Communication method and apparatus

Abstract

Embodiments of the present application relate to the technical field of communications, and provide a communication method and apparatus. The method comprises: acquiring first scene information, the first scene information being used for determining the scene where a first communication apparatus is located and / or the scene where a second communication apparatus is located; and inputting the first scene information into a first model to obtain a first path parameter set, the first path parameter set being used for determining at least one path for signal transmission between the first communication apparatus and the second communication apparatus. In this way, a method for predicting a path parameter set and selecting a signal transmission path on the basis of a predicted path parameter set is provided.

Description

A communication method and apparatus

[0001] Cross-reference to related applications

[0002] This application claims the priority of a Chinese patent application with the application number 202410208515.4 and the application title "A communication method and apparatus", which was filed with the National Intellectual Property Administration of the People's Republic of China on February 23, 2024. The entire content of this application is incorporated herein by reference in its entirety. Technical field

[0003] This application relates to the field of communication technologies, and in particular, to a communication method and apparatus. Background art

[0004] Multipath propagation refers to the situation where a signal reaches a receiving antenna after passing through two or more paths in a scenario. Reflections, diffractions, etc. of electromagnetic waves by objects in the scenario result in multiple paths (which can be referred to as multipaths). Signals passing through different paths have different time delays and phases, and the receiving antenna receives the superposition of these multipath signals. Multipath propagation increases the number of spatial division multiplexing streams in a communication system, may cause inter-symbol interference, and the cancellation caused by multipaths may lead to signal fading, etc. Predicting multiple paths in the environment is crucial for improving the service capabilities of a communication system. Predicting multiple paths means predicting information about possible multiple paths when a device communicates with another device at a certain spatial location. Information about multiple paths includes, for example, the number of paths and the time delay of the paths, etc. However, there is currently no corresponding solution for how to predict the information of paths.

[0005] Summary of the invention

[0006] This application provides a communication method and apparatus for providing a mechanism for predicting path information.

[0007] In a first aspect, an embodiment of this application provides a communication method. This method can be applied to a first communication device. The first communication device can be a terminal device, or a software module or hardware module (such as a chip) in a terminal device, or a network device, or a software module or hardware module (such as a chip) in a network device, etc., and no limitation is made thereto. A network device is, for example, a core network element, an access network element, or a third-party server, etc., and no specific limitation is made thereto. The method includes: obtaining first scenario information, where the first scenario information is used to determine the scenario where the first communication device is located and / or the scenario where the second communication device is located; inputting the first scenario information into a first model to obtain a first set of path parameters, where the first set of path parameters is used to determine at least one path for signal transmission between the first communication device and the second communication device.

[0008] The first scenario information may include one or more parameters related to the scenario. For example, the material or layout of objects in the scenario, etc. The objects may be people, things, or animals, etc. Things may be vehicles or buildings, etc. The first path parameter set may include one or more parameters related to the path. For example, the probability of the path existing, or the time delay of the path, etc. The first model may be described as a model for learning the mapping relationship between the scenario information and the path parameter set, or may be described as a model for determining the path parameter set.

[0009] In the embodiments of the present application, by predicting the path parameter set through the first model, a way to determine the path parameter set is provided. Moreover, with the real scenario information as the input of the first model, the accuracy of the predicted path parameter set can be improved. Further, the appropriate path for signal transmission can also be selected from multiple paths according to the determined path parameter set, improving the communication quality between the first communication device and the second communication device.

[0010] In a possible implementation manner, the first model includes at least one first neural network, a second neural network, and a third neural network, and the first scenario information includes at least one type of information; inputting the first scenario information into the first model to obtain the first path parameter set includes: respectively inputting each type of information in the at least one type of information into one first neural network in the at least one first neural network to obtain one type of feature, and a total of at least one type of feature is obtained; inputting the at least one type of feature into the second neural network to obtain a fused feature; inputting the fused feature into the third neural network to obtain the first path parameter set.

[0011] There can be various ways to divide the at least one type of information. For example, it can be divided according to different information sources, or it can also be divided according to the content of the information, etc. No specific limitation is made in this regard. Optionally, the second neural network includes a variant neural network based on the attention mechanism, where the attention mechanism is used to perform weighted fusion on the at least one type of feature to obtain the fused feature.

[0012] In the above implementation manner, at least one first neural network can be used to process different types of information included in the first scenario information. In this way, it is more targeted to process this information, which is beneficial to fully obtain the features of these types of information and facilitate obtaining a more accurate first path parameter set. Moreover, the first scenario information may include all aspects of information. Using at least one neural network to process this information separately makes the processing process more in line with the actual processing requirements.

[0013] In a possible implementation, the first scenario information includes at least one type of information, where the at least one type of information includes: environmental information, which indicates the environment where the first communication device and / or the second communication device is located; and / or communication information, which indicates parameters related to communication of the first communication device and / or the second communication device. Optionally, the environmental information may include the environmental information corresponding to the first communication device (such as referred to as the first environmental information), and / or the environmental information corresponding to the second communication device (such as referred to as the second environmental information). The communication information may include the communication information corresponding to the first communication device (such as referred to as the first communication information), and / or the communication information corresponding to the second communication device (such as referred to as the second communication information).

[0014] In the above implementation, the at least one type of information may include environmental information and / or communication information, etc., to obtain more comprehensive first scenario information. Moreover, the signal transmission path between communication devices is related to the environment and communication. Therefore, including such information in the first scenario information is conducive to obtaining a more accurate set of path parameters.

[0015] In a possible implementation, the environmental information indicates at least one of the following: the layout of objects in the environment; the material of objects in the environment; the position of objects in the environment; the layout of objects in the environment; the size of objects in the environment; or the speed of objects in the environment.

[0016] The object may be at least one of a person, a thing, or an animal, etc. A person may be a specific person or a group of people, etc. A thing may be a movable thing, such as a vehicle, or an immovable thing, such as a building.

[0017] In the above implementation, the environmental information can indicate various types of information of the object, facilitating a more accurate determination of the environment where the communication device is located, and thus enabling a more accurate prediction of the set of path parameters.

[0018] In a possible implementation, the communication information may indicate at least one of the following: the position of the first communication device; the configuration information of the first communication device; the position of the second communication device; or the configuration information of the second communication device.

[0019] The configuration information may include, for example, the antenna configuration of the communication device (such as at least one of the number, layout, or polarization direction of the antennas) and / or the time-frequency resource configuration, etc. The time-frequency resources may include time-domain resources and / or frequency-domain resources, etc.

[0020] In the above implementation, the communication information can indicate the position and / or configuration of the communication device, facilitating the determination of the position of the communication device and some configurations related to communication, and also conducive to a more accurate prediction of the set of path parameters.

[0021] In a possible implementation manner, the method further includes: obtaining a first loss value based on a first path parameter set and first measurement information, where the first measurement information includes a second path parameter set and / or information of a predicted channel, the first measurement information is obtained by channel measurement through a first communication device or a second communication device, and the channel refers to a channel (such as a wireless channel) between the first communication device and the second communication device; determining at least one of second scenario information, a third path parameter set, and a second model, where the second scenario information is obtained by updating first scenario information based on the first loss value, the third path parameter set is obtained by updating the first path parameter set based on the first loss value, and the second model is obtained by updating the first model based on the first loss value.

[0022] The first loss value is, for example, the global loss value or local loss value of the first model, etc., and there is no limitation thereto. Updating the first model may be to update some or all of the model parameters included in the first model. Optionally, the first model includes at least one of a first neural network, a second neural network, and a third neural network, and the second model may be obtained by updating at least one of the model parameters of the at least one first neural network, the model parameters of the second neural network, or the model parameters of the third neural network in the first model.

[0023] In the above implementation manner, it is possible to update at least one of the scenario information, the path parameter set, and the model by using the first loss value, so as to obtain at least one of more accurate scenario information, path parameter set, and model.

[0024] In a possible implementation manner, before obtaining the first loss value based on the first path parameter set and the first measurement information, the method further includes: performing a screening process on the second path parameter set, for example, using a preset first power value to perform a screening process on the second path parameter set; performing a normalization process on the screened second path parameter set to obtain a preprocessed second path parameter set. The first power value may be preconfigured or predefined in the first communication device, or determined by the first communication device itself, or determined through negotiation between the first communication device and the second communication device.

[0025] Both the screening process and the normalization process can be regarded as preprocessing of the second path parameter set.

[0026] In the above implementation manner, since preprocessing is performed on the second path parameter set to obtain a more reasonable path parameter set, and the preprocessed second path parameter set is used as the true value of the first model, it is beneficial to train the first model more accurately.

[0027] In a possible implementation, the method further includes: obtaining a first image, where the first image represents the scene where the first communication device is located and / or the scene where the second communication device is located; preprocessing the first image to obtain a second image; calibrating the position of the first communication device and / or the position of the second communication device based on the second image; and updating the second image based on the calibrated position of the first communication device and / or the position of the second communication device, where the first scene information is carried on the updated second image.

[0028] The images involved here can be point cloud images, aerial images, maps, etc., and no specific limitations are imposed on the form of the image.

[0029] In the above implementation, the first scene information can be carried by an image, which is convenient for the model to process the first scene information. Moreover, some calibration processing is performed on the collected first image and then used as the first scene information to input into the first model, improving the accuracy of the input of the first model, and thus facilitating the prediction of a more accurate set of path parameters.

[0030] In a possible implementation, the method further includes: preprocessing the first set of path parameters (for example, screening and / or normalizing the first set of path parameters based on a second power value) to obtain a fourth set of path parameters; obtaining information about the predicted channel based on the fourth set of path parameters, where the channel refers to the channel between the first communication device and the second communication device; and / or sending the fourth set of path parameters to the second communication device so that the second communication device can determine at least one path for signal transmission with the first communication device based on the fourth set of path parameters. The second power value and the first power value can be the same or different, and no limitation is imposed on this. The second power value can be preconfigured or predefined in the first communication device, or determined by the first communication device itself, or determined through negotiation between the first communication device and the second communication device.

[0031] The content of preprocessing the first set of path parameters can refer to the content of preprocessing the second set of path parameters in the previous text.

[0032] In the above implementation, preprocessing the first set of path parameters can obtain a more reasonable set of path parameters, enabling the second communication device to perform subsequent tasks, such as selecting a beam, based on the more reasonable set of path parameters.

[0033] In a possible implementation, the first set of path parameter information includes the path parameters of each path in at least one path.

[0034] In the above implementation, when updating the first set of path parameters, the path parameters corresponding to each path corresponding to the first set of path parameters can also be updated, thereby obtaining more comprehensive and accurate path parameters.

[0035] In a possible implementation, the path parameters include at least one of the following: the existence probability of the path, the departure angle information of the path, the arrival angle information of the path, the pitch angle information of the path, the azimuth angle information of the path, the loss information of the path, the time delay information of the path, the channel impulse response of the path, or the phase information of the path.

[0036] In the above implementation, the path parameters corresponding to each path or the updated path parameters may include information on all aspects of the path. That is, the first model can comprehensively and accurately predict information on all aspects of the path, and can also more comprehensively update information on all aspects of the path.

[0037] In a second aspect, an embodiment of the present application further provides a communication device. This device can be used to execute the method in the first aspect. This device can be the first communication device, or this device can be a component in the first communication device (such as a chip, or a chip system, or a circuit), or this device can be a corresponding logic module or software of the first communication device, or this device can be a device that can be used in matching with the first communication device.

