] For example, the protocol's preset transmission order configuration table has only one row (that is, the transmission order of modulation symbols / information bit streams / bit streams is fixed in the protocol). The terminal directly generates and transmits modulation symbols based on this transmission order, without needing to read the table index indication field carried in signaling such as DCI / RRC / MAC. For example, the preset transmission order could be 0 / 2 / 1 / 3. Or, the protocol could preset the transmission order of modulation symbols to be 0 / 2 / 1 / 3 / 1 / 3 / 0 / 2, etc. One possible approach is that if dynamic signaling / RRC signaling is not notified, the parameters used in the last transmission are used, or the protocol's default parameters are used. Some parameters can be configured to the terminal by the base station through semi-static signaling, and / or other parameters can be indicated to the terminal by the base station through MAC CE, and / or the remaining parameters can be indicated to the terminal by the base station through dynamic signaling.

[0234] Example 2: In some embodiments, a resource configuration method similar to SRS / CSI-RS can also be used, that is, the modulation symbols are configured to a certain resource location by the second device. And / or, the resource location of the modulation symbols is preset by the protocol. Alternatively, a combination of the above two methods can be used.

[0235] For example, whether to schedule the transmission of modulation symbols can be achieved using at least one of the following methods: aperiodic, semi-persistent, periodic, etc. One possible resource configuration method is shown in Figure 4B. Optionally, for the method of pre-setting the resource positions of modulation symbols in the protocol, different modulation methods can have different resource mapping modes. If resource configuration is performed through a gNB, at least one of the following needs to be determined through configuration parameters: time-domain symbol position, slot position (which can be determined through parameters such as slot offset and a certain calculation formula), period (the determination of slot and period applies to semi-persistent and periodic transmission methods), RE position, RB position (or, it can be determined through the starting RB position, the number of RBs, frequency domain density, etc.).

[0236] Optionally, there can be multiple different time-frequency domain resource modes within a slot / RB. Specifically, the mode used depends on the gNB configuration / instruction. It is conceivable that in this mode, if the time-frequency domain resources occupied by the uplink or downlink physical channel include those occupied by the modulation symbols, then the impact of this factor needs to be considered when performing TBS (Transport Block Size) calculation, rate-matching, and resource mapping. That is, the time-frequency domain resources occupied by the modulation symbols need to be excluded. The ordering method of modulation symbols such as M#0, M#1, etc., is shown in Example 1.

[0237] Example 3, based on a method similar to traditional uplink or downlink physical channels, performs periodic uplink transmission or downlink reception of modulation symbols based on semi-static resources configured / indicated by the base station. Some parameters for modulation symbol transmission can be configured by the gNB through semi-static signaling, such as MCS table selection and the transmission period of semi-static transmission, similar to PUSCH configured grant type 1 / SPS (PDSCH). Other signaling can be indicated through activation signaling, such as modulation order, modulation symbol transmission order, and time-frequency domain resources. The activation signaling can be DCI / MAC CE, etc. Optionally, when the terminal receives deactivation signaling (which can be DCI / MAC CE, etc.), it stops uplink modulation symbol transmission or downlink modulation symbol reception on the periodic uplink / downlink physical resources. The SPS / CG configuration can be one or more. Optionally, if there are multiple SPS / CGs, a group of SPS / CGs can be activated / deactivated through the same DCI / MAC CE. Another possible approach is that the parameters used for modulation symbol transmission are configured by the gNB via semi-static signaling. And / or, some parameters are preset by the protocol. For example, modulation order, time-frequency domain resources, etc., are configured by the gNB via semi-static signaling. Modulation symbol transmission order, etc., are preset by the protocol. In this approach, the SPS / CG can also be configured as one or more. Optionally, different SPS / CGs have different modulation order / modulation symbol transmission order configurations. Furthermore, the gNB can also stop transmitting uplink modulation symbols or receiving downlink modulation symbols via deactivation signaling.

[0238] Step 2102: The second device determines the first data.

[0239] Option 1: The first device is a terminal, and the second device is a network device.

[0240] In some embodiments, the second device determines the first data based on a protocol agreement. In some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, which is jointly agreed upon by the network device and the terminal, and the first symbol is a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, wherein the first bit stream is jointly agreed upon by the network device and the terminal, and the first bit stream is a bit stream that has not undergone channel coding; the first data is generated by performing channel coding on a second bit stream, rate matching and modulation on the coded bit stream, layer mapping, precoding, and resource mapping, wherein the second bit stream is jointly agreed upon by the network device and the terminal.

[0241] Option 2: The first device is a network device, and the second device is a terminal.

[0242] In some embodiments, the second device determines the first data based on network device instructions.

[0243] In some embodiments, the second device determines the first data based on the second information sent by the first device.

[0244] In some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, the first symbol being indicated by a network device, and the first symbol being a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, wherein the first bit stream is indicated by a network device, and the first bit stream is a bit stream that has not undergone channel coding; the first data is generated by channel coding a second bit stream, rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, wherein the second bit stream is indicated by a network device.

[0245] For example, the terminal generates the first data based on the parameters sent by the network device.

[0246] In some embodiments, the terms "precoding", "precoder", "weight", "precoding weight", "quasi-co-location (QCL)", "transmission configuration indication (TCI) status", "spatial relation", "spatial domain filter", "transmission power", "phase rotation", "antenna port", "antenna port group", "layer", "the number of layers", "rank", "resource", "resource set", "resource group", "beam", "beam width", "beam angular degree", "antenna", "antenna element", and "panel" can be used interchangeably.

[0247] For example, the physical layer process of node 1 may include at least one of the following:

[0248] Method 1: Node 1 first generates the bit stream corresponding to the modulation symbol stream to be transmitted, scrambles the bit stream, and then performs modulation to determine the modulation symbol to be transmitted. This method can achieve interference randomization and reduce interference from neighboring cells.

[0249] Method 2: Node 1 performs channel coding on the generated bitstream, rate-matches the encoded bits, and further modulates the rate-matched encoded bits according to the system modulation scheme. After passing through layer mapping, precoding, resource mapping, OFDM modulation, etc., it becomes the original data to be transmitted. The transmitting and receiving ends can agree on the system modulation scheme, modulation order, coding scheme, and the "0" and "1" information bitstreams, etc.

[0250] Method 3: Node 1 does not undergo channel coding; it directly agrees on the pre-modulation "0" and "1" bit stream and modulates the bit stream according to the system modulation method. The modulation method can be agreed upon by the transmitting and receiving ends, including the modulation order, coding method, and the "0" and "1" bit streams (the data content carried by each bit).

[0251] Step 2103: The second device sends the first data in a second manner, and the first device receives the second data in a second manner.

[0252] In some embodiments, the second method involves the second device not reusing the transmission of the first data and the transmission of the third data, wherein the third data is communication data not used for model training.

[0253] Option 1: The first device is a terminal, and the second device is a network device.

[0254] In some embodiments, the second device transmits first data on the first resource and third data on the second resource.

[0255] In some embodiments, the first device determines the first resource and the second resource based on the first information.

[0256] In some embodiments, the first device receives second data on a first resource and fourth data on a second resource, the fourth data being data received by the first device that corresponds to the third data sent by the second device.

[0257] In some embodiments, the first device receives second data corresponding to the first data sent by the second device, based on first information.

[0258] In some embodiments, the first device receives second data on a first resource and fourth data on a second resource based on first information.

[0259] In some embodiments, the first device determines the first resource and the second resource based on the first information sent by the second device.

[0260] Option 2: The first device is a network device, and the second device is a terminal.

[0261] In some embodiments, the second device determines the first resource and the second resource based on the first information sent by the first device.

[0262] For example, the terminal determines the first resource and the second resource based on the first information sent by the network device.

[0263] In some embodiments, the second device determines the first resource and the second resource based on the first information sent by the first device.

[0264] In some embodiments, "acquire," "get," "obtain," "receive," "transmit," "bidirectional transmission," and "send and / or receive" can be used interchangeably and can be interpreted as receiving from other entities, acquiring from protocols, acquiring from higher layers, obtaining through self-processing, or autonomous implementation. Protocols include, for example, at least one of the 3GPP protocol, Wi-Fi protocol, and audio and / or video protocols.

[0265] In some embodiments, terms such as “send,” “transmit,” “report,” “distribute,” “transfer,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0266] Step 2103a, the network device sends at least one of the second signaling, the first control information, and the first control unit to enable the terminal to determine the first resource and the second resource based on the instruction of the second device.

[0267] In some embodiments, determining the first resource and the second resource includes: receiving a second signaling, or receiving first control information or a first control unit, wherein the second signaling includes a first parameter, and the first control information or the first control unit includes a first field; determining first information based on the first parameter or the first field, and determining the first resource and the second resource according to the first information; wherein the first parameter or the first field is used to indicate at least one of the following: whether to schedule the first resource or the second resource; the resource location of the scheduled first resource; the number of resources of the scheduled first resource; the resource location of the scheduled second resource; and the number of resources of the scheduled second resource.

[0268] For example, if the transmitted symbols used for AI+Demodulation data collection are not multiplexed with ordinary data, then a semi-static scheduling method is used for scheduling ordinary data and the transmission of symbols related to AI+Demodulation data collection: During SPS / configured grant configuration, separate SPS / configured grant resources are configured for the transmission of ordinary data and modulation symbols. Resources corresponding to modulation symbols are distinguished from SPS / configured grant resources for ordinary data through RRC field names, modulation symbol transmission-related configuration parameters, and display indication fields.

[0269] In some embodiments, the network device sends a second signaling to the terminal, or sends first control information or a first control unit. The second signaling includes a first parameter, and the first control information or the first control unit includes a first field. The first parameter or the first field is used to indicate a first resource and a second resource.

[0270] For example, for dynamic scheduling, the scheduling can be distinguished in at least one of the following ways: whether it is for general data / and or CSI / and HA RQ feedback, or for transmission symbols related to AI+Demodulation data acquisition.

[0271] Method 1: Determine by using indicator fields related to AI data acquisition carried in the DCI. For example, distinguish them by the newly added fields in the DCI as described in Example 1 in step 2103, or by display enable fields related to dataset acquisition carried in the DCI.

[0272] Method 2: Determined through explicit configuration of RRC / MAC CE. For example, RRC may be configured to schedule data acquisition for AI+Demodulation within a specific time period, or within certain time units (slots / subframes / frames, etc.). Alternatively, RRC may be explicitly configured to use one or more HARQ processes for data acquisition, while avoiding the use of these HARQ processes when transmitting ordinary data during this period.

[0273] Method 3: If RRC or other signaling configurations include AI+Demodulation data acquisition parameters, or if AI+Demodulation data acquisition is enabled, the terminal can further determine which time units or HARQ processes are used for data acquisition through protocol presets. For example, determining the slot where mod(x, 10slot) = 1 / 3 is used for data acquisition.

[0274] Method 4: The separate DMRS pattern determines whether the scheduling is for general data / and or CSI / and / or HARQ feedback, or for transmission symbols related to AI+demodulation dataset acquisition. The DMRS pattern can be either the PDCCH DMRS pattern or the PDSCH DMRS pattern. The separate DMRS pattern is defined by the protocol and / or indicated by the gNB. For example, if this scheduling corresponds to the acquisition of AI sample data, then PDCCH / PDSCH uses DMRS pattern #1; otherwise, DMRS pattern #2 is used.

[0275] Method 5: By using separate search space and / or separate CORESET, that is, dynamic scheduling DCI for AI sample acquisition is transmitted using dedicated time-frequency domain resources.

[0276] Method 6: Determine resource configuration by separating RNTI, where the RNTI can be an independent RNTI configured by gNB for AI sample collection.

[0277] Step 2103b: The second device sends first data. The first data corresponds to the second data received by the first device, and the first and second data form a training data pair, which is used to train the first model.

[0278] In some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, which is agreed upon by both the network device and the terminal, or the first symbol is indicated by the network device, and the first symbol is a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, which is agreed upon by both the network device and the terminal, or the first bit stream is indicated by the network device, and the first bit stream is a bit stream that has not undergone channel coding; the first data is obtained by channel coding; the first data is generated by channel coding a second bit stream, rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, where the second bit stream is agreed upon by both the network device and the terminal, or the second bit stream is indicated by the network device.

[0279] Option 1: The first device is a terminal, and the second device is a network device.

[0280] In some embodiments, the first device receives second data, wherein the second data is data received by the first device that corresponds to the first data sent by the second device, and the first data and the second data form a training data pair, which is used to train a first model.

[0281] In some embodiments, the second device transmits the first data in a second manner. The second manner involves the second device not multiplexing the transmission of the first data and the transmission of the third data, where the third data is communication data not used for model training.

[0282] For example, the third data can be UCI (HARQ-ACK, CSI, SR, etc.) or ordinary business data.

[0283] In some embodiments, the second device transmits first data on the first resource and third data on the second resource.

[0284] In some embodiments, the first device receives second data on a first resource and fourth data on a second resource, the fourth data being data received by the first device that corresponds to the third data sent by the second device.

[0285] In some embodiments, the second device sends the first data to the first device on the first resource in a manner that does not reuse the third data. After receiving the second data, the first device determines the first resource and the second resource according to the instructions of the second device, thereby receiving the second data on the first resource.

[0286] For example, a network device sends first data on a first resource and third data on a second resource, while a terminal receives second data on the first resource and fourth data on the second resource.

[0287] In some embodiments, each data transmission requires transmitting the symbol resources corresponding to all constellation points under the modulation scheme; or, in one modulation symbol transmission, only the symbol resources corresponding to a portion of the constellation points under the modulation scheme may be transmitted.

[0288] Option 2: The first device is a network device, and the second device is a terminal.

[0289] In some embodiments, the first device receives second data, wherein the second data is data received by the first device that corresponds to the first data sent by the second device, and the first data and the second data form a training data pair, which is used to train a first model.

[0290] In some embodiments, the second device transmits the first data in a second manner. The second manner involves the second device not multiplexing the transmission of the first data and the transmission of the third data, where the third data is communication data not used for model training.

[0291] In some embodiments, the second device sends first data on a first resource and third data on a second resource based on the first information sent by the first device.

[0292] In some embodiments, the first device receives second data on a first resource and fourth data on a second resource, the fourth data being data received by the first device that corresponds to the third data sent by the second device.

[0293] In some embodiments, the second device sends first data to the first device on the first resource in a manner that does not reuse third data, and the first device receives second data on the first resource.

[0294] For example, the terminal sends first data on the first resource and third data on the second resource; the network device receives the first data on the first resource and fourth data on the second resource.

[0295] In some embodiments, each data transmission requires transmitting the symbol resources corresponding to all constellation points under the modulation scheme; or, in one modulation symbol transmission, only the symbol resources corresponding to a portion of the constellation points under the modulation scheme may be transmitted.

[0296] Step 2104: The first device determines the first data.

[0297] Option 1: The first device is a terminal, and the second device is a network device.

[0298] In some embodiments, the first device determines the first data based on the received second data according to a protocol agreement.

[0299] In some embodiments, the first data is the data demodulated by the first device.

[0300] In some embodiments, after receiving the second data based on the first information, the first device demodulates the data to obtain the first data.

[0301] In some embodiments, the first device combines the second data and the first data into a training data pair to train a first model.

[0302] For example, the receiving end transmits corresponding parameters according to the modulation symbol. It receives the AI ​​sample modulation symbol sent by node 1, and after processing by the receiving end (such as FFT, channel estimation, signal detection, etc.), it obtains at least one signal to be demodulated. At the same time, it sends the corresponding relevant parameters according to the modulation symbol agreed upon by both the transmitting and receiving parties, determines the modulation symbol corresponding to the sending end, and determines it as the label corresponding to the signal to be demodulated, thereby obtaining at least one set of data (the symbol to be demodulated, the label). The symbol to be demodulated is the input of the model training, and the label is the label corresponding to the model output. Based on this, multiple sets of data are collected to construct a candidate dataset.

[0303] In some embodiments, the first device determines the first data based on network device instructions, which may be based on second information sent by the second device and the received second data.

