Transmission method, first device, and second device

Abstract

The present application relates to a transmission method, a first device, a second device, a chip, a computer-readable storage medium, a computer program product, a computer program, and a communication system. The method comprises: a first device processes a plurality of sources on the basis of a first model to obtain first transmission information (S810); and the first device sends the first transmission information to a second device, the first transmission information being used by the second device to recover the plurality of sources on the basis of a second model (S820). According to the embodiments of the present application, multiplexing transmission for a plurality of sources can be implemented by using models, thereby saving transmission resources and improving data transmission efficiency.

Description

Transmission method, first device and second device TECHNICAL FIELD

[0001] The present application relates to the field of communication, and more particularly, to a transmission method, a first device, a second device, a chip, a computer readable storage medium, a computer program product, a computer program and a communication system. BACKGROUND

[0002] In a communication system, the use of multiplexing technology greatly improves the transmission efficiency of the channel. Multiplexing technology includes time division multiplexing, frequency division multiplexing, code division multiplexing, space division multiplexing, etc. In these multiplexing schemes, each information source transmits an information stream, and in the case where the amount of information transmitted by each information stream remains unchanged, as the number of multiplexed information streams increases, the transmission resources will also increase. How to save transmission resources and improve transmission efficiency is still a hot issue in the field of communication.

[0003] SUMMARY

[0004] Embodiments of the present application provide a transmission method, which can save transmission resources and improve transmission efficiency.

[0005] Embodiments of the present application provide a transmission method, comprising:

[0006] The first device processes a plurality of information sources based on a first model to obtain first transmission information;

[0007] The first device sends the first transmission information to a second device; wherein the first transmission information is used by the second device to recover the plurality of information sources based on a second model.

[0008] Embodiments of the present application provide a transmission method, comprising:

[0009] The second device receives first transmission information from a first device; wherein the first transmission information is obtained by processing a plurality of information sources based on a first model;

[0010] The second device processes the first transmission information based on a second model to recover the plurality of information sources.

[0011] Embodiments of the present application provide a first device, comprising:

[0012] A first processing module configured to process a plurality of information sources based on a first model to obtain first transmission information;

[0013] A first communication module configured to send the first transmission information to a second device; wherein the first transmission information is used by the second device to recover the plurality of information sources based on a second model.

[0014] Embodiments of the present application provide a second device, comprising:

[0015] The second communication module is configured to receive first transmission information from the first device, wherein the first transmission information is obtained by processing a plurality of signal sources based on a first model.

[0016] The second processing module is configured to process the first transmission information based on a second model to recover the plurality of signal sources.

[0017] The embodiments of the present application provide a first device, which comprises a transceiver, a processor and a memory. The memory is configured to store a computer program, the transceiver is configured to communicate with other devices, and the processor is configured to invoke and run the computer program stored in the memory, so that the first device performs the transmission method.

[0018] The embodiments of the present application provide a second device, which comprises a transceiver, a processor and a memory. The memory is configured to store a computer program, the transceiver is configured to communicate with other devices, and the processor is configured to invoke and run the computer program stored in the memory, so that the second device performs the transmission method.

[0019] The embodiments of the present application provide a chip, which is configured to implement the transmission method.

[0020] Specifically, the chip comprises a processor, which is configured to invoke and run a computer program from a memory, so that a device installed with the chip performs the transmission method.

[0021] The embodiments of the present application provide a computer readable storage medium, which is configured to store a computer program, and when the computer program is run by a device, the device performs the transmission method.

[0022] The embodiments of the present application provide a computer program product, which comprises computer program instructions, and the computer program instructions make a computer perform the transmission method.

[0023] The embodiments of the present application provide a computer program, which, when run on a computer, makes the computer perform the transmission method.

[0024] In the embodiments of the present application, a model is used to process a plurality of signal sources to obtain first transmission information, and the transmission of the plurality of signal sources is realized by sending the first transmission information, that is, the model is used to realize multiplexing transmission for the plurality of signal sources, so that the transmission resources of time domain, frequency domain, code domain and space domain can be saved, and the transmission efficiency of data is improved. BRIEF DESCRIPTION OF DRAWINGS

[0025] FIG. 1 is a schematic diagram of an application scenario according to the embodiments of the present application.

[0026] FIG. 2 is a schematic diagram of a basic working process in a wireless communication system.

[0027] Figure 3 is a schematic diagram of semantic communication.

[0028] Figure 4 is a schematic diagram of a structure of a neuron node.

[0029] Figure 5 is a schematic diagram of a fully connected neural network.

[0030] Figure 6 is a schematic diagram of a convolutional neural network.

[0031] Figure 7 is a schematic diagram of a structure of a basic LSTM unit.

[0032] Figure 8 is a schematic flowchart of a transmission method according to an embodiment of the application.

[0033] Figure 9 is a schematic flowchart of a transmission method according to another embodiment of the application.

[0034] Figure 10A is a schematic diagram of a transmission method in an application example of an embodiment of the application.

[0035] Figure 10B is a schematic diagram of a transmission method in another application example of an embodiment of the application.

[0036] Figure 11A is a schematic diagram of an application example of a model selection manner based on network configuration.

[0037] Figure 11B is a schematic diagram of another application example of a model selection manner based on network configuration.

[0038] Figure 12 is a schematic diagram of an initialization phase initiated by downlink in an application example of an embodiment of the application.

[0039] Figure 13 is a schematic diagram of an initialization phase initiated by uplink in an application example of an embodiment of the application.

[0040] Figure 14 is a schematic diagram of an automatic fallback mechanism for downlink transmission in an embodiment of the application.

[0041] Figure 15 is a schematic diagram of an automatic fallback mechanism for uplink transmission in an embodiment of the application.

[0042] Figure 16 is a schematic block diagram of a first device according to an embodiment of the application.

[0043] Figure 17 is a schematic block diagram of a second device according to an embodiment of the application.

[0044] Figure 18 is a schematic block diagram of a communication device according to an embodiment of the application.

[0045] Figure 19 is a schematic block diagram of a chip according to an embodiment of the application.

[0046] Figure 20 is a schematic block diagram of a communication system according to an embodiment of the application. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

[0048] The technical solutions in the embodiments of the present application can be applied to various communication systems, for example: a Long Term Evolution (LTE) system, an Advanced long term evolution (LTE-A) system, a New Radio (NR) system, an evolved system of the NR system, a LTE-based access to unlicensed spectrum (LTE-U) system, a NR-based access to unlicensed spectrum (NR-U) system, a Non-Terrestrial Networks (NTN) system, a Universal Mobile Telecommunication System (UMTS), a Wireless Local Area Networks (WLAN), a Wireless Fidelity (WiFi), a 5th-Generation (5G) system, a 6th-Generation (6G) system or other communication systems, etc.

[0049] Generally, a conventional communication system supports a limited number of connections, which is easy to implement. However, with the development of communication technology, a mobile communication system will not only support conventional communication, but also support, for example, Device to Device (D2D) communication, Machine to Machine (M2M) communication, Machine Type Communication (MTC), Vehicle to Vehicle (V2V) communication, or Vehicle to everything (V2X) communication, etc. The technical solutions in the embodiments of the present application can also be applied to these communication systems.

[0050] In an embodiment, the communication system in the embodiments of the present application can be applied to a carrier aggregation (CA) scenario, can also be applied to a dual connectivity (DC) scenario, and can also be applied to a standalone (SA) network deployment scenario.

[0051] In an embodiment, the communication system in the embodiments of the present application can be applied to an unlicensed spectrum, which can also be regarded as a shared spectrum, or can be applied to a licensed spectrum, which can also be regarded as a non-shared spectrum.

[0052] The embodiments of the present application combine network devices and terminal devices to describe various embodiments, wherein the terminal device can also be referred to as a user equipment (UE), an access terminal, a user unit, a user station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent or a user device, etc.

[0053] The terminal device can be a station (STA) in a WLAN, can be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system such as an NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0054] In the embodiments of the present application, the terminal device can be deployed on land, including indoor or outdoor, handheld, wearable or vehicle-mounted; can also be deployed on the water surface (such as ships, etc.); can also be deployed in the air (such as airplanes, balloons and satellites, etc.).

[0055] In the embodiments of the present application, the terminal device can be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self driving, a wireless terminal device in remote medical treatment, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, or a wireless terminal device in smart home, etc.

[0056] By way of example and not limitation, in the embodiments of the present application, the terminal device can also be a wearable device. The wearable device can also be referred to as a wearable smart device, which is a general term for devices that are designed and developed by applying wearable technology to daily wear, such as glasses, gloves, watches, clothing, and shoes, etc. The wearable device is a portable device that can be directly worn on the body or integrated into the clothes or accessories of the user. The wearable device is not only a hardware device, but also has powerful functions through software support and data interaction and cloud interaction. The general wearable smart device includes devices with full functions, large size, and the ability to realize complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and devices that focus on a certain type of application function and need to be used in cooperation with other devices, such as smart phones, such as various smart wristbands and smart jewelry for monitoring vital signs, etc.

[0057] In the embodiments of the present application, the network device can be a device for communicating with the mobile device, which can be an access point (AP) in a WLAN, an evolved node B (eNB or eNodeB) in LTE, or a relay station or an access point, or a vehicle-mounted device, a wearable device, and a network device in an NR network (gNB) or a future evolved PLMN network or a network device in an NTN network, etc.

[0058] By way of example and not limitation, in embodiments of the present application, a network device can have a mobile characteristic, for example, the network device can be a mobile device. Alternatively, the network device can be a satellite, a balloon station. For example, the satellite can be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. Alternatively, the network device can also be a base station disposed at a location on land, water, etc.

[0059] In embodiments of the present application, a network device can serve a cell, and a terminal device communicates with the network device through a transmission resource (for example, a frequency domain resource, or a spectrum resource) used by the cell. The cell can be a cell corresponding to the network device (for example, a base station), and the cell can belong to a macro base station or a base station corresponding to a small cell. The small cell can include a metro cell, a micro cell, a pico cell, a femto cell, etc., and these small cells have the characteristics of small coverage and low transmit power, and are suitable for providing high-speed data transmission services.

[0060] FIG. 1 illustrates a communication system 100. The communication system includes one network device 110 and two terminal devices 120. In an implementation, the communication system 100 can include multiple network devices 110, and each network device 110 can include other numbers of terminal devices 120 within its coverage, which is not limited in embodiments of the present application.

[0061] In an implementation, the communication system 100 can also include a mobility management entity (MME), an access and mobility management function (AMF), and other network entities, which are not limited in embodiments of the present application.

[0062] It should be understood that the devices with communication function in the network / system in the embodiments of the present application can be referred to as communication devices. For example, the communication system shown in FIG. 1, the communication devices can include network devices and terminal devices with communication function, which can be specific devices in the embodiments of the present application, and details are not described herein again; the communication devices can also include other devices in the communication system, such as network controllers, mobile management entities and other network entities, which are not limited in the embodiments of the present application.

[0063] It should be understood that the terms "system" and "network" are often used interchangeably herein. The term "and / or" herein is only used to describe the association relationship of the associated objects. For example, A and / or B can represent three cases: A exists alone, A and B exist together, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects.

[0064] It should be understood that the "indication" mentioned in the embodiments of the present application can be direct indication, indirect indication, or can represent an associated relationship. For example, A indicates B, which can mean that B can be obtained by A directly; or A indirectly indicates B, for example, A indicates C, and B can be obtained by C; or A and B have an associated relationship.

[0065] In the description of the embodiments of the present application, the term "corresponding" can represent a direct or indirect corresponding relationship between the two, or an associated relationship between the two, or an indication and being indicated, configuration and being configured relationship.

[0066] In order to facilitate the understanding of the technical solutions of the embodiments of the present application, the related technologies of the embodiments of the present application are described as follows. The following related technologies can be combined with the technical solutions of the embodiments of the present application in any way, and all belong to the protection scope of the embodiments of the present application.

[0067] (I) Overall description of wireless communication system

[0068] FIG. 2 is a schematic diagram of the basic working process in a wireless communication system. As shown in FIG. 2, in the wireless communication system, the basic working process can include the following steps.

[0069] Source coding: source coding compresses data by removing redundant information in the data to reduce the required bandwidth for transmission.

[0070] Channel Coding: To enhance error detection and correction capabilities during transmission, channel coding adds extra parity bits to improve data reliability. Common coding techniques include convolutional codes, Turbo codes, and Low-density Parity-check (LDPC) codes.

[0071] Modulation: Modulation maps the encoded data onto specific signal waveforms. Common modulation techniques include QAM (Quadrature Amplitude Modulation), PSK (Phase Shift Keying), and FSK (Frequency Shift Keying).