[0038] In a possible implementation, the device may include modules or units corresponding one by one to the methods / operations / steps / actions described in the first aspect. These modules or units can be hardware circuits, or software, or can be implemented by combining hardware circuits and software. In a possible implementation, the device may include a processing module (which can also be called a processing unit) and a transceiver module (which can also be called a transceiver unit). Among them, the transceiver module can be used to execute the functions of receiving and / or sending, and the processing module can be used to execute the method described in the above first aspect or any possible implementation in the first aspect.

[0039] In a third aspect, an embodiment of the present application provides a communication system, which may include: a first communication device and a second communication device. The first communication device is, for example, the first communication device involved in the first aspect, and the second communication device is, for example, the second communication device involved in the first aspect.

[0040] In a fourth aspect, an embodiment of the present application provides a communication device (which can also be called a processing device or a device, etc.). This device includes: a processor and an interface circuit. The interface circuit is used to receive signals from other communication devices outside the communication device and transmit them to the processor, or to send signals from the processor to other communication devices outside the communication device. The processor is used to implement the method provided in the above first aspect or any possible implementation thereof through logic circuits or by executing code instructions.

[0041] In the specific implementation process, the communication device can be a chip, and the processor can be transistors, gate circuits, flip-flops, and various logic circuits, etc. The embodiments of the present application do not limit the specific implementation manners of the processor.

[0042] In one implementation manner, the communication device can be a wireless communication device, that is, a computer device supporting wireless communication functions. Specifically, the wireless communication device can be a terminal device such as a smart phone, or a network device such as a wireless access network device (such as a base station).

[0043] In another implementation manner, the communication device can be some components in a wireless communication device, such as integrated circuit products such as a system chip or a communication chip. The system chip can also be referred to as a system on chip (SoC), or simply called an SoC chip for short. The communication chip can include a baseband processing chip and a radio frequency processing chip. The baseband processing chip is sometimes also called a modem or a baseband chip. The radio frequency processing chip is sometimes also called a radio frequency transceiver or a radio frequency chip. In a physical implementation, some or all of the chips in the communication chip can be integrated inside the SoC chip. For example, the baseband processing chip is integrated in the SoC chip, and the radio frequency processing chip is not integrated with the SoC chip. The interface circuit can be the radio frequency processing chip in the wireless communication device, and the processor can be the baseband processing chip in the wireless communication device. The interface circuit can be an input / output interface, an interface circuit, an output circuit, an input circuit, a pin, or a related circuit, etc. on the chip or chip system. The processor can also be embodied as a processing circuit or a logic circuit.

[0044] In yet another implementation manner, the communication device can be a chip system, and the chip system can be composed of chips, or can include chips and other discrete devices. The chip system can, for example, include a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a CPU, a network processor (NP), a DSP, a micro controller unit (MCU), a programmable logic device (PLD), or other integrated chips, etc.

[0045] Fifth aspect, an embodiment of the present application provides a communication device (or may be referred to as a processing device or a device, etc.). An embodiment of the present application provides a communication device, which includes: a processor; when the communication device runs, the processor executes any method described in the first aspect and any possible implementation manner. Optionally, the communication device further includes a memory, and one or more computer programs stored in the memory can be executed by the processor to implement any method described in the first aspect and any possible implementation manner.

[0046] Optionally, the communication device further includes other components, such as an antenna, an input / output module, an interface (such as a communication interface), etc. These components can be hardware, software, or a combination of software and hardware.

[0047] Sixth aspect, an embodiment of the present application provides a computer-readable storage medium. A computer program or instruction is stored in the storage medium, and when the computer program or instruction is executed by the computer, it can implement any method described in the first aspect and any possible implementation manner.

[0048] Seventh aspect, an embodiment of the present application further provides a computer program product including a computer program or instruction. When it runs on a computer, it causes any method described in the first aspect and any possible implementation manner to be executed.

[0049] Eighth aspect, an embodiment of the present application further provides a chip system. The chip system includes a processor for implementing any method described in the first aspect and any possible implementation manner.

[0050] In a possible design, the chip system may further include a memory for storing necessary program instructions and data for the loading device to execute. The chip system can be composed of chips or can include chips and other discrete devices.

[0051] The technical effects that can be achieved by the implementation manners of the second aspect to the eighth aspect can be correspondingly described by referring to the technical effects that can be achieved by the first aspect and any of its implementation manners, and will not be listed one by one here. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] FIG. 1 is a schematic diagram of a communication system applicable to an embodiment of the present application;

[0053] FIG. 2 is a schematic structural diagram of a first model provided by an embodiment of the present application;

[0054] FIG. 3 is a schematic structural diagram of another first model provided by an embodiment of the present application;

[0055] FIG. 4 is a schematic diagram of another communication system applicable to an embodiment of the present application;

[0056] FIG. 5 is a schematic diagram of another communication system applicable to the embodiment of the present application;

[0057] FIG. 6 is a schematic diagram of a communication method provided by the embodiment of the present application;

[0058] FIG. 7 is a schematic diagram of obtaining first scenario information provided by the embodiment of the present application;

[0059] FIG. 8 is a schematic diagram of obtaining a first path parameter set provided by the embodiment of the present application;

[0060] FIG. 9 is a schematic diagram of preprocessing a path parameter set provided by the embodiment of the present application;

[0061] FIG. 10 is a schematic diagram of updating parameters provided by the embodiment of the present application;

[0062] FIG. 11 is a schematic diagram of updating parameters provided by the embodiment of the present application;

[0063] FIG. 12 is a schematic diagram of another communication method provided by the embodiment of the present application;

[0064] FIG. 13 is a schematic diagram of processing a path parameter set provided by the embodiment of the present application;

[0065] FIG. 14 is a schematic diagram of processing a path parameter set provided by the embodiment of the present application;

[0066] FIG. 15 is a schematic diagram of another communication method provided by the embodiment of the present application;

[0067] FIG. 16 is a schematic diagram of yet another communication method provided by the embodiment of the present application;

[0068] FIG. 17 is a schematic diagram of the structure of a communication device provided by the embodiment of the present application;

[0069] FIG. 18 is a schematic diagram of the structure of a communication device provided by the embodiment of the present application;

[0070] FIG. 19 is a schematic diagram of the structure of a communication device provided by the embodiment of the present application. Detailed implementation manners

[0071] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0072] Hereinafter, some terms in the embodiments of the present application will be explained to facilitate the understanding of those skilled in the art.

[0073] 1. A communication device refers to a device with processing capabilities. The communication device may also have other functions such as communication functions, which are not limited herein. The communication device may be a device, multiple devices, a software module or a hardware module (such as a chip) in a device, a network element, a logical module, or a function, etc., and the specific implementation form thereof is not specifically limited. A signal may include information, signaling, or data, etc. The communication device may also be replaced by a device, an entity, a network entity, a communication device, a communication module, a node, or a communication node, etc.

[0074] The first communication device and the second communication device involved in the embodiments of the present application refer to two different devices, and the specific implementation forms and types of these two devices are not limited.

[0075] 2. A model refers to the ability to enable a computer to have intelligent behavior through learning. The model includes, for example, machine learning (ML) models, artificial intelligence (AI) models, algorithms, features, or functions, etc., and the specific implementation manner of the model is not limited. The AI model may include at least one of a linear regression model, a logistic regression model, a decision tree model, a support vector machine, a neural network model, a clustering model, or a generative adversarial network, etc., which is not limited herein. The neural network model may be, for example, a multilayer perceptron (MLP), a variant model (or a transformation network or a transformation model, etc.) (transformer network), a convolutional neural network (CNN), a recurrent neural network (RNN), or an attention mechanism, etc., in one or more forms, and the specific form thereof is not specifically limited. For example, the first model includes a variant model based on the attention mechanism, for example, a variant model based on the self-attention mechanism or a variant model based on the multi-head attention mechanism. The model involved in the embodiments of the present application may include one or more neural networks, which is not specifically limited. The implementation manners of the first model, the second model, or the third model, etc. involved in the embodiments of the present application may all refer to the implementation manner of the model herein. The structures of the first model, the second model, and the third model involved in the embodiments of the present application may be the same, and there may only be differences in the model parameters in the model. In addition, the first model, the second model, or the third model may be described as a model for learning the mapping relationship between scene information and a set of path parameters, or may be described as a model for determining a set of path parameters.

[0076] 3. A set of path parameters may also be referred to as a set of path parameters, a multi-path parameter set, multi-path information, or path information, etc., which is not limited herein.

[0077] A path parameter set is used to determine at least one path for signal transmission between two devices. When the at least one path includes multiple paths, the path parameter set can also be referred to as a multipath parameter set, multipath information, or multipath component (MPC), etc., and there is no limitation in this regard. The first path parameter set involved in the embodiments of the present application refers to at least one path for signal transmission between a first communication device and a second communication device. The first path parameter set includes path parameters of at least one path. For example, the first path parameter set includes path parameters of each path in at least one path. The path parameter of each path in the at least one path indicates at least one piece of information such as the number of paths, existence probability, phase, angle, intensity, loss (pathloss), delay, or channel impulse response. The channel impulse response can be understood as the channel time-domain response. For example, the angle of each path includes at least one of the departure angle (or also referred to as the direction of departure (DoD)), direction of arrival (DoA), elevation angle, or azimuth angle, etc.

[0078] (1) The number of paths can be understood as how many paths the signal can pass through from the first communication device to the second communication device in physical space, or it can be understood as how many paths the signal can pass through from the second communication device to the first communication device in physical space.

[0079] (2) The loss of a path can be understood as the loss of the power corresponding to the path.

[0080] Assume that there are N paths between the first communication device and the second communication device, where N is a positive integer. The longer the path the signal passes through, the greater the path loss usually is. After the signal undergoes reflection and diffraction, the path loss usually increases. Therefore, the path losses of the N paths are different, and each path corresponds to a path loss. The received power is the transmitted power minus the path loss. Therefore, each path has its own intensity.

[0081] (3) The departure angle of a path refers to the angle at which the path departs from the transmitting end. The departure angle includes the departure angle in the horizontal direction (also known as the azimuth angle) and the departure angle in the vertical direction (also known as the elevation angle).

[0082] (4) The arrival angle of a path refers to the angle at which the path arrives at the second communication device. Similarly, the arrival angle also includes the arrival angle in the horizontal direction (also known as the azimuth angle) and the arrival angle in the vertical direction (also known as the elevation angle).

[0083] (5) The delay of a path refers to the time consumed from the transmission of each path from the transmitting end to the reception by the second communication device, also known as the time of flight.

[0084] (6) The channel impulse response of a path refers to the changes in amplitude and phase experienced by a signal during propagation along that path.

[0085] (7) The existence probability of a path refers to the probability that such a path exists between the first communication device and the second communication device.

[0086] In the embodiments of the present application, the first path parameter set, the second path parameter set, and the third path parameter set all represent the path parameters between the first communication device and the second communication device. Among them, the parameters involved in the first path parameter set, the second path parameter set, and the third path parameter set, or the values of the parameters, may be different. The first path parameter set and the second path parameter set may be output by a model, and the third path parameter set is obtained by the first communication device or the second communication device through measuring the channel or the signal. The content of the first path parameter set, the second path parameter set, and the third path parameter set may all refer to the content of the path parameter set discussed above.