[0304] In some embodiments, the first device determines that the first data satisfies at least one of the following: the first data is generated based on a first symbol, the first symbol being jointly agreed upon by the network device and the terminal, or the first symbol being indicated by the network device, and the first symbol being a symbol that has not undergone channel coding; the first data is generated based on a first bit stream, by scrambling and modulating the first bit stream, the first bit stream being jointly agreed upon by the network device and the terminal, or the first bit stream being indicated by the network device, and the first bit stream being a bit stream before modulation without undergoing channel coding; the first data is generated based on a second bit stream, by channel coding the second bit stream, by rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, the second bit stream being jointly agreed upon by the network device and the terminal, or the second bit stream being indicated by the network device, and the second bit stream being a bit stream before performing channel coding.

[0305] Option 2: The first device is a network device, and the second device is a terminal.

[0306] In some embodiments, the network device determines the first data based on the received second data according to a protocol agreement.

[0307] In some embodiments, the first device determines that the first data satisfies at least one of the following: the first data is generated based on a first symbol, the first symbol being jointly agreed upon by the network device and the terminal, and the first symbol being a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, the first bit stream being jointly agreed upon by the network device and the terminal, and the first bit stream being a bit stream that has not undergone channel coding; the first data is generated by channel coding a second bit stream, rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, and the second bit stream being jointly agreed upon by the network device and the terminal.

[0308] For example, the first data sent by the second device is determined by the first device based on the received second data.

[0309] Method 1: Node 1 first generates the bit stream corresponding to the modulation symbol stream to be transmitted, scrambles the bit stream, and then performs modulation to determine the modulation symbol to be transmitted. This method can achieve interference randomization and reduce interference from neighboring cells.

[0310] Method 2: Node 1 performs channel coding on the generated bitstream, rate-matches the encoded bits, and further modulates the rate-matched encoded bits according to the system modulation scheme. After passing through layer mapping, precoding, resource mapping, OFDM modulation, etc., it becomes the original data to be transmitted. The transmitting and receiving ends can agree on the system modulation scheme, modulation order, coding scheme, and the "0" and "1" information bitstreams, etc.

[0311] Method 3: Node 1 does not undergo channel coding; it directly agrees on the pre-modulation "0" and "1" bit stream and modulates the bit stream according to the system modulation method. The modulation method can be agreed upon by the transmitting and receiving ends, including the modulation order, coding method, and the "0" and "1" bit streams (the data content carried by each bit).

[0312] Step 2105: The first device trains the first model.

[0313] In some embodiments, the first device trains a first model using a training data pair, wherein the training data pair includes first data and second data.

[0314] In some embodiments, the first model can be used for demodulation of signals or data, for example, taking input data or signals to be demodulated and outputting demodulated symbols. The first model can also be used for channel state information (CSI) compression, CSI recovery, beam management, etc., without limitation.

[0315] In some embodiments, the first device is a terminal and the second device is a network device. The terminal forms a training data pair based on the received second data and the determined first data to train the first model and obtain the trained first model.

[0316] In some embodiments, the first device is a network device and the second device is a terminal. The network device forms a training data pair based on the received second data and the determined first data to train the first model and obtain the trained first model.

[0317] In the above embodiments, the same node simultaneously performs data collection and data training. For example, in the case of the terminal side using the AI+demodulation model in downlink data transmission, the terminal performs data collection and training of the AI+demodulation model. In the case of the gNB using the AI+demodulation model in uplink data transmission, the network-side device performs training of the AI+demodulation model and receives AI sample data symbols.

[0318] In the above embodiments, the transmitting device sends the modulation symbols to the receiving device in a non-multiplexed manner. The receiving device demodulates the received data to obtain tags based on the protocol agreement or network device instructions, so as to form training data pairs of modulation symbols and tags, and trains the AI ​​model to improve the training accuracy and precision of the model.

[0319] The communication method involved in the embodiments of this disclosure may include at least one of steps 2101 to 2105. For example, step 2101 may be implemented as a standalone embodiment, step 2102 may be implemented as a standalone embodiment, and steps 2101+step 2102, 2101+step 2102+step 2103+step 2104, 2101+step 2102+step 2103+step 2104+step 2105 may be implemented as standalone embodiments, and so on, but are not limited thereto.

[0320] In some embodiments, steps 2101 and 2102 may be performed in an alternate order or simultaneously.

[0321] In some embodiments, the steps and their optional implementations in other embodiments described before or after this embodiment, as well as other related parts in the specification, can be referred to, and will not be repeated here.

[0322] In this implementation or embodiment, unless there is contradiction, each step can be independent, arbitrarily combined or exchanged in order, optional methods or optional examples can be arbitrarily combined, and can be arbitrarily combined with any steps of other implementations or other embodiments.

[0323] Figure 2B is an interactive schematic diagram of a communication method according to an embodiment of the present disclosure. As shown in Figure 2B, the embodiments of the present disclosure relate to a communication method. The method shown in Figure 2B involves a transmitting end multiplexing modulation symbols and data before transmitting them to a receiving end device. The first device is a terminal, and the second device is a network device; or, the first device is a network device, and the second device is a terminal. The above method includes:

[0324] Step 2201: Determine the first information and the second information.

[0325] In some embodiments, the first information is used for the second device to send first data and the first device to receive second data, and the second information is used for the second device to generate the first data and the first device to determine the first data.

[0326] In some embodiments, the optional implementation of determining the first information and the second information can be found in the optional implementation of step 2101 in Figure 2A, which will not be repeated here.

[0327] Step 2202: The second device determines the first data.

[0328] Option 1: The first device is a terminal, and the second device is a network device.

[0329] In some embodiments, the way the network device determines the first data can be referred to the optional implementation of Scheme 1 in step 2102 of FIG2A, which will not be repeated here.

[0330] Option 2: The first device is a network device, and the second device is a terminal.

[0331] In some embodiments, the way the terminal determines the first data can be referred to in the optional implementation of Scheme 2 in step 2102 of Figure 2A, which will not be repeated here.

[0332] Step 2203: The second device sends the first data in a first manner, and the first device receives the second data in a first manner.

[0333] In some embodiments, the second device transmits first data in a first manner, wherein the first manner is to multiplex the transmission of first data and the transmission of third data by the second device.

[0334] In some embodiments, the first device receives the second data in a first manner, wherein the first manner is to multiplex the sending of the first data and the sending of the third data by the second device.

[0335] In some embodiments, "acquire," "get," "obtain," "receive," "transmit," "bidirectional transmission," and "send and / or receive" can be used interchangeably and can be interpreted as receiving from other entities, acquiring from protocols, acquiring from higher layers, obtaining through self-processing, or autonomous implementation. Protocols include, for example, at least one of the 3GPP protocol, Wi-Fi protocol, and audio and / or video protocols.

[0336] In some embodiments, terms such as "send," "transmit," "report," "distribute," "transmit," "bidirectional transmission," and "send and / or receive" can be used interchangeably. In some embodiments, the first device and the second device determine a first quantity of first data to be sent by the second device and a first location from which the first data is sent; determine a second quantity of third data to be sent by the second device and a second location from which the third data is sent; based on the first quantity, the first location, the second quantity, and the second location, the second device sends the first data and the third data in a multiplexed manner, and the first device receives the second data and the third data in a multiplexed manner.

[0337] In some embodiments, determining the first quantity of the first data transmitted by the second device includes any one of the following: determining the first quantity as 1 / M based on a protocol, where M is the total number of available symbols, and M is determined by a protocol or indicated by a network device; determining the first quantity as Q based on a protocol, where Q is an integer multiple of all data corresponding to the current modulation order; or determining the first quantity based on at least one of the number of bits corresponding to the first data and the modulation order.

[0338] In some embodiments, determining the first quantity can be based on a protocol agreement, determining the number of bits Ns corresponding to the first data, and determining the first quantity based on the number of bits.

[0339] In some embodiments, for the above-described scheme one, the network device sends first data in a first manner, and the terminal receives the first data in a first manner.

[0340] In some embodiments, for the second scheme described above, the terminal sends the first data in a first manner, and the network device receives the first data in a first manner.

[0341] In some embodiments, the network device or terminal determines a first quantity of first data to be sent based on a protocol agreement.

[0342] For example, the transmitted symbols used for AI+Demodulation data collection can be multiplexed with ordinary data. Specifically, a similar approach to UCI multiplexing on PUSCH can be used to multiplex AI+Demodulation data and ordinary data. This includes determining the number of modulation symbols used for data acquisition, the transmission positions of coded bits / modulation symbols, the number of transmitted bits in UL-SCH, and the transmission bit positions in UL-SCH. Regarding the determination of the number of modulation symbols, one possible approach is that the protocol presets the number of transmitted modulation symbols to be 1 / M of the total number of available symbols, where the value of M is specified by the protocol or indicated by the gNB. The total number of available symbols is determined based on at least one of the following factors: the number of time-domain resources, the number of frequency-domain resources, and the number of layers.

[0343] For example, another possible way to determine the number of modulation symbols is that the protocol presets / base station configures / indicates that the number of symbols to be transmitted is Q or the number of bits to be transmitted is Ns, or the number of symbols to be transmitted or the number of bits to be transmitted is determined based on the protocol preset method. For example, the number of symbols to be transmitted is S times the total number of modulation symbols corresponding to the current modulation order (for 16QAM, the number of modulation symbols to be transmitted is 16*S), and the value of S is specified by the protocol / configured by the gNB.

[0344] Optionally, the number of symbols to be transmitted can be determined according to dynamic scheduling signaling, for example, DCI indicates that the number of symbols to be transmitted is Q, or DCI indicates the above-mentioned S value.

[0345] Another possible approach is for the terminal to determine the actual number of symbols to be transmitted based on a certain calculation formula. For example, it can be determined using the following formula, where Om is the number of information bits to be transmitted determined by any of the above methods, or the number of modulation symbols to be transmitted multiplied by the modulation order.

[0346] (Correspondingly, ACK / modulation symbol / Normal data can be multiplexed, and the transmission symbol of ACK is determined first). Optionally, CRC attachment can be performed, i.e., L is not 0, or CRC attachment can be performed, i.e., L is 0. This calculation formula corresponds to the second method in step 2202. The terminal processing method can be: first, perform CRC attachment based on O_m, encoding (optionally, LDPC encoding), rate matching, the rate matching is performed according to Qm' determined in the above formula, where the number of coded bits to be transmitted is Qm' * modulation order * layer number. Then, multiplexing of the coded bits of the symbol to be transmitted and normal data is performed.

[0347] The above calculation formula can be used to determine the number of transmission symbols for a layer. Correspondingly, M_UL_SC is the total number of resources available for PUSCH transmission.

[0348] In the above calculation formula, L is the CRC bit length. If the processing is based on method one or method three in step 2202, then L can be considered as 0. Of course, optionally, for method two, the value of L can also be 0.

[0349] In some embodiments, the second device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data respectively; and multiplexes the bit stream corresponding to the first data and the bit stream corresponding to the third data after the first processing.

[0350] In some embodiments, the first processing is used to perform data processing on the received bit stream, including but not limited to processes such as rate matching and / or code block concatenation, which are not limited in this disclosure.

[0351] For example, one possible way to multiplex data and AI samples is to use bit-level multiplexing. That is, after the bitstreams corresponding to ordinary data and the bitstreams corresponding to AI samples have each undergone rate matching and / or code block concatenation processes, the data bitstreams and the bitstreams of the AI ​​samples are then multiplexed together. Optionally, at least one of the following multiplexing methods can be used:

[0352] Method 1: tx = [modulation_bits, data_bits(0: length(data_bits) - length(modulation_bits) - 1], the output bitstream begins with the bitstream corresponding to the modulation symbols (AI samples), and the remaining positions contain the data coded bits, or vice versa. Alternatively, the position of modulation_bits in the bitstream is determined as follows: index of modulation bits = j * (total bit length / modulation bit length), j = 0, 1, ..., modulation bit length - 1. Where, modulation bit length = Q'm * number of layers * modulation order, where the number of layers is a positive integer. Total bit length is the total number of bits that can be transmitted across all layers (e.g., total bit length = total number of available time-frequency resources * number of layers * modulation order). In this method, when performing rate matching for ordinary data bits, it is based on the total number of available resources (Number of coded bits from...). Rate-matching (using RE * number of layers * modulation order) is performed for all PUSCH bits. During this multiplexing process, data bits are rate-matched. Alternatively, they can be rate-matched using puncture. That is, for data bits, the total number of bits that can be transmitted across all layers is determined, and then the data bits at the positions determined for the AI ​​samples are removed, and the corresponding positions are replaced with the coded bits corresponding to the AI ​​samples. For example, taking the bit corresponding to the AI ​​sample at the beginning of the multiplexing output bitstream, the following multiplexing method is used, where tx is the output bitstream after the multiplexing process is completed. The data_bits are the coded bits output after rate matching and code block concatenation.

[0353] Method 2: tx = [modulation_symbols, data_symbols(length(modulation_symbols), end)], Tx = [modulation_bits, data_bits]. In this method, when rate-matching ordinary data bits, the number of output coded bits for rate-matching is determined according to the number of available data coded bits (Number of data coded bits from rate-matching = all available REs of PUSCH * number of layers * modulation order - Q'm * number of layers * modulation order, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula). Based on this method, the transmission order of the modulation symbols can be configured by the gNB or preset by the protocol, or indicated by the scheduling DCI (existing fields can be reused, or new fields can be added). The modulation order of the modulation symbols can be the same as that of ordinary data, or a special modulation order indicated by DCI / MACCE or configured by RRC. Optionally, if no UL-SCH and / or CSI are generated, only AI+Dem odulation data collection will be performed.

[0354] In some embodiments, the second device multiplexes the bit stream corresponding to the first data and the bit stream corresponding to the third data; and performs layer mapping on the multiplexed bit stream.

[0355] In some embodiments, for the first scheme described above, the network device multiplexes the bit stream corresponding to the first data and the bit stream corresponding to the third data, and performs layer mapping on the multiplexed bit stream.

[0356] In some embodiments, for the second scheme described above, the terminal multiplexes the bit stream corresponding to the first data and the bit stream corresponding to the third data, and performs layer mapping on the multiplexed bit stream.

[0357] For example, multiplexing between symbols and data is performed during symbol-level processing.

[0358] For example, for methods one to three in step 2202, the terminal first calculates the number of modulation symbols to be transmitted (total number of modulation symbols to be transmitted = Q'm * number of layers) according to the above calculation formula, multiplexes the number of modulation symbols to be transmitted with the data after modulation, and then performs layer mapping and other processes.

[0359] `tx = [modulation_symbols, data_symbols(0: length(data_symbols) - length(modulation_symbols) - 1]` means that modulation symbols are placed at the beginning of the output symbol stream, and data symbols are placed at the end, or vice versa. Alternatively, the position of `modulation_symbols` in the symbol stream is determined as follows: `index of modulation symbol = j * (total symbol length / modulation symbol length)`, where `j = 0, 1, ..., modulation symbol length - 1`. Here, modulation symbol length = Q'm * number of layers, and the total number of symbols is the total number of symbols that can be transmitted across all layers. In this method, when rate matching is performed on ordinary data bits, rate matching is performed according to the total number of available resources (Number of coded bits from rate-matching g = PUSCH total available RE * number of layers * modulation order). In this multiplexing process, data symbols are rate-matched... The values ​​can be determined using the ching method. Alternatively, values ​​can be determined using the puncture method. That is, for data symbols, the total number of symbols that all layers can transmit is determined, and then the data symbols carried at the positions determined for the modulation symbols in this method are removed, and the corresponding positions are replaced with the corresponding modulation symbols. The corresponding puncture value determination method, taking the modulation symbol located at the beginning of the symbol stream as an example, has the following multiplexing formula, where tx is the output symbol stream after the multiplexing process is completed.