[0072] Signal Transmission: Signal transmission involves processing such as power amplification and filtering before transmission to ensure signal quality and compliance with transmission standards.

[0073] Through the Channel: During transmission, signals pass through various wireless propagation environments, causing attenuation, multipath effects, interference, and other effects.

[0074] Received Signal: The antenna and receiver at the receiving end capture the signals transmitted through the channel. These signals may be weak and accompanied by noise before reaching the receiver.

[0075] Channel Estimation: The receiver needs to estimate the characteristics of the channel in the received signal to support correct demodulation, often using known pilot signals.

[0076] Symbol Detection: Based on channel estimation results, the receiver performs symbol detection on the received analog signal, i.e., attempts to determine the digital data represented by each signal waveform.

[0077] Demodulation: Demodulation converts the detected symbols into a received bit stream for subsequent channel decoding.

[0078] Channel Decoding: Using the parity information of channel coding, the receiver can detect and correct errors that may occur during transmission.

[0079] Source Decoding: Restores data compressed by source encoding to recover the original data.

[0080] The above process is a simple illustration. There are other modules not listed in traditional communication systems, such as resource mapping, precoding, interference cancellation, and CSI (Channel-State Information) measurement. These modules are also designed and implemented independently, and after integration of each independent module, a complete wireless communication system can be formed.

[0081] (II) Semantic Communication and Physical Layer Implementation

[0082] Traditional communication mainly focuses on how to accurately and efficiently transmit each bit. Semantic communication focuses on the content and meaning of information, emphasizing the transmission of the meaning part of the information rather than simply transmitting raw data. The goal of this approach is to ensure that the receiver understands the sender's intention and the semantic content of the information, even in poor channel conditions or limited data transmission. The implementation of semantic communication usually relies on advanced AI algorithms to identify, extract and process semantic information in the source, so that even if there is a loss or miscommunication of transmitted bit information, the receiver can reconstruct and understand the basic meaning of the original information.

[0083] Implementing semantic communication at the physical layer means that not only the physical properties of the signal are considered in the signal processing and transmission process at the lowest level, but also the semantic content of the information. The main implementation is to add semantic encoding and decoding. Figure 3 is a schematic diagram of semantic communication. As shown in Figure 3, semantic communication mainly includes the following aspects:

[0084] Semantic encoding: the sender encodes the key information into a bit stream or directly into a transmittable symbol (modulation symbol) based on AI model semantic analysis and encoding of image, video or text data.

[0085] Traditional communication system transmission: transmit bit stream or modulation symbol through conventional wireless communication system.

[0086] Semantic decoding: the receiver decodes the received bit stream or symbol based on AI model to recover the core semantic content of the original information.

[0087] (Three) neural network

[0088] Neural network is an operation model composed of multiple neuron nodes connected to each other. Figure 4 is a schematic diagram of the structure of a neuron node. The connection between a node and other nodes represents the weighted value from input signal to output signal, called weight (such as w1, w2, w n Each node performs weighted summation (optionally also summed with bias information b) on different input signals (such as a1, a2, a n and outputs through a specific activation function f.

[0089] Figure 5 is a schematic diagram of a simple fully connected neural network. As shown in Figure 5, the neural network input layer, hidden layer and output layer can produce different outputs through different connection methods, weights and activation functions of multiple neurons, thereby fitting the mapping relationship from input to output. Each upper level node is connected to all lower level nodes.

[0090] Next, the CNN (Convolutional Neural Networks) is introduced. FIG. 6 is a schematic diagram of a convolutional neural network. The basic structure of the convolutional neural network includes an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer. Each neuron of the convolution kernel in the convolutional layer is locally connected to its input, and the maximum value or average value feature of a local layer is extracted by introducing the pooling layer, effectively reducing the parameters of the network and mining local features, so that the convolutional neural network can quickly converge and obtain excellent performance.

[0091] Finally, the RNN (Recurrent Neural Network) is introduced. The recurrent neural network is a neural network for modeling sequence data, and has achieved remarkable results in natural language processing fields such as machine translation and speech recognition. Specifically, the network memorizes information at past time and uses it in the calculation of the current output, that is, the nodes between the hidden layers are no longer unconnected but connected, and the input of the hidden layer includes not only the input layer but also the output of the previous hidden layer. Common RNNs include LSTM (Long Short-Term Memory), and FIG. 7 shows a basic LSTM cell structure. Unlike the recurrent neural network, which only considers the most recent state, the LSTM cell state determines which states should be retained and which states should be forgotten, solving the long-term memory defects of traditional recurrent neural networks.

[0092] In the time, frequency, code, and space division multiplexing scheme in the existing communication system, each information source transmits one information stream. In the case where the amount of information transmitted by each information stream remains unchanged, as the number of multiplexed information streams increases, the transmission resources also increase, and the utilization rate of the frequency spectrum has room for improvement. The embodiments of the present application utilize the semantic differences between different information sources and design a multiplexing transmission method based on AI, which brings a new multiplexing dimension beyond time, frequency, code, and space for the information multiplexing transmission of the communication system, and has high potential in saving transmission resources and improving transmission efficiency.

[0093] FIG. 8 is a schematic flowchart of a transmission method performed by a first device according to an embodiment of the present application. The method can be optionally applied to the system shown in FIG. 1, but is not limited thereto. The method includes:

[0094] S810, the first device processes the multiple information sources based on a first model to obtain first transmission information;

[0095] S820, the first device sends the first transmission information to a second device; wherein the first transmission information is used by the second device to recover the multiple information sources based on a second model.

[0096] Corresponding to the method described above, FIG. 9 is a schematic flowchart of a transmission method performed by a second device according to another embodiment of the present application. The method can optionally be applied to the system shown in FIG. 1, but is not limited thereto. The method comprises:

[0097] S910, the second device receives first transmission information from the first device; wherein the first transmission information is obtained by processing a plurality of information sources based on a first model;

[0098] S920, the second device processes the first transmission information based on a second model to recover the plurality of information sources.

[0099] In the embodiments of the present application, the first device is a sending end / sending device of the plurality of information sources; optionally, the first device can include a terminal device or a network device. The second device is a receiving end / receiving device of the plurality of information sources; optionally, the second device can include a network device or a terminal device. For example, the first device is a network device and the second device is a terminal device, i.e., the plurality of information sources can be downlink data; or the first device is a terminal device and the second device is a network device, i.e., the plurality of information sources can be uplink data. Optionally, the above method can also be applied to the transmission process between different terminal devices (i.e., the first device and the second device are both terminal devices) or the transmission process between different network devices (i.e., the first device and the second device are both network devices), which is not limited in the embodiments of the present application.

[0100] It can be understood that the information source refers to information to be transmitted, which can also be referred to as an information source. From the perspective of business, the information source can be any information to be transmitted, such as user data, measurement data, control information, scheduling information, configuration information, reporting information, etc.; from the perspective of information form, the information source can be a matrix (such as a CSI matrix, an image matrix, etc.), a voice sequence, a video sequence, a text sequence, etc. original data with semantic information, or a bit stream, such as a system information bit stream or an encoded bit stream obtained by encoding the original data information, etc.

[0101] In the embodiments of the present application, the first model is used to process the plurality of information sources to obtain the first transmission information. Optionally, processing the plurality of information sources based on the first model to obtain the first transmission information can include: inputting the plurality of information sources into the first model to obtain the first transmission information output by the first model. Since the first model jointly processes the plurality of information sources to obtain the first transmission information, it is different from processing each of the plurality of information sources to obtain different transmission information corresponding to each of the plurality of information sources, and therefore the processing of the plurality of first information sources based on the first model can be understood as multiplexing the plurality of first information sources.

[0102] Exemplarily, the first transmission information can include a bit stream or a symbol sequence (a sequence containing multiple modulation symbols). That is, the processing of the multiple sources based on the first model can include converting the multiple source processing into a bit stream, or converting the multiple sources into a symbol sequence. Exemplarily, the input interface of the first model can adapt one or more sources such as a CSI matrix, an image matrix, a speech sequence, a video sequence, a text sequence, a bit stream, etc., and the output interface can adapt a bit stream or a symbol sequence directly used for transmission.

[0103] Optionally, multiplexing the multiple sources based on the first model can implement or replace source coding of the multiple sources, and accordingly, the first transmission information is a bit stream, and the first device sending the first transmission information to the second device can include channel coding and modulation of the first transmission information before sending.

[0104] Optionally, multiplexing the multiple sources based on the first model can implement or replace source coding and channel coding of the multiple sources, and accordingly, the first transmission information is a bit stream, and the first device sending the first transmission information to the second device can include modulation of the first transmission information before sending.

[0105] Optionally, the processing or multiplexing of the multiple sources based on the first model can implement or replace source coding, channel coding and modulation of the multiple sources, and accordingly, the first transmission information is a symbol sequence, and the first device sending the first transmission information to the second device can include mapping the first transmission information to physical resources (such as antenna ports) for sending.

[0106] Optionally, the first model can include a first AI model, i.e., the first model processes the multiple sources based on AI technology, so that the multiplexing based on semantic information can be implemented by using the understanding and processing capabilities of AI for semantics. Optionally, the first AI model can be a model based on any neural network structure such as a fully connected neural network, a convolutional neural network or a recurrent neural network.

[0107] It can be understood that the first device sending the first transmission information to the second device, since the first transmission information is obtained based on the multiple sources, the second device can recover the multiple sources by using the first transmission information, in other words, the second device obtains the decoding result / recovery result of the multiple sources by decoding the first transmission information. Specifically, the second device can process the first transmission information based on a second model to recover the multiple sources.

[0108] Optionally, the second model can include a second AI model, i.e., the second model processes the first transmission information based on AI technology, so that the demultiplexing based on semantic information can be implemented by using the understanding and processing capability of AI for semantics. Optionally, the second AI model can be a model based on any neural network structure such as a fully connected neural network, a convolutional neural network, or a recurrent neural network.

[0109] It should be noted that in the embodiments of the present application, the AI model can be understood as an AI function or an AI feature, and accordingly, the first AI model can be understood as a first AI function or a first AI feature, the second AI model can be understood as a second AI function or a second AI feature, and so on. Specifically, in different scenarios or in different description manners, any form of model, function, and feature in a terminal device can be used to implement the processing of input information based on AI. For example, the first device is deployed with a first AI model, which can multiplex multiple signal sources of the input model; or the first device has a first AI function, and using the first AI function can multiplex multiple signal sources; or the first device has an AI feature, and because of the AI feature, it can multiplex multiple signal sources. In addition, in some application scenarios, ML (Machine Learning) can be used to implement the above AI processing. Therefore, in some examples or descriptions, the AI model can also be referred to as an ML model, an AI function, an ML function, an AI feature, or an ML feature, etc.

[0110] In actual applications, the first model is related to the second model, or the first model has a corresponding relationship with the second model, so that the multiplexing result of the first model can be demultiplexed based on the second model. For example, the first model and the second model are an encoder and a decoder in a self-encoder, and the first model and the second model are obtained by training the self-encoder. For another example, the second model is trained by using the output information of the first model and a corresponding label, and the label is the input information of the first model.

[0111] In order to facilitate understanding of the above technical solutions, FIG. 10A shows a schematic diagram of a transmission method in an application example of the embodiments of the present application. Taking an example of multiplexing two signal sources by the first model and the first transmission information being a symbol sequence, as shown in FIG. 10A, at the first device side, the first signal source and the second signal source can be input into the first model, the first model outputs a first symbol sequence, and the first device sends the first symbol sequence. At the second device side, the second device receives the first symbol sequence, inputs the first symbol sequence into the second model, and the second model can output the recovered first signal source and the recovered second signal source.

[0112] In the embodiments of the present application, a plurality of information sources are processed by using a model to obtain first transmission information, and the transmission of the plurality of information sources is realized by sending the first transmission information, that is, the model is used to realize multiplexing transmission for the plurality of information sources. For example, based on the first AI model, nonlinear multiplexing transmission is realized for the plurality of information sources by using the semantic difference between different information sources, and at the receiving end, based on the second AI model, the multiplexed different information sources are demultiplexed and recovered. This method can realize multiplexing transmission in a new dimension, i.e., semantic dimension, beyond time, frequency, code, and space dimensions, for a plurality of information sources, which can greatly save transmission resources in time, frequency, code, and space dimensions, and improve the transmission efficiency of data.

[0113] In some embodiments, a first information source in the plurality of information sources is an information source with semantic information; and a second information source in the plurality of information sources is a bit stream or an information source with semantic information.

