A communication method and apparatus

Abstract

A communication method and device, in which a first device transmits first information and second information. The first information is related to a first modulation order and a first spatial dimension, which are used by the first device to configure a first neural network used by the first device to process data to be transmitted to a second device. The second information is related to a first spatial dimension mapping manner used by the first device to map an output result of the first neural network. The second device configures a second neural network based on the first information, and determines the first spatial dimension mapping manner based on the second information, the second neural network being used by the second device to process data from the first device, and the first spatial dimension mapping manner being used by the second device to determine data input to the second neural network. The method enables the second device to configure the second neural network adapted to the first neural network, and also facilitates correct acquisition of the data input to the second neural network.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] End-to-end communication is flexible and scalable, and can cover a variety of new physical layer functions, including non-orthogonal multiple access, different antenna beamforming architectures, and smart metasurface technology.

[0003] In end-to-end communication, the transmitter can use a neural network to process the data to be transmitted, and the receiver can use a neural network to process the received data. However, how to configure the receiver's neural network to be compatible with the transmitter's neural network in order for the receiver to correctly demodulate the data from the transmitter requires further research. Summary of the Invention

[0004] This application provides a communication method and apparatus that enables the receiver to be configured with a receiving neural network adapted to the transmitting neural network.

[0005] In a first aspect, embodiments of this application provide a communication method. The method includes: a first device determining first information and second information; the first device sending the first information and the second information to a second device. The first information is related to a first modulation order and a first spatial dimension, which are used by the first device to configure a first neural network. The first neural network is used by the first device to process data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the first device to map the output of the first neural network.

[0006] In an optional implementation, the method further includes: a first device transmitting third information, the third information including one or more indices and a modulation order and spatial dimension associated with each index. The first information includes a first index, which is an index among one or more indices associated with a first modulation order and a first spatial dimension.

[0007] In an optional implementation, the method further includes: a first device transmitting fourth information, the fourth information including one or more indices and a modulation order associated with each index. The first information includes: a second index and a first spatial dimension, the second index being an index among the one or more indices associated with the first modulation order.

[0008] In one alternative implementation, the first information includes: a first modulation order and a first spatial dimension.

[0009] In one optional implementation, the method further includes: a first device sending fifth information, the fifth information including one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The second information includes a first indication, which is the indication in the fifth information associated with a first spatial dimension mapping method.

[0010] In one alternative implementation, multiple spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

[0011] In one alternative implementation, the second information is used to indicate the first spatial dimension mapping method.

[0012] Secondly, embodiments of this application provide a communication method. The method includes: a second device receiving first information and second information from a first device, wherein the first information is related to a first modulation order and a first spatial dimension, and the second information is related to a first spatial dimension mapping method. Based on the first information, the second device configures a second neural network, which is used by the second device to process data from the first device. Based on the second information, the second device determines the first spatial dimension mapping method, which is used by the second device to determine the data input to the second neural network.

[0013] In one optional implementation, the method further includes: a second device receiving third information, the third information including one or more indices and a modulation order and spatial dimension associated with each index. The first information includes a first index. The second device configures a second neural network based on the first information, including: configuring input information of the second neural network based on a first spatial dimension associated with the first index in the third information; and configuring output information of the second neural network based on a first modulation order associated with the first index in the third information.

[0014] In an optional implementation, the method further includes: a second device receiving fourth information, the fourth information including one or more indices and a modulation order associated with each index. The first information includes: a second index and a first spatial dimension. The second device configures a second neural network based on the first information, including: configuring input information of the second neural network based on the first spatial dimension in the first information; and configuring output information of the second neural network based on the first modulation order associated with the second index in the fourth information.

[0015] In one optional implementation, the first information includes: a first modulation order and a first spatial dimension. The second device configures a second neural network based on the first information, including: configuring the input information of the second neural network based on the first spatial dimension in the first information; and configuring the output information of the second neural network based on the first modulation order in the first information.

[0016] In one optional implementation, the method further includes: a second device receiving fifth information, the fifth information including one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The second information includes a first indication. The second device determines a first spatial dimension mapping method based on the second information, including: the second device determining the first spatial dimension mapping method associated with the first indication from the fifth information.

[0017] In one alternative implementation, multiple spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

[0018] In one alternative implementation, the second information is used to indicate the first spatial dimension mapping method.

[0019] Thirdly, embodiments of this application provide a communication device, the communication device comprising:

[0020] The processing unit is used to determine the first information and the second information.

[0021] The communication unit is used to send first information and second information to the second device.

[0022] The first information is related to the first modulation order and the first spatial dimension. The first modulation order and the first spatial dimension are used by the communication device to configure the first neural network, and the first neural network is used by the communication device to process the data to be sent to the second device. The second information is related to the mapping method of the first spatial dimension, and the mapping method of the first spatial dimension is used by the communication device to map the output of the first neural network.

[0023] In addition, other alternative implementations of the communication device in this regard can be found in the relevant content of the first aspect above, and will not be described in detail here.

[0024] Fourthly, embodiments of this application provide a communication device, the communication device comprising:

[0025] The communication unit is used to receive first information and second information from the first device. The first information is related to the first modulation order and the first spatial dimension, and the second information is related to the mapping method of the first spatial dimension.

[0026] The processing unit is configured to configure a second neural network based on the first information, the second neural network being used by the communication device to process data from the first device.

[0027] The processing unit is also configured to determine a first spatial dimension mapping method based on the second information, the first spatial dimension mapping method being used by the communication device to determine the data input to the second neural network.