[0360] tx=[modulation_symbols,data_symbols(length(modulation_symbols),end)]

[0361] Tx = [modulation_symobls, data_symbols]. In this mode, when performing rate matching for ordinary data bits, the number of output coded bits for rate matching is determined according to the number of data coded bits available from rate-matching (Number of data coded bits from rate-matching = all available REs from PUSCH * number of layers * modulation order - Q'm * number of layers * modulation order, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0362] It should be noted that in this method, length(modulation_symbols) and length(data_symbols) are the total number of modulation symbols and data symbols of all layers before multiplexing is performed; or, first perform layer mapping for data symbols and modulation symbols respectively in the above manner, and then perform multiplexing for modulation symbols and data symbols of each layer.

[0363] Opt.1:tx = [modulation_symbols, data_symbols(0:length(data_symbols)-length(modulation_symbols)-1], meaning that modulation symbols are placed at the beginning of the output symbol stream, and data symbols are placed at the end, or vice versa. Alternatively, the position of modulation_symbols in the symbol stream is determined as follows: index of modulation symbols = j * (total symbol length / modulation symbol length), j = 0, 1, ..., modulation symbol length - 1, where the modulation symbol length = Q'm, which can be determined by the previously mentioned formula. The total number of symbols is the total number of symbols that can be transmitted in one layer. In this mode, when ordinary data bits are rate-matched, rate matching is performed according to the total number of available resources (Number of coded bits from rate-matching = PUSCH all available REs * modulation order * number of layers). In this multiplexing process, Data symbols are valued using a rate-matching method. Alternatively, they can be valued using a puncture method. That is, for data symbols, layer mapping is performed according to the total number of symbols that a layer can transmit. Then, the data symbol carried by the symbol position determined for the modulation symbol in this method is removed and replaced with the modulation symbol.

[0364] Opt.2: Tx = [modulation_symobls, data_symbols], In this mode, when performing rate matching for ordinary data bits, the number of output coded bits for rate matching is determined according to the number of data coded bits from rate-matching = all available RE * modulation order * number of layers - Q'm * modulation order * number of layers, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0365] It should be noted that in this method, length(modulation_symbols) and length(data_symbols) represent the total number of modulation symbols and data symbols determined for each layer after the layer mapping is completed.

[0366] Alternatively, data symbols and modulation symbols can be encoded independently (method three in step 2202), undergoing processes such as CRC attachment, rate matching, code block concatenation, layer mapping, and precoding. Finally, resource mapping is performed based on Q'm calculated using the aforementioned formula. For example, Q'm resource units can be reserved for modulation symbols. These resources can be indexed in a frequency-domain priority or time-domain priority manner, such as reserving the first Q'm resource units, or reserving the last Q'm resource units, or reserving the j*(total number of symbols / Q'm), where j = 0, 1, ..., Q'm-1, and the total number of symbols is the total number of symbols that a layer can carry. Based on this, the modulation symbols of each layer are mapped to the reserved resource units using either interleaving or non-interleaving methods, with the remaining parts mapped to data symbols. Furthermore, in this method, rate matching of data symbols can be performed in the following way:

[0367] Opt.1: When rate matching is performed on the encoded bits of ordinary data, rate matching is performed according to the total number of data encoded bits from rate-matching = all available REs in PUSCH * modulation order * number of layers. Furthermore, in this method, the mapped data symbols can be mapped using puncturing or rate-matching. Punching involves mapping all data symbols of each layer onto the resource grid according to a certain resource mapping order; then, the data symbols in the resource positions reserved for modulation symbols in the above method are removed and replaced with the corresponding modulation symbols. Alternatively, data symbol resource mapping can also be performed using rate matching. That is, data_symbols(0:length(data_symbols)-length(modulation_symbols)-1) are mapped to the positions of the remaining data symbols according to a certain resource mapping order.

[0368] Opt.2: When performing rate matching for ordinary data bits, the number of output coded bits is determined according to the number of data coded bits from rate-matching = all available RE * modulation order * number of layers - Q'm * modulation order * number of layers, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0369] Another possible implementation is to add a bit stream corresponding to the modulation symbol, or add a modulation symbol, to the end of the data information bits or the end of the data symbol; or, for the padding bits constituting the MAC PDU, replace them with a bit stream corresponding to the modulation symbol or the modulation symbol itself. In this implementation, to identify that the padding bit carried by the MAC SDU has been replaced with a bit stream corresponding to the modulation symbol, a new LCID can be used, or the LCID of the padding bit can be reused. The number of modulation symbols is determined by the above formula, or by the terminal through RRC signaling or dynamic signaling, or the size of the bit stream corresponding to the modulation symbol is determined by the number of remaining bits to be transmitted, which is the total number of bits that the current resources can carry - UL - SCH MAC SDU - MAC CE - other overhead (such as MAC subheader overhead, etc.). Optionally, for dynamic signaling indication mode, one of the values ​​of this field can also indicate that the modulation symbol occupies "0" bits, or 0 symbols, that is, no multiplexing, or it can also indicate a ratio, or the protocol can support multiplexing or non-multiplexing modes at the same time. Further, the gNB's display indication can be used to determine whether only AI+Demodulation data acquisition is performed, or AI+Demodulation data acquisition and UL-SCH and / or CSI (including HARQ feedback, etc.) transmission are performed simultaneously.

[0370] Step 2204: The first device determines the first data.

[0371] Option 1: The first device is a terminal, and the second device is a network device.

[0372] In some embodiments, the first device performs corresponding processing based on the received second data to determine the first data. That is, the terminal performs corresponding processing based on the received second data to determine the first data.

[0373] In some embodiments, the first device demultiplexes the second data to obtain a bit stream corresponding to the first data and a bit stream corresponding to the third data. The second device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data before sending them.

[0374] In some embodiments, the first processing is used to perform data processing on the received bit stream, including but not limited to processes such as rate matching and / or code block concatenation, which are not limited in this disclosure.

[0375] For example, the terminal demultiplexes the second data to obtain the bit stream corresponding to the first data and the bit stream corresponding to the third data. Before sending the bit stream corresponding to the first data and the bit stream corresponding to the third data, the network device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data, respectively.

[0376] For example, the transmitting end uses bit-level multiplexing between the data and the AI ​​samples. That is, after the bit streams corresponding to ordinary data and the bit streams corresponding to AI samples have each undergone processes such as rate matching and / or code block concatenation, the data bit stream and the bit stream between the AI ​​samples are multiplexed together. The receiving end demultiplexes the received data to obtain the restored original data.

[0377] In some embodiments, the first device performs de-mapping on the second data to obtain a de-mapping bitstream; and demultiplexes the de-mapping bitstream to obtain a bitstream corresponding to the first data and a bitstream corresponding to the third data.

[0378] For example, the terminal performs de-mapping on the second data to obtain a de-mapping bitstream; the de-mapping bitstream is then demultiplexed to obtain the bitstream corresponding to the first data and the bitstream corresponding to the third data.

[0379] For example, the transmitting end uses symbol-level processing to multiplex the data and AI samples. That is, it first calculates the number of modulation symbols to be transmitted (total number of modulation symbols to be transmitted = Q'm * number of layers), multiplexes the number of modulation symbols to be transmitted with the data after modulation, and then performs layer mapping and other processes. The receiving end performs de-layer mapping on the received data to obtain the de-layer mapped bit stream, and demultiplexes the de-layer mapped bit stream to obtain the restored original data.

[0380] Option 2: The first device is a network device, and the second device is a terminal.

[0381] In some embodiments, the first device performs corresponding processing based on the received second data to determine the first data. That is, the terminal performs corresponding processing based on the received second data to determine the first data.

[0382] In some embodiments, the first device demultiplexes the second data to obtain a bit stream corresponding to the first data and a bit stream corresponding to the third data. The second device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data before sending them.

[0383] For example, the network device demultiplexes the second data to obtain the bit stream corresponding to the first data and the bit stream corresponding to the third data. Before sending the bit stream corresponding to the first data and the bit stream corresponding to the third data, the terminal performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data, respectively.

[0384] For example, the transmitting end uses bit-level multiplexing between the data and the AI ​​samples. That is, after the bit streams corresponding to ordinary data and the bit streams corresponding to AI samples have each undergone processes such as rate matching and / or code block concatenation, the data bit stream and the bit stream between the AI ​​samples are multiplexed together. The receiving end demultiplexes the received data to obtain the restored original data.

[0385] In some embodiments, the first device performs de-mapping on the second data to obtain a de-mapping bitstream; and demultiplexes the de-mapping bitstream to obtain a bitstream corresponding to the first data and a bitstream corresponding to the third data.

[0386] For example, the network device performs de-mapping on the second data to obtain a de-mapping bitstream; and demultiplexes the de-mapping bitstream to obtain the bitstream corresponding to the first data and the bitstream corresponding to the third data.

[0387] For example, the transmitting end uses symbol-level processing to multiplex the data and AI samples. That is, it first calculates the number of modulation symbols to be transmitted (total number of modulation symbols to be transmitted = Q'm * number of layers), multiplexes the number of modulation symbols to be transmitted with the data after modulation, and then performs layer mapping and other processes. The receiving end performs de-layer mapping on the received data to obtain the de-layer mapped bit stream, and demultiplexes the de-layer mapped bit stream to obtain the restored original data.

[0388] Step 2205: The first device trains the first model.

[0389] In some embodiments, the first device trains a first model using a training data pair, wherein the training data pair includes first data and second data.

[0390] In some embodiments, the first model can be used for demodulation of signals or data, for example, taking input data or signals to be demodulated and outputting demodulated symbols. The first model can also be used for channel state information (CSI) compression, CSI recovery, beam management, etc., without limitation.

[0391] In some embodiments, the first device is a terminal and the second device is a network device. The terminal forms a training data pair based on the received second data and the determined first data to train the first model and obtain the trained first model.

[0392] In some embodiments, the first device is a network device and the second device is a terminal. The network device forms a training data pair based on the received second data and the determined first data to train the first model and obtain the trained first model.

[0393] In the above embodiments, the same node simultaneously performs data collection and data training. For example, in the case of the terminal side using the AI+demodulation model in downlink data transmission, the terminal performs data collection and training of the AI+demodulation model. In the case of the gNB using the AI+demodulation model in uplink data transmission, the network-side device performs training of the AI+demodulation model and receives AI sample data symbols.

[0394] In the above embodiments, the transmitting device sends the modulation symbols to the receiving device in a non-multiplexed manner. The receiving device demodulates the data received through the communication channel according to the protocol to obtain tags, which are used to form training data pairs of modulation symbols and tags to train the AI ​​model, thereby improving the training accuracy and precision of the model.

[0395] The communication method involved in the embodiments of this disclosure may include at least one of steps 2201 to 2205. For example, step 2201 may be implemented as a standalone embodiment, step 2202 may be implemented as a standalone embodiment, and steps 2201+step 2202, 2201+step 2202+step 2203+step 2204, 2201+step 2202+step 2203+step 2204+step 2205 may be implemented as standalone embodiments, and so on, but are not limited thereto.

[0396] In some embodiments, steps 2201 and 2202 may be performed in an alternate order or simultaneously.

[0397] In some embodiments, the steps and their optional implementations in other embodiments described before or after this embodiment, as well as other related parts in the specification, can be referred to, and will not be repeated here.

[0398] In this implementation or embodiment, unless there is contradiction, each step can be independent, arbitrarily combined or exchanged in order, optional methods or optional examples can be arbitrarily combined, and can be arbitrarily combined with any steps of other implementations or other embodiments.

[0399] Figure 2C is an interactive schematic diagram of a communication method according to an embodiment of the present disclosure. As shown in Figure 2C, the embodiments of the present disclosure relate to a communication method, which includes:

[0400] Step 2301: The second device sends the first data, and the first device receives the second data.

[0401] In some embodiments, the first data sent by the second device may be determined based on a protocol agreement.

[0402] In some embodiments, the method by which the second device determines the first data based on the protocol can be referred to the method of the second device determining the first data in Figures 2A and 2B, and will not be repeated here.

[0403] In some embodiments, the second device may send the first data in a first manner or in a second manner, and correspondingly, the first device may receive the second data in a first manner or in a second manner.

[0404] In some embodiments, the method by which the second device sends the first data can refer to the method of the second device sending the first data in FIG2A or FIG2B, which will not be repeated here.

[0405] In some embodiments, the manner in which the second device sends the first data and the manner in which the first device receives the second data may be other data transmission methods, which are not limited in this disclosure.

[0406] In some embodiments, the second device may send the first data and the third data to the first device in a multiplexed manner, or it may send the first data and the third data to the first device in a non-multiplexed manner.

[0407] In some embodiments, "acquire," "get," "obtain," "receive," "transmit," "bidirectional transmission," and "send and / or receive" can be used interchangeably and can be interpreted as receiving from other entities, acquiring from protocols, acquiring from higher layers, obtaining through self-processing, or autonomous implementation. Protocols include, for example, at least one of the 3GPP protocol, Wi-Fi protocol, and audio and / or video protocols.

[0408] In some embodiments, terms such as “send,” “transmit,” “report,” “distribute,” “transfer,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0409] Step 2302: The first device determines the first data.

[0410] In some embodiments, the first device determines the first data based on the received second data, either according to a protocol or an instruction from the network device.

[0411] For example, in a scenario where the first device is a terminal and the second device is a network device, the first device determines the first data based on the received second data, according to the protocol agreement or the network device's instruction.

[0412] For example, in a scheme where the first device is a network device and the second device is a terminal, the first device determines the first data based on the received second data according to the protocol agreement, or determines the first data based on the second data and the instruction sent to the second device.

[0413] In some embodiments, the way the first device determines the first data based on the protocol agreement may correspond to the way the second device determines the first data based on the protocol agreement, that is, the first device determines the first data based on the received data and with reference to the method for determining the first data agreed upon in the protocol.

[0414] In some embodiments, the method by which the first device determines the first data can refer to the method of the first device determining the first data in FIG2A or FIG2B, which will not be repeated here.

[0415] Step 2303: The first device trains the first model.

[0416] In some embodiments, the first device uses the first data and the second data as training data pairs to train the first model.

[0417] In some embodiments, the first model can be used for demodulation of signals or data, for example, taking input data or signals to be demodulated and outputting demodulated symbols. The first model can also be used for channel state information (CSI) compression, CSI recovery, beam management, etc., without limitation.

[0418] In some embodiments, the first device training the first model can refer to the implementation of the first device training the first model in FIG2A or FIG2B, which will not be repeated here.

[0419] In the above embodiments, the first device collects training data pairs and uses the collected training data pairs to train the model, thereby improving the accuracy and precision of the model training.

[0420] Figure 2D is an interactive schematic diagram of a communication method according to an embodiment of the present disclosure. As shown in Figure 2D, the embodiments of the present disclosure relate to a communication method. The method shown in Figure 2D involves devices acting as both a sender and a receiver, with one side collecting training data and the other side performing model training. The first device is a terminal, and the second device is a network device; or the first device is a network device, and the second device is a terminal. The above method includes:

[0421] Step 2401: The second device sends the first data, and the first device receives the second data.

[0422] Option 1: The first device is a terminal, and the second device is a network device.

[0423] In some embodiments, the second device determines the first data based on a protocol agreement. That is, the network device determines the first data based on a protocol agreement.

[0424] In some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, wherein the first symbol is jointly agreed upon by the network device and the terminal, and the first symbol is a symbol that has not undergone channel coding; the first data is generated by scrambling and modulating a first bit stream, wherein the first bit stream is jointly agreed upon by the network device and the terminal, and the first bit stream is a bit stream that has not undergone channel coding; the first data is generated by channel coding a second bit stream, rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, wherein the second bit stream is jointly agreed upon by the network device and the terminal.

[0425] In some embodiments, the second device may send the first data in a first manner or in a second manner, and correspondingly, the first device may receive the second data in a first manner or in a second manner.

[0426] In some embodiments, the second device sends the first data in a second manner, and the first device receives the second data in a second manner. The second manner is that the second device does not reuse the sending of the first data and the sending of the fourth data, and the fourth data is communication data that is not used for model training.

[0427] In some embodiments, the second device sends first data in a first manner, and the first device receives second data in a first manner. The first manner is to multiplex the sending of the first data and the sending of the third data by the second device.