[0114] It can be understood that the plurality of information sources, i.e., at least two information sources, can include only the first information source and the second information source, or can include the first information source, the second information source, and one or more other information sources. Among them, at least one information source (for example, the first information source or the first information source and the second information source) is an information source with semantic information. Optionally, at most one information source (for example, the second information source) in the plurality of information sources is a bit stream. In this way, the understanding and processing capability of the AI model for semantic information can be used to multiplex and demultiplex different semantic information or semantic information and a bit stream, and the correctness of information transmission can be ensured.

[0115] In some embodiments, the information source with semantic information includes CSI or user data. For example, the first information source can include first CSI or first user data; and the second information source can include a bit stream, second CSI, or second user data.

[0116] In some embodiments, the bit stream includes a system information bit stream or an encoded bit stream obtained based on user data. For example, the first information source can include first CSI or first user data; and the second information source can include a system information bit stream or an encoded bit stream obtained based on second user data.

[0117] In some embodiments, the system information bit stream comprises information for control or scheduling. Exemplarily, the system information bit stream can comprise HARQ (Hybrid Automatic Repeat Request), scheduling request, CQI (Channel Quality Indicator), RI (Rank Indication), response message in random access procedure, resource block allocation, time allocation information, transport format and modulation scheme, power control information, and other control information or scheduling information for supporting the operation of the communication system. Optionally, the control information or scheduling information can be information transmitted through broadcasting.

[0118] In some embodiments, the user data can comprise image, voice, text or video. Exemplarily, the first source can comprise CSI, image, voice, text or video, and the second source can comprise CSI, image, voice, text, video or bit stream.

[0119] In some embodiments, the first source and the second source have different semantic information. Exemplarily, when both the first source and the second source are sources with semantic information, the first source and the second source are sources with different semantic information. For example, the first source is a first image, and the second source can be CSI, a second image, voice, text or video, etc. For another example, the first source is a first text, and the second source can be CSI, image, voice, a second text or video, etc.

[0120] In practical applications, for uplink transmission (i.e. the first device is a terminal device and the second device is a network device), the combination of the first source and the second source can be as follows:

[0121] a. the first source is a CSI matrix, and the second source is system information bits (e.g. comprising HARQ feedback, scheduling request, CQI, RI, response message in random access procedure, etc.);

[0122] b. the first source is a CSI matrix, and the second source is a coded bit stream of user data (e.g. text, video, voice data, etc.);

[0123] c. the first source is a CSI matrix, and the second source is user data (e.g. text, video, voice data, etc.);

[0124] d. the first source is user data (e.g. text, video, voice data, etc.), and the second source is system information bits (e.g. comprising HARQ feedback, scheduling request, CQI and RI, etc.);

[0125] e. The first source is first user data, and the second source is second user data; wherein the first user data and the second user data can be different data types (such as multiplexing between text, video, and voice), or can be the same data type but different types of data content (such as pictures describing landscapes and pictures describing animals).

[0126] It can be understood that, for uplink transmission, different sources in the plurality of sources in the embodiments of the present application can also be combinations of other source forms. Details can be referred to the foregoing embodiments and actual application scenarios, system conventions, protocol conventions, and the like. For brevity, they are not listed one by one here.

[0127] For downlink transmission (i.e. the first device is a network device, and the second device is a terminal device), the combination of the first source and the second source can be as follows:

[0128] a. The first source is first user data, and the second source is second user data; wherein the first user data and the second user data can be different data types (such as multiplexing between text, video, and voice), or can be the same data type but different types of data content (such as pictures describing landscapes and pictures describing animals);

[0129] b. The first source is user data (such as text, video, and voice data), and the second source is system information bits (such as including resource block allocation, time allocation information, transmission format and modulation scheme, power control information, and the like);

[0130] It can be understood that, for downlink transmission, different sources in the plurality of sources in the embodiments of the present application can also be combinations of other source forms. Details can be referred to the foregoing embodiments and actual application scenarios, system conventions, protocol conventions, and the like. For brevity, they are not listed one by one here.

[0131] In some embodiments, the second model includes N models, an i-th model in the N models is used to recover a third source in the plurality of sources based on the first transmission information, N is an integer greater than or equal to 2, and i is a positive integer less than or equal to N.

[0132] The third source can be any one of the plurality of sources, such as the first source or the second source in the foregoing embodiments. Alternatively, the i-th model can be used to recover one or more third sources, and the i-th model can also be any one of the N models.

[0133] Optionally, N can be equal to the number of sources processed by the first model (i.e. the number of sources in the plurality of sources), and the N models in the second model correspond to the plurality of sources one by one, and each model is used to recover the source corresponding to the model based on the first transmission information. FIG. 10B is a schematic diagram of a transmission method in another application example of the embodiments of the present application. Taking an example of multiplexing two sources by the first model and the first transmission information being a symbol sequence, as shown in FIG. 10B, on the first device side, the first source and the second source can be input into the first model, the first model outputs a first symbol sequence, and the first device transmits the first symbol sequence. On the second device side, the second device receives the first symbol sequence, inputs the first symbol sequence into model A in the second model, and the model A outputs the recovered first source; inputs the first symbol sequence into model B in the second model, and the model B outputs the recovered first source and the second source.

[0134] Optionally, N can be less than the number of sources processed by the first model, and each model included in the second model is used to recover at least one source based on the first transmission information. For example, the plurality of sources includes source 1, source 2 and source 3, model A in the second model is used to recover source 1 and source 2, and model B is used to recover source 3.

[0135] In some embodiments, the transmission method performed by the first device further includes: the first device selecting a first model from M models, and transmitting a first message to the second device; wherein M is an integer greater than or equal to 1; and the first message is used by the second device to determine the second model corresponding to the first model.

[0136] Correspondingly, in some embodiments, the transmission method performed by the second device further includes: the second device receiving the first message from the first device; wherein the first message is used by the second device to determine the second model corresponding to the first model selected by the first device from M models, and M is an integer greater than or equal to 1.

[0137] In the above embodiments, the M models are pre-configured. That is, the first device can select the first model currently used for source multiplexing from the pre-configured at least one model, and indicate the second device through the first message, so that the second device can determine the second model corresponding to the first model, and thus process the received first transmission information by using the second model. The above embodiments support the first device to flexibly select a suitable first model, thereby facilitating to improve the accuracy of source multiplexing and recovery and improving the system performance.

[0138] In some embodiments, the first device selecting a first model from M models includes: the first device selecting the first model from the M models according to the source type of each source in the plurality of sources.

[0139] Exemplarily, M models corresponding to different source types respectively can be preconfigured, and the first device selects a model corresponding to a source type currently to be transmitted from the M models as the first model. In this way, appropriate models can be configured and selected for different source types, thereby facilitating improvement of source multiplexing and recovery accuracy and system performance.

[0140] In some embodiments, the first message comprises at least one of the following information:

[0141] a source quantity of the plurality of sources;

[0142] a source type of each source in the plurality of sources.

[0143] Exemplarily, M models corresponding to different source quantities can be preconfigured, and the first device selects a first model according to a source quantity from the M models and indicates the source quantity to the second device through the first message, so that the second device can also determine a corresponding second model from the preconfigured models.

[0144] Exemplarily, M models corresponding to different source types respectively can be preconfigured, and the first device selects a model corresponding to a source type currently to be transmitted from the M models as the first model, and indicates the source type to the second device through the first message, so that the second device can also determine a corresponding second model from the preconfigured models.

[0145] Exemplarily, M models corresponding to different source quantities and different source types can be preconfigured, and the first device selects a model corresponding to a source quantity and a source type currently to be transmitted from the M models as the first model, and indicates the source quantity and the source type to the second device through the first message, so that the second device can also determine a corresponding second model from the preconfigured models. For example, the M models can include X models corresponding to two sources, and the X models correspond to different combination manners of two source types; the M models also include Y models corresponding to three sources, and the Y models correspond to different combination manners of three source types. In this way, appropriate models can be configured and selected for different source quantities and source types, thereby facilitating improvement of source multiplexing and recovery accuracy and system performance.

[0146] In some embodiments, the first message is sent through a control channel. Exemplarily, the first message can include downlink control information (DCI) or uplink control information (UCI), which is transmitted through a control channel to ensure that the receiving end (the second device) is ready to use the correct demultiplexing model (the second model) before the first transmission information arrives.

[0147] In some embodiments, the M models are configured by the network device by sending a second message to the terminal device.

[0148] Exemplarily, in the initialization phase, the network device configures the at least one terminal device with the M models that are potentially available through the second message; in the usage phase, the sender (the first device, which can be the network device or the terminal device) selects a suitable first model for source multiplexing according to the number and / or type of sources, and transmits an indication related to the demultiplexing model to the receiver (the second device) through the first message, and the second device uses the corresponding second model to demultiplex / recover the sources according to the first message.

[0149] In some embodiments, the second message is sent through broadcasting. Exemplarily, the network device broadcasts the information of the M models to each terminal device within the coverage range through the second message, and the second message can include a master information block (MIB), a system information block (SIB), a paging message, a dedicated signaling message, a radio resource control (RRC) reconfiguration message, or a radio resource control (RRC) connection establishment message, etc.

[0150] In some embodiments, the second message includes at least one of the following:

[0151] an ID of each of the M models;

[0152] information for determining the parameters and / or structure of each model;

[0153] a source type corresponding to each model.

[0154] Exemplarily, the second message can include an ID of each of the M models. Specifically, the M models can be all or part of the models in a predefined model list, each model having a unique ID and corresponding detailed parameters, structure, or training method. The network device can indicate the IDs of the potentially available models within the range of the predefined model list through the second message, so that the terminal devices within the coverage range can download the complete model parameters and structure according to the IDs to obtain the required M models. Optionally, the second message can also include a source type corresponding to each model, so that the terminal device can select the first model from the M models according to the source type. It can be understood that since the information of the M models is configured by the network device, the network device can also select the first model from the M models according to the source type, and since the terminal device and the network device understand the information of the M models in alignment, the receiver can select the first model according to the source type regardless of whether the terminal device or the network device sends the first message and the first transmission information.

[0155] Exemplarily, the second message can comprise information for determining the parameters and / or structures of each model. The network device indicates the information for determining the parameters and / or structures of each model through the second message, so that the terminal device in the coverage range can determine the parameters and / or structures of each model according to the information, thereby obtaining the required M models. Optionally, the second message can further comprise a source type corresponding to each model, so that the terminal device can select a first model from the M models according to the source type.

[0156] It should be noted that the second message can only comprise a source type corresponding to each model, or comprise an ID of each model, information for determining the parameters and / or structures of each model, and a source type corresponding to each model. In actual applications, the message content that can be contained in the second message can be set according to scene requirements, system conventions or protocol conventions, and is not listed one by one here for the sake of brevity.

[0157] In some embodiments, the information for determining the parameters and / or structures of each model comprises at least one of the following:

[0158] compressed information of the parameters and / or structures of each model;

[0159] information for downloading the parameters and / or structures of each model;

[0160] a training manner of each model.

[0161] Exemplarily, the information for determining the parameters and / or structures of each model comprises compressed information of the parameters and / or structures of each model, so that the terminal device can decompress based on the compressed information to obtain the parameters and / or structures of each model.

[0162] Exemplarily, the information for determining the parameters and / or structures of each model comprises information for downloading the parameters and / or structures of each model, such as a download link or an identifier, so that the terminal device can further obtain the parameters and / or structures of each model from a specified server.

[0163] Exemplarily, the information for determining the parameters and / or structures of each model comprises a training manner of each model, such as a training data set and a training iteration round, so that the terminal device can obtain each model by training using the training manner.

[0164] In actual applications, the information for determining the parameters and / or structures of each model can also be any combination of the above information, which can be set according to scene requirements, system conventions or protocol conventions, and is not listed one by one here for the sake of brevity.

[0165] To facilitate understanding of the above network configuration-based model selection method, specific application examples are provided below in conjunction with the drawings. Among them, FIG. 11A takes the first device as a terminal device and the second device as a network device as an example, and shows a schematic diagram of one application example of the network configuration-based model selection method in the transmission method of the embodiments of the present application. FIG. 11B takes the first device as a network device and the second device as a terminal device as an example, and shows a schematic diagram of another application example of the network configuration-based model selection method in the transmission method of the embodiments of the present application. In the above application examples, the transmission method includes an initialization phase and a use phase.

[0166] Initialization phase:

[0167] A. Network device preparation phase: When the network starts or restarts, the network management system retrieves the list of all approved multiplexing and demultiplexing models from the model management database. These models can include models customized by the operator as well as pre-defined models that comply with industry standards. Each model has a unique model ID and can contain detailed parameters of the model itself, structural description and training method.