[0028] In addition, other alternative implementations of the communication device in this regard can be found in the relevant content of the second aspect above, and will not be described in detail here.

[0029] Fifthly, embodiments of this application provide a communication device, which includes a memory and a processor; optionally, the communication device further includes a communication interface.

[0030] Memory, used to store computer programs;

[0031] A communication interface used to receive or send data;

[0032] A processor is used to call program instructions stored in memory.

[0033] In one alternative implementation, the processor invokes a computer program to perform the following operations:

[0034] The system determines first information and second information; it then sends the first information and second information to the second device. The first information is related to a first modulation order and a first spatial dimension. These two parameters are used by the communication device to configure a first neural network, which in turn is used by the communication device to process data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the communication device to map the output of the first neural network.

[0035] In addition, other optional implementations of the communication device in this manner can be found in the relevant content of the first aspect above, and will not be described in detail here.

[0036] In another alternative implementation, the processor invokes a computer program to perform the following operations:

[0037] The system receives first information and second information from a first device. The first information is related to a first modulation order and a first spatial dimension, and the second information is related to a mapping method for the first spatial dimension. Based on the first information, a second neural network is configured. This second neural network is used by the communication device to process data from the first device. Based on the second information, the first spatial dimension mapping method is determined. This first spatial dimension mapping method is used by the communication device to determine the data input to the second neural network.

[0038] In addition, other optional implementations of the communication device in this manner can be found in the relevant content of the second aspect above, and will not be described in detail here.

[0039] Sixthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being used to receive or send data.

[0040] In one alternative implementation, the processor is configured to cause the chip to perform:

[0041] The system determines first information and second information; it then sends the first information and second information to the second device. The first information is related to a first modulation order and a first spatial dimension. These two parameters are used by the chip to configure a first neural network, which in turn is used by the chip to process the data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the chip to map the output of the first neural network.

[0042] In addition, other optional implementation methods of the chip in this approach can be found in the relevant content of the first aspect above, and will not be described in detail here.

[0043] In another alternative implementation, the processor is configured to cause the chip to perform:

[0044] The chip receives first information and second information from a first device. The first information is related to a first modulation order and a first spatial dimension, and the second information is related to a mapping method for the first spatial dimension. Based on the first information, a second neural network is configured. The second neural network is used by the chip to process data from the first device. Based on the second information, the first spatial dimension mapping method is determined. The first spatial dimension mapping method is used by the chip to determine the data to be input to the second neural network.

[0045] In addition, other optional implementation methods of the chip in this approach can be found in the relevant content of the second aspect above, and will not be described in detail here.

[0046] In a seventh aspect, embodiments of this application provide a module device, the module device including a communication module, a power module, a storage module, and a chip, wherein:

[0047] The power module is used to provide electrical energy to the module device;

[0048] The storage module is used to store data and instructions;

[0049] The communication module is used for internal communication within the module device, or for communication between the module device and external devices.

[0050] The chip is used to perform the method described in the first or second aspect above.

[0051] Eighthly, embodiments of this application provide a computer-readable storage medium for storing computer software instructions used by the aforementioned terminal, including a program for executing the methods described in the first or second aspect.

[0052] Ninthly, embodiments of this application also provide a computer program product, which, when run on a processor, enables the implementation of the method flow described in the first or second aspect above.

[0053] In this embodiment, a first device determines first information and second information, and sends the first information and second information to a second device. The first information is related to a first modulation order and a first spatial dimension, which are used by the first device to configure a first neural network. The first neural network is used by the first device to process data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the first device to map the output of the first neural network. The second device configures a second neural network based on the first information, and the second neural network is used by the second device to process data from the first device. The second device determines the first spatial dimension mapping method based on the second information, and the first spatial dimension mapping method is used by the second device to determine the data input to the second neural network.

[0054] As can be seen, the first device informs the second device of the modulation order and spatial dimension used to configure the first neural network (i.e., the transmitting neural network), so that the second device can configure a second neural network (i.e., the receiving neural network) that is compatible with the first neural network based on the obtained modulation order and spatial dimension. Furthermore, the first device also informs the second device of the spatial dimension mapping method used for the output of the first neural network, so that the second device can correctly extract data from the signal sent by the first device based on the obtained spatial dimension mapping method, and then input the correctly extracted data into the configured second neural network for processing. Attached Figure Description

[0055] Figure 1 A schematic diagram of a communication system provided in an embodiment of this application;

[0056] Figure 2 A schematic diagram illustrating a neural network application scenario provided in an embodiment of this application;

[0057] Figure 3 A schematic diagram illustrating an end-to-end modulation implementation as provided in an embodiment of this application;

[0058] Figure 4 A flowchart illustrating a communication method provided in an embodiment of this application;

[0059] Figure 5 A schematic diagram illustrating another communication method provided in an embodiment of this application;

[0060] Figure 6 A schematic diagram illustrating another communication method provided in an embodiment of this application;

[0061] Figure 7 A schematic diagram illustrating another communication method provided in an embodiment of this application;

[0062] Figure 8 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0063] Figure 9 This is a schematic diagram of another communication device provided in an embodiment of this application;

[0064] Figure 10 This is a schematic diagram of the structure of a module device provided in an embodiment of this application. Detailed Implementation

[0065] The embodiments of this application will now be described in conjunction with the accompanying drawings.