[0428] In some embodiments, the method by which the second device sends the first data can refer to the method of the second device sending the first data in FIG2A or FIG2B, which will not be repeated here.

[0429] In some embodiments, the manner in which the second device sends the first data and the manner in which the first device receives the second data may be other data transmission methods, which are not limited in this disclosure.

[0430] In some embodiments, the second device may send the first data and the third data to the first device in a multiplexed manner, or it may send the first data and the third data to the first device in a non-multiplexed manner.

[0431] For example, a network device sends first data to a terminal, and the terminal receives second data. The first data sent by the network device can be sent by multiplexing it with third data, or it can be sent by not multiplexing it with third data.

[0432] Option 2: The first device is a network device, and the second device is a terminal.

[0433] In some embodiments, the second device determines the first data based on network device instructions.

[0434] In some embodiments, the second device determines the first data based on the second information sent by the first device.

[0435] For example, the terminal generates the first data based on the parameters for generating the first data indicated by the network device.

[0436] In some embodiments, the first data satisfies at least one of the following: the first data is generated based on a first symbol, the first symbol being indicated by a network device, and the first symbol being a symbol that has not undergone channel coding; the first data is generated based on a first bit stream, by scrambling and modulating the first bit stream, wherein the first bit stream is indicated by a network device, and the first bit stream is a bit stream directly modulated without undergoing channel coding; the first data is generated based on a second bit stream, by channel coding the second bit stream, by rate matching and modulation, layer mapping, precoding, and resource mapping of the coded bit stream, wherein the second bit stream is indicated by a network device, and the second bit stream is a bit stream before performing channel coding.

[0437] In some embodiments, the terms "precoding", "precoder", "weight", "precoding weight", "quasi-co-location (QCL)", "transmission configuration indication (TCI) status", "spatial relation", "spatial domain filter", "transmission power", "phase rotation", "antenna port", "antenna port group", "layer", "the number of layers", "rank", "resource", "resource set", "resource group", "beam", "beam width", "beam angular degree", "antenna", "antenna element", and "panel" can be used interchangeably.

[0438] For example, the physical layer process of node 1 may include at least one of the following:

[0439] Method 1: Node 1 first generates the bit stream corresponding to the modulation symbol stream to be transmitted, scrambles the bit stream, and then performs modulation to determine the modulation symbol to be transmitted. This method can achieve interference randomization and reduce interference from neighboring cells.

[0440] Method 2: Node 1 performs channel coding on the generated bitstream, rate-matches the encoded bits, and further modulates the rate-matched encoded bits according to the system modulation scheme. After passing through layer mapping, precoding, resource mapping, OFDM modulation, etc., it becomes the original data to be transmitted. The transmitting and receiving ends can agree on the system modulation scheme, modulation order, coding scheme, and the "0" and "1" information bitstreams, etc.

[0441] Method 3: Node 1 does not undergo channel coding; it directly agrees on the pre-modulation "0" and "1" bit stream and modulates the bit stream according to the system modulation method. The modulation method can be agreed upon by the transmitting and receiving ends, including the modulation order, coding method, and the "0" and "1" bit streams (the data content carried by each bit).

[0442] In some embodiments, the second device may send the first data in a first manner or in a second manner, and correspondingly, the first device may receive the second data in a first manner or in a second manner.

[0443] In some embodiments, the method by which the second device sends the first data can refer to the method of the second device sending the first data in FIG2A or FIG2B, which will not be repeated here.

[0444] In some embodiments, the manner in which the second device sends the first data and the manner in which the first device receives the second data may be other data transmission methods, which are not limited in this disclosure.

[0445] In some embodiments, the second device may send the first data and the third data to the first device in a multiplexed manner, or it may send the first data and the third data to the first device in a non-multiplexed manner.

[0446] For example, the terminal generates first data according to the instructions of the network device, and sends the first data to the network device in either a first method or a second method. The first method involves multiplexing the first data and the third data, while the second method involves not multiplexing the first data and the third data.

[0447] In some embodiments, "acquire," "get," "obtain," "receive," "transmit," "bidirectional transmission," and "send and / or receive" can be used interchangeably and can be interpreted as receiving from other entities, acquiring from protocols, acquiring from higher layers, obtaining through self-processing, or autonomous implementation. Protocols include, for example, at least one of the 3GPP protocol, Wi-Fi protocol, and audio and / or video protocols.

[0448] In some embodiments, terms such as “send,” “transmit,” “report,” “distribute,” “transfer,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0449] Step 2402: The first device determines the fifth data.

[0450] Option 1: The first device is a terminal, and the second device is a network device.

[0451] In some embodiments, the first device performs a second processing on the second data to obtain the fifth data, which is data that has not undergone demodulation and bitstream processing. Alternatively, the fifth data is data obtained by quantizing the sixth data obtained after the second processing in a manner agreed upon by the protocol, which is data that has not undergone demodulation.

[0452] Non-quantization scheme: In some embodiments, the first device performs a second processing on the second data to obtain the fifth data. The fifth data is the data before the first device demodulates the second data; the fifth data has not undergone bit-level processing.

[0453] In some embodiments, the second processing may be channel estimation and signal detection of the data during the reception process, which is not limited in this disclosure.

[0454] For example, the terminal receives downlink data sent by the gNB, performs channel estimation, signal detection, and stores the symbols before demodulation.

[0455] Quantization scheme: In some embodiments, the first device performs a second processing on the second data to obtain the sixth data, which is the data before the first device demodulates; based on the agreed method, the sixth data is quantized to obtain the fifth data.

[0456] For example, firstly, based on the agreed-upon method, the symbol to be demodulated is quantized, and then the quantized values ​​of the I and Q channels of the modulation symbol are transmitted using "0" and "1" bits. In this method, a decoding process needs to be performed. After the "0" and "1" bits are successfully decoded, the symbol to be demodulated is reconstructed (converted into specific values ​​of the I and Q channels) to obtain the input for training the AI ​​model.

[0457] Option 2: The first device is a network device, and the second device is a terminal.

[0458] In some embodiments, the first device performs a second processing on the second data to obtain the fifth data, which is data that has not undergone demodulation and bitstream processing. Alternatively, the fifth data is data obtained by quantizing the sixth data obtained after the second processing in a manner agreed upon by the protocol, which is data that has not undergone demodulation.

[0459] Non-quantization scheme: In some embodiments, the first device performs a second processing on the second data to obtain the fifth data. The fifth data is the data before demodulation by the first device and has not undergone bit-level processing.

[0460] For example, the gNB receives uplink data sent by the terminal, performs channel estimation, signal detection, and stores the symbols before demodulation.

[0461] Quantization scheme: In some embodiments, the first device performs a second processing on the second data to obtain the sixth data, which is the data before the first device demodulates; based on the agreed method, the sixth data is quantized to obtain the fifth data.

[0462] For example, firstly, based on the agreed-upon method, the symbol to be demodulated is quantized, and then the quantized values ​​of the I and Q channels of the modulation symbol are transmitted using "0" and "1" bits. In this method, the network-side device needs to perform a decoding process. After successfully decoding the "0" and "1" bits, the symbol to be demodulated is reconstructed (converted into specific values ​​of the I and Q channels) to obtain the input for training the AI ​​model.

[0463] Step 2403: The first device sends the fifth data.

[0464] Option 1: The first device is a terminal, and the second device is a network device.

[0465] In some embodiments, the first device sends fifth data to the second device via a third signaling or via a second control unit.

[0466] For example, the terminal stores the symbols before demodulation and sends them directly to the gNB via the uplink physical channel; that is, the symbols to be demodulated are transmitted directly during this process. Specifically, in this process, the symbols to be demodulated undergo layer mapping, precoding, resource mapping, OFDM modulation, and other processes before uplink transmission. There are no bit-level processing procedures such as LDPC coding and rate matching. For the uplink transmission of downlink modulated symbols, the base station does not need to configure / indicate the corresponding uplink channel transmission modulation order to the terminal again.

[0467] For example, the transmission of the symbols to be demodulated can be carried out by reporting the data symbols to be demodulated through RRC signaling (that is, the symbols to be demodulated are uploaded in RRC signaling, which is applicable to the quantization scheme mentioned above). Specifically, the data symbols to be demodulated can be reported through MDT, UAI or new RRC signaling, etc.

[0468] For example, the transmission of the symbol to be demodulated can be similar to the transmission of a CSI report, with data being reported directly via PUSCH / PUCCH.

[0469] For example, the transmission of the symbols to be demodulated can be an uplink transmission based on MAC CE.

[0470] In some embodiments, the first device uses a third resource to send fifth data to the second device, wherein the third resource is a periodic resource or a dynamic resource.

[0471] For example, the terminal uses a third resource to send fifth data to the network device, where the third resource is a periodic resource or a dynamic resource.

[0472] For example, the transmission of symbols to be demodulated can be performed using physical layer resources for uplink transmission.

[0473] For example, physical layer resources can be configured periodic resources, dynamic resources, or a combination of periodic and dynamic resources.

[0474] Option 2: The first device is a network device, and the second device is a terminal.

[0475] In some embodiments, the first device sends fifth data to the second device via a third signaling or via a second control unit.

[0476] For example, the gNB stores the symbols before demodulation and sends them directly to the terminal via the downlink physical channel; that is, the symbols to be demodulated are transmitted directly during this process. Specifically, the symbols to be demodulated undergo layer mapping, precoding, resource mapping, OFDM modulation, and other processes before downlink transmission. There are no bit-level processing steps such as LDPC coding and rate matching. For downlink transmission of uplink modulated symbols, the base station does not need to configure / indicate the corresponding downlink channel transmission modulation order to the terminal again.

[0477] For example, the transmission of the symbols to be demodulated can be carried out through downlink transmission of the data symbols to be demodulated via RRC signaling (that is, the symbols to be demodulated are transmitted in RRC signaling, which is applicable to the quantization scheme mentioned above). Specifically, the data symbols to be demodulated can be sent out through MDT, UAI, or new RRC signaling.

[0478] For example, the transmission of the symbols to be demodulated can be similar to the transmission of a CSI report, with data being directly transmitted via PDSCH / PDCCH.

[0479] For example, the transmission of the symbol to be demodulated can be a downlink transmission based on MAC CE.

[0480] In some embodiments, the first device uses a third resource to send fifth data to the second device, wherein the third resource is a periodic resource or a dynamic resource.

[0481] For example, the terminal uses a third resource to send fifth data to the network device, where the third resource is a periodic resource or a dynamic resource.

[0482] For example, the transmission of the symbols to be demodulated can be performed using physical layer resources for downlink transmission.

[0483] For example, physical layer resources can be configured periodic resources, dynamic resources, or a combination of periodic and dynamic resources.

[0484] For example, for periodic transmissions, specific periodic resources need to be configured. The gNB and the terminal jointly agree on which downlink physical channel's demodulated symbol corresponds to the demodulated symbol received by the CG uplink physical channel. For instance, the demodulated symbols transmitted via the CG uplink physical channel are: (the last available CG uplink physical channel time-domain position in the previous period + 1 time-domain unit, the first available CG uplink physical channel time-domain position in the current period - N time-domain units) of all downlink physical channel symbols received within this interval, or the first M symbols to be demodulated. M corresponds to the available resources of the CG uplink physical channel (i.e., M is determined based on the number of available resources of the CG uplink physical channel), and N is determined by the terminal's PDSCH and / or PUSCH channel processing capabilities. M is ordered within a slot in a frequency-domain-first, time-domain-second order, and between slots in a time-domain-first order. Alternatively, the CG uplink physical channel transmits the demodulated signal received on the downlink physical channel in the most recent slot. Optionally, this could be M demodulated symbols received on the downlink physical channel in the most recent slot, where M could be the first M symbols, the last M symbols, or any M symbols determined by a protocol preset rule / gNB configuration / indication. Alternatively, the gNB activates the terminal to transmit downlink demodulated symbols on the semi-static resource via dynamic DCI signaling.

[0485] For example, for dynamic scheduling, scheduling can be performed through uplink DCI signaling, or the downlink scheduling DCI can be extended, while also supporting uplink transmission of the signal to be demodulated based on uplink PUSCH.

[0486] For uplink DCI, or downlink DCI indicating both uplink and downlink, optionally, at least one of the following fields is included:

[0487] Uplink time-frequency domain resource allocation field. Optionally, this field may not exist for downlink DCI with joint uplink and downlink indication, and the uplink resources are implicitly determined by the downlink time-frequency domain resources. For example, for time-domain resources, the time-domain location of the uplink transmission is the same as the time-domain location of the downlink transmission. For frequency-domain resources, for schemes with the same uplink and downlink BWP size and SCS, the relative frequency-domain location within the BWP occupied by the uplink transmission is the same as the relative frequency-domain location within the downlink BWP occupied by the downlink transmission; for schemes with different uplink and downlink BWP sizes, the relative start number or relative end number of the frequency-domain resources within the BWP occupied by the uplink transmission is the same, etc.

[0488] Uplink transmission indication field for demodulated symbols. Optionally, this field exists only if the gNB enables the collection of downlink demodulated symbols or enables the uplink transmission of downlink demodulated symbols via RRC signaling.

[0489] MCS. For non-quantized modes, the modulation order is directly determined by the modulation order indicated by the downlink scheduling DCI, and the modulation order in the MCS is ignored; for quantized modes, this field takes effect.

[0490] For example, when using uplink DCI for independent scheduling, a certain mapping or association relationship needs to be considered between downlink DCI and uplink DCI to ensure that the terminal can determine which downlink received demodulated symbol to transmit to the gNB. This can be achieved by considering relationships such as: the same search space; related search spaces; related monitoring occasions within the same search space; related CORES ET; or the uplink DCI carrying information such as the downlink DCI's HPN and timestamp; or both uplink and downlink DCIs carrying a flag with the same value. For instance, one possible implementation is that the uplink transmission indication field (if present) of the demodulated symbol carried in the uplink DCI has the same value as the downlink demodulated symbol collection field carried in the downlink DCI (optionally, multiple values ​​can be used, such as 16, 32, etc.). Accordingly, the uplink transmission then transmits the corresponding downlink received demodulated symbol.

[0491] Furthermore, if uplink DCI is independently scheduled, a method similar to that used in semi-static transmission can be employed to determine the downlink reception symbols to be demodulated (e.g., all downlink PDSCH transmission symbols to be demodulated from the N time domain positions before the start of the current scheduled PUSCH transmission to the last uplink scheduled PUSCH transmission time domain position + 1, or the first M symbols to be demodulated, are transmitted uplink; or at least M symbols to be demodulated from the most recent PDSCH are transmitted). Alternatively, the transmission method of the demodulated symbols can be the same as the transmission method of the modulation symbols corresponding to AI sample acquisition in Figures 2A-2C (cases where model training and model inference are performed on the same side).

[0492] Optionally, for downlink scheduling DCI, the following indication field may also exist: demodulation symbol collection field (optionally, this field exists only if the downlink demodulation symbol collection function is enabled by the gNB).

[0493] For example, if the gNB indicates that the field value is "0", the terminal will not perform the collection of downlink demodulated symbols; if the gNB indicates that the field value is "1", the terminal will perform the collection of downlink demodulated symbols. Alternatively, whether the terminal performs the collection of downlink demodulated symbols and / or the uplink transmission of uplink demodulated symbols can also be determined by parameters configured by the gNB through RRC signaling.

[0494] In some embodiments, the terms "uplink", "uplink", and "physical uplink" can be used interchangeably, as can the terms "downlink", "downlink", and "physical downlink", as well as the terms "sidelink", "sidelink", "sidelink communication", "sidelink communication", "direct connection", "direct link", "direct communication", and "direct link communication".

[0495] In some embodiments, the terms “downlink control information (DCI),” “downlink (DL) assignment,” “DL DCI,” “uplink (UL) grant,” and “UL DCI” can be used interchangeably.

[0496] In some embodiments, terms such as "physical downlink shared channel (PDSCH)" and "DL data" can be used interchangeably, as can terms such as "physical uplink shared channel (PUSCH)" and "UL data".