[0168] B. Model broadcast: The network device broadcasts the information of these models to all terminal devices within the coverage range through a second message. The broadcast message includes the model ID of each model, and optionally includes the compressed representation of the model parameters and structure. For those model parameters and structure information that are too complex or have large data volume, the network device will provide a download link or identifier so that the terminal device can further obtain the model details from the designated server. The second message can be of the following types:

[0169] Master Information Block (MIB): MIB can be used to inform terminal devices about the model supported by the base station and guide the device to obtain more detailed model information;

[0170] System Information Block (SIB): MIB can be used to inform terminal devices about the model supported by the base station and guide the device to obtain more detailed model information;

[0171] Paging message: In specific cases, such as when the network side needs to deliver urgent updates or specific model configurations to specific terminals, the paging message can be used. It is usually used to wake up the device when it is in idle mode and inform it to receive important information;

[0172] Dedicated signaling message: For terminal devices that have established a connection, the network can directly send model configuration information through a dedicated signaling message. It is usually used for personalized services or quick update of model information, such as when the terminal device starts a specific task or application;

[0173] Wireless Resource Control (RRC) reconfiguration message: In the case of an established RRC connection, the network device can specify or update the model configuration through the RRC reconfiguration message;

[0174] Wireless Resource Control (RRC) connection establishment message: In the RRC connection establishment process, the network device can embed model configuration information.

[0175] C. Terminal device processing:

[0176] After the terminal device receives the first message, it selects whether to download the complete model parameters and structure according to its processing capacity and application requirements. For models that need to be downloaded, the terminal device can download through pre-set network services to ensure that the latest and most suitable model version is obtained.

[0177] D. Model synchronization and update: The network device regularly checks for model updates and notifies of updates through subsequent system information broadcasts.

[0178] Usage phase:

[0179] A. Model selection: The sending end (first device) first selects a suitable multiplexing model (first model) according to the type of source to be multiplexed, such as text, image, video, audio, bit stream, or CSI, before preparing to send data. This selection is based on a pre-configured model library that contains multiple models for different media types.

[0180] B. Multiplexing: After selecting the model, the sending end (first device) uses the selected model to multiplex the original source data and converts it into a symbol sequence (first transmission information) that can be directly transmitted through time-frequency resources.

[0181] C. Building the first message: The sending end builds the first message according to the transmission direction and specific network protocol. The first message contains control information, which includes detailed indications of the semantic demultiplexing model used, such as model ID or key parameters. For example, Table 1 below is a mapping table of model ID and source type, which can be used to associate different source types with corresponding model IDs.

[0182] Table 1

[0183] D. Signaling transmission: The first message can be transmitted through control channels, such as downlink control information (DCI) or uplink control information (UCI), to ensure that the receiving end is ready to use the correct demultiplexing model before the data arrives.

[0184] E. Receiver processing: After the receiver (the second device) receives the control information containing the model indication, it parses the information and calls the corresponding demultiplexing model (the second model) to prepare to demultiplex the information that is about to arrive;

[0185] F. Transmission: Once the receiver sets up the demultiplexing model, the sender starts transmitting the first transmission information. The receiver uses the previously configured demultiplexing model (the second model) to process the received data and recover the original source content.

[0186] As shown in FIG. 11A, after the network device completes the configuration of the model through the second message, the terminal device can select the first model according to the source type before preparing to transmit data, and then send the first message to indicate that the network device is ready for the second model; then the terminal device implements source multiplexing based on the first model, transmits the first transmission information, and the network device demultiplexes the received first transmission information based on the second model to recover the original source content.

[0187] As shown in FIG. 11B, after the network device completes the configuration of the model through the second message, the network device can select the first model according to the source type before preparing to transmit data, and then send the first message to indicate that the terminal device is ready for the second model; then the network device implements source multiplexing based on the first model, transmits the first transmission information, and the terminal device demultiplexes the received first transmission information based on the second model to recover the original source content.

[0188] In some embodiments, the transmission method performed by the first device further includes: the first device sending a third message to the second device; wherein the third message is used to indicate a third model and / or related information of the third model, and the third model includes a model supported by the first device for multiplexing multiple sources to be transmitted in the first device.

[0189] Correspondingly, in some embodiments, the transmission method performed by the second device further includes: the second device receiving the third message from the first device; wherein the third message is used to indicate a third model and / or related information of the third model, and the third model includes a model supported by the first device for multiplexing multiple sources to be transmitted in the first device.

[0190] According to the above embodiments, the first device indicates the multiplexing model (the third model) it supports to the second device by transmitting the third message. Optionally, the number of third models can be greater than or equal to 1. Then, the first device can determine the first model from the supported multiplexing models. That is, the sender device actively negotiates the available models by sending the third message to the receiver device.

[0191] In some embodiments, the transmission method performed by the second device further comprises: the second device sending a fourth message to the first device; wherein the fourth message is used to indicate whether the second device adapts to the third model.

[0192] Correspondingly, in some embodiments, the transmission method performed by the first device further comprises: the first device receiving a fourth message from the second device; wherein the fourth message is used to indicate whether the second device adapts to the third model.

[0193] According to the above embodiments, the second device indicates to the first device whether to adapt to the multiplexing model (third model) supported by the first device by transmitting the fourth message. In this way, the two-side negotiation is completed to finally determine the multiplexing model used.

[0194] In some embodiments, the transmission method performed by the first device further comprises:

[0195] If the fourth message indicates that the second device adapts to the third model, the first device determines the first model based on the third model; and / or,

[0196] If the fourth message indicates that the second device does not adapt to the third model, the first device determines the M models configured by the network device to select the first model from the M models.

[0197] For example, if the fourth message indicates that the second device adapts to the third model, the first device determines the first model based on the third model, for example, takes the third model as the first model, or selects one from a plurality of third models supported by the first device according to the source type as the first model. If the fourth message indicates that the second device adapts to the third model, the first device can fallback to transmit multiple sources in other ways, or select the model based on network configuration, for example, determine the first model by referring to the application examples shown in FIG. 11A or FIG. 11B.

[0198] In some embodiments, the third message comprises at least one of the following: an ID of the third model; a source type corresponding to the third model; and description information of a latent variable distribution associated with the third model. The ID of the third model and the source type can be used by the second device to determine the specific details of the third model supported by the first device according to the pre-defined model information, and then determine whether to adapt to the third model; the latent variable distribution associated with the third model can be used by the second device to determine whether to adapt to the third model. In actual application, the third message can include one or a combination of multiple information as above, which can be set according to scene requirements, system conventions or protocol conventions. For the sake of brevity, they are not listed one by one here.

[0199] In some embodiments, the transmission method performed by the second device further includes: the second device sending a fifth message to the first device; wherein the fifth message is used to indicate the fourth model and / or related information of the fourth model, and the fourth model includes a model supported by the second device for multiplexing the multiple sources to be transmitted in the first device.

[0200] Correspondingly, in some embodiments, the transmission method performed by the first device further includes: the first device receiving the fifth message from the second device; wherein the fifth message is used to indicate the fourth model and / or related information of the fourth model, and the fourth model includes a model supported by the second device for multiplexing the multiple sources to be transmitted in the first device.

[0201] According to the above embodiments, the second device indicates the multiplexing model (the fourth model) supported by the second device to the first device by transmitting the fifth message. Optionally, the number of the fourth model can be greater than or equal to 1. Then, the first device can determine the first model from the multiplexing model supported by the second device. That is, the receiving end device transmits the fifth message to the sending end device to negotiate the available model.

[0202] In some embodiments, the transmission method performed by the first device further includes: the first device sending a sixth message to the second device; wherein the sixth message is used to indicate whether the first device adapts the fourth model.

[0203] Correspondingly, in some embodiments, the transmission method performed by the second device further includes: the second device receiving the sixth message from the first device; wherein the sixth message is used to indicate whether the first device adapts the fourth model.

[0204] According to the above embodiments, the first device indicates whether to adapt the multiplexing model (the fourth model) supported by the second device to the second device by transmitting the sixth message. In this way, the two sides negotiate to finally determine the multiplexing model to be used.

[0205] In some embodiments, the transmission method performed by the first device further includes:

[0206] If the first device adapts the fourth model, the first device determines the first model based on the fourth model; and / or,

[0207] If the first device does not adapt the third model, the first device determines the M models configured by the network device to select the first model from the M models.

[0208] Exemplarily, if the first device adapts to the third model, the first device determines the first model based on the fourth model, for example, takes the fourth model as the first model, or selects one of the plurality of fourth models supported by the first device as the first model according to the source type. If the first device does not adapt to the third model, in addition to informing the second device, the first device can also fallback to transmit the plurality of sources in other ways, or select the model based on network configuration, for example, determine the first model by referring to the application examples shown in FIG. 11A or FIG. 11B.

[0209] In some embodiments, the fifth message includes at least one of the following: an ID of the fourth model; a source type corresponding to the fourth model; and description information of a latent variable distribution associated with the fourth model. The ID of the fourth model and the source type can be used by the first device to determine the specific details of the fourth model supported by the second device according to the predefined model information, and then determine whether to adapt to the fourth model; the latent variable distribution associated with the fourth model can be used by the first device to determine whether to adapt to the fourth model. In actual applications, the fifth message can include one or a combination of the above information, which can be set according to the scene requirements, system conventions or protocol conventions. For the sake of brevity, they are not listed one by one here.

[0210] In order to facilitate understanding of the above-mentioned model selection method based on bilateral negotiation, specific application examples are provided below. In this application example, the transmission method includes an initialization phase and a use phase.

[0211] In the initialization phase, the network device can inform the terminal device of the third model supported by the network device through the third message; then, the terminal device feeds back whether to adapt to the third model through the fourth message based on the received third message; if it is confirmed that the terminal device adapts, data transmission can be started; otherwise, network configuration-based initialization is started.

[0212] Alternatively, the terminal device can inform the network device of the fourth model supported by the terminal device through the fifth message; the network device can issue whether to adapt to the fourth model through the sixth message based on the received fifth message, and if it adapts, data transmission can be started; otherwise, network configuration-based initialization is started.

[0213] The above starting data transmission can be source multiplexing transmission based on a model by the network device to the terminal device, that is, the network device as the first device (sending end) can actively send the third message to the terminal device to complete the model selection based on the double-sided negotiation through the interaction of the third message and the fourth message; the network device can also receive the fifth message sent by the terminal device to complete the model selection based on the double-sided negotiation through the interaction of the fifth message and the sixth message. It can be understood that the source processing of the terminal device can also adopt a similar manner. That is, the terminal device as the first device (sending end) can actively send the third message to the network device to complete the model selection based on the double-sided negotiation through the interaction of the third message and the fourth message; or receive the fifth message sent by the network device to complete the model selection based on the double-sided negotiation through the interaction of the fifth message and the sixth message.

[0214] The indication form of the model supported by the network device and the model supported by the terminal device can include the ID of the model and the source type. Optionally, the model information notified by the network device to the terminal device can include a description of the latent variable distribution, and the description of the latent variable distribution can be a standard pre-defined ID.

[0215] In the use phase, on the basis of the above double-sided negotiation, the sending end (the first device) selects a suitable multiplexing model (the first model) according to the source type, and transmits an indication related to the demultiplexing model (the second model) to the receiving end through the first message; the receiving end (the second device) uses the corresponding demultiplexing model for demultiplexing according to the received indication.

[0216] Specifically, FIG. 12 shows a schematic diagram of an initialization phase initiated by downlink. As shown in FIG. 12, the initialization phase initiated by downlink includes the following processing:

[0217] A. Model library preparation: Constructing a multiplexing and demultiplexing model library. These models are classified according to different application scenarios (such as text, image, video, etc.), and are assigned a unique model ID. Each model is associated with its corresponding latent variable distribution, which can be a pre-defined ID based on industry standards.

[0218] B. Constructing and sending a message for indicating supported models: The network device constructs a message for indicating supported models. The message includes a list of all supported models, the ID of each model, and a description of the associated latent variable distribution. The message can be:

[0219] MIB: The MIB can be used to notify the terminal device of the information about the model supported by the network device, and guide the terminal device to obtain more detailed model information;

[0220] SIB: The MIB can be used to inform the terminal device about the network device support model and guide the terminal device to obtain more detailed model information;

[0221] Paging message: In certain cases, such as when the network device needs to deliver an emergency update or a specific model configuration to a specific terminal device, a paging message can be used. It is usually used to wake up the terminal device when it is in idle mode and inform it to receive important information;

[0222] Dedicated signaling message: For terminal devices that have established a connection, the network can send model configuration information directly through a dedicated signaling message. It is usually used for personalized services or quick updates of model information, such as when the terminal device is performing a specific task or application startup;

[0223] RRC reconfiguration message: In the case of an established RRC connection, the network side can specify or update the model configuration through an RRC reconfiguration message;

[0224] RRC connection establishment message: During the RRC connection establishment process, the network side can embed model configuration information.