[0066] In this application, the terms "first" and "second," etc., used in the specification, claims, and drawings are used to distinguish different objects, not to describe a specific order. "First" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include one or more of that feature. In the description of this embodiment, unless otherwise stated, "a plurality of" means two or more.

[0067] Furthermore, in the description of this embodiment, the use of terms such as "first" and "second" does not limit the naming convention. For example, "first information" can be named using other names, and "second information" can also be named using other names.

[0068] Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0069] It should be understood that in this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "...when" and "if" both indicate that a corresponding action will be taken under certain objective circumstances, not a time limit, nor do they require a judgment action during implementation, nor do they imply any other limitations.

[0070] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0071] To better understand the embodiments of this application, the system architecture involved in the embodiments of this application will first be introduced:

[0072] This application is applicable to 5th generation mobile communication (5G) systems, as well as 4th generation mobile communication (4G) systems, 3rd generation mobile communication (3G) systems, and various new communication systems in the future, such as 6th generation (6G) mobile communication, 7th generation (7G) mobile communication, etc., but the embodiments of this application are not limited thereto.

[0073] Figure 1 This is a schematic diagram of a communication system provided in an embodiment of this application. The solution in this application can be applied to this communication system, which may include, but is not limited to, a network device and a terminal device. Figure 1 The number and form of devices shown are for illustrative purposes and do not constitute a limitation on the embodiments of this application. In actual applications, more than one network device and more than one terminal device may be included. Figure 1 The communication system illustrated uses one network device and one terminal device as an example. In this embodiment, the communication system may also include other devices that communicate with the network device or the terminal device; this embodiment does not limit the scope of the application.

[0074] In this application embodiment, the terminal device can be a mobile phone, tablet personal computer, laptop computer or also called a notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, vehicle user equipment (VUE), pedestrian user equipment (PUE), smart home (home devices with wireless communication functions, such as refrigerators, televisions, washing machines or furniture, etc.), game console, personal computer (PC), ATM or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. It should be noted that the specific type of terminal is not limited in the embodiments of this application.

[0075] Network devices can be devices with wireless transceiver capabilities or chips that can be configured in such devices. A network device can be a base station (BS), also known as a base station device, which is a device deployed in a wireless access network (RAN) to provide wireless communication capabilities. For example, in 2G networks, devices providing base station functionality include base transceiver stations (BTS). In 3G networks, devices providing base station functionality include NodeBs (NodeBs). In 4G networks, devices providing base station functionality include evolved NodeBs (eNBs). In wireless local area networks (WLANs), devices providing base station functionality are access points (APs). In 5G new radio (NR), the base station includes a gNB (ggle node) providing base station functionality and an evolved Node B (ng-eNB). The gNB communicates with terminal devices using NR technology, while the ng-eNB communicates with terminals using evolved universal terrestrial radio access (E-UTRA) technology. Both the gNB and ng-eNB can connect to the 5G core network. The base station in this embodiment also includes equipment providing base station functionality in future new communication systems, such as base stations in 6G mobile communication systems.

[0076] Artificial intelligence (AI) is a broad field of science encompassing various areas such as machine learning (ML) and computer vision. Machine learning, a crucial branch of AI, primarily studies how to enable computers to learn independently. Machine learning algorithms include neural networks (NN), decision trees (DT), support vector machines (SVM), and genetic algorithms (GA), among others. Neural networks consist of numerous nodes called neurons. The constituent information of a neuron includes: input (a... l a2, ..., a KThe weights / multiplicative coefficients (w), the bias / additive coefficients (b), and the activation function (δ(.)) are all used. Common activation functions include the sigmoid function, the hyperbolic tangent function (tanh function), the rectified linear unit (ReLU), and the modified linear unit.

[0077] like Figure 2 As shown, neural networks can be applied to end-to-end communication scenarios. Deep learning techniques can be used to jointly optimize transceiver modules with the goal of minimizing message transmission error probability. End-to-end communication can break down the modular barriers of traditional communication systems, enabling joint optimization of the transmitter and receiver to achieve globally optimal bit error rate performance in the communication link.

[0078] End-to-end communication offers flexible scalability, encompassing various new physical layer functionalities, including non-orthogonal multiple access, different antenna beamforming architectures, and smart metasurface technologies. For example, through joint optimization of source coding / decoding and the end-to-end physical layer, semantic communication can be achieved with the goal of minimizing semantic error rates. By using neural networks to learn the semantic features of specific transmitted content and matching them with channel characteristics, higher transmission rates can be achieved with fewer communication resources. Joint transmitter / receiver design in end-to-end communication is a guarantee of globally optimal performance.

[0079] Combination Figure 3 , Figure 3 This is a schematic diagram of end-to-end modulation. The transmitter performs trainable constellation processing, normalization processing, and orthogonal frequency division multiplexing (OFDM) modulation on the data to be transmitted to obtain the signal, which is then transmitted. This signal is transmitted to the receiver via the channel. The receiver performs OFDM demodulation and neural demapping on the received signal.

[0080] The process of the transmitter processing the data to be transmitted can be implemented using a neural network, and the process of the receiver processing the received data can also be implemented using a neural network. For the receiver to correctly demodulate the data from the transmitter, a receiving-end neural network adapted to the transmitting-end neural network needs to be configured. This application provides a communication method that enables the receiver to configure an adapted receiving-end neural network; furthermore, this method also facilitates the receiver in correctly acquiring the data input to the receiving-end neural network.