[0497] In some embodiments, the terms “frame”, “radio frame”, “subframe”, “slot”, “sub-slot”, “mini-slot”, “symbol”, “symbol”, and “transmission time interval (TTI)” can be used interchangeably.

[0498] Step 2404: The second device sends the trained first model.

[0499] In some embodiments, the first model can be used for demodulation of signals or data, for example, taking input data or signals to be demodulated and outputting demodulated symbols. The first model can also be used for channel state information (CSI) compression, CSI recovery, beam management, etc., without limitation.

[0500] In some embodiments, the first device receives a trained first model sent by the second device, wherein the trained first model is trained by the second device, or the trained first model is provided by the second device to a third device with training data pairs and obtained from the third device.

[0501] In some embodiments, the second device trains the first model using a training data pair consisting of the first data and the fifth data; and sends the trained first model to the first device.

[0502] In some embodiments, the second device sends a training data pair consisting of first data and fifth data to the third device; and receives a trained first model sent by the third device, wherein the trained first model is trained by the third device based on the training data pair.

[0503] Option 1: The first device is a terminal, and the second device is a network device.

[0504] In some embodiments, the network device trains the first model using a training data pair consisting of first data and fifth data; and sends the trained first model to the terminal.

[0505] In some embodiments, the network device sends a training data pair consisting of first data and fifth data to other network devices; receives a trained first model sent by other network devices, and sends the trained first model to the terminal.

[0506] For example, while performing downlink transmission, the network side stores the bits before modulation or the modulated symbols after modulation as tags. Simultaneously, the network-side device performs uplink reception to obtain the downlink symbols to be demodulated transmitted by the terminal. These demodulated symbols can be used as input for model training. This achieves the collection of training data (received demodulated symbols, transmitted modulated symbols / transmitted coded bit stream or modulated symbols), that is, (model input data, tags).

[0507] For example, the node that performs model training can be a gNB or a different entity from the gNB. The gNB also needs to send (the receiving end demodulation symbol, the label) to the model training node through the network device interface.

[0508] Option 2: The first device is a network device, and the second device is a terminal.

[0509] In some embodiments, the terminal uses a training data pair consisting of first data and fifth data to train the first model; and sends the trained first model to the network device.

[0510] In some embodiments, the terminal sends a training data pair consisting of first data and fifth data to other terminals or network devices; receives a trained first model sent by other terminals or network devices, and sends the trained first model to the network device.

[0511] For example, while performing uplink transmission, the terminal stores the bits before modulation or the modulated symbols after modulation as tags. Simultaneously, the terminal device performs downlink reception to obtain the uplink symbols to be demodulated sent by the network device. These demodulated symbols can be used as input for model training. This achieves the collection of training data (received demodulated symbols, transmitted modulated symbols / transmitted coded bit stream or modulated symbols), that is, (model input data, tags).

[0512] For example, the node that performs model training can be a terminal or other terminals.

[0513] In the above embodiments, the receiving device sends the received data to the sending device without knowing the data sent by the sending device, so that the sending device can obtain the trained model. The receiving device then uses the trained model for model inference, which improves the accuracy and precision of model training.

[0514] The communication method involved in the embodiments of this disclosure may include at least one of steps 2401 to 2404. For example, step 2401 may be implemented as a standalone embodiment, and steps 2401+2402, 2401+2402+2403, 2401+2402+2403+2404 may be implemented as standalone embodiments, and so on, but are not limited thereto.

[0515] In some embodiments, the steps and their optional implementations in other embodiments described before or after this embodiment, as well as other related parts in the specification, can be referred to, and will not be repeated here.

[0516] In this implementation or embodiment, unless there is contradiction, each step can be independent, arbitrarily combined or exchanged in order, optional methods or optional examples can be arbitrarily combined, and can be arbitrarily combined with any steps of other implementations or other embodiments.

[0517] Figure 3 is an interactive schematic diagram of the communication method provided in this disclosure. As shown in Figure 3, the embodiments of this disclosure relate to a communication method, which includes:

[0518] Step 3101: The second device sends the first data, and the first device receives the second data.

[0519] The second data is the data received by the first device that corresponds to the first data sent by the second device. The first data and the second data form a training data pair, which is used to train the first model.

[0520] The optional implementations of step 3101 can be found in the optional implementations of step 2103 in Figure 2A, step 2203 in Figure 2B, step 2301 in Figure 2C, and step 2401 in Figure 2D, as well as other related parts in the embodiments involved in Figures 2A, 2B, 2C, and 2D, which will not be repeated here.

[0521] The following are specific implementation schemes of the communication method provided according to embodiments of this disclosure:

[0522] For Case 1, where model training and inference are performed on the same node, the following design approach can be adopted:

[0523] Key Point 1: Node 1 generates various modulation symbols based on the 6G modulation scheme and the corresponding constellation diagrams, and sends the corresponding modulation symbols to Node 2. Specifically, the physical layer process may include at least one of the following:

[0524] Method 1: Without channel coding, the modulation symbols to be transmitted are directly agreed upon and sent to Node 2. It can be imagined that in order to achieve interference randomization and reduce interference from neighboring cells, Node 1 can first generate the bit stream corresponding to the modulation symbol stream to be transmitted, scramble the bit stream, perform modulation, and then send it to Node 2.

[0525] Method 2: Analogous to the processing flow of the terminal baseband in a 5G system, the bit stream undergoes channel coding, the encoded bits are rate matched, and the coded bits output by rate matching are further modulated according to the 6G system modulation method. After passing through layer mapping, precoding, resource mapping, OFDM modulation and other processes, the symbols are sent to the base station. The terminal and the base station agree on the 6G system modulation method, modulation order, coding method, and "0" and "1" information bit streams, etc.

[0526] Method 3: Without channel coding, the "0" and "1" bit streams before modulation are directly agreed upon. The bit streams are modulated according to the 6G system modulation method and sent to the base station. The terminal and the base station agree on the 6G system modulation method, modulation order, and the "0" and "1" bit streams (the data content carried by each bit).

[0527] Key Point 2: Based on Key Point 1, the parameters related to the transmission of the modulation symbols or the bit stream shall be jointly agreed upon by the base station and the terminal through at least one of the following methods:

[0528] Specifically, the parameters used for transmitting modulation symbols / bit streams include at least one of the following:

[0529] For Method 1 of Point 1, the parameters include at least one of the following: time-domain resource allocation (location and number), frequency-domain resource allocation parameters (the frequency-domain resource location and number are determined based on these parameters and other parameters such as resource allocation method), modulation method, modulation order, precoding matrix (uplink), antenna port, number of layers, modulation symbol transmission order, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency-domain resource allocation method (continuous or discrete), resource mapping method, etc.

[0530] For Method 2 of Point 1, the parameters include at least one of the following: time / frequency domain resource allocation parameters, modulation scheme, modulation order, precoding matrix (uplink), antenna port, number of layers, "0" and "1" information bit streams, encoding scheme, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency domain resource allocation scheme (continuous or discrete), resource mapping scheme (downlink, whether interleaving is allowed under continuous frequency domain resource allocation), code rate, etc.

[0531] For Method 3 of Point 1, the parameters include at least one of the following: time / frequency domain resource allocation parameters, modulation method, modulation order, precoding matrix (uplink), antenna port, number of layers, "0" and "1" bit streams, waveform (uplink, or uplink and downlink), PRB bundling size (downlink), frequency domain resource allocation method, resource mapping method (downlink, whether interleaving is enabled), redundancy version, etc.

[0532] Specifically, the above parameters can be agreed upon by the base station and both parties in at least one of the following ways:

[0533] Method 1: The base station configures the terminal device via semi-static signaling. One possible method is that the semi-static signaling can be RRC, such as RRCRecofiguration, RRCsetup, RRCResume, RRCestablishment, RRCelease, etc.

[0534] Method 2: The base station instructs the terminal device via dynamic signaling. One possible approach is that the dynamic signaling can be DCI or MAC CE;

[0535] Method 3: Combining Method 1 and Method 2, the base station first configures a list of values ​​for relevant parameters. Then, dynamic signaling indicates a row in the table to indicate the specific values ​​of the relevant parameters to the terminal device.

[0536] Method 4: The parameter values ​​are determined jointly by the base station and the terminal based on the protocol agreement;

[0537] Method 5: In addition, some parameters can be reused from those used in normal data transmission, such as pdsch-TimeDomainAllocationList, datascramblingIdentityPDSCH, etc.

[0538] It is conceivable that different parameters can have different conventions, and the above-mentioned different conventions can be combined with each other.

[0539] Specifically, for parameter indication using DCI, existing fields or newly added fields can be used. For indication using existing fields, such as in method 1 / 3 of point 1, since LDPC encoding and rate matching are no longer performed, fields such as RV / MCS / NDI / DAI / PDSCH-to-HARQ_feedback timing indicator can be used to indicate the aforementioned parameters. For example, this could be used to indicate the index in the bitstream configuration table in Example 1 below. The MCS field can be used in conjunction with other parameters to jointly indicate the modulation order, such as indicating the bitstream order / modulation symbol stream order, or using some bits to indicate the modulation order and others to indicate the bitstream order.

[0540] Specifically, the parameters related to the transmission of the modulation symbols or the bit stream shall be agreed upon by the base station and the terminal through at least one of the following methods:

[0541] Example 1 (Dynamic Scheduling Based on Traditional Uplink or Downlink Physical Channel Transmission): At least some parameters are indicated by the gNB via dynamic signaling. Optionally, other parameters can be configured by the gNB or preset by the protocol. The protocol presets the modulation scheme; for example, the set of preset modulation schemes can be {QPSK, 16QAM, 64QAM, 256QAM, 1024QAM}, etc., where the QAM modulation can be rectangular QAM modulation. The base station sends a configuration table of modulation symbol / information bit stream / bit stream transmission order to the terminal via semi-static signaling; or the protocol presets a configuration table of modulation symbol / information bit stream / bit stream transmission order, and different modulation schemes can have different configuration tables. The transmitting and receiving ends determine the transmission order of modulation symbols / information bit streams / bit streams based on a row in the configuration table, and generate all modulation symbols / information bit streams / bit streams transmitted over the air interface based on the order. Furthermore, during each dynamic scheduling / semi-static configuration, the gNB can indicate the modulation order and table index, thereby indicating the transmission order or bit stream corresponding to the corresponding modulation order to the terminal device.

[0542] For example, regarding Method 1 as described in Point 1, assuming the modulation scheme is QPSK, the four modulation symbols are numbered sequentially. For instance, they can be numbered according to the order of quadrants 1 / 2 / 3 / 4 (modulation symbols #0 / 1 / 2 / 3 in sequence). Alternatively, they can be numbered according to the bitstream value corresponding to each modulation symbol; then, the modulation symbols corresponding to "00", "01", "10", and "11" can be numbered as symbols #0 / 1 / 2 / 3 respectively.

[0543] Furthermore, the protocol presets or the gNB configures a transmission order table for the modulation symbols. For example, the preset transmission order table is shown in Figure 4A. When performing dynamic scheduling / semi-static configuration, the gNB can indicate one row index in the table below, and the terminal determines the transmission order of the modulation symbols based on this indication. Modulation symbols are then generated based on this order and transmitted over the air interface. Additionally, if the table index is dynamically indicated, it can be based on existing fields in the DCI (RRC / DCI additionally displays an indication that AI+ demodulation dataset collection is enabled), or newly added fields (based on the newly added fields, it determines whether AI+ demodulation dataset collection is enabled, and determines the transmission order based on the indication information of the newly added fields).

[0544] Specifically, determining the transmission order of modulation symbols based on the index indication includes: the indication index is the transmission order of modulation symbols. For example, if the index is 0, then the number of modulation symbols is determined according to the resources (time-frequency spatial domain resources) allocated by the gNB, and then the modulation symbols are generated according to the order indication. For example, if the number of modulation symbols is 10, the transmitted QSPK symbols are: 0, 1, 2, 3, 0, 1, 2, 3, 0, 1; or, the starting transmission order is determined based on the index indication. If the number of symbols determined based on the resources is greater than the number of symbols under a certain modulation order, then the transmission is cyclically performed based on the index corresponding to the starting transmission order. For example, assuming the number of symbols is 40 and the starting transmission order indication is 2, then the transmission will be cyclically performed with indices 2, 3, 0, 1, meaning the order of modulation symbols to be transmitted is: 0, 2, 1, 3, 1, 3, 0, 2, 0, 1, 2, 3, 3, 2, 1, 0, 0, 2, 1, 3, 1, 3, 0, 2…

[0545] Alternatively, the protocol may pre-define the transmission order configuration table to have only one row (that is, the transmission order of modulation symbols / information bit streams / bit streams is fixed in the protocol). The terminal directly generates and transmits modulation symbols based on the transmission order, without needing to read the table index indication field carried in signaling such as DCI / RRC / MAC. For example, the preset transmission order could be 0 / 2 / 1 / 3. Or, the protocol may pre-define the transmission order of modulation symbols as 0 / 2 / 1 / 3 / 1 / 3 / 0 / 2, etc.

[0546] One possible approach is to use the parameters employed in the last transmission, or the protocol default parameters, if no notification is received via dynamic signaling / RRC signaling.

[0547] It is conceivable that some parameters can be configured to the terminal by the base station through semi-static signaling, and / or other parameters can be indicated to the terminal by the base station through MAC CE, and / or the remaining parameters can be indicated to the terminal by the base station through dynamic signaling.

[0548] Example 2: A resource configuration method similar to SRS / CSI-RS can also be used, where the modulation symbol is configured to a specific resource location by the gNB. And / or, the resource location of the modulation symbol is preset by the protocol (or a combination of the above two methods). Specifically, whether to schedule the transmission of the modulation symbol can be determined using at least one of the following methods: aperiodic, semi-persistent, periodic, etc. A schematic diagram of a possible resource configuration method is shown in Figure 4B.

[0549] Optionally, for the method of pre-setting the resource location of the modulation symbol in the protocol, different modulation methods can have different resource mapping patterns. In this method, if the resource is configured through a gNB, at least one of the following needs to be determined by the configuration parameters: time-domain symbol location, slot location (which can be determined by parameters such as time slot offset and certain calculation formulas), period (the determination of slot and period is applicable to semi-persistent and periodic transmission methods), RE location, RB location (or, it can be determined by the starting RB location, the number of RBs, frequency domain density, etc.).

[0550] Optionally, there can be multiple different time-frequency domain resource patterns within a slot / RB. Specifically, the choice of which pattern to use depends on the gNB configuration / instruction. In this mode, if the time-frequency domain resources occupied by the uplink physical channel or downlink physical channel include the time-frequency domain resources occupied by the modulation symbols, then the influence of this factor needs to be considered when performing TBS calculation, rate-matching, and resource mapping. That is, the time-frequency domain resources occupied by the modulation symbols need to be excluded.

[0551] The ordering of the modulation symbols such as M#0 and M#1 is as shown in Example 1.

[0552] Example 3 (Periodic transmission based on a method similar to traditional uplink or downlink physical channels):

[0553] Based on semi-static resources configured / indicated by the base station, periodic uplink transmission or downlink reception of modulation symbols is performed.

[0554] Some parameters used for modulation symbol transmission can be configured by the gNB through semi-static signaling, such as MCS table selection and transmission period of semi-static transmission, similar to PUSCH configured grant type 1 / SPS(PDSCH). Other signaling can be indicated by activation signaling, such as modulation order, modulation symbol transmission order, and time-frequency domain resources.

[0555] The activation signaling can be DCI / MAC CE, etc.

[0556] Optionally, when the terminal receives a deactivation signaling (which may be DCI / MAC CE, etc.), it stops transmitting uplink modulation symbols or receiving downlink modulation symbols on the periodic uplink / downlink physical resources.

[0557] The SPS / CG configuration can be one or more. Optionally, if there are multiple SPS / CGs, a group of SPS / CGs can be activated / deactivated through the same DCI / MAC CE.

[0558] Another possible approach is that the parameters used for modulated symbol transmission are configured by the gNB via semi-static signaling. And / or, some parameters are preset by the protocol.