[0087] 4. Measurement information refers to the information about the path and / or the measured channel obtained by a device through measuring the channel. The first measurement information involved in the embodiments of the present application can be used as an example of the measurement information. The first measurement information refers to the information obtained by the first communication device or the second communication device through measuring the channel. The first measurement information includes, for example, the third path parameter set(s) and / or the information of the measured channel. The third path parameter set can be understood as being obtained by the first communication device or the second communication device through measuring the channel. The information of the measured channel can be understood as being obtained by the first communication device or the second communication device through measuring the channel. The channels involved in the embodiments of the present application include the channel (such as a wireless channel) between the first communication device and the second communication device.

[0088] The third path parameter set is obtained by the first communication device or the second communication device through measuring the channel. The third path parameter set is used to determine at least one path between the first communication device and the second communication device.

[0089] Among them, the information of the channel represents the information related to the channel between the first communication device and the second communication device, and includes, for example, at least one of channel state information, channel precoding information, beam information, beam angle information, beam power information, beam indication information, channel eigenvector, channel eigenvalue, amplitude information of the channel, or phase information of the channel. The channels involved in the embodiments of the present application may be an uplink channel, a downlink channel, a sidelink channel, etc., and are not limited thereto.

[0090] Channel state information is used to indicate the state of a channel. Channel precoding information is used to indicate a channel precoding matrix, etc. Beam information is used to indicate a beam for transmitting or receiving a signal, etc., for example, including an index of the beam. Beam angle information includes, for example, at least one of a beam direction, a beam width, or a beamforming method. The beam direction includes, for example, the main lobe direction formed by beamforming. The beam width refers to the degree of broadening of the main lobe formed by beamforming in space. The beamforming method refers to a method of beamforming, for example, a numerical method, etc. Beam power information is used to indicate the power of a beam. Beam indication information refers to parameters required for beamforming. A channel eigenvector is a vector used to represent channel transmission characteristics. A channel eigenvalue refers to an eigenvalue of a channel matrix. The amplitude information of a channel refers to the amplitude change of a signal during transmission. The phase information of a channel refers to the phase change of a signal during transmission. <> <>

[0091] The content of the information of the measured channel and the predicted channel involved in each embodiment of this application can refer to the content of the information of the channel discussed here. However, the information of the measured channel can be measured by a communication device (such as a first communication device or a second communication device), and the information of the predicted channel refers to the channel information predicted through a model, etc. <> }<>

[0092] 5. Scenario information, which can also be referred to as a scenario parameter set, a scenario parameter collection, or a trainable parameter set (trainable parameters), etc. <> <>

[0093] Scenario information is used to determine the scenario where at least one of two devices communicating with each other is located. For example, the first scenario information, the second scenario information, and the third scenario information, etc., involved in the embodiments of this application can all be regarded as an example of a scenario parameter set. The first scenario information, the second scenario information, and the third scenario information, etc., are used to determine the scenario where the second communication device is located, and / or the scenario where the first communication device is located. Among them, the parameters or the values of the parameters involved in the first scenario information, the second scenario information, and the third scenario information may be different. A scenario refers to a specific environment or situation, including a series of factors and conditions related to information transmission and reception. Alternatively, scenario information can also represent the scenario characteristics or parameters, etc., where at least one of two devices communicating with each other is located. <> <>

[0094] The format of the scenario information can be various. For example, it can be one or more of a table, an information stream, an array, a matrix, a vector, an image, etc., and there is no limitation in this regard. The first scenario information, the second scenario information, etc. involved in the embodiments of this application can all be used as an example of the scenario information. Hereinafter, the first scenario information will be taken as an example for introduction. The first scenario information includes environmental information and / or communication information. The environmental information and the communication information can be regarded as different types of information included in the first scenario information, that is, the first scenario information includes multiple types of information, and these multiple types of information can include environmental information and / or communication information.

[0095] The environmental information includes the information of the environment corresponding to the first communication device (hereinafter referred to as the first environmental information), and / or the information of the environment corresponding to the second communication device (hereinafter referred to as the second environmental information). The communication information includes the communication information corresponding to the first communication device (hereinafter referred to as the first communication information), and / or the communication information corresponding to the second communication device (hereinafter referred to as the second communication information).

[0096] (1) The first communication information indicates the parameters related to communication of the first communication device. For example, it includes the position of the first communication device and / or the configuration information of the first communication device, etc. The configuration information of the first communication device, for example, includes the antenna configuration of the first communication device and / or the time-domain resources of the first communication device, etc.

[0097] (1-1) The position of the first communication device can be the position of the first communication device in the world coordinate system, and / or can be the position of the first communication device in the reference coordinate system, and there is no limitation in this regard.

[0098] (1-2) The antenna configuration of the first communication device, for example, includes at least one of the number of antennas of the first communication device, the arrangement mode of the antennas, the gain of the antennas, the type of the antennas, the radiation pattern of the antennas, the polarization of the antennas, or the operating frequency band of the antennas, etc. The arrangement mode of the antennas, for example, includes the antenna orientation, etc.

[0099] (1-3) The time-domain resources of the first communication device can include at least one of the time-domain resources used by the first communication device for communicating with the second communication device, the frequency-domain resources used by the first communication device for communicating with the second communication device, the time-domain resources that the first communication device can use for communication, the frequency-domain resources that the first communication device can use for communication, etc. The frequency-domain resources of the first communication device, for example, include communication subcarriers, etc.

[0100] (2) The second communication information indicates the parameters related to communication of the second communication device. For example, it includes the position of the second communication device and / or the configuration information, etc. The content of the position and configuration information of the second communication device can refer to the content of the position and configuration information of the first communication device, and the repeated parts will not be elaborated here.

[0101] (3) The first environmental information indicates (or is used to determine) the environment where the first communication device is located. The first environmental information includes, for example, object information in the environment where the first communication device is located, such as at least one of the number, material, layout, location information, contour, size, or speed of the objects in the environment where the first communication device is located, etc. The object is, for example, at least one of a person, an animal, or a thing, such as at least one of a crowd, a vehicle, a building, or a plant, etc. Optionally, the first environmental parameters include at least one of the layout of the building, the material of the building, the layout of the vegetation, the layout of an object (such as a vehicle), the material of an object (such as a vehicle), the point cloud description of an object (such as a vehicle), the location of an object (such as a vehicle), the size of an object (such as a vehicle), the orientation of an object (such as a vehicle), or the moving speed of an object (such as a vehicle), the location of a pedestrian, or the distribution of a crowd, etc.

[0102] (4) The second environmental information indicates (or is used to determine) the environment where the second communication device is located. The second environmental parameters include, for example, object information in the environment where the second communication device is located. The content items included in the second environmental parameters can refer to the content of the first environmental parameter set described above and will not be listed here. For example, the second environmental parameters include the quantity information, contour information, and / or material property information of the objects in the environment where the second communication device is located, etc., and there is no limitation on this.

[0103] The content of the second scenario parameter set and the third scenario parameter set can refer to the content of the first scenario parameter set described above and will not be listed one by one.

[0104] 6. A reference signal (RS), which can also be referred to as a pilot signal or pilot, is a known signal. For example, a reference signal can be a signal provided by a transmitting end to a receiving end for channel estimation, channel sounding, or data demodulation, etc. Reference signals include uplink reference signals and downlink reference signals. Uplink reference signals such as demodulation reference signals (DMRS) and sounding reference signals (SRS). DMRS can, for example, include DMRS for demodulating the physical uplink control channel (PUCCH) (which can be abbreviated as DMRS for PUCCH) and DMRS for demodulating the physical uplink shared channel (PUSCH) (which can be abbreviated as DMRS for PUCCH). Downlink reference signals such as channel state information reference signals (CSI-RS), cell-specific reference signals (C-RS / CRS), and positioning reference signals (P-RS / PRS). There are various reference signals. As the standard continues to evolve, the names of reference signals may change, and there may also be more reference signals emerging. No specific limitations are imposed on this.

[0105] 7. A beam can be understood as a spatial filter or spatial parameters. The beam used for transmitting signals can be called a transmitting beam, a transmission beam (Tx beam), a spatial domain transmit filter, or spatial transmit parameters (spatial Tx parameters). The transmitting beam can also refer to the distribution of signal strength formed in different directions in space after the signal is transmitted by the antenna. From this perspective, the transmitting beam can also be a spatial transmission angle (such as Azimuth (also called horizontal angle), Zenith (also called elevation angle)) or a spatial transmission angle range (such as azimuth center angle and offset, azimuth uncertainty, azimuth protection range, zenith center angle and offset, zenith uncertainty, zenith protection range), etc. Correspondingly, the beam used for receiving signals can be called a receiving beam (Rx beam), a spatial domain receive filter, or spatial receive parameters (spatial Rx parameters). The receiving beam can also refer to the distribution of signal strength of the wireless signals received from the antenna in different directions in space. From this perspective, the receiving beam can also be a spatial reception angle (such as Azimuth, Zenith) or a spatial reception angle range (such as azimuth center angle and offset, azimuth uncertainty, azimuth protection range, zenith center angle and offset, zenith uncertainty, zenith protection range), etc.

[0106] In the 5th generation new radio (NR) protocol, a beam can be a spatial filter. It should be understood that with the continuous evolution of the standard, it does not exclude the possibility of defining other terms in future protocols to represent the same or similar meaning as a beam.

[0107] Each of the above-mentioned terms (such as scenario information, path parameter set, measurement information, reference signal, or beam, etc.) may have other names, or with the continuous evolution of the standard, other terms may appear, and no specific limitations are made in this regard.

[0108] In various embodiments of the present application, unless otherwise specified, for the number of nouns, it means "singular noun or plural noun", that is, "one or more". "At least one" means one or more, and "multiple" means two or more. "And / or" describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. For example, A / B means: A or B. "At least one of the following" or its similar expressions refer to any combination of these items, including any combination of single item(s) or plural item(s). For example, at least one of a, b, or c means: a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0109] In the embodiments of the present application, "indication" may include direct indication, indirect indication, display indication, and implicit indication. When it is described that a certain indication information is used to indicate A, it can be understood that the indication information carries A, directly indicates A, or indirectly indicates A. In the present application, the information indicated by the indication information is called the information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated. For example, but not limited to, it can directly indicate the information to be indicated, such as the information to be indicated itself or the index of the information to be indicated, etc. It can also indirectly indicate the information to be indicated by indicating other information, where there is an association relationship between the other information and the information to be indicated. It can also only indicate a part of the information to be indicated, while the other parts of the information to be indicated are known or pre-agreed. For example, it can also rely on the arrangement order of each pre-agreed (such as protocol-defined) information to achieve the indication of specific information, thereby reducing the indication overhead to a certain extent. In addition, the information to be indicated can be sent as a whole, or divided into multiple sub-information and sent separately, and the sending periods and / or sending timings of these sub-information can be the same or different.

[0110] In the embodiments of the present application, "transmission" and "reception" indicate the direction of signal transmission. For example, "transmitting information to XX" can be understood as the destination of the information being XX, which may include direct transmission via the air interface, or indirect transmission via the air interface by other units or modules. "Receiving information from YY" can be understood as the source of the information being YY, which may include direct reception from YY via the air interface, or indirect reception from YY via the air interface by other units or modules. "Transmission" can also be understood as the "output" of the chip interface, and "reception" can also be understood as the "input" of the chip interface. In other words, transmission and reception can occur between devices, for example, between a network device and a terminal device, or within a device, for example, transmission or reception between components, modules, chips, software modules, or hardware modules within a device via a bus, trace, or interface.