[0081] The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0082] Please see Figure 4 , Figure 4 This is a flowchart illustrating a communication method provided in an embodiment of this application. The method is illustrated using a first device and a second device as the executing entities. For example, the first device may be a network device, and the second device may be a terminal device. Alternatively, the first device may be a terminal device, and the second device may be a network device. Or, the first device and the second device may be different network devices. Or, the first device and the second device may be different terminal devices. However, this application does not limit the executing entity of the method. For example, the device in the method can also be a chip or software that supports the implementation of the corresponding method. The communication method includes the following steps.

[0083] S101, The first device determines the first information and the second information.

[0084] The first information is related to the first modulation order and the first spatial dimension. The first modulation order and the first spatial dimension are used by the first device to configure the first neural network, and the first neural network is used by the first device to process the data to be sent to the second device. The second information is related to the mapping method of the first spatial dimension, and the mapping method of the first spatial dimension is used by the first device to map the output of the first neural network.

[0085] Optionally, the first modulation order is used by the first device to configure the input information of the first neural network, and the first spatial dimension is used by the first device to configure the output information of the first neural network. That is, the first device can configure the input information of the first neural network based on the first modulation order and configure the output information of the first neural network based on the first spatial dimension. The input information of the first neural network may include, for example, the input dimension of the first neural network, and the output information of the first neural network may include, for example, the output dimension of the first neural network.

[0086] Optionally, the first spatial dimension mapping method can be one of the following: time-domain mapping, frequency-domain mapping, or spatial-domain mapping. Understandably, the first device can map the output of the first neural network onto the time domain, or onto the frequency domain, or onto the spatial domain, thereby transmitting the output of the first neural network to the second device through the generated signal. For example, if the output of the first neural network is a multi-dimensional matrix, the first device can map the multi-dimensional matrix onto the time domain (e.g., a time slot), or onto the frequency domain (e.g., a subcarrier or carrier), or onto the spatial domain (e.g., an antenna).

[0087] S102, the first device sends first information and second information to the second device; correspondingly, the second device receives the first information and second information from the first device.

[0088] S103. The second device configures a second neural network based on the first information. The second neural network is used by the second device to process data from the first device.

[0089] S104. The second device determines the first spatial dimension mapping method based on the second information. The first spatial dimension mapping method is used by the second device to determine the data input to the second neural network.

[0090] Understandably, the second device configures a second neural network based on the first information. It can acquire data from the signal of the first device and input the acquired data into the configured second neural network for processing according to a determined first spatial dimension mapping method. For example, taking the first spatial dimension mapping method as a time-domain mapping, the first device maps the multidimensional matrix generated by the first neural network onto the time domain and sends the resulting signal to the second device; the second device extracts data from the received signal, sorts the data according to the time-domain method, and inputs it into the configured second neural network for processing.

[0091] The first information is described below, as described in the following optional implementation methods 1.1 to 1.3.

[0092] In implementation method 1.1, the first information includes a first index, which is an index among one or more indices associated with a first modulation order and a first spatial dimension. The method further includes: a first device sending third information to a second device, the third information including one or more indices and a modulation order and spatial dimension associated with each index. Correspondingly, the second device receives the third information. Then, the second device can determine that the modulation order associated with the first index in the third information is the first modulation order, and the spatial dimension associated with the first index is the first spatial dimension, and thus configure the second neural network based on the first modulation order and the first spatial dimension.

[0093] Optionally, the third information is modulation scheme (MS) table information, which includes one or more MS indices and the modulation order and spatial dimension associated with each MS index. Therefore, the first index in the first information is the MS index in the MS table information associated with the first modulation order and the first spatial dimension.

[0094] Table 1 is an exemplary MS table. The MS table shown in Table 1 includes the MS index and the modulation order, spatial dimension information, and spectral efficiency associated with each MS index. The spatial dimension information is represented as: 1 / spatial dimension × 2, with the symbol " / " representing division.

[0095] Table 1

[0096]

[0097]

[0098] For example, the first device uses the MS corresponding to MS index 2 in Table 1. Therefore, the first modulation order used to configure the first neural network is 2, and the first spatial dimension is 6. The first device sends MS table information to the second device, which indicates the MS table described in Table 1. The first device also sends the MS index, which is 2. The second device determines from the MS table described in Table 1 that the modulation order associated with MS index 2 is 2 and the associated spatial dimension is 6, and then configures the second neural network according to the first modulation order of 2 and the first spatial dimension of 6.

[0099] In addition to the aforementioned method of the first device sending third information (which includes one or more indices and the modulation order and spatial dimension associated with each index) to the second device, the second device can also obtain the third information in other ways. For example, the third information can also be pre-configured in the second device, without limitation.

[0100] In implementation method 1.2, the first information includes a second index and a first spatial dimension, wherein the second index is an index associated with the first modulation order among one or more indices. The method further includes: a first device sending fourth information, the fourth information including one or more indices and a modulation order associated with each index. Correspondingly, a second device receives the fourth information. Then, the second device can determine that the modulation order associated with the first index in the fourth information is the first modulation order, and subsequently configure the second neural network based on the first modulation order and the first spatial dimension included in the first information.

[0101] Optionally, the fourth information is an MS table information, which includes one or more MS indices and the modulation order associated with each MS index. This MS table information may be the same as or different from the MS table information described in Implementation 1.1. For example, if this MS table information is different from the MS table information described in Implementation 1.1, each MS index in this MS table information may not be associated with a spatial dimension.

[0102] Table 2 is an exemplary MS table, which includes MS indices and the modulation order, target code rate, and spectral efficiency associated with each MS index.