[0559] For example, parameters such as modulation order and time-frequency domain resources are configured by the gNB through semi-static signaling. The transmission order of modulation symbols is preset by the protocol.

[0560] In this approach, the SPS / CG configuration can also be one or more.

[0561] Optionally, different SPS / CG configurations have different modulation order / modulation symbol transmission order.

[0562] In addition, the gNB can also stop transmitting uplink modulation symbols or receiving downlink modulation symbols by deactivation signaling.

[0563] Example 4: Place the known modulation symbol at a preset resource location and transmit it along with ordinary data using multiplexing.

[0564] Example 4-1: Reuse the DMRS location and perform multiplexing transmission of ordinary data.

[0565] Example 4-2: Known modulation symbols are placed at preset positions and transmitted in multiplexing mode with ordinary data. The preset positions are agreed upon by the terminal and the gNB through a certain negotiation method, including: the protocol presets the positions of the known symbols; the gNB configures / indicates the positions of the known symbols.

[0566] The positions of known modulation symbols are determined by combining the two methods described above.

[0567] Based on the above two methods (Examples 4-1 / 4-2), the transmission of the uplink or downlink physical channel does not need to carry the DMRS (demodulation reference signal) again. The receiving end can perform channel estimation and signal detection of the uplink or downlink physical channel based on the above-mentioned known modulation symbol.

[0568] Key Point 3: Based on Key Points 1 and 2, regarding whether the transmitted symbols used for AI+demodulation data acquisition should be multiplexed with ordinary data / HARQ feedback / CSI, there are two approaches:

[0569] Direction 1: If the transmitted symbols used for AI+Demodulation data collection are not multiplexed with ordinary data, then the scheduling of ordinary data and the scheduling of symbol transmissions related to AI+Demodulation data collection shall be performed in at least one of the following ways:

[0570] Semi-static scheduling: During SPS / configured grant configuration, separate SPS / configured grant resources are configured for the transmission of ordinary data and modulation symbols. The resources corresponding to modulation symbols are distinguished from the SPS / configured grant resources for ordinary data through RRC field name, modulation symbol transmission-related configuration parameters, display indication fields, etc.

[0571] Dynamic scheduling: For dynamic scheduling, the scheduling can be distinguished in at least one of the following ways as to whether it is for ordinary data and / or CSI and / or HARQ feedback, or for transmission symbols related to AI+Demodulation data acquisition.

[0572] Method 1: Determine the data collection information through the indicator fields carried in the DCI. For example, distinguish them by the newly added fields in the DCI as described in Example 1 of Point 2, or by the display enable fields related to dataset collection carried in the DCI.

[0573] Method 2: Determined through explicit configuration of RRC / MAC CE. For example, RRC may be configured to schedule data acquisition for AI+Demodulation within a specific time period, or within certain time units (slots / subframes / frames, etc.). Alternatively, RRC may be explicitly configured to use one or more HARQ processes for data acquisition, while avoiding the use of these HARQ processes when transmitting ordinary data during this period.

[0574] Method 3: If RRC or other signaling configurations include parameters related to AI+Demodulation data acquisition, or if AI+Demodulation data acquisition is enabled, the terminal can further determine which time units or HARQ processes are used for data acquisition through protocol presets. For example, determining the slot where mod(x, 10slot) = 1 / 3 is used for data acquisition.

[0575] Method 4: The separate DMRS pattern determines whether the scheduling is for general data and / or CSI and / or HARQ feedback, or for transmitted symbols related to AI+demodulation dataset acquisition. The DMRS pattern can be either the PDCCH DMRS pattern or the PDSCH DMRS pattern. The separate DMRS pattern is defined by the protocol and / or indicated by the gNB. For example, if this scheduling corresponds to the acquisition of AI sample data, then the PDCCH / PDSCH uses DMRS pattern #1; otherwise, it uses DMRS pattern #2.

[0576] Method 5: By using separate search space and / or separate CORESET, that is, dynamic scheduling DCI for AI sample acquisition is transmitted using dedicated time-frequency domain resources.

[0577] Method 6: Determine that the RNTI can be an independent RNTI configured by gNB for AI sample acquisition by separating the RNTI.

[0578] Direction 2: Transmit symbols used for AI+Demodulation data collection can be multiplexed with ordinary data.

[0579] Specifically, a similar approach to UCI multiplexing on PUSCH can be used for multiplexing AI+Demodulation data and ordinary data. This includes determining the number of modulation symbols used for data acquisition, the transmission positions of coded bits / modulation symbols, the number of transmitted bits in UL-SCH, and the positions of the transmitted bits in UL-SCH.

[0580] One possible approach to determining the number of modulation symbols is that the protocol predetermines the number of modulation symbols to be transmitted as 1 / M of the total number of available symbols, where the value of M is specified by the protocol or indicated by the gNB. The total number of available symbols is determined based on at least one of the following factors: the number of time-domain resources, the number of frequency-domain resources, and the number of layers.

[0581] Alternatively, another possible way to determine the number of modulation symbols is to preset / configure / indicate the number of symbols to be transmitted as Q or the number of bits to be transmitted as Ns, or to determine the number of symbols to be transmitted as Q or the number of bits to be transmitted as Ns based on the preset method of the protocol. For example, the number of symbols to be transmitted is S times the total number of modulation symbols corresponding to the current modulation order (for 16QAM, the number of modulation symbols to be transmitted is 16*S), and the value of S is specified by the protocol / configured by the gNB.

[0582] Optionally, the number of symbols to be transmitted can be determined according to dynamic scheduling signaling, for example, DCI indicates that the number of symbols to be transmitted is Q, or DCI indicates the above-mentioned S value.

[0583] Another possible approach is for the terminal to determine the actual number of symbols to be transmitted based on a certain calculation formula. For example, it can be determined using the following formula, where O_m is the number of information bits to be transmitted determined by any of the above methods, or the number of modulation symbols to be transmitted multiplied by the modulation order.

[0584] (Correspondingly, ACK / modulation symbol / Normal data can be multiplexed, and the transmission symbol for ACK is determined first.) Optionally, CRC attachment can be performed, i.e., L is not 0, or CRC attachment can be performed, i.e., L is 0.

[0585] As just one possible implementation, this calculation formula can correspond to Method 2 in Point 1. The terminal processing method can be as follows: First, perform CRC attachment based on O_m, encoding (optionally, LDPC encoding), and rate matching. The rate matching is performed according to Qm' determined in the above formula, where the number of coded bits to be transmitted is Qm' * modulation order * layer number. Then, perform multiplexing of the coded bits of the symbol to be transmitted and the normal data.

[0586] The above calculation formula can be used to determine the number of transmission symbols for a layer. Correspondingly, M_UL_SC is the total number of resources available for PUSCH transmission.

[0587] In the above calculation formula, L is the CRC bit length. If the processing is based on method 1 / 3 in point 1, then L can be considered as 0. Of course, alternatively, for method 2, the value of L can also be 0.

[0588] For multiplexing data and AI samples, one possible approach is to use bit-level multiplexing. That is, after the bitstreams corresponding to ordinary data and the bitstreams corresponding to AI samples have each undergone rate matching and / or code block concatenation processes, the data bitstreams and the bitstreams of the AI ​​samples are then multiplexed together. Optionally, at least one of the following multiplexing methods can be used:

[0589] tx = [modulation_bits, data_bits(0: length(data_bits) - length(modulation_bits) - 1], meaning that the bitstream corresponding to the modulation symbols (AI samples) is placed at the beginning, and the remaining positions are filled with data coded bits, or vice versa. Alternatively, the position of modulation_bits in the bitstream is determined as follows: index of modulation bits = j * (total bit length / modulation bit length), j = 0, 1, ..., modulation bit length - 1. Where, modulation bit length = Q'm * number of layers * modulation order, where the number of layers is a positive integer. Total bit length is the total number of bits that can be transmitted across all layers (e.g., total bit length = total number of available time-frequency resources * number of layers * modulation order). In this method, when performing rate matching for ordinary data bits, it is based on the total number of available resources (Number of coded bits from...). Rate-matching (PUSCH can be rate-matched using RE * number of layers * modulation order) is performed. During this multiplexing process, data bits are rate-matched. Alternatively, they can be rate-matched using puncture. That is, for data bits, the total number of bits that can be transmitted across all layers is determined, and then the data bits at the bit positions determined for the AI ​​samples are removed, and the corresponding positions are replaced with the coded bits corresponding to the AI ​​samples. For example, taking the bit corresponding to the AI ​​sample at the beginning of the multiplexing output bitstream, the following multiplexing method is used, where tx is the output bitstream after the multiplexing process is completed. The data_bits are the coded bits output after rate matching and code block concatenation.

[0590] tx=[modulation_symbols,data_symbols(length(modulation_symbols),end)]

[0591] Tx = [modulation_bits, data_bits]. In this mode, when performing rate matching for ordinary data bits, the number of output coded bits for rate matching is determined according to the number of data coded bits from rate-matching = all available REs from PUSCH * number of layers * modulation order - Q'm * number of layers * modulation order, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0592] Based on this method, the transmission order of the modulation symbols can be configured by the gNB or preset by the protocol, or indicated by the scheduling DCI (existing fields can be reused, or new fields can be added). The modulation order of the modulation symbols can be the same as that of ordinary data, or a special modulation order indicated by DCI / MAC CE or configured by RRC.

[0593] Optionally, if no UL-SCH and / or CSI are generated, only AI+Demodulation data acquisition will be performed.

[0594] Another possible approach is to perform multiplexing between symbols and data during symbol-level processing. For example, for methods 1 / 2 / 3, the terminal first calculates the number of modulation symbols to be transmitted (total number of modulation symbols to be transmitted = Q'm * number of layers) according to the above calculation formula. It then performs multiplexing of the number of modulation symbols to be transmitted with the normal data after modulation, and then performs layer mapping and other processes. `tx = [modulation_symbols, data_symbols(0: length(data_symbols) - length(modulation_symbols) - 1]` means that modulation symbols are placed at the beginning of the output symbol stream, and data symbols are placed at the end, or vice versa. Alternatively, the position of `modulation_symbols` in the symbol stream is determined as follows: `index of modulation symbol = j * (total symbol length / modulation symbol length)`, where `j = 0, 1, ..., modulation symbol length - 1`. Here, modulation symbol length = Q'm * number of layers, and the total number of symbols is the total number of symbols that can be transmitted by all layers. In this method, when performing rate matching for ordinary data bits, it follows the total number of available resources (Number of coded bits from...). Rate-matching is performed using the formula (all PUSCH symbols can be rate-matched using RE * number of layers * modulation order). In this multiplexing process, data symbols are valued using rate-matching. Alternatively, values ​​can be determined based on puncture. That is, for each data symbol, the total number of symbols that can be transmitted across all layers is determined. Then, the data symbol carrying the position determined for the modulation symbol in this method is removed, and the corresponding position is replaced with the corresponding modulation symbol. The corresponding puncture value method, taking the modulation symbol at the beginning of the symbol stream as an example, has the following multiplexing formula, where tx is the output symbol stream after the multiplexing process is completed:

[0595] tx=[modulation_symbols,data_symbols(length(modulation_symbols),end)]

[0596] Tx = [modulation_symobls, data_symbols]. In this mode, when performing rate matching for ordinary data bits, the number of output coded bits for rate matching is determined according to the number of data coded bits available from rate-matching (Number of data coded bits from rate-matching = all available REs from PUSCH * number of layers * modulation order - Q'm * number of layers * modulation order, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0597] It should be noted that in this method, length(modulation_symbols) and length(data_symbols) are the total number of modulation symbols and data symbols of all layers before multiplexing is performed; or, first perform layer mapping for data symbols and modulation symbols respectively in the above manner, and then perform multiplexing for modulation symbols and data symbols of each layer.

[0598] Opt.1:tx = [modulation_symbols, data_symbols(0:length(data_symbols)-length(modulation_symbols)-1], meaning that modulation symbols are placed at the beginning of the output symbol stream, and data symbols are placed at the end, or vice versa. Alternatively, the position of modulation_symbols in the symbol stream is determined as follows: index of modulation symbols = j * (total symbol length / modulation symbol length), j = 0, 1, ..., modulation symbol length - 1, where the modulation symbol length = Q'm, which can be determined by the previously mentioned formula. The total number of symbols is the total number of symbols that a layer can transmit. In this mode, when ordinary data bits are rate matched, the data is matched according to the total number of available resources (Number of coded bits from...). Rate-matching (PUSCH can be rate-matched using RE * modulation order * number of layers) is also possible. In this multiplexing process, data symbols are rate-matched. Alternatively, values ​​can be obtained using puncture; that is, for data symbols, layer mapping is performed according to the total number of symbols that a layer can transmit. Then, the data symbol carrying the position determined for the modulation symbol in this method is removed and replaced with the modulation symbol.

[0599] Opt.2: Tx = [modulation_symobls, data_symbols], In this mode, when performing rate matching for ordinary data bits, the number of output coded bits for rate matching is determined according to the number of data coded bits from rate-matching = all available RE * modulation order * number of layers - Q'm * modulation order * number of layers, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0600] It should be noted that in this method, length(modulation_symbols) and length(data_symbols) represent the total number of modulation symbols and data symbols determined for each layer after the layer mapping is completed.

[0601] Alternatively, data symbols and modulation symbols can be encoded independently (for method two), undergoing processes such as CRC attachment, rate matching, code block concatenation, layer mapping, and precoding. Finally, resource mapping is performed based on Q'm calculated using the aforementioned formula. For example, Q'm resource units can be reserved for modulation symbols. These resources can be indexed in a frequency-domain priority or time-domain priority manner, such as reserving the first Q'm resource units, the last Q'm resource units, or the j-th (total number of symbols / Q'm), where j = 0, 1, ..., Q'm-1, and the total number of symbols is the total number of symbols that a layer can carry. Based on this, the modulation symbols of each layer are mapped to the reserved resource units using either interleaving or non-interleaving methods, with the remaining parts mapped to data symbols. Furthermore, in this method, rate matching of data symbols can be performed in the following way:

[0602] Opt.1: When rate-matching the encoded bits of ordinary data, rate matching is performed according to the total number of data coded bits from rate-matching = all available REs in PUSCH * modulation order * number of layers. Furthermore, in this method, the mapped data symbols can be mapped using puncturing or rate-matching. Punching involves mapping all data symbols of each layer onto the resource grid according to a certain resource mapping order; then, the data symbols in the resource positions reserved for modulation symbols in the above method are removed and replaced with the corresponding modulation symbols. Alternatively, data symbol resource mapping can also be performed using rate matching. That is, data_symbols(0:length(data_symbols)-length(modulation_symbols)-1) are mapped to the positions of the remaining data symbols according to a certain resource mapping order.

[0603] Opt.2: When performing rate matching for ordinary data bits, the number of output coded bits is determined according to the number of data coded bits from rate-matching = all available PUSCH RE * modulation order * number of layers - Q'm * modulation order * number of layers, where Q'm is the number of modulation symbols that can be transmitted per layer calculated in the aforementioned formula).

[0604] Another possible implementation is to add a bit stream corresponding to the modulation symbol, or to add a modulation symbol, to the end of the data information bits or the end of the data symbol; or to replace the padding bits constituting the MAC PDU with a bit stream corresponding to the modulation symbol or a modulation symbol. In this implementation, to identify that the padding bit carried by the MAC SDU has been replaced with a bit stream corresponding to the modulation symbol, a new LCID can be used, or the LCID of the padding bit can be reused. The number of modulation symbols is determined by the above formula, or by the terminal through RRC signaling or dynamic signaling, or the size of the bit stream corresponding to the modulation symbol is determined by the number of remaining bits to be transmitted, which is the total number of bits that the current resources can carry - UL - SCH MAC SDU - MAC CE - other overhead (such as MAC subheader overhead, etc.). Optionally, for dynamic signaling indication mode, one of the values ​​of this field can also indicate that the modulation symbol occupies "0" bits, or 0 symbols, that is, no multiplexing, or it can also indicate a ratio, or the protocol can simultaneously support Direction 1 / 2, and further determine through the gNB display indication whether only AI+Demodulation data acquisition is performed, or AI+Demodulation data acquisition and UL-SCH and / or CSI (including HARQ feedback, etc.) transmission are performed simultaneously.