[0111] The predicted path can assist the device in selecting a beam to improve the communication quality between devices. One method for predicting multiple paths is ray tracing. In this method, the real environment is modeled in a virtual physical world, and the size, position, and material of the objects in the real world are restored as much as possible. Next, the signal transmitter and receiver are placed at the positions where the multipath is to be predicted in the virtual physical world, and then the multipath between them is simulated using the ray-tracing method. The principle of ray tracing is usually to emit X rays from the transmitter in the modeled environment, and after interacting with the objects in the environment (reflection, diffraction, scattering, etc.), the number of rays received at the receiver is counted. Assuming that Y rays are received at the receiver, it means that Y rays have been simulated. Both X and Y are positive integers. There are always differences between the virtual environment simulated by this method and the real environment, which results in relatively large differences in the predicted multipath.

[0112] In view of this, the embodiments of the present application provide a communication method. In this method, a mechanism for predicting paths is provided by using a model to predict a set of path parameters. Moreover, the first scene information corresponding to the real scene is input into the first model to obtain the first set of path parameters. Since the first model directly uses the real scene information as input, the accuracy of the predicted set of path parameters can be improved, that is, the accuracy of the predicted multipath is relatively high.

[0113] The solutions involved in the various embodiments of the present application can be applied to various communication networks (or systems) including a first communication device and a second communication device. Various communication networks, for example, the fifth-generation (5 thgeneration, 5G), or new radio (NR) systems, long term evolution (LTE) systems, frequency division duplex (FDD) systems, time division duplex (TDD) systems, and multiple-input multiple-output (MIMO) systems, etc. The technical solutions provided in this application can also be applied to future communication systems, such as the sixth generation (6 th generation, 6G) mobile communication systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, sidelink (SL) systems, machine-to-machine (M2M) communication, machine type communication (MTC), and internet of things (IoT) communication systems. SL can also be referred to as a sidelink communication link, sidelink, sidelink, direct link, side link, or secondary link, etc. SL can include links for device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, or sidelink on unlicensed spectrum (SL-U) communication, etc. V2X communication can include vehicle-to-vehicle (V2V) communication, vehicle-to-infrastructure (V2I) communication, vehicle-to-pedestrian (V2P) communication, and vehicle-to-network (V2N) communication. Each embodiment of this application can also be applied to non-terrestrial network (NTN) systems such as inter-satellite communication and satellite communication.

[0114] Exemplarily, the first communication device can be deployed with a first model and use the first model to output a first set of path parameters. Optionally, the second communication device can obtain the first set of path parameters from the first communication device, etc., to facilitate processing subsequent services.

[0115] Please refer to FIG. 1, which is a schematic diagram of a communication system applicable to an embodiment of the present application. FIG. 1 schematically shows a first communication device, a second communication device, a training device, an initiating device, a storage device, a data acquisition device, etc. As shown in FIG. 1, the data acquisition device can acquire data to obtain a training set and send the training set to the storage device. The storage device can store the training set. The training device can pre-train or train a model based on the training set to obtain a first model. The training device configures the configuration file of the first model for the first communication device. Thus, when the initiating device needs to process a service, it can send a request for processing the service to the first communication device. Based on this request, the first communication device predicts a path parameter set using the first model. Furthermore, the first communication device can send the path parameter set to the initiating device and / or the second communication device.

[0116] In a possible implementation, the data acquisition device and the storage device can be the same device. The initiating device and the first communication device can be the same device. The training device and the first communication device can be the same device. The initiating device and the second communication device can also be the same device, and no specific limitation is made thereto.

[0117] Any device involved in FIG. 1 can be a terminal device or a network device, and no limitation is made thereto. For example, in various embodiments of the present application, the first communication device involved is, for example, a terminal device, and the second communication device is, for example, a network device. Or, in various embodiments of the present application, the second communication device involved is, for example, a network device, and the first communication device is, for example, a terminal device. Or, both the first communication device and the second communication device are terminal devices, etc., and no specific limitation is made thereto. Or, both the first communication device and the second communication device are network devices, etc.

[0118] The terminal device includes various devices with wireless communication functions, which can be used to connect people, objects, machines, etc. The terminal device can be widely applied to various scenarios, such as: cellular communication, D2D, V2X, peer-to-peer (P2P), M2M, MTC, IoT, virtual reality (VR), augmented reality (AR), industrial control, autonomous driving, remote medical treatment, smart grid, smart furniture, smart office, smart wearables, smart transportation, smart city drones, robots, remote sensing, passive sensing, positioning, navigation and tracking, autonomous delivery, etc. The terminal device can be a terminal in any of the above scenarios, such as an MTC terminal, an IoT terminal, etc. The terminal device can be a 3rd Generation Partnership Project (3 rdUser equipment (UE), terminal, fixed device, mobile station device or mobile device, subscriber unit, handheld device, vehicle-mounted device, wearable device, cellular phone, smart phone, SIP phone, wireless data card, personal digital assistant (PDA), computer, tablet computer, laptop computer, wireless modem, handset, laptop computer, computer with wireless transceiver function, smart book, vehicle, satellite, global positioning system (GPS) device, target tracking device, aircraft (such as drone, helicopter, or airplane, etc.), ship, remote control device, smart home device, industrial device, or a device built into the above devices (such as a communication module, modem, or chip in the above devices), or other processing devices connected to a wireless modem.

[0119] It should be understood that in some scenarios, the terminal device can also be used as a base station. For example, the terminal device can act as a scheduling entity that provides sidelink signals between terminal devices in scenarios such as V2X, D2D, or P2P.

[0120] In the embodiments of this application, the device for implementing the functions of the terminal device, that is, the terminal device, can be the terminal device or a device capable of supporting the terminal device to implement the functions, such as a chip system or a chip, and this device can be installed in the terminal device. In the embodiments of this application, the chip system can be composed of chips or can also include chips and other discrete devices.

[0121] A network device can be a device for communicating with a terminal device. This network device can also be referred to as an access network device or a radio access network device. For example, the network device can be a base station. The network device in the embodiments of this application can refer to a radio access network (RAN) node (or device) that connects the terminal device to a wireless network. A base station can generally cover various names as follows, or be replaced with the following names. For example: Node B, evolved Node B (eNB), next generation Node B (gNB), relay station, reception point (RP), transmission reception point (TRP), transmission point (TP), master station, secondary station, multi-mode radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), radio head (RH), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, etc. A base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. A base station can also refer to a communication module, a modem or a chip used in the foregoing device or apparatus. A base station can also be a mobile switching center and a device that undertakes the function of a base station in D2D, V2X, M2M communications, a network-side device in a 6G network, a device that undertakes the function of a base station in a future communication system, etc. A base station can support networks with the same or different access technologies. The specific technologies and specific device forms adopted by the network device in the various embodiments of this application are not limited.

[0122] A base station can be fixed or mobile. For example, a helicopter or a drone can be configured to act as a mobile base station, and one or more cells can move according to the position of the mobile base station. In other examples, a helicopter or a drone can be configured to be used as a device for communicating with another base station.

[0123] In some deployments, the network devices mentioned in various embodiments of the present application may be devices including a CU, or a DU, or a device including a CU and a DU, or a control plane CU node (Central Unit Control Plane (CU-CP)) and a user plane CU node (Central Unit User Plane (CU-UP)) and a DU node.

[0124] In some deployments, multiple RAN nodes cooperate to assist a terminal in achieving wireless access, and different RAN nodes respectively implement partial functions of a base station. For example, the RAN node may be a CU, a DU, a CU-CP, a CU-UP, or a radio unit (RU), etc. The CU and the DU may be separately provided, or may also be included in the same network element, such as a BBU. The RU may be included in a radio frequency device or a radio frequency unit, such as included in an RRU, an AAU, or an RRH.

[0125] In different systems, the CU (or CU-CP and CU-UP), DU, or RU may also have different names, but those skilled in the art can understand their meanings. For example, the radio access network may also be an Open Radio Access Network (O-RAN) architecture. In the ORAN system, the CU may also be referred to as an Open CU (O-CU), the DU may also be referred to as an Open DU (O-DU), the CU-CP may also be referred to as an Open (O-CU-CP), the CU-UP may also be referred to as an Open (O-CU-UP), and the RU may also be referred to as an Open RU (O-RU). Any one of the CU (or CU-CP, CU-UP), DU, and RU in the present application may be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0126] The network device and the terminal device may be deployed on land, including indoor or outdoor, handheld or vehicle-mounted; may also be deployed on water; may also be deployed on airplanes, balloons, and satellites in the air. The scenarios where the network device and the terminal device are located in the embodiments of the present application are not limited.

[0127] The first model involved in each embodiment of this application may be a network or may include multiple networks, and no specific limitation is made thereto. The following will give an example introduction in combination with the schematic architecture diagram of the first model shown in FIG. 2. As shown in FIG. 2, the first model includes at least one first neural network. Optionally, the first model further includes a second neural network and at least one third neural network. FIG. 2 schematically shows multiple first neural networks and one third neural network, and the number of the first neural networks and the third neural network is actually not limited. In the following, the introduction will be mainly given by taking the example that the first model includes one third neural network. At least one first neural network may also be referred to as at least one backbone, backbone network, feature extraction module or feature extraction network. The second neural network may also be referred to as a neck, neck network, feature fusion module or feature fusion network, etc. The third neural network may be referred to as a head, head network, output module, feature inference module or feature inference network, etc.

[0128] As shown in FIG. 2, at least one first neural network can respectively input information of different modalities related to the task and extract features of different modalities related to the task. The second neural network can fuse the features extracted by at least one network to obtain a fusion result. At least one third neural network can respectively obtain outputs related to the task according to the fusion result.

[0129] Any one of the at least one first neural network, the second neural network or any one of the at least one third neural network, etc. may include at least one of ResNet (residual network), MLP network, CNN network, attention mechanism-based network, Transformer network, attention mechanism-based Transformer network, or parts of these neural networks (such as MLP network, CNN network, attention mechanism-based network, Transformer network, or attention mechanism-based Transformer network), etc., and no specific limitation is made thereto. For example, the second neural network may be an encoder in the Transformer network, and the third neural network is, for example, a decoder in the Transformer network, etc.

[0130] Optionally, the first model may also only include at least one first neural network, or only include a second neural network and at least one third neural network, etc., and no specific limitation is made thereto.

[0131] In one possible implementation, at least one first neural network, a second neural network, and at least one third neural network may be deployed in different devices or all deployed in the same device, without specific limitation thereto. For example, at least one first neural network may be deployed in a second communication device, and the second neural network and at least one third neural network may be deployed in a first communication device. Alternatively, at least one first neural network, the second neural network, and at least one third neural network are all deployed in the first communication device.

[0132] FIG. 2 is an example of partitioning the structure of the first model. In fact, there are various other partitioning methods for the first model. Below, with reference to the schematic structural diagram of the first model shown in FIG. 3, another partitioning method of the first model will be introduced. The first model includes at least one module, and any one of the at least one module may include the functions of one or more of a network, a formula, an algorithm, a neuron, a function, etc., without specific limitation thereto.