[0103] Table 2

[0104]

[0105]

[0106] For example, the first device uses the MS corresponding to MS index 2 in Table 2, indicating that the first modulation order used to configure the first neural network is 2. Additionally, the first spatial dimension used to configure the first neural network is 6. The first device sends MS table information to the second device, which indicates the MS table described in Table 2. The first device also sends MS index and spatial dimension information to the second device; the MS index is 2, and the spatial dimension information indicates that the first spatial dimension is 6. The second device determines from the MS table described in Table 2 that the modulation order associated with MS index 2 is 2, and then configures the second neural network according to the first modulation order of 2 and the first spatial dimension of 6.

[0107] In addition to the aforementioned method of the first device sending the fourth information (which includes one or more indices and the modulation order associated with each index) to the second device, the second device can also obtain the fourth information in other ways. For example, the fourth information can also be pre-configured in the second device, without limitation.

[0108] In implementation method 1.3, the first information includes: a first modulation order and a first spatial dimension. Therefore, the first device can directly send the first modulation order and the first spatial dimension to the second device. In this case, the second device configures the second neural network based on the first modulation order and the first spatial dimension in the first information.

[0109] In one optional implementation, the second device configures the second neural network based on the first modulation order and the first spatial dimension, including: configuring the input information of the second neural network based on the first spatial dimension, and configuring the output information of the second neural network based on the first modulation order. The input information of the second neural network includes, for example, the input dimension of the second neural network, and the output information of the second neural network includes, for example, the output dimension of the second neural network.

[0110] The second information will now be described in the following optional implementation methods 2.1 and 2.2.

[0111] In implementation method 2.1, the second information includes a first indication, which is an indication associated with the first spatial dimension mapping method in the fifth information. The fifth information includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The method further includes: a first device sending the fifth information. Correspondingly, a second device receiving the fifth information. Then, the second device can determine that the spatial dimension mapping method associated with the first indication in the fifth information is the first spatial dimension mapping method. The fifth information can also be referred to as spatial mapping information, for example. Optionally, the various spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping. For details on frequency domain mapping, time domain mapping, and spatial domain mapping, please refer to the foregoing descriptions, which will not be repeated here.

[0112] For example, the fifth piece of information is represented as: {00: frequency domain mapping; 01: time domain mapping; 10: spatial domain mapping}, that is, "00" is associated with frequency domain mapping, "01" is associated with time domain mapping, and "10" is associated with spatial domain mapping. If the first spatial dimension mapping method is frequency domain mapping, and the first indication sent by the first device is "00", the second device can determine from the fifth piece of information that the first spatial dimension mapping method is frequency domain mapping associated with "00". If the first spatial dimension mapping method is time domain mapping, and the first indication sent by the first device is "01", the second device can determine from the fifth piece of information that the first spatial dimension mapping method is time domain mapping associated with "01". If the first spatial dimension mapping method is spatial domain mapping, and the first indication sent by the first device is "10", the second device can determine from the fifth piece of information that the first spatial dimension mapping method is spatial domain mapping associated with "10".

[0113] In addition to the aforementioned method of the first device sending the fifth information (which includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method) to the second device, the second device may also obtain the fifth information in other ways. For example, the fifth information may be pre-configured in the second device, without limitation.

[0114] In implementation 2.2, the second information is used to indicate the first spatial dimension mapping method. Therefore, the second information can directly indicate the first spatial dimension mapping method, which eliminates the need for the first device to send the fifth information (the fifth information includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method).

[0115] It is understood that the above-described embodiments 1.1 to 1.3 can be arbitrarily combined with embodiments 2.1 and 2.2. That is, the second device can configure the second neural network based on any of embodiments 1.1 to 1.3, and the second device can determine the first spatial dimension mapping method based on any of embodiments 2.1 and 2.2. Exemplary examples are given below, as described in Examples 1, 2, and 3.

[0116] Example 1: Combination Figure 5 The first device sends third and fifth information to the second device. The first device also sends a first index and a first indication to the second device. The third information includes one or more indices and the modulation order and spatial dimension associated with each index. The fifth information includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The first index is the index in the third information associated with the first modulation order and the first spatial dimension. The first indication is the indication in the fifth information associated with the first spatial dimension mapping method.

[0117] The second device configures a second neural network based on the first modulation order and the first spatial dimension associated with the first index in the third information; and determines the mapping method of the first spatial dimension associated with the first indication from the fifth information.

[0118] Example 2: Combination Figure 6 The first device sends fourth and fifth information to the second device. The first device also sends a second index, spatial dimension information, and a first indication to the second device. The fourth information includes one or more indices and a modulation order associated with each index. The fifth information includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The second index is the index in the fourth information associated with the first modulation order. The spatial dimension information indicates the first spatial dimension. The first indication is the indication in the fifth information associated with the first spatial dimension mapping method.

[0119] The second device configures the second neural network based on the first modulation order associated with the second index in the fourth information and the first spatial dimension indicated by the spatial dimension information; and determines the mapping method of the first spatial dimension associated with the first indication from the fifth information.

[0120] Example 3: Combination Figure 7 The first device sends a first modulation order, spatial dimension information, and second information to the second device. The spatial dimension information indicates the first spatial dimension, and the second information indicates the mapping method for the first spatial dimension. Based on the first modulation order and the first spatial dimension indicated by the spatial dimension information, the second device configures a second neural network; and based on the second information, determines the mapping method for the first spatial dimension.