[0605] Key Point 4: Node 2 (the receiving end) transmits the corresponding relevant parameters according to the modulation symbols, receives the AI ​​sample modulation symbols sent by Node 1, and obtains at least one signal to be demodulated through the receiving end processing flow (such as FFT, channel estimation, signal detection, etc.). Simultaneously, it transmits the corresponding relevant parameters according to the modulation symbols agreed upon by both the transmitting and receiving parties, determines the modulation symbol corresponding to the sending end, and identifies it as the label corresponding to the signal to be demodulated, thus obtaining at least one set of data (the demodulated symbol, the label), where the demodulated symbol is the input for model training, and the label is the label corresponding to the model output. Based on this, multiple sets of data are collected to construct a candidate dataset.

[0606] Key Point 5: Optionally, each modulation symbol transmission requires transmitting all the symbol resources corresponding to all constellation points under the modulation method; or, each modulation symbol transmission may only transmit the symbol resources corresponding to a portion of the constellation points under the modulation method.

[0607] Key Point 6: Node 1 can be a terminal, and Node 2 can be a base station or network-side equipment; alternatively, Node 1 can be a network-side equipment, and Node 2 can be a terminal. This approach primarily considers cases where the same node simultaneously performs data collection and data training. For example, in the case of using an AI+Demodulation model on the terminal side during downlink data transmission, the terminal performs data collection and simultaneously trains the AI+Demodulation model. In the case of using an AI+Demodulation model on the gNB during uplink data transmission, the network-side equipment trains the AI+Demodulation model and simultaneously receives AI sample data symbols.

[0608] Case 2: The opposite node performs model training, while the local node performs model inference.

[0609] Key Point 1: For cases where the network side performs AI+Demodulation (downlink) model training for the terminal (the AI+demodulation model is used for downlink channel demodulation on the terminal side), training data can also be collected in the following ways:

[0610] The terminal receives downlink data sent by the gNB, performs channel estimation and signal detection, stores the symbols before demodulation, and sends them back to the gNB.

[0611] Non-quantization scheme: Specifically, the method can be to store the symbols before demodulation and send them directly to the gNB via the uplink physical channel. That is, in this process, the symbols to be demodulated are directly transmitted (similar to the direct transmission of the modulation symbols corresponding to AI sample acquisition in Case 1). Specifically, this process involves directly performing layer mapping, Discrete Fourier Transform (DFT), precoding, resource mapping, OFDM modulation, etc., on the symbols to be demodulated before uplink transmission, without LDPC coding, rate matching, or other bit-level processing. In this scheme, for the uplink transmission of downlink modulation symbols, the base station does not need to reconfigure / indicate the corresponding uplink channel transmission modulation order to the terminal.

[0612] Quantization scheme: Alternatively, first, based on the agreed-upon protocol, quantize the symbol to be demodulated, and then use "0" and "1" bits to transmit the quantized values ​​of the I and Q channels of the modulation symbol. In this method, the network-side device needs to perform a decoding process to successfully decode the "0" and "1" bits before reconstructing the symbol to be demodulated (converting it into specific values ​​for the I and Q channels) to obtain the input for training the AI ​​model.

[0613] Specifically, for at least one of the quantization schemes and non-quantization methods, the transmission of the symbols to be demodulated can be carried out in the following manner:

[0614] It is possible to report the data symbols to be demodulated via RRC signaling (that is, to upload the symbols to be demodulated via RRC signaling, which is applicable to quantization schemes).

[0615] The data symbols to be demodulated can be reported through MDT, UAI, or new RRC signaling.

[0616] Alternatively, similar to the transmission of CSI reports, data can be reported directly via PUSCH / PUCCH.

[0617] Alternatively, uplink transmission of the symbols to be demodulated can be performed based on MAC CE.

[0618] Furthermore, physical layer resources can be configured periodic resources, and / or, can be dynamic resources.

[0619] Specifically, for periodic transmissions, the following considerations are taken into account:

[0620] Specific periodic resources need to be configured; in addition, the gNB and the terminal agree on which demodulation symbol is received on the downlink physical channel corresponding to the demodulation symbol transmitted on the uplink physical channel of the CG.

[0621] For example, the demodulated symbols transmitted on the CG uplink physical channel are the demodulated symbols of all downlink physical channels received within the interval (the time domain position of the last available CG uplink physical channel in the previous cycle + 1 time domain unit, the time domain position of the first available CG uplink physical channel in the current cycle - N time domain units), or the first M demodulated symbols. M corresponds to the available resources of the CG uplink physical channel (i.e., M is determined based on the number of available resources of the CG uplink physical channel), and N is determined by the terminal's PDSCH and / or PUSCH channel processing capabilities. M is ordered within a slot in a frequency domain-first, time domain-second order, and between slots in a time domain-order order.

[0622] Alternatively, the CG uplink physical channel transmits the demodulated signal received on the downlink physical channel in the most recent slot. Optionally, it can be one of the M demodulated symbols received on the downlink physical channel in the most recent slot, where M can be the first M symbols, the last M symbols, or which M symbols are determined by the protocol preset rules / gNB configuration / indication.

[0623] Alternatively, the gNB can activate the terminal to transmit downlink demodulated symbols on the semi-static resources via dynamic DCI signaling.

[0624] Furthermore, physical layer resources can be configured periodic resources, and / or, can be dynamic resources.

[0625] Specifically, for dynamic scheduling, scheduling can be performed through uplink DCI signaling, or the downlink scheduling DCI can be extended, while also supporting uplink transmission of the signal to be demodulated based on uplink PUSCH.

[0626] For uplink DCI, or downlink DCI indicating both uplink and downlink, optionally, at least one of the following fields is included:

[0627] Uplink Time-Frequency Domain Resource Allocation Field: Optionally, this field may not exist for downlink DCIs with joint uplink and downlink indications, and the uplink resources are implicitly determined through downlink time-frequency domain resources. In one possible embodiment, for time-domain resources, the time-domain location of the uplink transmission is the same as the time-domain location of the downlink transmission. For frequency-domain resources, for cases where the uplink and downlink BWP sizes and SCS are the same, the relative frequency-domain location within the BWP occupied by the uplink transmission is the same as the relative frequency-domain location within the downlink BWP occupied by the downlink transmission; for cases where the uplink and downlink BWP sizes are different, the relative start number or relative end number of the frequency-domain resources within the BWP occupied by the uplink transmission is the same, or so on.

[0628] Uplink transmission indication field for demodulated symbols: Optionally, this field exists only if the gNB enables the collection of downlink demodulated symbols or enables the uplink transmission of downlink demodulated symbols via RRC signaling.

[0629] MCS: For non-quantized mode, the modulation order is directly determined by the modulation order indicated by the downlink scheduling DCI, and the modulation order in MCS is ignored; for quantized mode, this field is effective.

[0630] In addition, if the uplink DCI is scheduled independently, a certain mapping or correlation relationship needs to be considered between the downlink DCI and the uplink DCI to ensure that the terminal can determine which downlink received demodulation symbol to transmit to the gNB.

[0631] The same search space; or related search spaces; related monitoring occasions within the same search space; related CORESETs; or the uplink DCI carrying information such as the downlink DCI's HPN and timestamp; or both the uplink and downlink DCIs carrying a flag with the same value. For example, one possible implementation is that the uplink transmission indication field for the demodulated symbol carried in the uplink DCI (if it exists) has the same value as the downlink demodulated symbol collection field carried in the downlink DCI (optionally, multiple values ​​can be taken, such as 16, 32, etc.), and correspondingly, the uplink transmission transmits the demodulated symbol from the downlink reception.

[0632] In addition, if the uplink DCI is independently scheduled, a method similar to that used in semi-static transmission can be adopted to determine the downlink reception demodulation symbols (for example, all downlink PDSCH transmission symbols to be demodulated from the N time domain positions before the start of the current scheduled PUSCH transmission time domain position to the PUSCH transmission time domain position +1 of the last uplink scheduled PUSCH transmission time domain position, or the first M demodulation symbols, are transmitted uplink; or at least M demodulation symbols of the most recent PDSCH are transmitted).

[0633] Alternatively, the transmission method of the symbols to be demodulated can be the same as the transmission method of the modulation symbols corresponding to the AI ​​sample acquisition in Case 1 (the case where model training and model inference are performed on the same side).

[0634] Optionally, for downlink scheduling DCI, the following indication field may also exist: a field for collecting downlink demodulated symbols (optionally, this field exists only if the gNB has enabled the downlink demodulated symbol collection function). One possible implementation is as follows: if the gNB indicates that this field is "0", the terminal does not perform downlink demodulated symbol collection; if the gNB indicates that this field is "1", the terminal performs downlink demodulated symbol collection. Alternatively, whether the terminal performs downlink demodulated symbol collection, and / or uplink transmission of uplink demodulated symbols, can also be determined by parameters configured by the gNB through RRC signaling.

[0635] Key Point 2:

[0636] 2-1: While performing downlink transmission, the network side stores the bits before modulation or the modulation symbols after modulation as labels;

[0637] 2-2: Simultaneously, the network-side equipment performs UL reception to obtain the downlink symbols to be demodulated sent by the terminal. It is conceivable that these demodulated symbols can be used as input for model training.

[0638] 2-3: Based on 2-1 / 2, the collection of training data (the receiving end of the symbol to be demodulated, the transmitting end of the modulation symbol / the transmitting end of the (encoded) bit stream or modulation symbol) can be realized, that is, (model input data, label).

[0639] In addition, if the node performing model training and the gNB are not the same entity, the gNB also needs to send (the receiving end symbol to be demodulated, label) to the model training node through the network device interface.

[0640] It is important to note that the receiving end and sending end mentioned in this scheme refer to the original data transmission, which is relative to the downlink.

[0641] In the embodiments disclosed herein, some or all of the steps and their optional implementations may be arbitrarily combined with some or all of the steps in other embodiments, or may be arbitrarily combined with the optional implementations in other embodiments.

[0642] This disclosure also proposes an apparatus (also referred to as a communication device, etc.) for implementing any of the above methods. For example, an apparatus is proposed that includes units or modules for implementing the steps performed by the first device in any of the above methods. Furthermore, another apparatus is proposed that includes units or modules for implementing the steps performed by the second device (e.g., an access network device, a core network functional node, a core network device, etc.) in any of the above methods.

[0643] It should be understood that the division of units or modules in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units or modules in the device can be implemented by a processor calling software: for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units or modules in the above device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuits. The functionality of some or all of the units or modules can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the units or modules is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby achieving the functionality of some or all of the units or modules. All units or modules of the above device can be implemented entirely through processor-called software, entirely through hardware circuits, or partially through processor-called software with the remaining parts implemented through hardware circuits.

[0644] In this embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units or modules. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a Neural Network Processing Unit (NPU), a Tensor Processing Unit (TPU), or a Deep Learning Processing Unit (DPU).

[0645] Figure 5A is a schematic diagram of the structure of a first device according to an embodiment of this disclosure. The first device 5100 is used to perform any of the above methods. In some embodiments, as shown in Figure 5A, the first device 5100 may include at least one of a transceiver module 5101, a processing module 5102, etc.

[0646] In some embodiments, the transceiver module is used to receive second data corresponding to the first data sent by the second device, wherein the first data and the second data form a training data pair, and the training data pair is used to train the first model.

[0647] Optionally, the transceiver module is used to perform at least one of the communication steps such as sending and / or receiving performed by the first device 5100 in any of the above methods (e.g., steps 2103, 2203, 2301, 2401, 2403, 2404, 3101, but not limited thereto), which will not be elaborated here.

[0648] Optionally, the above processing module is used to execute at least one of the other communication steps executed by the first device 5100 in any of the above methods (e.g., steps 2101, 2104, 2105, 2201, 2204, 2205, 2302, 2303, 2402, but not limited thereto), which will not be elaborated here.

[0649] Figure 5B is a schematic diagram of the structure of the second device proposed in an embodiment of this disclosure. The second device 5200 is used to perform any of the above methods. In some embodiments, as shown in Figure 5B, the second device 5200 may include at least one of a transceiver module 5201, a processing module 5202, etc.

[0650] In some embodiments, the transceiver module is used to send first data, wherein the first data corresponds to the second data received by the first device, and the first data and the second data form a training data pair, which is used to train the first model.

[0651] Optionally, the transceiver module is used to perform at least one of the communication steps such as sending and / or receiving performed by the second device 5200 in any of the above methods (e.g., steps 2103, 2203, 2301, 2401, 2403, 2404, 3101, but not limited thereto), which will not be elaborated here.

[0652] Optionally, the above processing module is used to execute at least one of the other communication steps (e.g., steps 2101, 2102, 2201, 2202, but not limited thereto) executed by the second device 5200 in any of the above methods, which will not be elaborated here.

[0653] In some embodiments, the transceiver module may include a transmitting module and / or a receiving module, which may be separate or integrated. Optionally, the transceiver module may be interchangeable with a transceiver.

[0654] In some embodiments, the processing module may be a single module or may include multiple sub-modules. Optionally, the multiple sub-modules may each perform all or part of the steps required by the processing module.

[0655] In some embodiments, the processing module can be replaced by the processor, and the transceiver module can be replaced by the transceiver.

[0656] Figure 6A is a schematic diagram of the structure of the communication device 6100 proposed in an embodiment of this disclosure. The communication device 6100 can be a network device (e.g., access network device, core network device, etc.), a terminal (e.g., user equipment, etc.), a chip, chip system, or processor that supports the network device in implementing any of the above methods, or a chip, chip system, or processor that supports the terminal in implementing any of the above methods. The communication device 6100 can be used to implement the methods described in the above method embodiments; for details, please refer to the descriptions in the above method embodiments.

[0657] As shown in Figure 6A, the communication device 6100 is used to execute any of the above methods. In some embodiments, the communication device 6100 includes one or more processors 6101. The processor 6101 may be a general-purpose processor or a special-purpose processor, such as a baseband processor or a central processing unit. The baseband processor may be used to process communication protocols and communication data, and the central processing unit may be used to control communication devices (e.g., base stations, baseband chips, terminal devices, terminal device chips, DUs or CUs, etc.), execute programs, and process program data. Optionally, the communication device 6100 is used to execute any of the above methods. Optionally, one or more processors 6101 are used to invoke instructions to cause the communication device 6100 to execute any of the above methods.

[0658] In some embodiments, the communication device 6100 further includes one or more transceivers 6102. When the communication device 6100 includes one or more transceivers 6102, the transceiver 6102 performs at least one of the communication steps such as sending and / or receiving in the above-described method (e.g., steps 2103, 2203, 2301, 2401, 2403, 2404, 3101, but not limited thereto), and the processor 6101 performs at least one of other steps (e.g., steps 2101, 2102, 2104, 2105, 2201, 2202, 2204, 2205, 2302, 2303, 2402, but not limited thereto). In optional embodiments, the transceiver may include a receiver and / or a transmitter, which may be separate or integrated together. Optionally, terms such as transceiver, transceiver unit, transceiver, transceiver circuit, interface circuit, and interface can be used interchangeably; terms such as transmitter, transmitting unit, transmitter, and transmitting circuit can be used interchangeably; and terms such as receiver, receiving unit, receiver, and receiving circuit can be used interchangeably.

[0659] In some embodiments, the communication device 6100 further includes one or more memories 6103 for storing data and / or instructions. Optionally, one or more processors 6101 are used to invoke instructions stored in the memory 6103 to cause the communication device 6100 to perform any of the above methods. Optionally, all or part of the memory 6103 may also be located outside the communication device 6100. In an optional embodiment, the communication device 6100 may include one or more interface circuits 6104. Optionally, the interface circuit 6104 is connected to the memory 6102 and can be used to receive data and / or instructions from the memory 6102 or other devices, and can be used to send data and / or instructions to the memory 6102 or other devices. For example, the interface circuit 6104 can read data and / or instructions stored in the memory 6102 and send the data and / or instructions to the processor 6101.