[0133] As shown in FIG. 3, the first model (or at least one module) includes a feature extraction module and a path parameter output module. Optionally, the first model further includes at least one of a loss obtaining module, a scene parameter update module, a path parameter update module, a model parameter update model, a scene preprocessing module, a path parameter preprocessing module, a path parameter query module, a path-to-channel conversion module, a beam selection module, a path matching module, a mobility management module, a positioning module, a precoding parameter output module, a transmission parameter output module, or a movement route planning output module. Of course, the first model (or at least one module) may further include more modules, without limitation thereto. Below, the basic functions of each module will be introduced.

[0134] A1. The scene preprocessing module may also be referred to as a scene information preprocessing module, an input preprocessing module, etc. The scene preprocessing module is used to preprocess scene information. Optionally, the scene preprocessing module includes an environment preprocessing unit and / or a communication preprocessing unit. The environment preprocessing unit is used to preprocess environment information. The communication preprocessing unit is used to preprocess communication information.

[0135] A2. The feature extraction module is used to extract features required for a task. It is used to extract features of scene information. Optionally, the feature extraction module involved in A2 may be an example of at least one first neural network involved in FIG. 2. Alternatively, the feature extraction module involved in A2 may be an example of at least one first neural network and a second neural network.

[0136] A3. The path parameter output module, which can also be called the output module, or the path parameter set output module, etc. The path parameter output module is used to output a path parameter set based on features. Optionally, the path parameter output module shown in A3 can be an example of one of the at least one third neural network.

[0137] A4. The loss obtaining module can obtain a loss value based on the path parameter set output by the output module.

[0138] A5. The scene parameter update module updates the scene parameter set output by the scene parameter extraction module based on the loss value.

[0139] A6. The path parameter update module is used to update the path parameter set.

[0140] A7. The model parameter update module is used to update the model, for example, the model can be updated.

[0141] A8. The path parameter preprocessing module is used to preprocess the path parameter set.

[0142] A9. The path parameter query module is used to query path parameters.

[0143] A10. The path-to-channel conversion module is used to convert the path parameter set into information of the predicted channel. Optionally, the path-to-channel conversion module shown in A10 can be an example of one of the at least one third neural network.

[0144] A11. The beam selection module is used to select a suitable beam based on the path parameter set or the information of the predicted channel, etc. Optionally, the beam selection module shown in A11 can be an example of one of the at least one third neural network.

[0145] A12. The path matching module is used to match the predicted path with the real path and screen the suitable path. Optionally, the path matching module shown in A12 can be an example of one of the at least one third neural network.

[0146] A13. The mobility management module performs mobility management of the communication device (such as management of cell handover) based on the path parameter set or the information of the predicted channel. The communication device includes the first communication device and / or the second communication device. Optionally, the mobility management module shown in A13 can be an example of one of the at least one third neural network.

[0147] A14. A positioning module that can be used to determine the location of a communication device based on information in a path parameter set or a predicted channel. The location of the communication device is, for example, the positioning of the communication device in a non-line-of-sight (NLoS) scenario. The communication device includes a first communication device and / or a second communication device. Optionally, the positioning module shown in A14 can be an example of one of at least one third neural network.

[0148] A15. A precoding parameter output module that can output precoding information based on information in a path parameter set or a predicted channel. The precoding information includes, for example, a precoding matrix and / or precoding matrix indication information, etc. Optionally, the precoding parameter output module shown in A15 can be an example of one of at least one third neural network.

[0149] A16. A transmission parameter output module that can output transmission parameters for communication based on information in a path parameter set or a predicted channel. The transmission parameters include at least one of, for example, modulation and coding scheme (MCS) indication, beam information, transmission resource indication, or resource scheduling information, etc. Optionally, the transmission parameter output module shown in A16 can be an example of one of at least one third neural network.

[0150] A17. A movement route planning output module that can output a movement route plan for a communication device based on information in a path parameter set or a predicted channel. The movement route plan, for example, recommends that a vehicle move on a path with high throughput and few obstructions. Optionally, the movement route planning output module shown in A17 can be an example of one of at least one third neural network.

[0151] Optionally, each of these at least one module can also be subdivided into at least one unit, and these at least one unit is used to implement the function of the module. The implementation manner of the unit can also refer to the implementation manner of the foregoing module, and no specific limitation is made thereto. For example, the path parameter update module includes a phase update unit, and the phase update unit is used to update the phase of the path. Of course, the other modules shown in FIG. 3 can also be divided into multiple units, and no specific limitation is made thereto.

[0152] FIG. 2 or FIG. 3 is an example of the structure of the first model. In fact, the structure of the first model can be various, and no specific limitation is made thereto.

[0153] The embodiments of the present application can be applied not only to the communication system shown in FIG. 1 above, but also to the communication systems shown in FIG. 4 or FIG. 5 below, which will be introduced separately below.

[0154] Refer to Figure 4, which is a schematic diagram of a communication system applicable to an embodiment of the present application. As shown in Figure 4, the communication system includes a wireless access network 400. The wireless access network 400 can be a next-generation (e.g., 6G or higher) wireless access network, or a traditional (e.g., 5G, 4G, 3G, or 2G) wireless access network. One or more terminal devices (420a-420j, collectively referred to as 420) can be connected to each other or to one or more network devices (410a, 410b, collectively referred to as 410) in the wireless access network 400. Network elements in the wireless communication system are connected through interfaces (e.g., NG, Xn) or air interfaces. Any terminal device in Figure 4 can be used as an example of a second communication device, and any network device in the one or more network devices in Figure 4 can be used as an example of a first communication device. Alternatively, one terminal device in Figure 4 can be used as an example of a first communication device, and another terminal device in Figure 4 can be used as an example of a second communication device.

[0155] FIG4 is only a schematic diagram. The wireless communication system may further include other devices, such as core network devices, wireless relay devices and / or wireless backhaul devices, which are not shown in FIG4 .

[0156] FIG5 illustrates a communication system applicable to an embodiment of the present application. FIG5 is a schematic diagram of an SL communication scenario, for example. The two terminal devices in FIG5 can communicate with each other. One terminal device involved in FIG5 can be used as an example of a first communication device, and the other terminal device can be used as an example of a second communication device.

[0157] The embodiments of the present application are also applicable to satellite communication systems. According to the communication mode of the satellite, the satellite system can be divided into transparent mode and regenerative mode. Transparent mode can also be called transparent forwarding mode. In transparent mode, the satellite can also be called a satellite base station, etc. The satellite is used to convert the frequency and forward the signal, which is generated and sent by the satellite ground station. In regenerative mode, the satellite can be equipped with (or coupled with) a base station or a DU in the base station. The satellite can parse and process the signals received from the ground and send the processed signals to the terminal device to achieve signal regeneration. The satellite can serve as a base station or a terminal device. The satellite can refer to a drone, a hot air balloon, a low-orbit satellite, a medium-orbit satellite, or a high-orbit satellite. The satellite can also refer to a non-ground base station or non-ground device. In a satellite communication system, a satellite or a satellite ground station can be used as an example of a first communication device, and a terminal device can be used as an example of a second communication device. Alternatively, the terminal device can be used as an example of a first communication device, and a satellite or a satellite ground station can be used as an example of a second communication device.

[0158] Figures 1, 4 and 5 are examples of communication systems applicable to the embodiments of the present application, and do not actually limit the communication systems or scenarios to which the embodiments of the present application can be applied.

[0159] The method provided in the embodiments of the present application is described below with reference to the accompanying drawings.

[0160] In the accompanying drawings corresponding to the various embodiments of the present application, all steps represented by dotted lines are optional steps. The first communication device involved in the various embodiments of the present application is, for example, the first communication device involved in Figure 1, the terminal equipment involved in Figure 4 or Figure 5, the network equipment involved in Figure 4, or a satellite or ground station, etc. The second communication device involved in the various embodiments of the present application is, for example, the second communication device involved in Figure 1, the network equipment involved in Figure 4, the terminal equipment involved in Figure 4 or Figure 5, or a satellite or ground station, etc. The structure of the first model involved in the various embodiments of the present application may be, for example, the structure of the first model involved in Figure 2 or Figure 3. And, the various modules included in the first model involved in the embodiment of the present application can be understood as the various modules involved in Figure 3. In addition, if the technical solutions provided in the various embodiments of the present application are applied to other communication systems, the name and / or function of the device may change, and there is no limitation on this.

[0161] Please refer to Figure 6, which is a schematic diagram of a communication method provided in an embodiment of the present application. The following describes the various steps involved in Figure 6.

[0162] S601: A first communication device obtains first scene information.

[0163] The first communication device can independently collect the first scene information. The content of the first scene information can refer to the content of the first scene information discussed above, and any repetitions are not listed here. For example, if the first scene information includes the location of the first communication device, the first communication device can independently obtain the location of the first communication device.

[0164] For example, the first communication device is a communication device, a radar (such as an optical radar (LiDAR)), a satellite, an airplane or a drone, and the first scene information may be captured or sensed by the first communication device. Taking the first communication device as LiDAR as an example, the first scene information sensed by the first communication device may be in the data format of a 3D point cloud. Taking the first communication device as an airplane as an example, the first scene information may include an aerial photo. Taking the first communication device as a satellite as an example, the first scene information includes a satellite photo. Taking the first communication device as a communication device as an example, the first scene information includes information sensed using the transmission path between communication devices. The first scene information includes information obtained by the first communication device through a map server. For example, the first scene information includes an environmental picture or 3D environmental information around the base station downloaded from the map server.

[0165] Alternatively, the first communication device collects source data, and the source data is used to determine the first scene information. The first communication device obtains the first scene information after pre-processing the source data.

[0166] For example, please refer to FIG7 (1), which is a schematic diagram of obtaining the first scene information provided in an embodiment of the present application. As shown in FIG7 (1), the first communication device collects source data, and the first communication device inputs the source data into the scene preprocessing module in the first model to obtain the first scene information. The scene preprocessing module is, for example, the content of the scene preprocessing module involved in FIG3, which is not listed here.

[0167] For example, a first communication device captures a first image. The first image may be a two-dimensional image or an image of three dimensions or greater, without limitation. The first image may be one or more images, and the first image represents, reflects, or includes a scene in which the first communication device is located and / or a scene in which the second communication device is located. For example, the first image includes at least one image pair, where one image in the image pair represents the scene in which the first communication device is located, and the other image in the image pair represents the scene in which the second communication device is located.

[0168] Optionally, the first image can serve as the first scene information, or can be described as the first scene information including the first image, or can be described as the first image carrying the first scene information. Alternatively, the first communication device can preprocess the first image to obtain a second image, and the second image can serve as the first scene information, or can be described as the first scene information including the second image, or can be described as the second image carrying the second scene information. Preprocessing, for example, includes at least one of scaling, enhancing, cropping, splicing, rotating, or flipping. Alternatively, the first communication device calibrates the position of the first communication device and / or the position of the second communication device based on the second image, and updates the calibrated position of the first communication device and / or the calibrated position of the second communication device in the second image, thereby obtaining an updated second image. The updated second image (hereinafter referred to as the third image for ease of description) can serve as the first scene information, or can be described as the first scene information being carried on the third image.