[0121] In summary, in this communication method, the first device determines first information and second information, and sends the first information and second information to the second device. The first information is related to a first modulation order and a first spatial dimension, which are used by the first device to configure a first neural network. The first neural network is used by the first device to process data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the first device to map the output of the first neural network. The second device configures a second neural network based on the first information, and the second neural network is used by the second device to process data from the first device. The second device determines the first spatial dimension mapping method based on the second information, and the first spatial dimension mapping method is used by the second device to determine the data input to the second neural network.

[0122] As can be seen, the first device informs the second device of the modulation order and spatial dimension used to configure the first neural network (i.e., the transmitting neural network), so that the second device can configure a second neural network (i.e., the receiving neural network) that is compatible with the first neural network based on the obtained modulation order and spatial dimension. Furthermore, the first device also informs the second device of the spatial dimension mapping method used for the output of the first neural network, so that the second device can correctly extract data from the signal sent by the first device based on the obtained spatial dimension mapping method, and then input the correctly extracted data into the configured second neural network for processing.

[0123] See Figure 8 , Figure 8 This is a schematic diagram of the structure of a communication device 800 provided in an embodiment of the present invention. The communication device 800 may be a first device, or a component of the first device (e.g., an integrated circuit, a chip, etc.); or, the communication device 800 may be a second device, or a component of the second device (e.g., an integrated circuit, a chip, etc.). The communication device 800 may include a processing unit 801. Optionally, the communication device 800 may further include a communication unit 802, and the processing unit 801 may be used to control the communication unit 802 to perform data / signaling transmission and reception. Optionally, the communication device 800 may further include a storage unit.

[0124] In one implementation, the communication device 800 is used to perform the functions of the first device in the aforementioned method embodiments:

[0125] Processing unit 801 is used to determine the first information and the second information.

[0126] The communication unit 802 is used to send first information and second information to the second device.

[0127] The first information is related to the first modulation order and the first spatial dimension. The first modulation order and the first spatial dimension are used by the communication device 800 to configure the first neural network. The first neural network is used by the communication device 800 to process data to be sent to the second device. The second information is related to the mapping method of the first spatial dimension. The mapping method of the first spatial dimension is used by the communication device 800 to map the output of the first neural network.

[0128] In an alternative implementation, the communication unit 802 is further configured to transmit third information, the third information including one or more indices and a modulation order and spatial dimension associated with each index. The first information includes a first index, which is an index among one or more indices associated with a first modulation order and a first spatial dimension.

[0129] In an alternative implementation, the communication unit 802 is further configured to transmit fourth information, the fourth information including one or more indices and a modulation order associated with each index. The first information includes a second index and a first spatial dimension, the second index being an index among one or more indices associated with the first modulation order.

[0130] In one alternative implementation, the first information includes: a first modulation order and a first spatial dimension.

[0131] In an optional implementation, the communication unit 802 is further configured to transmit fifth information, the fifth information including one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The second information includes a first indication, which is an indication in the fifth information associated with a first spatial dimension mapping method.

[0132] In one alternative implementation, multiple spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

[0133] In one alternative implementation, the second information is used to indicate the first spatial dimension mapping method.

[0134] In another implementation, the communication device 800 is used to perform the functions of the second device in the aforementioned method embodiments:

[0135] The communication unit 802 is used to receive first information and second information from the first device. The first information is related to the first modulation order and the first spatial dimension, and the second information is related to the mapping method of the first spatial dimension.

[0136] The processing unit 801 is configured to configure a second neural network based on the first information, the second neural network being used by the communication device 800 to process data from the first device.

[0137] The processing unit 801 is also configured to determine a first spatial dimension mapping method based on the second information, the first spatial dimension mapping method being used by the communication device 800 to determine the data input to the second neural network.

[0138] In an optional implementation, the communication unit 802 is further configured to receive third information, the third information including one or more indices and a modulation order and spatial dimension associated with each index. The first information includes a first index. The processing unit 801 configures a second neural network based on the first information, specifically configured to: configure the input information of the second neural network based on the first spatial dimension associated with the first index in the third information; and configure the output information of the second neural network based on the first modulation order associated with the first index in the third information.

[0139] In an optional implementation, the communication unit 802 is further configured to receive fourth information, the fourth information including one or more indices and a modulation order associated with each index. The first information includes: a second index and a first spatial dimension. The processing unit 801 configures a second neural network based on the first information, specifically configured to: configure the input information of the second neural network based on the first spatial dimension in the first information; and configure the output information of the second neural network based on the first modulation order associated with the second index in the fourth information.

[0140] In one optional implementation, the first information includes: a first modulation order and a first spatial dimension. The processing unit 801 configures a second neural network based on the first information, specifically for: configuring the input information of the second neural network based on the first spatial dimension in the first information; and configuring the output information of the second neural network based on the first modulation order in the first information.

[0141] In an optional implementation, the communication unit 802 is further configured to receive fifth information, the fifth information including one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method. The second information includes a first indication. The processing unit 801 determines a first spatial dimension mapping method based on the second information, specifically configured to: determine the first spatial dimension mapping method associated with the first indication from the fifth information.

[0142] In one alternative implementation, multiple spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

[0143] In one alternative implementation, the second information is used to indicate the first spatial dimension mapping method.

[0144] The embodiments of this application and the above-described method embodiments are based on the same concept and have the same technical effects. For the specific principles, please refer to the description of the above-described method embodiments, which will not be repeated here.