[0225] C. Adaptation analysis of terminal device: After the terminal device receives the message indicating the supported model, it will parse the message content to identify the model supported by the network device. The terminal device matches the model ID and implicit variable distribution provided in the message according to the internally pre-stored model information or capabilities. If the terminal device supports one or more models mentioned in the broadcast, it will perform internal configuration and prepare to use these models for subsequent data communication.

[0226] D. Adaptation feedback: For the identified and supported model, the terminal device feeds back its model adaptation status to the network device through a feedback message. This includes confirmation of the adapted model ID and any related configuration. The feedback message can be: uplink control information (UCI) or an RRC message.

[0227] E. Fallback mechanism: The network device makes subsequent instructions according to the feedback of the terminal device, such as confirming adaptation, then starting data transmission, specifically, it can be: otherwise, start network-based initialization. Here, data transmission can be the network device sending data to the terminal device, or the terminal device sending data to the network device. It can be understood that the aforementioned message indicating the supported model can be the third message or the fifth message in the aforementioned embodiments; the aforementioned feedback message can be the fourth message or the sixth message in the aforementioned embodiments.

[0228] Figure 13 shows a schematic diagram of an initialization phase initiated by uplink. As shown in Figure 13, the initialization phase initiated by uplink includes the following processes:

[0229] A. Model library preparation: the terminal device loads the multiplexing and demultiplexing model information in its model library at initialization, including the model ID, source form, and related hidden variable distribution description. Each model is assigned a unique identifier (ID), and the model ID can be based on standard predefinition.

[0230] B. Building and sending a message for indicating supported models: the terminal device builds a message for indicating supported models and sends it to the network device, which contains the model information supported by the terminal, including the model ID, source form, and hidden variable distribution description. In addition, the message can also contain the capability information of the terminal device, indicating the model type that can be processed, the maximum number of models, or the version, etc. The message for indicating supported models can be: an RRC message (such as an RRC connection reconfiguration complete message or an RRC connection setup complete message), a UCI message.

[0231] C. Network device model adaptation analysis: after receiving the message for indicating supported models, the network device first checks the model information supported by the terminal device and compares it with the model library of the network device.

[0232] D. Adaptation feedback: the network device builds and sends a feedback message for feedback on the model adaptation status. If the adaptation is successful, the message will contain the confirmed model ID and related configuration information; otherwise, the terminal device is instructed to enter the network configuration-based model selection process. The feedback message can be: a DCI message, a dedicated signaling message, an RRC reconfiguration message: an RRC connection setup message.

[0233] E. Backoff mechanism: if the terminal device confirms the model adaptation after receiving the feedback message, the two ends start using the common semantic model for data transmission. Otherwise, the network configuration-based initialization is started. Here, the data transmission can be the network device sending data to the terminal device, or the terminal device sending data to the network device. It can be understood that the above-mentioned message for indicating supported models can be the third message or the fifth message in the foregoing embodiments; the above-mentioned feedback message can be the fourth message or the sixth message in the foregoing embodiments.

[0234] It can be seen that the embodiments of the present application give a two-side negotiation-based initialization mode, which ensures the effective use of multiplexing models.

[0235] In some embodiments, the above transmission method further comprises: the first device and / or the second device monitoring a first performance indicator; wherein the first performance indicator is related to the communication state and / or semantic transmission; the first performance indicator is used to determine whether to trigger the fallback mechanism of source multiplexing.

[0236] That is, in the communication system, the first performance indicator can be monitored by either or both of the data transmission ends, so that the fallback mechanism of the source multiplexing is triggered based on the first performance indicator, to avoid continuous use of the low-performance multiplexing mode in the case of performance degradation of the semantic-based source multiplexing mode, thereby guaranteeing the communication performance.

[0237] Exemplarily, the fallback mechanism of the source multiplexing includes stopping multiplexing the multiple sources based on the first model. Optionally, the fallback mechanism can further include multiplexing the to-be-transmitted sources using a first multiplexing mode; wherein the first multiplexing mode is a pre-configured or default multiplexing mode, or the first multiplexing mode can also be configured by the network device.

[0238] Exemplarily, the first performance indicator can include at least one of a semantic error rate, a transmission delay, and semantic integrity. These performance indicators can reflect whether the processing of the multiple sources based on the model can accurately restore the semantics of the multiple sources, and whether it can effectively save transmission resources and improve transmission efficiency. Therefore, determining whether to trigger the fallback mechanism of the source multiplexing based on these performance indicators can guarantee the communication performance.

[0239] Exemplarily, the first performance indicator can be monitored by the receiving end, i.e., the second device. For example, for downlink transmission, the first performance indicator is monitored by the receiving end (terminal device); for uplink transmission, the first performance indicator is monitored by the receiving end (network device), so that the accuracy and availability of the first performance indicator can be ensured.

[0240] In some embodiments, the receiving end (second device) includes a terminal device; and the transmission method further includes: the terminal device sends a seventh message to the network device; wherein the seventh message is used to report the first performance indicator, so that the network device determines whether to trigger the fallback mechanism of the source multiplexing.

[0241] According to the above embodiment, if the terminal device is the receiving end, the first performance indicator is monitored by the terminal device, and the seventh message can be sent by the terminal device to the network device to report the first performance indicator, so that the network device determines whether to trigger the fallback mechanism of the source multiplexing.

[0242] In some embodiments, the transmission method performed by the second device further includes: the terminal device, in the case of receiving an eighth message from the network device, uses the first multiplexing mode to multiplex the to-be-transmitted sources; wherein the eighth message is used to indicate that the fallback mechanism of the source multiplexing is triggered.

[0243] Specifically, the eighth message is sent by the network device when it determines that the fallback mechanism needs to be triggered based on the first performance indicator reported by the terminal device (the second device). The eighth message can be understood as a fallback indication issued by the network device. The eighth message triggers the terminal device to fallback to using the first multiplexing manner, which can include at least one of time division multiplexing, frequency division multiplexing, code division multiplexing, and space division multiplexing. By timely fallback to the traditional scheme, the communication performance can be guaranteed under the condition of performance degradation of semantic consumption.

[0244] The above embodiments are based on monitoring of the first performance indicator by the receiving end (the second device), and introduce the processing manner of the corresponding receiving end for the case where the receiving end is a terminal device or a network device (i.e., downlink transmission or uplink transmission). The following embodiments provide the processing manner of the sending end.

[0245] In some embodiments, the first device includes a terminal device; and the transmission method further includes: in a case where the terminal device receives the eighth message from the network device, the terminal device falls back to multiplexing the to-be-transmitted source using the first multiplexing manner; and the eighth message is used to indicate the fallback mechanism of triggering source multiplexing. That is, in the case where the sending end is a terminal device, i.e., for the case of uplink transmission, the sending end can wait for the eighth message issued by the network device, and falls back to using the first multiplexing manner in a case where the eighth message is received.

[0246] In some embodiments, the first device includes a network device; and the transmission method further includes: the network device receives the seventh message from the terminal device; the seventh message is used to report the first performance indicator; the first performance indicator is related to the communication state and / or semantic transmission; and the first performance indicator is used by the network device to determine whether to trigger the fallback mechanism of source multiplexing. That is, for downlink transmission, the network device can determine whether to trigger the fallback mechanism of source multiplexing based on the first performance indicator reported by the terminal device.

[0247] In some embodiments, the transmission method further includes: in a case where the network device determines to trigger the fallback mechanism of source multiplexing, the network device sends the eighth message to the terminal device. The eighth message can be understood as a fallback indication issued by the network device. Through the indication of the network device, the communication system can timely fallback to using the traditional scheme for multiplexing.

[0248] As described above, the first performance indicator can also be monitored by the receiving end or by both ends. In a case where the sending end monitors the first performance indicator, if the first device includes a terminal device, the terminal device can also send the seventh message to the network device to indicate the first performance indicator, so that the network device determines whether to trigger the fallback mechanism and whether to issue the eighth message; and if the first device includes a network device, the network device can directly determine whether to trigger the fallback mechanism and issue the eighth message according to the monitored first performance indicator.

[0249] Optionally, the seventh message reported by the terminal device can comprise uplink control information (UCI).

[0250] Optionally, the eighth message issued by the network device can comprise downlink control information (DCI).

[0251] To facilitate understanding of the above technical solutions, specific application examples are provided below in conjunction with the accompanying drawings. In the following application examples, the first performance indicator is monitored by the receiving end (second device).

[0252] FIG. 14 is a schematic diagram of the automatic fallback mechanism for downlink transmission in the embodiments of the present application. As shown in FIG. 14, for downlink transmission:

[0253] (a) The terminal device monitors the multiplexing communication state and the first performance indicator in real time, and periodically reports the current multiplexing communication state and the first performance indicator to the base station through the seventh message;

[0254] (b) The network device analyzes the content of the seventh message to determine whether the fallback mechanism needs to be triggered;

[0255] (c) Once it is determined that fallback is needed, the network device sends a switching instruction to the terminal device through the eighth message to fallback to the traditional multiplexing scheme.

[0256] FIG. 15 is a schematic diagram of the automatic fallback mechanism for uplink transmission in the embodiments of the present application. As shown in FIG. 15, for uplink transmission:

[0257] (a) The network device monitors the multiplexing communication state and the performance indicator in real time to determine whether the fallback mechanism needs to be triggered;

[0258] (b) Once it is determined that fallback is needed, the network device sends a switching instruction to the terminal device through the eighth message to fallback to the traditional multiplexing scheme.

[0259] The above application examples further consider using the fallback mechanism to ensure smooth transition to the traditional multiplexing scheme when the nonlinear multiplexing efficiency is reduced. For downlink transmission, the terminal device periodically reports the current communication state and performance indicator to the base station through uplink control information (UCI). The network device analyzes the received UCI data to determine whether the fallback mechanism needs to be triggered. Once it is decided to fallback, all related terminal devices are notified through DCI instructions. For uplink transmission, the network device monitors the communication state and performance indicator in real time to determine whether the fallback mechanism needs to be triggered. Once it is decided to fallback, all related terminal devices are notified through DCI instructions.

[0260] In the above application examples, the first performance indicator involved in semantic effect monitoring can be a key indicator for the system to monitor the semantic transmission in real time. For example, the first performance indicator can include at least one of a semantic error rate, a transmission delay, and information integrity. Thresholds are set for each indicator based on historical data and expected performance. When the monitored indicator continuously exceeds the threshold, for example, the semantic error rate exceeds 5% for 5 consecutive minutes, a fallback mechanism can be triggered.

[0261] In the above application examples, the collected semantic evaluation data will be analyzed to determine whether fallback is needed. For example, the distance between the hidden space vectors of the transmitted semantics and the received semantics is used as the semantic similarity;

[0262] In the above application examples, once it is decided to fallback, the network device can send an indication of encoding switching (multiplexing mode switching) to the user equipment through an eighth message, for example, downlink control information (DCI).

[0263] It can be seen that the embodiments of the present application first use the semantic differences between different information sources based on AI to realize nonlinear multiplexing transmission for multiple information sources, and then use AI at the receiving end to demultiplex and recover the different information sources after multiplexing. This method can perform multiplexing transmission in a new dimension beyond time, frequency, code, and space, i.e., the semantic dimension, for multiple information sources, which can greatly save transmission resources in time, frequency, code, and space, and improve the transmission efficiency of data. Further, the AI / ML model / function / feature used for multiplexing transmission at the sending end and the receiving end needs to be adapted according to different multiplexing contents. The embodiments of the present application provide initialization based on network configuration, initialization based on negotiation between both sides, dynamic selection of subsequent AI / ML model / function / feature, and fallback mechanism, which ensures the effective use of the multiplexing model.

[0264] FIG. 16 is a schematic block diagram of a first device 1600 according to an embodiment of the present application. The first device 1600 can include:

[0265] The first processing module 1610 is configured to process the multiple information sources based on a first model to obtain first transmission information.

[0266] The first communication module 1620 is configured to send the first transmission information to a second device; wherein the first transmission information is used by the second device to recover the multiple information sources based on a second model.

[0267] In some embodiments, the first model includes a first AI model.

[0268] In some embodiments, a first information source in the multiple information sources is an information source with semantic information; and a second information source in the multiple information sources is a bit stream or an information source with semantic information.

[0269] In some embodiments, the sources with semantic information include CSI or user data.