[0169] For example, taking the first image as a two-dimensional environmental map, since the first model input may require a uniform image size, the first image can be processed to conform to this size requirement. Therefore, the first communication device can scale the first image and perform preprocessing such as data enhancement on the first image to obtain a second image. Based on the second image, the first communication device can redetermine the relative positions of the first and second communication devices in the second image and update the redetermined relative positions of the first and second communication devices in the second image in the second image. In this manner, a third image is obtained.

[0170] The following takes the third image carrying the first scene information as an example to introduce the manner in which the third image carries the first scene information.

[0171] For example, the third image may be a map of the environment in which the second communication device is located and the environment in which the first communication device is located, with the locations of the first communication device and the second communication device marked on the third image. Alternatively, the environmental information, the locations of the first communication device and the second communication device may be carried on different channels of the third image.

[0172] Exemplarily, environmental information (e.g., indicating a building outline) in the first scene information is carried in the first channel of the third image, information on the location of the first communication device is carried in the second channel of the third image, and information on the location of the second communication device is carried in the third channel of the third image. For example, if the third image is a red, green, and blue (RGB) image, the environmental information can be carried in the channel corresponding to R in the third image, information on the location of the first communication device can be carried in the channel corresponding to G in the third image, and information on the location of the second communication device can be carried in the channel corresponding to B in the third image.

[0173] If the first model does not limit the number of channels for input, for example, the first model includes a neural network (such as a CNN-type neural network) that can support an arbitrary number of input channels, such as 3 or more, then the first scene information can be carried on any number of channels of the third image. For example, the environmental information in the first scene information can be carried on three channels of the third image. In this case, the environmental information can be regarded as a color picture, such as a satellite image, a map, or an aerial photo. Furthermore, the second communication device carries the information of the location of the first communication device on the fourth channel of the third image, and the information of the location of the second communication device on the fifth channel of the third image.

[0174] Of course, the above is an example of how the third image carries the first scene information, and does not actually limit the way the third image carries the first scene information, nor does it limit the specific form of the first scene information.

[0175] Alternatively, the first communication device may also obtain the first scene information from other devices (such as the second communication device), for example, the second communication device sends the first information to the first communication device, and the first information indicates at least one of the first environment information, the first communication information, the second environment information or the second communication information.

[0176] Optionally, please refer to FIG7 (2), which is a schematic diagram of the first communication device according to an embodiment of the present application obtaining the first scene information. As shown in FIG7 (2), the second communication device can be configured with a fourth model, the second communication device can collect source data, and the second communication device uses the fourth model to process the source data to obtain the first scene information. The implementation method of the fourth model can refer to the content of the first model above, and will not be listed one by one here.

[0177] Alternatively, the first communication device may collect a portion of the first scene information on its own and obtain another portion of the first scene information from another device, thereby obtaining the first scene information. For example, the first scene information includes first environmental information and second environmental information. The first communication device may collect the first environmental information on its own and obtain the second environmental information from the second communication device.

[0178] In the case where the first scene information includes the first environmental parameter and / or the second environmental parameter, the first environmental information can be used to determine the environment within the first distance range where the first communication device is located, or it can be understood that the first environmental parameter indicates the environment within part or all of the range within the environment where the first communication device is located, and the second environmental information can be used to determine the environment within the second distance range where the second communication device is located, or it can be understood that the second environmental information indicates the environment within part or all of the range within the environment where the second communication device is located.

[0179] The implementation of the first model can refer to the implementation of the model discussed above. The first model can be an untrained model or a pre-trained model. The first model can be pre-configured or pre-defined in the first communication device, or can be obtained by the first communication device from another device (such as a second communication device), without limitation. The following example describes how the first communication device pre-trains the first model.

[0180] Exemplarily, the first communication device inputs sample scenario information from a sample set into a first model to obtain a sample path parameter set. After calculating the error between these sample path parameter sets and a true path parameter set (e.g., a path parameter set measured by the first communication device or the second communication device), the gradients are back-propagated. These back-propagated gradients are used, for example, to adjust the model parameters (e.g., weight values ​​or weights) of the first model. After multiple rounds of updates in the above process, a first model, i.e., a pre-trained first model, is obtained.

[0181] S602: The first communication device inputs the first scenario information into the first model to obtain a first path parameter set.

[0182] The first path parameter set can be regarded as local data or final output of the first model, etc., and is not specifically limited thereto. The structure of the first model is different, so the method of obtaining the first path parameter set may also be different, as described below with examples.

[0183] B1. The first model includes a feature extraction module and a path parameter output module.

[0184] Under B1, the first communication device may input the first scene information into the feature extraction module to obtain relevant features of the scene, and input the relevant features of the scene into the path parameter output module, thereby obtaining a first path parameter set.

[0185] B2. The first model includes at least one first neural network, a second neural network, and at least one third neural network. The following description uses the example of at least one first neural network including multiple networks.

[0186] In B2, the first communication device may input multiple types of information included in the first scene information into multiple first neural networks. The multiple first neural networks may extract features of the multiple types of information, thereby obtaining multiple types of features. The multiple types of features may jointly represent features corresponding to the first scene information. The second neural network may fuse the multiple types of features to obtain a fused feature. At least one third neural network may input the first path parameter set based on the fused feature.

[0187] Please refer to Figure 8, which is a schematic diagram of obtaining a path parameter set provided in an embodiment of the present application. Figure 8 takes multiple types of information including environmental information and communication information as an example. As shown in Figure 8, the first communication device inputs environmental information into a first neural network to obtain environmental features, and inputs communication information into another first neural network to obtain communication features. The first communication device can input environmental features and communication features into a second neural network, and the second neural network fuses the environmental features and communication features to obtain scene features. Furthermore, the first communication device can input scene features into a third neural network to obtain a first path parameter set.

[0188] In one possible implementation, the first communications device obtains a first loss value based on the first path parameter set and the first measurement information. Based on the first loss value, the first communications device determines at least one of the second scenario information, the third path parameter set, and the second model. The content of the first measurement information can refer to the content of the first measurement information discussed above.

[0189] The second scene parameter set refers to the result of updating the first scene parameter set. For example, it can be the result of updating some or all parameters in the first scene parameter set, such as the result of updating the position of the first communication device in the first scene parameter set. For example, the environmental information in the scene parameter set can be updated, such as the outline of the building in the environment (if the environment is indoors, the outline of the indoor objects) and the corresponding material properties. It can also be updated in the scene parameter set. It can also be updated in the scene parameter set. It can also be updated in the antenna orientation of the first communication device. It can also be updated in the scene parameter set. Two or more scene parameters can also be updated simultaneously, such as the outline of the building in the environment and the position of the first communication device. The third path parameter set refers to the result of updating the first path parameter set. For example, it can be the result of updating some or all parameters in the first path parameter set. The second model refers to the result of updating the first model. For example, it can be the result of updating some or all model parameters in the first model.

[0190] Optionally, when the first measurement information includes second path parameters, the first communications device may further preprocess the second path parameter set before determining the first loss value based on the second path parameter set. This allows for a more reasonable second path parameter set to be obtained, thereby obtaining a relatively more accurate first loss value, facilitating more accurate training of the first model.

[0191] Exemplarily, preprocessing includes screening the second path parameter set. For example, the first communication device may be preconfigured or predefined with a first power value, or the first communication device may determine the first power value independently or through negotiation with the second communication device. The first communication device may filter out paths with power less than or equal to a third power value from at least one path corresponding to the second path parameter set, and determine path parameters corresponding to the at least one path other than the filtered-out paths. Optionally, the first communication device may use the determined path parameters of these paths as the preprocessed second path parameter set. The third power value may be the difference between a preset maximum power value and the first power value.

[0192] For example, the second path parameter set corresponds to 10 paths. At this time, according to the maximum power value Pmax, the first power value, such as 25 decibels (dB), is subtracted. Among the multiple paths, paths with power weaker than Pmax-25dB are removed, and the preprocessed second path parameter set is obtained based on the filtered path information.

[0193] The preprocessing also includes normalizing the second path parameter set. This prevents excessively large values ​​for a parameter on a particular path, optimizes that parameter, and prevents subsequent training of the first model from overemphasizing that parameter and neglecting other parameters in the path parameter set. Alternatively, the first communication device may only normalize the third path parameter set.

[0194] For example, the first communication device may normalize the path parameter corresponding to each path in the at least one path in the second path parameter set to a value between 0 and 1. For example, the second path parameter set includes the azimuth and elevation angles of the departure angle of the at least one path, and the azimuth and elevation angles of the arrival angle, with the initial range being -pi to pi. The angles may be normalized to a range between 0 and 1 by dividing by 2*pi and then adding 0.5.

[0195] For example, the second path parameter set includes the loss of at least one path. In outdoor scenarios, path loss typically ranges from several tens of decibels (dB) to over 100 dB. Normalization can be performed based on the data distribution of the actual dataset. For example, if the multipath path loss data in the dataset ranges from 50 dB to 150 dB, the loss can be subtracted by 50 and then divided by 100 to normalize to the range of 0 to 1.

[0196] For another example, the second path parameter set includes the delay of at least one path, and the delay is generally tens to hundreds of nanoseconds. The first communication device may also perform normalization processing according to the data distribution of the actual data set.

[0197] Alternatively, please refer to Figure 9, which is a schematic diagram of a preprocessed path parameter set provided in an embodiment of the present application. As shown in Figure 9, the first communication device can input the second path parameter set into the path parameter preprocessing module to obtain a preprocessed second path parameter set. Of course, the path parameter preprocessing module can also filter and / or normalize the second path parameter set. The content of the filtering and normalization processing involved can refer to the content of the filtering and normalization processing discussed above and will not be listed here. The path parameter preprocessing module can, for example, be the path parameter preprocessing module involved in Figure 3.

[0198] The following describes a method for obtaining the second scene parameter set, the second model, or the third path parameter set.

[0199] C1. Obtain a second scene parameter set.

[0200] The first communication device may update the first scenario parameter set based on the first loss value, which is equivalent to using the first scenario parameter set as an adjustable parameter, thereby obtaining updated first scenario parameters, namely, the second scenario parameter set.

[0201] For example, please refer to FIG10 (1), which is a schematic diagram of obtaining a second scenario parameter set provided in an embodiment of the present application. As shown in FIG10 (1), the first loss value of the first communication device and the first scenario parameter are input into a scenario parameter update module in the first model, and the scenario parameter update module can output a second scenario parameter set.

[0202] C2. Obtain the second model.

[0203] The first communication device may update the first model based on the first loss value to obtain a second model.

[0204] Of course, the first communication device can synchronously update the first scenario parameter set and the first model based on the first loss value, so as to obtain the second scenario parameter set and the second model.

[0205] For example, please refer to (2) in Figure 10, which is a schematic diagram of obtaining the second model provided in an embodiment of the present application. As shown in (2) in Figure 10, the first loss value and the first model of the first communication device are input into the scene parameter update module in the first model, and the model update module can output the second model. Specifically, the first communication device can input the first loss value and the model parameters of the first model into the scene update module, and the model update module can output the parameters of the second model.

[0206] When the first model includes multiple networks, the second model may be obtained by updating the model parameters of some or all of the multiple networks in the first model. For example, when the first model includes at least one first neural network, a second neural network, and a third neural network, the second model may be obtained by updating at least one of the model parameters of at least one first neural network, the model parameters of the second neural network, or the model parameters of the third neural network in the first model, without specific limitation.