[0145] Please see Figure 9 , Figure 9 This is a schematic diagram of another communication device provided in an embodiment of this application. The communication device 900 may include a memory 901 and a processor 902. Optionally, it may also include a communication interface 903. The memory 901, processor 902, and communication interface 903 are connected via one or more communication buses. The communication interface 903 is controlled by the processor 902 for sending and receiving information.

[0146] Memory 901 may include read-only memory and random access memory, and provides instructions and data to processor 902. A portion of memory 901 may also include non-volatile random access memory.

[0147] Communication interface 903 is used to receive or send data.

[0148] Processor 902 can be a central processing unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor; optionally, processor 902 can also be any conventional processor. Wherein:

[0149] Memory 901 is used to store program instructions.

[0150] Processor 902 is used to call program instructions stored in memory 901.

[0151] In one optional implementation, when the processor 902 invokes the computer program, it performs the functions of the first device in the aforementioned method embodiments:

[0152] Processor 902 is used to determine the first information and the second information.

[0153] Communication interface 903 is used to send first information and second information to the second device.

[0154] The first information is related to the first modulation order and the first spatial dimension. The first modulation order and the first spatial dimension are used by the communication device 900 to configure the first neural network. The first neural network is used by the communication device 900 to process data to be sent to the second device. The second information is related to the mapping method of the first spatial dimension. The mapping method of the first spatial dimension is used by the communication device 900 to map the output of the first neural network.

[0155] In another optional implementation, when the processor 902 invokes the computer program, it performs the functions of the second device in the aforementioned method embodiments:

[0156] The communication interface 903 is used to receive first information and second information from the first device. The first information is related to the first modulation order and the first spatial dimension, and the second information is related to the mapping method of the first spatial dimension.

[0157] The processor 902 is configured to configure a second neural network based on the first information, the second neural network being used by the communication device 900 to process data from the first device.

[0158] The processor 902 is also used to determine a first spatial dimension mapping method based on the second information, the first spatial dimension mapping method being used by the communication device 900 to determine the data input to the second neural network.

[0159] The embodiments of this application and the above-described method embodiments are based on the same concept and have the same technical effects. For the specific principles, please refer to the description of the above-described method embodiments, which will not be repeated here.

[0160] This application also provides a chip that can perform the relevant steps of the first or second device in the foregoing method embodiments. The chip includes a processor and a communication interface, the communication interface being used to receive or send data.

[0161] In one embodiment, the chip performs the relevant steps of the first device in the aforementioned method embodiments:

[0162] The processor is configured to cause the chip to perform the following operations: determine first information and second information; and send the first information and second information to a second device. The first information is related to a first modulation order and a first spatial dimension, which are used by the chip to configure a first neural network. The first neural network is used by the chip to process the data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by the chip to map the output of the first neural network.

[0163] In another approach, the chip performs the relevant steps of the second device in the aforementioned method embodiments:

[0164] The processor is configured to cause the chip to perform the following operations: receiving first information and second information from a first device, wherein the first information is related to a first modulation order and a first spatial dimension, and the second information is related to a first spatial dimension mapping method. Based on the first information, configuring a second neural network for the chip to process data from the first device. Based on the second information, determining the first spatial dimension mapping method, which is used by the chip to determine the data input to the second neural network.

[0165] The embodiments of this application and the above-described method embodiments are based on the same concept and have the same technical effects. For the specific principles, please refer to the description of the above-described method embodiments, which will not be repeated here.

[0166] For each device or product applied to or integrated into a chip, each of its modules can be implemented using hardware methods such as circuits, or at least some modules can be implemented using software programs that run on a processor integrated inside the chip, while the remaining (if any) modules can be implemented using hardware methods such as circuits.

[0167] like Figure 10 As shown, Figure 10 This is a schematic diagram of the structure of a module device provided in an embodiment of this application. The module device 1000 can perform the relevant steps of the first device or the second device in the aforementioned method embodiments. The module device 1000 includes: a communication module 1001, a power module 1002, a storage module 1003, and a chip 1004.

[0168] The power module 1002 is used to provide power to the module device; the storage module 1003 is used to store data and instructions; and the communication module 1001 is used for internal communication within the module device or for communication between the module device and external devices.

[0169] In one embodiment, chip 1004 is used to execute the method performed by the first device in the above method embodiment: determining first information and second information; and sending the first information and second information to a second device. The first information is related to a first modulation order and a first spatial dimension, which are used by chip 1004 to configure a first neural network. The first neural network is used by chip 1004 to process data to be sent to the second device. The second information is related to a first spatial dimension mapping method, which is used by chip 1004 to map the output of the first neural network.

[0170] In another embodiment, chip 1004 performs the method executed by the second device in the above method embodiments: receiving first information and second information from the first device, wherein the first information is related to a first modulation order and a first spatial dimension, and the second information is related to a first spatial dimension mapping method. Based on the first information, a second neural network is configured, which is used by chip 1004 to process data from the first device. Based on the second information, the first spatial dimension mapping method is determined, which is used by chip 1004 to determine the data input to the second neural network.

[0171] The implementation method of this module device can be found in the relevant content of the above method embodiments, and will not be described in detail here.

[0172] The embodiments of this application and the above-described method embodiments are based on the same concept and have the same technical effects. For the specific principles, please refer to the description of the above-described method embodiments, which will not be repeated here.

[0173] This application also provides a computer-readable storage medium storing instructions that, when executed on a processor, enable the implementation of the method flow described in the above method embodiments.