[0270] In some embodiments, the bitstream includes a system information bitstream or an encoded bitstream based on user data.

[0271] In some embodiments, the system information bitstream includes information for control or scheduling.

[0272] In some embodiments, the system information bitstream includes broadcast information.

[0273] In some embodiments, the user data includes images, voice, text or video.

[0274] In some embodiments, the first source and the second source have different semantic information.

[0275] In some embodiments, the second model includes N models, an i-th model in the N models is used to recover a third source in the plurality of sources based on the first transmission information, N is an integer greater than or equal to 2, and i is a positive integer less than or equal to N.

[0276] In some embodiments, the first communication module 1620 is further configured to:

[0277] select a first model from the M models, and send a first message to the second device, wherein the M models are configured by the network device, and the first message is used for the second device to determine a second model corresponding to the first model.

[0278] In some embodiments, the first processing module 1610 is further configured to:

[0279] select a first model from the M models according to a source type of each source in the plurality of sources.

[0280] In some embodiments, the first message includes at least one of the following information:

[0281] a number of the plurality of sources;

[0282] a source type of each source in the plurality of sources.

[0283] In some embodiments, the first message is sent through a control channel.

[0284] In some embodiments, the M models are configured by the network device by sending a second message to the terminal device.

[0285] In some embodiments, the second message is sent through broadcasting.

[0286] In some embodiments, the second message includes at least one of the following information:

[0287] an identification ID of each of the M models;

[0288] information for determining a parameter and / or a structure of each model;

[0289] a source type corresponding to each model.

[0290] In some embodiments, the information for determining the parameter and / or the structure of each model comprises at least one of:

[0291] compressed information of the parameter and / or the structure of each model;

[0292] information for downloading the parameter and / or the structure of each model;

[0293] a training manner of each model.

[0294] In some embodiments, the first communication module 1620 is further configured to:

[0295] send a third message to the second device, wherein the third message is used to indicate a third model and / or related information of the third model, and the third model comprises a model supported by the first device and used for multiplexing the multiple sources to be transmitted in the first device.

[0296] In some embodiments, the first communication module 1620 is further configured to:

[0297] receive a fourth message from the second device, wherein the fourth message is used to indicate whether the second device adapts to the third model.

[0298] In some embodiments, the first processing module 1610 is further configured to:

[0299] if the fourth message indicates that the second device adapts to the third model, determine the first model based on the third model; and / or,

[0300] if the fourth message indicates that the second device does not adapt to the third model, determine the M models configured by the network device, and select the first model from the M models.

[0301] In some embodiments, the third message comprises at least one of:

[0302] an ID of the third model;

[0303] a source type corresponding to the third model;

[0304] description information of a latent variable distribution associated with the third model.

[0305] In some embodiments, the first communication module 1620 is further configured to:

[0306] receive a fifth message from the second device; wherein the fifth message is used to indicate the fourth model and / or related information of the fourth model, the fourth model comprising a model supported by the second device for multiplexing a plurality of sources to be transmitted in the second device.

[0307] In some embodiments, the first communication module 1620 is further configured to:

[0308] send a sixth message to the second device; wherein the sixth message is used to indicate whether the first device adapts the fourth model.

[0309] In some embodiments, the fifth message comprises at least one of:

[0310] an ID of the fourth model;

[0311] a source type corresponding to the fourth model;

[0312] description information of a latent variable distribution associated with the fourth model.

[0313] In some embodiments, the first processing module 1610 is further configured to:

[0314] monitor a first performance indicator; wherein the first performance indicator is related to a communication state and / or semantic transmission; and the first performance indicator is used to determine whether to trigger a fallback mechanism of source multiplexing.

[0315] In some embodiments, the first device comprises a terminal device; and the first communication module 1620 is further configured to:

[0316] send a seventh message to a network device; wherein the seventh message is used to report the first performance indicator, so that the network device determines whether to trigger the fallback mechanism of source multiplexing.

[0317] In some embodiments, the first processing module 1610 is further configured to:

[0318] in a case where an eighth message from the network device is received, fallback to multiplexing the sources to be transmitted using a first multiplexing manner; wherein the eighth message is used to indicate to trigger the fallback mechanism of source multiplexing.

[0319] In some embodiments, the first device comprises a network device; and the first processing module 1610 is further configured to:

[0320] receive a seventh message from a terminal device; wherein the seventh message is used to report the first performance indicator; the first performance indicator is related to a communication state and / or semantic transmission; and the first performance indicator is used by the network device to determine whether to trigger the fallback mechanism of source multiplexing.

[0321] In some embodiments, the first communication module 1620 is further configured to:

[0322] The eighth message is sent to the terminal device in a case where a fallback mechanism for determining a trigger source of multiplexing is determined.

[0323] In some embodiments, the first performance indicator includes at least one of a semantic error rate, a transmission delay, and semantic integrity.

[0324] In some embodiments, the first transmission information is a bit stream or a symbol sequence.

[0325] The first device 1600 of the embodiments of the present application can realize the corresponding functions of the first device in the foregoing method embodiments. The processes, functions, implementation manners, and beneficial effects of the respective modules (sub-modules, units, or components, etc.) in the first device 1600 can be referred to the corresponding descriptions in the foregoing method embodiments, which will not be described here again. It should be noted that the functions described with respect to the respective modules (sub-modules, units, or components, etc.) in the first device 1600 of the embodiments of the present application can be realized by different modules (sub-modules, units, or components, etc.), or can be realized by the same module (sub-module, unit, or component, etc.).

[0326] FIG. 17 is a schematic block diagram of a second device 1700 according to an embodiment of the present application. The second device 1700 can include:

[0327] The second communication module 1710 is configured to receive first transmission information from the first device, wherein the first transmission information is obtained by processing a plurality of sources based on a first model.

[0328] The second processing module 1720 is configured to process the first transmission information based on a second model to recover the plurality of sources.

[0329] In some embodiments, the second model includes a second AI model.

[0330] In some embodiments, a first source of the plurality of sources is a source with semantic information; and a second source of the plurality of sources is a bit stream or a source with semantic information.

[0331] In some embodiments, the source with semantic information includes CSI or user data.

[0332] In some embodiments, the bit stream includes a system information bit stream or an encoded bit stream based on user data.

[0333] In some embodiments, the system information bit stream includes information for control or scheduling.

[0334] In some embodiments, the system information bit stream includes broadcast information.

[0335] In some embodiments, the user data includes images, voice, text or video.

[0336] In some embodiments, the first source and the second source have different semantic information.

[0337] In some embodiments, the second model includes N models, an i-th model in the N models is used to recover a third source in the multiple sources based on the first transmission information, N is an integer greater than or equal to 2, and i is a positive integer less than or equal to N.

[0338] In some embodiments, the second communication module 1710 is further configured to:

[0339] receive a first message from the first device; wherein the first message is used by the second device to determine a second model corresponding to a first model selected by the first device from M models; wherein the M models are configured by the network device.

[0340] In some embodiments, the first message includes at least one of the following information: a number of sources in the multiple sources; a source type of each source in the multiple sources.

[0341] In some embodiments, the first message is sent through a control channel.

[0342] In some embodiments, the M models are configured by the network device by sending a second message to the terminal device.

[0343] In some embodiments, the second message is sent through broadcasting.

[0344] In some embodiments, the second message includes at least one of the following: an identification ID of each model in the M models; information used to determine parameters and / or structures of each model; a source type corresponding to each model.

[0345] In some embodiments, the information used to determine parameters and / or structures of each model includes at least one of the following: compressed information of parameters and / or structures of each model; information used to download parameters and / or structures of each model; a training method of each model.

[0346] In some embodiments, the second communication module 1710 is further configured to:

[0347] receive a third message from the first device; wherein the third message is used to indicate a third model and / or related information of the third model, and the third model includes a model supported by the first device for multiplexing the multiple sources to be transmitted in the first device.

[0348] In some embodiments, the second communication module 1710 is further configured to:

[0349] sending a fourth message to the first device; wherein the fourth message is used to indicate whether the second device adapts the third model.

[0350] In some embodiments, the third message comprises at least one of: an ID of the third model; a source type corresponding to the third model; and description information of a latent variable distribution associated with the third model.

[0351] In some embodiments, the second communication module 1710 is further configured to:

[0352] sending a fifth message to the first device; wherein the fifth message is used to indicate the fourth model and / or related information of the fourth model, and the fourth model comprises a model supported by the second device for multiplexing a plurality of sources to be transmitted in the second device.

[0353] In some embodiments, the second communication module 1710 is further configured to:

[0354] receiving a sixth message from the first device; wherein the sixth message is used to indicate whether the first device adapts the fourth model.

[0355] In some embodiments, the fifth message comprises at least one of: an ID of the fourth model; a source type corresponding to the fourth model; and description information of a latent variable distribution associated with the fourth model.

[0356] In some embodiments, the second processing module 1720 is further configured to:

[0357] monitoring a first performance indicator; wherein the first performance indicator is related to a communication state and / or semantic transmission; and the first performance indicator is used to determine whether to trigger a fallback mechanism of source multiplexing.

[0358] In some embodiments, the second device comprises a terminal device; and the second communication module 1710 is further configured to:

[0359] sending a seventh message to a network device; wherein the seventh message is used to report the first performance indicator, so that the network device determines whether to trigger the fallback mechanism of source multiplexing.

[0360] In some embodiments, the second processing module 1720 is further configured to:

[0361] in a case where an eighth message from the network device is received, fallback to multiplexing the sources to be transmitted using a first multiplexing manner; wherein the eighth message is used to indicate to trigger the fallback mechanism of source multiplexing.

[0362] In some embodiments, the second device comprises a network device; and the second processing module 1720 is further configured to:

[0363] determining whether to trigger the fallback mechanism of source multiplexing based on the first performance indicator.

[0364] In some embodiments, the second communication module 1710 is further configured to:

[0365] In a case where it is determined that the fallback mechanism of triggering source multiplexing, send an eighth message to the terminal device.

[0366] In some embodiments, the first performance indicator includes at least one of a semantic error rate, a transmission delay, and semantic integrity.

[0367] The second device 1700 of the embodiments of the present application can realize the corresponding functions of the second device in the foregoing method embodiments. The processes, functions, implementation manners, and beneficial effects of the various modules (sub-modules, units, or components, etc.) in the second device 1700 can be referred to the corresponding descriptions in the foregoing method embodiments, which will not be described here. It should be noted that the functions described with respect to the various modules (sub-modules, units, or components, etc.) in the second device 1700 of the embodiments of the present application can be realized by different modules (sub-modules, units, or components, etc.), or can be realized by the same module (sub-module, unit, or component, etc.).

[0368] FIG. 18 is a schematic structural diagram of a communication device 1800 according to an embodiment of the present application. The communication device 1800 includes a processor 1810, which can call and run a computer program from a memory to enable the communication device 1800 to implement the method in the embodiments of the present application.

[0369] In an implementation manner, the communication device 1800 can further include a memory 1820. The processor 1810 can call and run a computer program from the memory 1820 to enable the communication device 1800 to implement the method in the embodiments of the present application.

[0370] The memory 1820 can be a separate device independent of the processor 1810, or can be integrated in the processor 1810.

[0371] In an implementation manner, the communication device 1800 can further include a transceiver 1830, and the processor 1810 can control the transceiver 1830 to communicate with other devices, specifically, to send information or data to other devices, or receive information or data sent by other devices.

[0372] The transceiver 1830 can include a transmitter and a receiver. The transceiver 1830 can further include an antenna, and the number of antennas can be one or more.

[0373] In an embodiment, the communication device 1800 can be a first device of the embodiments of the present application, and the communication device 1800 can implement the corresponding procedures implemented by the first device in each method of the embodiments of the present application. For brevity, details are not repeated here.

[0374] In an embodiment, the communication device 1800 can be a second device of the embodiments of the present application, and the communication device 1800 can implement the corresponding procedures implemented by the second device in each method of the embodiments of the present application. For brevity, details are not repeated here.

[0375] FIG. 19 is a schematic structural diagram of a chip 1900 according to an embodiment of the present application. The chip 1900 includes a processor 1910, which can call and run a computer program from a memory to implement the method in the embodiments of the present application.

[0376] In an embodiment, the chip 1900 can further include a memory 1920. The processor 1910 can call and run a computer program from the memory 1920 to implement the method performed by the first device or the second device in the embodiments of the present application.

[0377] The memory 1920 can be a separate device independent of the processor 1910, or can be integrated in the processor 1910.

[0378] In an embodiment, the chip 1900 can further include an input interface 1930. The processor 1910 can control the input interface 1930 to communicate with other devices or chips, and specifically, can obtain information or data sent by other devices or chips.