[0207] In the case that the first model includes multiple modules, the second model can be obtained by updating model parameters of some or all of the multiple modules in the first model.

[0208] The first communication device can execute at least one of C1 and / or C2 in each training without limitation. For example, the first communication device can fix the model and only update the scene parameter set through gradient feedback; it can also fix the scene parameter set and only update the model; it can also not fix both and update both at the same time. Simultaneous updating also includes: when updating, the two modes use different learning rates, such as the update of the scene parameter set uses a learning rate of 0.1, that is, each update changes faster, while the update of the model uses a learning rate of 0.001, that is, each update changes slower. It can also be updated in turn, such as the scene parameter set is updated once and then the model is updated once, and only one mode is working at a time, and a rotating working method is adopted, which is not limited to this.

[0209] C3. Obtain a third path parameter set.

[0210] The first communication device may obtain at least one of the second scenario parameter set and the second model based on C1 or C2. The first communication device may input the second scenario parameter set into the first model to obtain a third path parameter set. Alternatively, the first communication device may input the second scenario parameter set into the second model to obtain a third path parameter set. Alternatively, the first communication device may input the second scenario parameter set into the second model to obtain a third path parameter set, etc. Alternatively, the first communication device may input the first path parameter set and the first loss value into a path parameter update module in the first model to obtain a third path parameter set, without limitation.

[0211] For example, please refer to (3) in Figure 10, which is a schematic diagram of obtaining a third path parameter set provided in an embodiment of the present application. As shown in (3) in Figure 10, the first loss value and the first path parameter set of the first communication device are input into the path parameter update module in the first model, and the path parameter update module can output the third path parameter set.

[0212] In one possible design, the first communication device may also obtain a second loss value based on the third path parameter set, and determine at least one of a third model, a third scenario parameter set, or a fifth path parameter set based on the second loss value. The content of determining the second loss value may refer to the content of determining the first loss value discussed above. The third model is obtained by updating the second model based on the second loss value, wherein the method for obtaining the third model may refer to the content of obtaining the second model discussed above. The third scenario parameter set is obtained by updating the second scenario parameter set based on the second loss value, and the method for obtaining the third scenario parameter set may refer to the content of obtaining the second scenario parameter set discussed above. The fifth path parameter set is obtained by updating the third path parameter set based on the second loss value, and the method for obtaining the fifth path parameter set may refer to the content of obtaining the third path parameter set discussed above.

[0213] For example, please refer to (1) in Figure 11, which is a schematic diagram for obtaining the third model. As shown in (1) in Figure 11, the first communication device inputs the second loss value and the second model into the model update module in the second model to obtain the third model. Please refer to (2) in Figure 11, which is a schematic diagram for obtaining the third scenario parameter set. As shown in (2) in Figure 11, the first communication device inputs the second loss value and the second scenario parameter into the model update module in the second model to obtain the third scenario parameter set. As shown in (3) in Figure 11, the first communication device inputs the second loss value and the third path parameter set into the model update module in the second model to obtain the fourth path parameter set.

[0214] In this way, the first communication device can perform multiple rounds of iterative updates on at least one of the first model, the first scenario parameter set, or the first path parameter set.

[0215] Optionally, the first communication device may preprocess the first path parameter set to obtain a fourth path parameter set, and the first communication device may send the fourth path parameter set to the second communication device. For example, the first communication device may filter the first path parameter set based on the second power value and normalize the filtered first path parameter set. For example, the first communication device may input the fourth path parameter set into a path-to-channel conversion module to obtain predicted channel information. The method for preprocessing the first path parameter set can refer to the method for preprocessing the second path parameter set discussed above, and any repetitions are not listed here. Optionally, the first communication device may also process the fourth path parameter set to obtain predicted channel information.

[0216] Optionally, the first communication device may send second information to the second communication device. The second information indicates part or all of the third path parameter set and / or part or all of the second scenario parameter set. In this way, the second communication device can obtain a more accurate scenario parameter set and / or path parameter set.

[0217] In the case where the second scenario parameter includes the second environmental parameter corresponding to the second communication device, the second communication device may optionally further fuse the second environmental parameters from the second scenario parameters of multiple first communication devices (which may be referred to as the second environmental parameters corresponding to the multiple first communication devices). Fusion of the second environmental parameters corresponding to the multiple first communication devices may be a weighted summation of the second environmental parameters corresponding to the multiple first communication devices, such as averaging, etc., which is not specifically limited. The second communication device may use the fusion result as the second environmental parameter, which is equivalent to calibrating the second environmental parameter in the second scenario parameter from a certain first communication device. In this way, by combining the second environmental parameters corresponding to the multiple first communication devices, a more accurate second environmental parameter is obtained.

[0218] The following takes the case where the first model includes at least one first neural network, a second neural network and a third neural network, and at least one first neural network includes three first neural networks, one third neural network among the three first neural networks is ResNet, one third neural network is MLP, and one third neural network is MLP, the second neural network is the encoder in the Transformer, the third neural network is the decoder in the Transformer, and the first scene information includes environmental information and communication information, and the communication information includes the position of the first communication device and the position of the second communication device as an example, to exemplify the communication method involved in Figure 6.

[0219] Please refer to Figure 12, which is a schematic diagram of a communication method provided in an embodiment of the present application. The following describes the various steps involved in Figure 12.

[0220] S1201. The first communication device inputs first scene information into at least one first neural network in the first model.

[0221] Optionally, S1201 includes S1201a, S1201b and S1201c. Among them, S1201a is: inputting the environmental information in the first scene information into one of the at least one first neural networks (such as ResNet) to obtain environmental features. S1201b is: the first communication device inputs the position of the first communication device into one of the at least one first neural networks (such as MLP) to obtain first position features. S1201c is: the first communication device inputs the position of the second communication device into one of the at least one first neural networks (such as MLP) to obtain second position features. In other words, each type of information in the first scene information is input into different networks for feature extraction.

[0222] For example, the environmental information involved in the embodiments of the present application includes information about the environment in which the second communication device is located, specifically, for example, picture information indicating the position and height of various objects (such as buildings, trees, etc.) and material information of the objects. Optionally, the environmental information may also include information about the environment in which the first communication device is located. Environmental features are used to represent features corresponding to the environment. Environmental features may be in the form of a vector, matrix, or sequence, and are not limited to this. The first position feature may also be in the form of a vector, matrix, or sequence, and are not limited to this. The second position feature may also be in the form of a vector, matrix, or sequence, and are not limited to this.

[0223] S1202. The first communication device inputs at least one feature output by the first neural network into the second neural network to obtain a fusion feature.

[0224] S1202 includes S1202a, S1202b, and S1202c. S1202a involves the first communication device inputting environmental features into a second neural network (e.g., an encoder). S1202b involves the first communication device inputting first position features into a second neural network (e.g., an encoder). S1202c involves the first communication device inputting second position features into a second neural network (e.g., an encoder). In this manner, the encoder performs feature fusion on the environmental features, the first position features, and the second position features to obtain a fused feature.

[0225] For example, environmental features, first position features, and second position features form a sequence of B*256*100 dimensions (for ease of description, this sequence is referred to as the first sequence). The length of the first sequence is 100, and each element of the first sequence can be called a token, that is, the first sequence includes 100 tokens, and each token is a vector of length 256, that is, the dimension of the first sequence is B*100*256. B is the batch size, and the maximum batch size is the total number of samples. At this point, it is equivalent to converting the first scene information into the first sequence. The first communication device can input the first sequence into the second neural network. Optionally, the second neural network includes a self-attention mechanism. The first communication device copies the first sequence three times to obtain three vectors. These three vectors are represented as Q, K, and V, respectively. For ease of distinction, the three vectors are described as Q1, K1, and V1 below. The first communication device inputs these three vectors into the self-attention mechanism for processing to obtain the attention output feature A1, and the dimension of feature A1 is the same as that of the first sequence. Since the order of the sequence is ignored during self-attention processing, a position embedding P1 can be configured by inputting the first sequence, Q1, K1, V1, or (Q1 and K1). This ensures that the second neural network learns in an orderly manner. The position embedding P1 may be different for different inputs. Feature A1 can be considered an example of a fused feature.

[0226] S1203. The first communication device inputs the fusion feature into the third neural network to obtain a first path parameter set.

[0227] For example, the first communication device inputs the fused feature (such as feature A1) into a third neural network, and obtains the first path parameter set through the third neural network. Optionally, the third neural network can be a Transformer decoder. Optionally, the Transformer decoder includes a self-attention mechanism and a cross-attention mechanism.

[0228] For example, a sequence of length X of the first communication device (hereinafter referred to as the second sequence for ease of description) is input into the decoder, and the dimension of the second sequence can be expressed as B*X*256. The first communication device can copy the second sequence three times to obtain Q, K and V. For ease of distinction, K, Q and V here will be represented as Q2, K2, V2 below. The first communication device can input Q2, K2, V2 into the self-attention mechanism module in the third neural network for processing to obtain the output feature A2. The dimension of feature A2 is B*X*256. Optionally, a position vector P2 can also be configured for the second sequence, or Q2, or K2, or V2 or (Q2 and K2). For second sequences of different lengths, the position vector P2 may be different.

[0229] The first communication device copies feature A1 twice to obtain K3 and V3, and uses feature A2 as Q3. It uniformly inputs Q3, K3, and V3 into the cross-attention module in the third neural network for processing to obtain attention output feature A3. The dimension of feature A3 can be B*X*256. Optionally, a position vector P2 can be configured for A2 or Q3, and similarly, a position vector P1 can be configured for K3, V3, or (V3 and K3). After feature A3 is processed by the fully connected network in the third neural network, a sequence (such as a third sequence) is obtained. The dimension of the third sequence is, for example, B*X*5. X represents the X paths predicted by the network (hereinafter referred to as X predicted paths), and the dimension 5 represents the 5 values ​​of each multipath (1 multipath existence probability and 4 multipath information). The third sequence is an example of the first path parameter set. The first path parameter set here is equivalent to indicating X predicted paths.

[0230] S1204. The first communication device determines beam information according to the first path parameter set.

[0231] The beam information indicates a beam used for communication, and includes, for example, an index of the beam used for communication.

[0232] In one possible implementation, the first communication device may directly input the first path parameter set into the beam selection module to obtain beam information. In another possible implementation, the first communication device may match the path indicated by the first path parameter set with the actual path. Based on the successfully matched path, the path parameter set corresponding to the successfully matched path (e.g., a sixth path parameter set) may be obtained. The first communication device may then input the sixth path parameter set into the beam selection module to obtain beam information.

[0233] For example, a first path parameter set indicates X predicted paths, and a second path parameter set indicates Y actual paths. The first communication device may match the X predicted paths with the Y actual paths to obtain Y1 predicted paths that best match the Y actual paths. The Y actual paths may be pre-acquired or obtained through actual measurement, without limitation. Please refer to FIG13 , which is a schematic diagram of processing a path parameter set according to an embodiment of the present application. As shown in FIG13 , the first communication device may input the X predicted paths into a path matching module and perform a binary matching of the X predicted paths with the Y actual paths, such as using the Hungarian algorithm. Y1 predicted paths are selected from the X predicted paths, where Y1 and Y may be the same or different. For example, the first communication device may use binary matching to ensure that the Y1 predicted paths in X best match the Y actual paths. Thus, the parameters corresponding to the Y1 predicted paths are an example of a sixth path parameter set.