[0174] This application also provides a computer program product, which, when run on a processor, enables the implementation of the method flow described in the above method embodiments.

[0175] Regarding the modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or a combination of both. For example, for various devices and products applied to or integrated into a chip, all of their modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on the chip's integrated processor, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into a chip module, all of their modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same part (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units... It can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, the modules / units they contain can all be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0176] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some operations can be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0177] The descriptions of the various embodiments provided in this application can be referenced mutually. Each embodiment has its own emphasis, and parts not described in detail in a certain embodiment can be referred to the relevant descriptions of other embodiments. For the sake of convenience and brevity, for example, the functions and operations of the various devices and equipment provided in the embodiments of this application can be referred to the relevant descriptions of the method embodiments of this application. The method embodiments and the device embodiments can also be referenced, combined or cited from each other.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A communication method, characterized in that, Applied to a first device, the method includes: Determine the first and second information; Send the first information and the second information to the second device; Wherein, the first information is related to the first modulation order and the first spatial dimension, the first modulation order and the first spatial dimension are used by the first device to configure the first neural network, and the first neural network is used by the first device to process the data to be sent to the second device; The second information is related to the first spatial dimension mapping method, which is used by the first device to map the output of the first neural network. The first spatial dimension mapping method is one of the following: time domain mapping, frequency domain mapping, and spatial domain mapping.

2. The method according to claim 1, characterized in that, The method further includes: Send a third message, which includes one or more indices and the modulation order and spatial dimension associated with each index; The first information includes a first index, which is one of the one or more indices associated with the first modulation order and the first spatial dimension.

3. The method according to claim 1, characterized in that, The method further includes: Send a fourth message, which includes one or more indices and the modulation order associated with each index; The first information includes: a second index and the first spatial dimension, wherein the second index is one of the one or more indices that is associated with the first modulation order.

4. The method according to claim 1, characterized in that, The first information includes: the first modulation order and the first spatial dimension.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Send a fifth message, which includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method; The second information includes a first indication, which is an indication associated with the first spatial dimension mapping method in the fifth information.

6. The method according to claim 5, characterized in that, The various spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

7. The method according to any one of claims 1 to 4, characterized in that, The second information is used to indicate the first spatial dimension mapping method.

8. A communication method, characterized in that, Applied to a second device, the method includes: Receive first information and second information from a first device, wherein the first information is related to a first modulation order and a first spatial dimension, and the second information is related to the mapping method of the first spatial dimension; Based on the first information, a second neural network is configured, which is used by the second device to process data from the first device. Based on the second information, the first spatial dimension mapping method is determined, and the first spatial dimension mapping method is used by the second device to determine the data input to the second neural network; The first spatial dimension mapping method is one of the following: time domain mapping, frequency domain mapping, or spatial domain mapping.

9. The method according to claim 8, characterized in that, The method further includes: Receive third information, which includes one or more indices and the modulation order and spatial dimension associated with each index; The first information includes a first index; The step of configuring the second neural network based on the first information includes: configuring the input information of the second neural network based on the first spatial dimension associated with the first index in the third information; and configuring the output information of the second neural network based on the first modulation order associated with the first index in the third information.

10. The method according to claim 8, characterized in that, The method further includes: Receive fourth information, which includes one or more indices and the modulation order associated with each index; The first information includes: the second index and the first spatial dimension; The step of configuring the second neural network based on the first information includes: configuring the input information of the second neural network based on the first spatial dimension in the first information; and configuring the output information of the second neural network based on the first modulation order associated with the second index in the fourth information.

11. The method according to claim 8, characterized in that, The first information includes: the first modulation order and the first spatial dimension; The step of configuring the second neural network based on the first information includes: configuring the input information of the second neural network based on the first spatial dimension in the first information; and configuring the output information of the second neural network based on the first modulation order in the first information.

12. The method according to any one of claims 8 to 11, characterized in that, The method further includes: Receive fifth information, which includes one or more spatial dimension mapping methods and an indication associated with each spatial dimension mapping method; The second information includes the first instruction; Determining the first spatial dimension mapping method based on the second information includes: determining the first spatial dimension mapping method associated with the first indication from the fifth information.

13. The method according to claim 12, characterized in that, The various spatial dimension mapping methods include: frequency domain mapping, time domain mapping, and spatial domain mapping.

14. The method according to any one of claims 8 to 11, characterized in that, The second information is used to indicate the first spatial dimension mapping method.

15. A communication device, characterized in that, The apparatus includes modules or units for implementing the method of any one of claims 1 to 7, or includes modules or units for implementing the method of any one of claims 8 to 14.

16. A communication device, characterized in that, The communication device includes a processor and a memory interconnected thereto, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to perform the method as described in any one of claims 1 to 7, or to perform the method as described in any one of claims 8 to 14.

17. A chip, characterized in that, The device includes a processor and a communication interface, wherein the processor is configured to cause the chip to perform the method as described in any one of claims 1 to 7, or to perform the method as described in any one of claims 8 to 14.

18. A module device, characterized in that, The module device includes a communication module, a power module, a storage module, and a chip, wherein: The power module is used to provide electrical energy to the module device; The storage module is used to store data and instructions; The communication module is used for internal communication within the module device, or for communication between the module device and external devices; The chip is used to perform the method as described in any one of claims 1 to 7, or to perform the method as described in any one of claims 8 to 14.

19. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions that, when executed on a communication device, cause the communication device to perform the method according to any one of claims 1 to 7, or to perform the method according to any one of claims 8 to 14.

Images