[0379] In an embodiment, the chip 1900 can further include an output interface 1940. The processor 1910 can control the output interface 1940 to communicate with other devices or chips, and specifically, can output information or data to other devices or chips.

[0380] In an embodiment, the chip can be applied to the first device in the embodiments of the present application, and the chip can implement the corresponding procedures implemented by the first device in each method of the embodiments of the present application. For brevity, details are not repeated here.

[0381] In an embodiment, the chip can be applied to the second device in the embodiments of the present application, and the chip can implement the corresponding procedures implemented by the second device in each method of the embodiments of the present application. For brevity, details are not repeated here.

[0382] The chip applied to the first device and the second device can be the same chip or different chips.

[0383] It should be appreciated that the chip mentioned in the embodiments of the present application can also be referred to as a system chip, a system chip, a chip system or a system on chip, etc.

[0384] The processor mentioned above can be a general purpose processor, a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC) or other programmable logic device, a transistor logic device, a discrete hardware component, etc. Among them, the general purpose processor mentioned above can be a microprocessor or any conventional processor, etc.

[0385] The memory mentioned above can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable ROM (EPROM), an electrically EPROM (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM).

[0386] It should be understood that the above-mentioned memory is an example but not a limiting description, for example, the memory in the embodiments of the present application can also be a static RAM (SRAM), a dynamic RAM (DRAM), a synchronous DRAM (SDRAM), a double data rate SDRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a synch link DRAM (SLDRAM) and a direct memory bus random access memory (Direct Rambus RAM, DRRAM), etc. That is, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable type of memory.

[0387] FIG. 20 is a schematic block diagram of a communication system 2000 according to an embodiment of the present application. The communication system 2000 includes a first device 2110 and a second device 2120.

[0388] The first device 2110 is configured to process the multiple sources based on the first model to obtain first transmission information.

[0389] The first device 2110 is further configured to send the first transmission information to the second device.

[0390] The second device 2120 is configured to receive the first transmission information from the first device, and process the first transmission information based on the second model to recover the multiple sources.

[0391] The first device 2110 can be configured to implement the corresponding functions of the first device in the above method, and the second device 2120 can be configured to implement the corresponding functions of the second device in the above method. For brevity, details are not repeated here.

[0392] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed by a computer, the computer instructions generate the processes or functions in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes one or more available media sets. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), optical medium (for example, DVD), or semiconductor medium (for example, solid state disk (SSD)) and the like.

[0393] It should be understood that in various embodiments of the present application, the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0394] Those skilled in the art can clearly understand the specific working process of the system, device and unit described above for the convenience and brevity of description, which can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0395] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A transmission method, comprising: The first device processes multiple information sources based on the first model to obtain the first transmission information; The first device sends the first transmission information to the second device; wherein the first transmission information is used by the second device to recover the plurality of information sources based on the second model.

2. The method according to claim 1, wherein, The first model includes the first artificial intelligence (AI) model.

3. The method according to claim 1 or 2, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

4. The method according to claim 3, wherein, The information sources with semantic information include Channel State Information (CSI) or user data.

5. The method according to claim 3 or 4, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

6. The method according to claim 5, wherein, The system information bitstream includes information used for control or scheduling.

7. The method according to any one of claims 4-6, wherein, The user data includes images, voice, text, or video.

8. The method according to any one of claims 3-7, wherein, The first information source and the second information source have different semantic information.

9. The method according to any one of claims 1-8, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source among the multiple sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

10. The method according to any one of claims 1-9, wherein, The method further includes: The first device selects the first model from M models and sends a first message to the second device; where M is an integer greater than or equal to 1; the first message is used by the second device to determine the second model corresponding to the first model.

11. The method according to claim 10, wherein, The first device selects the first model from M models, including: The first device selects the first model from the M models based on the source type of each of the plurality of sources.

12. The method according to claim 10 or 11, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

13. The method according to any one of claims 10-12, wherein, The first message is sent via the control channel.

14. The method according to any one of claims 10-13, wherein, The M models are configured by the network device by sending a second message to the terminal device.

15. The method according to claim 14, wherein, The second message was sent via broadcast.

16. The method according to claim 14 or 15, wherein, The second message includes at least one of the following: The identifier ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

17. The method according to claim 16, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

18. The method according to any one of claims 1-17, wherein, The method further includes: The first device sends a third message to the second device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

19. The method according to claim 18, wherein, The method further includes: The first device receives a fourth message from the second device; wherein the fourth message is used to indicate whether the second device is compatible with the third model.

20. The method according to claim 19, wherein, The method further includes: If the fourth message indicates that the second device adapts to the third model, then the first device determines the first model based on the third model; and / or, If the fourth message indicates that the second device is not compatible with the third model, then the first device determines M models configured by the network device to select the first model from the M models.

21. The method according to any one of claims 18-20, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

22. The method according to any one of claims 1-21, wherein, The method further includes: The first device receives a fifth message from the second device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

23. The method according to claim 22, wherein, The method further includes: The first device sends a sixth message to the second device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

24. The method according to claim 22 or 23, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

25. The method according to any one of claims 1-24, wherein, The method further includes: The first device monitors a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

26. The method according to claim 25, wherein, The first device includes a terminal device; the method further includes: The terminal device sends a seventh message to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

27. The method according to any one of claims 1-24, wherein, The first device includes a terminal device, and the method further includes: Upon receiving an eighth message from the network device, the terminal device reverts to using the first multiplexing method to multiplex the source to be transmitted; wherein the eighth message is used to indicate the revert mechanism that triggers source multiplexing.

28. The method according to any one of claims 1-24, wherein, The first device includes a network device; the method further includes: The network device receives a seventh message from a terminal device; wherein the seventh message is used to report a first performance indicator; the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used by the network device to determine whether to trigger the fallback mechanism of source multiplexing.

29. The method according to claim 28, wherein, The method further includes: When the network device determines that the fallback mechanism for triggering source multiplexing has been activated, it sends an eighth message to the terminal device.

30. The method according to any one of claims 25-29, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

31. The method according to any one of claims 1-30, wherein, The first transmitted information is a bit stream or a symbol sequence.

32. A transmission method, comprising: The second device receives first transmission information from the first device; wherein the first transmission information is obtained by processing multiple information sources based on a first model; The second device processes the first transmitted information based on the second model to recover the multiple information sources.

33. The method according to claim 32, wherein, The second model includes the second AI model.

34. The method according to claim 32 or 33, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

35. The method according to claim 34, wherein, The information sources with semantic information include CSI or user data.

36. The method according to claim 34 or 35, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

37. The method of claim 36, wherein, The system information bitstream includes information used for control or scheduling.

38. The method according to any one of claims 35-37, wherein, The user data includes images, voice, text, or video.

39. The method according to any one of claims 34-38, wherein, The first information source and the second information source have different semantic information.

40. The method according to any one of claims 32-39, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source among the multiple sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

41. The method according to any one of claims 32-40, wherein, The method further includes: The second device receives a first message from the first device; wherein the first message is used by the second device to determine the relationship with the... The second model is the first model selected by the first device from M models, where M is an integer greater than or equal to 1.

42. The method according to claim 41, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

43. The method according to claim 41 or 42, wherein, The first message is sent via the control channel.

44. The method according to any one of claims 41-43, wherein, The M models are configured by the network device by sending a second message to the terminal device.

45. The method according to claim 44, wherein, The second message was sent via broadcast.

46. ​​The method according to claim 44 or 45, wherein, The second message includes at least one of the following: The ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

47. The method according to claim 46, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

48. The method according to any one of claims 32-47, wherein, The method further includes: The second device receives a third message from the first device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

49. The method according to claim 48, wherein, The method further includes: The second device sends a fourth message to the first device; wherein the fourth message is used to indicate whether the second device is compatible with the third model.

50. The method according to claim 48 or 49, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

51. The method according to any one of claims 32-50, wherein, The method further includes: The second device sends a fifth message to the first device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

52. The method according to claim 51, wherein, The method further includes: The second device receives a sixth message from the first device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

53. The method according to claim 51 or 52, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

54. The method according to any one of claims 32-53, wherein, The method further includes: The second device monitors a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

55. The method according to claim 54, wherein, The second device includes a terminal device; the method further includes: The terminal device sends a seventh message to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

56. The method according to claim 55, wherein, The method further includes: Upon receiving an eighth message from the network device, the terminal device reverts to using the first multiplexing method to multiplex the source to be transmitted; wherein the eighth message is used to indicate the revert mechanism that triggers source multiplexing.

57. The method according to claim 54, wherein, The second device includes a network device; the method further includes: Based on the first performance metric, the network device determines whether to trigger the fallback mechanism for source multiplexing.

58. The method according to claim 57, wherein, The method further includes: When the network device determines that the fallback mechanism for triggering source multiplexing has been activated, it sends an eighth message to the terminal device.

59. The method according to any one of claims 54-58, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

60. A first device, comprising: The first processing module is used to process multiple information sources based on the first model to obtain the first transmission information; A first communication module is used to send the first transmission information to a second device; wherein the first transmission information is used by the second device to recover the plurality of information sources based on a second model.

61. The first device according to claim 60, wherein, The first model includes the first AI model.

62. The first device according to claim 60 or 61, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

63. The first device according to claim 62, wherein, The information sources with semantic information include CSI or user data.

64. The first device according to claim 62 or 63, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

65. The first device according to claim 64, wherein, The system information bitstream includes information used for control or scheduling.

66. The first device according to any one of claims 63-65, wherein, The user data includes images, voice, text, or video.

67. The first device according to any one of claims 62-66, wherein, The first information source and the second information source have different semantic information.

68. The first device according to any one of claims 60-67, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source of the plurality of sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

69. The first device according to any one of claims 60-68, wherein, The first communication module is further configured to: The first model is selected from M models, and a first message is sent to the second device; where M is an integer greater than or equal to 1; the first message is used by the second device to determine the second model corresponding to the first model.

70. The first device according to claim 69, wherein, The first processing module is further configured to: Based on the source type of each of the plurality of sources, the first model is selected from the M models.

71. The first device according to claim 69 or 70, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

72. The first device according to any one of claims 69-71, wherein, The first message is sent via the control channel.

73. The first device according to any one of claims 69-72, wherein, The M models are configured by the network device by sending a second message to the terminal device.

74. The first device according to claim 73, wherein, The second message was sent via broadcast.

75. The first device according to claim 73 or 74, wherein, The second message includes at least one of the following: The ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

76. The first device according to claim 75, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

77. The first device according to any one of claims 60-76, wherein, The first communication module is further configured to: Send a third message to the second device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

78. The first device according to claim 77, wherein, The first communication module is further configured to: A fourth message is received from the second device; wherein the fourth message is used to indicate whether the second device is compatible with the third model.

79. The first device according to claim 78, wherein, The first processing module is further configured to: If the fourth message indicates that the second device adapts to the third model, then the first model is determined based on the third model; and / or, If the fourth message indicates that the second device is not compatible with the third model, then M models configured by the network device are determined to select the first model from the M models.

80. The first device according to any one of claims 77-79, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

81. The first device according to any one of claims 60-80, wherein, The first communication module is further configured to: Receive a fifth message from the second device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

82. The first device according to claim 81, wherein, The first communication module is further configured to: A sixth message is sent to the second device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

83. The first device according to claim 81 or 82, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

84. The first device according to any one of claims 60-83, wherein, The first processing module is further configured to: The first device monitors a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

85. The first device according to claim 84, wherein, The first device includes a terminal device; the first communication module is further configured to: A seventh message is sent to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

86. The first device according to any one of claims 60-83, wherein, The first device includes a terminal device, and the first processing module is further configured to: Upon receiving an eighth message from the network device, the fallback mechanism reuses the source to be transmitted using a first multiplexing method; wherein the eighth message is used to indicate the fallback mechanism that triggers source multiplexing.

87. The first device according to any one of claims 60-83, wherein, The first device includes a network device; the first communication module is further configured to: Based on the first performance metric, determine whether to trigger the fallback mechanism for source multiplexing.

88. The first device according to claim 87, wherein, The first communication module is further configured to: The network device receives a seventh message from a terminal device; wherein the seventh message is used to report a first performance indicator; the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used by the network device to determine whether to trigger the fallback mechanism of source multiplexing.

89. The first device according to any one of claims 84-88, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

90. The first device according to any one of claims 60-89, wherein, The first transmitted information is a bit stream or a symbol sequence.

91. A second device, comprising: The second communication module is used to receive first transmission information from the first device; wherein the first transmission information is obtained by processing multiple information sources based on a first model; The second processing module is used to process the first transmitted information based on the second model in order to recover the multiple information sources.