[0660] The communication device 6100 described in the above embodiments may be a network device or a terminal, but the scope of the communication device 6100 described in this disclosure is not limited thereto, and the structure of the communication device 6100 may not be limited by FIG. 6A. The communication device may be a standalone device or a part of a larger device. For example, the communication device may be: (1) a standalone integrated circuit IC, or chip, or chip system or subsystem; (2) a collection of one or more ICs, optionally, the IC collection may also include storage components for storing data, programs and / or instructions; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, terminal device, smart terminal device, cellular phone, wireless device, handheld device, mobile unit, vehicle device, network device, cloud device, artificial intelligence device, etc.; (6) others, etc.

[0661] Figure 6B is a schematic diagram of the structure of chip 6200 according to an embodiment of this disclosure. For cases where the communication device 6100 can be a chip or a chip system, please refer to the schematic diagram of chip 6200 shown in Figure 6B, but it is not limited thereto.

[0662] Chip 6200 includes one or more processors 6201. Chip 6200 is used to perform any of the methods described above.

[0663] In some embodiments, chip 6200 further includes one or more interface circuits 6202. Optionally, terms such as interface circuit, interface, and transceiver pin can be used interchangeably. In some embodiments, chip 6200 further includes one or more memories 6203 for storing data and / or instructions. Optionally, all or part of the memories 6203 may be located outside of chip 6200. Optionally, interface circuit 6202 is connected to memory 6203, and interface circuit 6202 can be used to receive data and / or instructions from memory 6203 or other devices, and interface circuit 6202 can be used to send data and / or instructions to memory 6203 or other devices. For example, interface circuit 6202 can read data and / or instructions stored in memory 6203 and send the data and / or instructions to processor 6201.

[0664] In some embodiments, the interface circuit 6202 performs at least one of the communication steps such as sending and / or receiving in the above-described method (e.g., steps 2103, 2203, 2301, 2401, 2403, 2404, 3101, but not limited thereto). The interface circuit 6202 performing the communication steps such as sending and / or receiving in the above-described method refers, for example, to the interface circuit 6202 performing data and / or instruction interaction between the processor 6201, the chip 6200, the memory 6203, or the transceiver device. In some embodiments, the processor 6201 performs at least one of other steps (e.g., steps 2101, 2102, 2104, 2105, 2201, 2202, 2204, 2205, 2302, 2303, 2402, but not limited thereto).

[0665] The modules and / or devices described in the various embodiments, such as virtual devices, physical devices, and chips, can be combined or separated arbitrarily as needed. Optionally, some or all steps can also be performed collaboratively by multiple modules and / or devices, which is not limited here.

[0666] This disclosure also proposes a storage medium storing instructions that, when executed on a communication device, cause the communication device to perform any of the above methods. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but not limited thereto; it may also be a storage medium readable by other devices. Optionally, the storage medium may be a non-transitory storage medium, but not limited thereto; it may also be a temporary storage medium.

[0667] This disclosure also proposes a program product, including a program and / or instructions, which, when executed by a communication device, cause the communication device to perform any of the above methods. Optionally, the program product is a computer program product. Optionally, the program product is stored on the storage medium.

[0668] This disclosure also proposes a computer program that, when run on a computer, causes the computer to perform any of the above methods.

Claims

1. A communication method characterized by comprising: The method is performed by a first device, and the method includes: Receive second data corresponding to first data sent by a second device, wherein the first data and the second data form a training data pair, and the training data pair is used to train a first model.

2. The method of claim 1, wherein, The method further includes: Based on the protocol agreement or network device instruction, first information and second information are determined. The first information is used for the second device to send the first data and for the first device to receive the second data. The second information is used for the second device to generate the first data and for the first device to determine the first data.

3. The method of claim 2, wherein, The first data satisfies at least one of the following: The first data is generated based on a first symbol, which is agreed upon by both the network device and the terminal, or the first symbol is indicated by the network device. The first symbol is a symbol that has not undergone channel coding. The first data is generated by scrambling and modulating the first bit stream. The first bit stream is agreed upon by both the network device and the terminal, or the first bit stream is indicated by the network device. The first bit stream is a bit stream that has not undergone channel coding. The first data is generated by channel coding of the second bit stream, rate matching and modulation of the encoded bit stream, layer mapping, precoding, and resource mapping. The second bit stream is agreed upon by both the network device and the terminal, or the second bit stream is indicated by the network device.

4. The method according to claim 2 or 3, characterized in that, The first information and the second information include at least one of the following: Time-domain resource allocation parameters; Frequency domain resource allocation parameters; Frequency domain resource allocation method; Modulation method; Modulation order; Precoding matrix; Antenna port; Number of floors; Information related to the modulation symbol; Waveform; Physical Resource Block (PRB) binding size; Resource mapping method; The first bit stream corresponding to the first data; Encoding method; Bitrate; Redundant version; Modulation sequence.

5. The method according to any one of claims 2 to 4, characterized in that, The determination of the first information and the second information based on protocol agreement or network device indication includes at least one of the following: Receive at least one of the first information and the second information indicated by the second device through the first signaling, wherein the first signaling includes any one of dynamic signaling and semi-static signaling; The system receives a first list sent by the second device, the first list including at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and receives dynamic signaling sent by the second device, the dynamic signaling being used to indicate at least one of all or part of the index of the first information and the index of the second information, and all or part of the value of the first information and the value of the second information. The protocol pre-defined first list is determined, the first list including at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and dynamic signaling is received from the second device. The signaling is used to indicate at least one of all or part of the indices of the first information and the second information, and all or part of the values ​​of the first information and the second information; Based on the agreement, at least one of the first information and the second information is determined; The device receives third data sent by the second device. The third data includes a first field, which is used to indicate at least one of the first information and the second information, or the first field is used to indicate whether the parameters for sending the third data can be reused as the parameters for sending the first data. The third data is communication data that is not used for model training.

6. The method according to any one of claims 2 to 5, characterized in that, The first device is a terminal, the second device is a network device, and the determination of the first information and the second information based on protocol agreement or network device indication includes: The first information and the second information are determined based on the agreement, or the first information and the second information are received from the network device.

7. The method according to any one of claims 2 to 5, characterized in that, The first device is a network device, and the second device is a terminal. The determination of the first and second information based on protocol agreements or network device instructions includes: The first information and the second information are determined based on the agreement, or the first information and the second information are sent to the terminal.

8. The method according to any one of claims 2 to 7, characterized in that, The receipt of the second data includes: The second data is received in either a first method or a second method, wherein the first method involves the second device multiplexing the first data and the third data, and the second method involves the second device not multiplexing the first data and the third data, wherein the third data is communication data not used for model training.

9. The method of claim 8, wherein, Receiving the second data in the second manner includes: Based on the first information, the first resource and the second resource are determined; The second data is received on the first resource, and the fourth data is received on the second resource, wherein the fourth data is data corresponding to the third data sent by the second device.

10. The method of claim 9, wherein, The determination of the first resource and the second resource includes: Receive a second signaling message, or receive first control information or a first control unit, wherein the second signaling message includes a first parameter, and the first control information or the first control unit includes a first field; Based on the first parameter or the first field, determine the first information, and determine the first resource and the second resource according to the first information; Wherein, the first parameter or the first field is used to indicate at least one of the following: Whether to schedule the first resource or the second resource; Schedule the resource location of the first resource; The number of resources to schedule the first resource; Schedule the resource location of the second resource; The number of resources to be scheduled for the second resource.

11. The method of claim 8, wherein, Receiving the second data in the first manner includes: Determine the first quantity of the first data sent by the second device and the first location where the first data was sent; Determine a second quantity of the third data sent by the second device and a second location where the third data is sent; Based on the first quantity, the first location, the second quantity, and the second location, the second data is received, and the second device transmits the first data and the third data in a multiplexed manner.

12. The method of claim 11, wherein, Determining the first quantity of the first data sent by the second device includes any one of the following: Based on the protocol agreement, the first quantity is determined to be 1 / M, where M is the total number of available symbols, and M is determined by the protocol or indicated by the network device. Based on the protocol agreement or network device indication, the first quantity is determined to be Q, where Q is an integer multiple of all data corresponding to the current modulation order; The first quantity is determined based on at least one of the number of bits corresponding to the first data and the modulation order.

13. The method according to claim 11 or 12, characterized in that, The method further includes: The second data is demultiplexed to obtain the bit stream corresponding to the first data and the bit stream corresponding to the third data. Before sending the bit stream corresponding to the first data and the bit stream corresponding to the third data, the second device performs a first processing on the bit stream corresponding to the first data and the bit stream corresponding to the third data, respectively.

14. The method of claim 11 or 12, wherein, The method further includes: Perform de-mapping on the second data to obtain the de-mapping bitstream; The demultiplexed bitstream is demultiplexed to obtain the bitstream corresponding to the first data and the bitstream corresponding to the third data.

15. The method of claim 1, wherein, The method further includes: The second data is processed in a second way to obtain the fifth data, which is data that has not undergone demodulation and bitstream processing, or the fifth data is data obtained by quantizing the sixth data obtained after the second processing in a way agreed upon by the protocol, which is data that has not undergone demodulation. The fifth data is sent to the second device, and the first data and the fifth data are used to train the first model.

16. The method according to any one of claims 1 to 15, characterized in that, The method further includes: The system receives a trained first model sent by the second device, wherein the trained first model is trained by the second device, or the trained first model is provided by the second device to the third device and obtained from the third device.

17. A method of communication, comprising: The method is performed by a second device, and the method includes: Send first data, wherein the first data corresponds to the second data received by the first device, and the first data and the second data form a training data pair, the training data pair being used to train a first model.

18. The method of claim 17, wherein, The method further includes: Based on the protocol agreement or network device instruction, first information and second information are determined. The first information is used for the second device to send the first data and for the first device to receive the second data. The second information is used for the second device to generate the first data and for the first device to determine the first data.

19. The method of claim 18, wherein, The first data satisfies at least one of the following: The first data is generated based on a first symbol, which is agreed upon by both the network device and the terminal, or the first symbol is indicated by the network device. The first symbol is a symbol that has not undergone channel coding. The first data is generated by scrambling and modulating the first bit stream. The first bit stream is agreed upon by both the network device and the terminal, or the first bit stream is indicated by the network device. The first bit stream is a bit stream that has not undergone channel coding. The first data is generated by channel coding of the second bit stream, rate matching and modulation of the encoded bit stream, layer mapping, precoding, and resource mapping. The second bit stream is agreed upon by both the network device and the terminal, or the second bit stream is indicated by the network device.

20. The method of claim 18 or 19, wherein, The first information and the second information include at least one of the following: Time-domain resource allocation parameters; Frequency domain resource allocation parameters; Frequency domain resource allocation method; Modulation method; Modulation order; Precoding matrix; Antenna port; Number of floors; Information related to the modulation symbol; Waveform; Physical Resource Block (PRB) binding size; Resource mapping method; The first bit stream corresponding to the first data; Encoding method; Bitrate; Redundant version; Modulation sequence.

21. The method of any one of claims 18-20, wherein, The determination of the first information and the second information based on protocol agreement or network device indication includes at least one of the following: Send a first signaling to the first device, wherein at least one of the first information and the second information is included in the first signaling, and the first signaling is either dynamic signaling or semi-static signaling; Send a first list to the first device, the first list including at least one of the index of the first information, the value of the first information, the index of the second information, and the value of the second information; and send dynamic signaling to the first device, the dynamic signaling being used to indicate at least one of all or part of the index of the first information and the index of the second information, and all or part of the value of the first information and the value of the second information. A first list preset by the protocol is determined, the first list including the index of the first information, the value of the first information, the index of the second information, the value of the second information, and, by dynamic signaling, indicating to the first device at least one of all or part of the index of the first information and the index of the second information, and all or part of the value of the first information and the value of the second information; Based on the agreement, at least one of the first information and the second information is determined; Send third data to the first device. The third data includes a first field, which is used to indicate at least one of the first information and the second information, or the first field is used to indicate whether the parameters for sending the third data can be reused as the parameters for sending the first data. The third data is communication data that is not used for model training.

22. The method according to any one of claims 18 to 21, characterized in that, The first device is a terminal, the second device is a network device, and the determination of the first information and the second information includes: The first information and the second information are determined based on the agreement, or the first information and the second information are sent to the terminal; Alternatively, if the first device is a network device and the second device is a terminal, determining the first information and the second information includes: The first information and the second information are determined based on the agreement, or the first information and the second information are received from the network device.

23. The method of any one of claims 18-22, wherein, The sending of the first data includes: The first data is sent in either a first method or a second method, wherein the first method involves the second device multiplexing the sending of the first data and the sending of the third data, and the second method involves the second device not multiplexing the sending of the first data and the sending of the third data, wherein the third data is communication data not used for model training.

24. The method of claim 23, wherein, Sending the first data in the second manner includes: Based on the first information, the first resource and the second resource are determined; The first data is sent on the first resource, and the third data is sent on the second resource.

25. The method of claim 24, wherein, The method further includes: Send a second signaling to the first device, or send first control information or a first control unit, wherein the second signaling includes a first parameter, the first control information or the first control unit includes a first field, the first parameter or the first field is used to indicate the first information, and the first information is used to determine the first resource and the second resource; Wherein, the first parameter or the first field is used to indicate at least one of the following: Whether to schedule the first resource or the second resource; Schedule the resource location of the first resource; The number of resources to schedule the first resource; Schedule the resource location of the second resource; The number of resources to be scheduled for the second resource.

26. The method of claim 23, wherein, Sending the first data in the first manner includes: Determine the first quantity of the first data sent by the second device and the first location where the first data was sent; Determine a second quantity of the third data sent by the second device and a second location where the third data is sent; Based on the first quantity, the first position, the second quantity, and the second position, the first data and the third data are sent in a reused manner.

27. The method of claim 26, wherein, Determining the first quantity of the first data sent by the second device includes at least one of the following: Based on the protocol agreement, the first quantity is determined to be 1 / M, where M is the total number of available symbols, and M is determined by the protocol or indicated by the network device. Based on the protocol agreement, the first quantity is determined to be Q, where Q is an integer multiple of all data corresponding to the current modulation order. The first quantity is determined based on at least one of the number of bits corresponding to the first data and the modulation order.

28. The method of claim 26 or 27, wherein, The method further includes: The bit stream corresponding to the first data and the bit stream corresponding to the third data are respectively subjected to a first processing, and the bit stream corresponding to the first data and the bit stream corresponding to the third data after the first processing are multiplexed; or, the bit stream corresponding to the first data and the bit stream corresponding to the third data are multiplexed and layer mapped.

29. The method of claim 17, wherein, The method further includes: The system receives fifth data sent by the first device, which is obtained by the first device performing a second processing on the second data. The fifth data is used to train the first model and is data that has not undergone demodulation and bit-level processing. Alternatively, the fifth data is data obtained by quantizing the sixth data according to the protocol agreement. The sixth data is data obtained by the first device performing a second processing on the second data and is data that has not undergone demodulation.

30. The method of any one of claims 17-29, wherein, The method further includes: The first model is trained using training data consisting of the first data and the fifth data, or the training data consisting of the first data and the fifth data is sent to a third device, and the trained first model obtained by the third device based on the training data is received. The trained first model is sent to the first device.

31. A communications device, characterized by The communication device is used to perform the communication method according to any one of claims 1 to 16, 17 to 30.

32. A communication system, characterized by include: A first device and a second device, wherein the first device is configured to implement the method of any one of claims 1 to 16, and the second device is configured to implement the method of any one of claims 17 to 30.

33. A storage medium, the storage medium storing instructions, wherein, When the instructions are executed on a communication device, the communication device performs the method as described in any one of claims 1 to 16, 17 to 30.

34. A program product comprising at least one of a program, instructions, characterized in that When at least one of the programs or instructions is executed by a communication device, it implements the steps of the method described in claims 1 to 16, 17 to 30.

Images