92. The second device according to claim 91, wherein, The second model includes the second AI model.

93. The second device according to claim 91 or 92, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

94. The second device according to claim 93, wherein, The information sources with semantic information include CSI or user data.

95. The second device according to claim 93 or 94, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

96. The second device according to claim 95, wherein, The system information bitstream includes information used for control or scheduling.

97. The second device according to any one of claims 94-96, wherein, The user data includes images, voice, text, or video.

98. The second device according to any one of claims 93-97, wherein, The first information source and the second information source have different semantic information.

99. The second device according to any one of claims 91-98, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source among the multiple sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

100. The second device according to any one of claims 91-99, wherein, The second communication module is also used for: Receive a first message from the first device; wherein the first message is used by the second device to determine the second model corresponding to the first model selected by the first device from M models, where M is an integer greater than or equal to 1.

101. The second device according to claim 100, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

102. The second device according to claim 100 or 101, wherein, The first message is sent via the control channel.

103. The second device according to any one of claims 100-102, wherein, The M models are configured by the network device by sending a second message to the terminal device.

104. The second device according to claim 103, wherein, The second message was sent via broadcast.

105. The second device according to claim 103 or 104, wherein, The second message includes at least one of the following: The ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

106. The second device according to claim 105, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

107. The second device according to any one of claims 91-106, wherein, The second communication module is also used for: Receive a third message from the first device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

108. The second device according to claim 107, wherein, The second communication module is also used for: A fourth message is sent to the first device; wherein the fourth message is used to indicate whether the second device is compatible with the third model.

109. The second device according to claim 107 or 108, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

110. The second device according to any one of claims 91-109, wherein, The second communication module is also used for: Send a fifth message to the first device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

111. The second device according to claim 110, wherein, The second communication module is also used for: A sixth message is received from the first device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

112. The second device according to claim 110 or 111, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

113. The second device according to any one of claims 91-112, wherein, The second processing module is further configured to: Monitor a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

114. The second device according to claim 113, wherein, The second device includes a terminal device; the second communication module is further configured to: A seventh message is sent to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

115. The second device according to claim 114, wherein, The second processing module is further configured to: Upon receiving an eighth message from the network device, the fallback mechanism reuses the source to be transmitted using a first multiplexing method; wherein the eighth message is used to indicate the fallback mechanism that triggers source multiplexing.

116. The second device according to claim 113, wherein, The second device includes a network device; the second processing module is further configured to: Based on the first performance metric, determine whether to trigger the fallback mechanism for source multiplexing.

117. The second device according to claim 116, wherein, The second communication module is also used for: If the fallback mechanism that triggers source multiplexing is determined, the eighth message is sent to the terminal device.

118. The second device according to any one of claims 113-117, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

119. A first device, comprising: The transceiver, processor, and memory, wherein the memory stores a computer program, the transceiver is used to communicate with other devices, and the processor invokes and runs the computer program stored in the memory to cause the first device to perform: Based on the first model, multiple information sources are processed to obtain the first transmission information; The first transmission information is sent to the second device; wherein the first transmission information is used by the second device to recover the plurality of information sources based on the second model.

120. The first device according to claim 119, wherein, The first model includes the first AI model.

121. The first device according to claim 119 or 120, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

122. The first device according to claim 121, wherein, The information sources with semantic information include CSI or user data.

123. The first device according to claim 121 or 122, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

124. The first device according to claim 123, wherein, The system information bitstream includes information used for control or scheduling.

125. The first device according to any one of claims 122-124, wherein, The user data includes images, voice, text, or video.

126. The first device according to any one of claims 121-125, wherein, The first information source and the second information source have different semantic information.

127. The first device according to any one of claims 119-126, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source among the multiple sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

128. The first device according to any one of claims 119-127, wherein, The processor is also configured to cause the first device to perform: Select the first model from the M models and send a first message to the second device; wherein the first message is used by the second device to determine the second model corresponding to the first model.

129. The first device according to claim 128, wherein, The processor is also configured to cause the first device to perform: Based on the source type of each of the plurality of sources, the first model is selected from the M models.

130. The first device according to claim 128 or 129, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

131. The first device according to any one of claims 128-130, wherein, The first message is sent via the control channel.

132. The first device according to any one of claims 128-131, wherein, The M models are configured by the network device by sending a second message to the terminal device.

133. The first device according to claim 132, wherein, The second message was sent via broadcast.

134. The first device according to claim 132 or 133, wherein, The second message includes at least one of the following: The ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

135. The first device according to claim 134, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

136. The first device according to any one of claims 119-135, wherein, The processor is also configured to cause the first device to perform: Send a third message to the second device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

137. The first device according to claim 136, wherein, The processor is also configured to cause the first device to perform: Receive a fourth message from the second device; wherein the fourth message is used to indicate whether the second device is compatible with the third... Model.

138. The first device according to claim 137, wherein, The processor is also configured to cause the first device to perform: If the fourth message indicates that the second device adapts to the third model, then the first model is determined based on the third model; and / or, If the fourth message indicates that the second device is not compatible with the third model, then M models configured by the network device are determined to select the first model from the M models.

139. The first device according to any one of claims 136-138, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

140. The first device according to any one of claims 119-139, wherein, The processor is also configured to cause the first device to perform: Receive a fifth message from the second device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

141. The first device according to claim 140, wherein, The processor is also configured to cause the first device to perform: A sixth message is sent to the second device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

142. The first device according to claim 140 or 141, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

143. The first device according to any one of claims 119-142, wherein, The processor is also configured to cause the first device to perform: Monitor a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

144. The first device according to claim 143, wherein, The first device includes a terminal device; the processor is further configured to cause the terminal device to perform: A seventh message is sent to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

145. The first device according to any one of claims 119-142, wherein, The first device includes a terminal device; the processor is further configured to cause the terminal device to perform: Upon receiving an eighth message from the network device, the fallback mechanism reuses the source to be transmitted using a first multiplexing method; wherein the eighth message is used to indicate the fallback mechanism that triggers source multiplexing.

146. The first device according to any one of claims 119-142, wherein, The first device includes a network device; the processor is further configured to cause the network device to perform: The network device receives a seventh message from a terminal device; wherein the seventh message is used to report a first performance indicator; the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used by the network device to determine whether to trigger the fallback mechanism of source multiplexing.

147. The first device according to claim 146, wherein, The processor is also configured to enable the network device to perform: If the fallback mechanism that triggers source multiplexing is determined, the eighth message is sent to the terminal device.

148. The first device according to any one of claims 143-147, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

149. The first device according to any one of claims 119-148, wherein, The first transmitted information is a bit stream or a symbol sequence.

150. A second device, comprising: The transceiver, processor, and memory, wherein the memory stores a computer program, the transceiver is used to communicate with other devices, and the processor invokes and runs the computer program stored in the memory to cause the second device to perform: Receive first transmission information from a first device; wherein the first transmission information is obtained by processing multiple information sources based on a first model; The first transmitted information is processed based on the second model to recover the multiple information sources.

151. The second device according to claim 150, wherein, The second model includes the second AI model.

152. The second device according to claim 150 or 151, wherein, The first source among the plurality of information sources is a source with semantic information; the second source among the plurality of information sources is a bit stream or a source with semantic information.

153. The second device according to claim 152, wherein, The information sources with semantic information include CSI or user data.

154. The second device according to claim 152 or 153, wherein, The bitstream includes system information bitstreams or encoded bitstreams obtained based on user data.

155. The second device according to claim 154, wherein, The system information bitstream includes information used for control or scheduling.

156. The second device according to any one of claims 153-155, wherein, The user data includes images, voice, text, or video.

157. The second device according to any one of claims 152-156, wherein, The first information source and the second information source have different semantic information.

158. The second device according to any one of claims 150-157, wherein, The second model includes N models, and the i-th model among the N models is used to recover the third source among the multiple sources based on the first transmission information, where N is an integer greater than or equal to 2 and i is a positive integer less than or equal to N.

159. The second device according to any one of claims 150-158, wherein, The processor is also configured to cause the second device to perform: Receive a first message from the first device; wherein the first message is used by the second device to determine the second model corresponding to the first model selected by the first device from M models.

160. The second device according to claim 159, wherein, The first message includes at least one of the following: The number of sources among the multiple information sources; The source type of each of the multiple sources.

161. The second device according to claim 159 or 160, wherein, The first message is sent via the control channel.

162. The second device according to any one of claims 159-161, wherein, The M models are configured by the network device by sending a second message to the terminal device.

163. The second device according to claim 162, wherein, The second message was sent via broadcast.

164. The second device according to claim 162 or 163, wherein, The second message includes at least one of the following: The ID of each of the M models; Information used to determine the parameters and / or structure of each model; The source type corresponding to each model.

165. The second device according to claim 164, wherein, The information used to determine the parameters and / or structure of each model includes at least one of the following: The compression information of the parameters and / or structure of each model; Used to download the parameters and / or structural information of each model; The training method for each model.

166. The second device according to any one of claims 150-165, wherein, The processor is also configured to cause the second device to perform: Receive a third message from the first device; wherein the third message is used to indicate a third model and / or related information of the third model, the third model including a model supported by the first device for multiplexing multiple information sources to be transmitted in the first device.

167. The second device according to claim 166, wherein, The processor is also configured to cause the second device to perform: A fourth message is sent to the first device; wherein the fourth message is used to indicate whether the second device is compatible with the third model.

168. The second device according to claim 166 or 167, wherein, The third message includes at least one of the following: The ID of the third model; The source type corresponding to the third model; The description information of the latent variable distribution associated with the third model.

169. The second device according to any one of claims 150-168, wherein, The processor is also configured to cause the second device to perform: Send a fifth message to the first device; wherein the fifth message is used to indicate a fourth model and / or related information of the fourth model, the fourth model including a model supported by the second device for multiplexing multiple information sources to be transmitted in the first device.

170. The second device according to claim 169, wherein, The processor is also configured to cause the second device to perform: A sixth message is received from the first device; wherein the sixth message is used to indicate whether the first device is compatible with the fourth model.

171. The second device according to claim 169 or 170, wherein, The fifth message includes at least one of the following: The ID of the fourth model; The source type corresponding to the fourth model; The descriptive information of the latent variable distribution associated with the fourth model.

172. The second device according to any one of claims 150-171, wherein, The processor is also configured to cause the second device to perform: Monitor a first performance indicator; wherein the first performance indicator is related to the communication status and / or semantic transmission; the first performance indicator is used to determine whether the fallback mechanism of source multiplexing is triggered.

173. The second device according to claim 172, wherein, The second device includes a terminal device; the processor is further configured to cause the terminal device to perform: A seventh message is sent to the network device; wherein the seventh message is used to report the first performance indicator so that the network device can determine whether to trigger the fallback mechanism of source multiplexing.

174. The second device according to claim 173, wherein, The processor is also configured to enable the terminal device to perform: Upon receiving an eighth message from the network device, the fallback mechanism reuses the source to be transmitted using a first multiplexing method; wherein the eighth message is used to indicate the fallback mechanism that triggers source multiplexing.

175. The second device according to claim 172, wherein, The second device includes a network device; the processor is further configured to cause the network device to perform: Based on the first performance metric, determine whether to trigger the fallback mechanism for source multiplexing.

176. The second device according to claim 175, wherein, The processor is also configured to enable the network device to perform: If the fallback mechanism that triggers source multiplexing is determined, the eighth message is sent to the terminal device.

177. The second device according to any one of claims 172-176, wherein, The first performance metric includes at least one of semantic error rate, transmission latency, and semantic integrity.

178. A chip, comprising: A processor for retrieving and running a computer program from memory, causing a device on which the chip is mounted to perform the method as described in any one of claims 1 to 31.

179. A chip, comprising: A processor for retrieving and running a computer program from memory, causing a device on which the chip is mounted to perform the method as described in any one of claims 32 to 59.

180. A computer-readable storage medium for storing a computer program that, when run by a device, causes the device to perform the method as claimed in any one of claims 1 to 31.

181. A computer-readable storage medium for storing a computer program that, when run by a device, causes the device to perform the method as described in any one of claims 32 to 59.

182. A computer program product comprising computer program instructions that cause a computer to perform the method as described in any one of claims 1 to 31.

183. A computer program product comprising computer program instructions that cause a computer to perform the method as described in any one of claims 32 to 59.

184. A computer program that causes a computer to perform the method as claimed in any one of claims 1 to 31.

185. A computer program that causes a computer to perform the method as described in any one of claims 32 to 59.

186. A communication system, comprising: A first device for performing the method as described in any one of claims 1 to 31; A second device for performing the method as described in any one of claims 32 to 59.

Images