[0234] Optionally, when binary matching is performed on the X predicted paths and the Y real paths, a path parameter query module may be used to query the predicted paths that match the Y real paths.

[0235] A representation of Y1 prediction paths can refer to the following formula (1).

[0236] Where Ri represents the i-th true path, Rpxi represents the i-th predicted path, N represents the total number of Y1 selected predicted paths, and x represents the set of Y1 selected predicted paths. i is a positive integer.

[0237] In one possible design, referring again to FIG. 13 , the first communication device inputs Y1 predicted paths and Y actual paths into a loss acquisition module to obtain a first loss value. The loss acquisition module may separately calculate classification loss and path parameter loss. The first loss value may be obtained, for example, by taking a weighted sum of the classification loss and path parameter loss.

[0238] A formula for calculating the first loss value can refer to the following formula (2).

[0239] Among them, P xi represents the classification loss between the i-th predicted path and the true path, L ray (R i ,R pxi ) represents the path parameters of the i-th predicted path and the i-th multipath information loss Lray.

[0240] For example, the first communication device can take the existence probability of each predicted path and calculate the cross entropy with the category of the actual path to obtain the classification loss. Optionally, the existence probability of the path can only include 1 or 0, that is, either the path exists or the path does not exist. In this case, the predicted path parameter set contains only one value related to the path category: that is, the existence probability of the path. Alternatively, the path can be divided into more categories: such as indoor NLoS, indoor line of sight (LoS), outdoor LoS or outdoor NLoS. In this case, the dimension of the predicted path parameter set is also increased accordingly. For example, when the path category only includes the existence probability, the dimension of the path parameter of a path in the first path parameter set can be 1+4=5 dimensions. When the path category also includes indoor LoS, indoor line of sight LoS, outdoor LoS or outdoor NLoS, the dimension of the path parameter set of a path in the first path parameter set can be 4+4=8 dimensions.

[0241] For example, the first communication device can calculate the deviation of path parameters between the predicted path and the actual path, including the deviation of values ​​such as departure angle, arrival angle, path loss, and delay, and obtain the final path parameter loss by weighted calculation of some or all of these deviations.

[0242] A calculation formula for path parameter loss can refer to the following formula (3). ray (R i ,R pxi )=diff([dod, doa, pathloss, delay]|x) (3)

[0243] Among them, diff represents deviation, dod, doa, pathloss, and delay represent the departure angle, arrival angle, path loss, and delay of the path (such as the predicted path or the actual path), respectively.

[0244] Optionally, the first communication device may calculate a gradient based on the first loss value, and then transmit the gradient back to update the model parameters of some or all networks in the first model. For example, at least one of the model parameters of the first neural network, the model parameters of the second neural network, and the model parameters of the third neural network may be updated. Optionally, the first communication device may also update the position vector P1 and the position vector P2 based on the first loss value to obtain the second model.

[0245] Optionally, referring to FIG. 13 , the first communication device inputs Y1 predicted paths into the signal conversion module to obtain predicted channel information. The content of the predicted channel information can be referenced above, and any repetitions are omitted. Of course, the predicted channel information here refers to channel information predicted based on the model.

[0246] Optionally, refer to FIG14 , which is a schematic diagram of processing a first path parameter set according to an embodiment of the present application. As shown in FIG14 , the first communication device may input information about the first path parameter set or the predicted channel into a mobility management module to obtain mobility management (such as cell handover management) information of the communication device, such as a mobility management policy.

[0247] Optionally, referring to FIG. 14 , the first communication device may input the first path parameter set or the predicted channel information into a positioning module to obtain the position of the communication device (e.g., positioning in an NLoS (non-line-of-sight) scenario). The position of the communication device may be, for example, the position of the first communication device and / or the second communication device.

[0248] Optionally, referring to FIG14 , the first communication device may input the first path parameter set or the predicted channel information into the precoding parameter output module to obtain precoding information (which may be a precoding matrix, precoding matrix indication information, etc.).

[0249] Optionally, please continue to refer to Figure 14. The first communication device can input the first path parameter set or the predicted channel information into the transmission parameter output module to obtain the transmission parameters of the communication, such as at least one of the MCS indication, beam information, transmission resource indication or resource scheduling information.

[0250] Optionally, please continue to refer to Figure 14. The first communication device can input the first path parameter set or the predicted channel information into the mobile route planning output module to obtain the mobile route planning of the communication device, such as suggesting that the vehicle move on a path with high throughput and less obstruction.

[0251] Optionally, regardless of the structure of the first model, the first communication device can use the first model to determine at least one of beam information, predicted channel information, mobility management information, the location of the communication device, precoding information, transmission parameters or mobile route planning, without specific limitation.

[0252] The following takes the case where the first model includes a first neural network, a second neural network and a third neural network, and the first neural network includes one first neural network, one third neural network in the one first neural network is ResNet, the second neural network is the encoder in the Transformer, the third neural network is the decoder in the Transformer, and the first scene information includes environmental information and communication information, and the communication information includes the position of the first communication device and the position of the second communication device as an example, to exemplify the communication method involved in Figure 6.

[0253] Please refer to Figure 15, which is a schematic diagram of a communication method provided in an embodiment of the present application. The following describes the various steps involved in Figure 15.

[0254] S1501. The first communication device inputs first scene information into the first neural network in the first model to obtain scene features.

[0255] In the embodiment of the present application, the first scenario information includes environmental information, the position of the first communication device, and the position of the second communication device.

[0256] For example, if the first scene information is carried on the third image, which has dimensions of 3*128*128, the first model's input dimensions are B*3*128*128. B is the batch dimension, indicating that B samples are fed into the first model; 3 is the number of channels; and 128*128 means that the third image has a height and width of 128 pixels each.

[0257] After the first neural network's forward computation, the input of the first model becomes dimensional to B*256*10*10, where B is the batch dimension, 256 is the number of channels, and 10*10 represents the original image's length and width after being compressed by the first neural network. This shows that after the third image is processed by the first neural network, the number of channels increases, while the length and width decrease. After the first neural network's input of the first model is transformed from these two 10*10 dimensions into a single dimension of 100, thus obtaining scene features of B*256*100 dimensions.

[0258] In one possible implementation, the first neural network involved in S1501 may be an untrained network, or the first neural network involved in S1501 may be a pre-trained first neural network. The following describes a method for pre-training the first neural network by the first communication device.

[0259] Exemplarily, the first communication device can perform dimensionality reduction processing on the scene features output by the first neural network, for example, directly perform dimensionality reduction processing on the scene features through a network (such as MLP) to obtain a fifth path parameter set. The first communication device can obtain a third loss value based on the fifth path parameter set and the second measurement information, adjust the model parameters of the first neural network based on the third loss value, and pre-train the first neural network. The content of the fifth path parameter set can refer to the content of the first path parameter set discussed above, and will not be listed here. The content of the second measurement information can refer to the content of the first measurement information discussed above, and the content of determining the third loss value and adjusting the model parameters of the first neural network can refer to the content of the third loss value discussed above, and the content of adjusting the model parameters of the first neural network can refer to the content of the third loss value discussed above, and the content of adjusting the model parameters of the first neural network, and the repeated parts will not be repeated.

[0260] In this case, when pre-training the first model, there is no need to introduce the second neural network and the third neural network, thereby reducing the processing load of the pre-trained first model.

[0261] For example, the dimension of the scene features output by the first neural network is B*256*10*10, where B is the batch size and 256 is the number of channels. After being transformed, 10*10 becomes a dimension of 100, that is, B*256*100. The first communication device uses a small network (which can be MLP or ResNet) to transform the dimension of 256 into a dimension of 4, that is, B*4*100, which is the first path parameter set. The first communication device adjusts the position of each dimension to obtain an output B*100*4, where the dimension of 100 contains the 100 paths predicted by the network for each sample, and the dimension of 4 contains the information of each path (departure angle, arrival angle, path loss and delay). If you want to increase the predicted information, you can simply add the dimension of 4. For example, if you add the departure angle and arrival angle, then the departure angle and arrival angle also contain the azimuth and pitch angle respectively, and the dimension of 4 becomes 6.

[0262] Alternatively, the output of the first neural network is B*100*2*2, where B is the batch dimension and 100 is the number of channels. After being transformed, 2*2 becomes a dimension of 4, that is, B*100*4. In this case, B is still the batch dimension (i.e., B samples). The dimension 100 contains the 100 paths predicted by the network for each sample, and the dimension 4 contains the parameters of each path (departure angle, arrival angle, path loss, and delay). To increase the predicted information, just add the dimension 4. For example, if the departure angle and arrival angle also include the azimuth and pitch angles, respectively, the dimension 4 becomes 6.

[0263] S1502: The first communication device inputs the scene features into the second neural network to obtain fused features. In this case, the fused features can be regarded as being obtained by further feature extraction and fusion of the scene features.

[0264] For example, the first communication device may regard the scene features as a sequence (such as called the fourth sequence). The fourth sequence and the first sequence discussed in Figure 12 above may be the same or different sequences, and there is no limitation on this. The length of the fourth sequence is 100, and each element of the fourth sequence can be called a token, that is, the fourth sequence includes 100 tokens, and each token is a vector of length 256, that is, the dimension of the fourth sequence is B*100*256. So far, it is equivalent to converting the first scene information into the fourth sequence. The first communication device can input the fourth sequence into the second neural network. Optionally, the second neural network includes a self-attention mechanism. The first communication device copies the first sequence three times to obtain three vectors. These three vectors are represented as Q, K, and V respectively. For ease of distinction, the three vectors are described as Q1, K1, and V1 below. The first communication device inputs these three vectors into the self-attention mechanism for processing to obtain the attention output feature A3, and the dimension of feature A3 is the same as that of the first sequence. Since the order of the sequence is ignored during self-attention processing, a position embedding P1 can be configured by inputting the first sequence, Q1, K1, V1, or (Q1 and K1). This ensures that the second neural network learns in an orderly manner. The position embedding P1 may be different for different inputs. Feature A3 can be considered an example of a fused feature.

[0265] S1503. The first communication device inputs the fusion feature into the third neural network to obtain a first path parameter set.

[0266] For example, the first communication device inputs the fused feature (e.g., feature A3) into a third neural network, and obtains the first path parameter set through the third neural network. Optionally, the third neural network can be a Transformer decoder. Optionally, the Transformer decoder includes a self-attention mechanism and a cross-attention mechanism.

[0267] The process of the third neural network processing the fusion features can refer to the process of the third neural network processing the fusion features discussed in Figure 12 above, and the repeated parts are not listed here.

[0268] S1504. The first communication device determines a first loss value according to the first path parameter set and the second path parameter set.

[0269] The content of determining the first loss value can also refer to the content of determining the first loss value discussed in Figure 12 above, and the repeated parts are not listed again.

[0270] In another possible embodiment, the first model only includes the first neural network, and the first communication device can perform dimensionality reduction processing on the scene features output by the first neural network. For example, the scene features can be directly reduced in dimension through a network (such as MLP or ResNet) to obtain the second path scene parameter set. That is, in this case, when training the first model, there is no need to introduce the second neural network and the third neural network, thereby reducing the processing load of training the first model. In this way, the first communication device can obtain the first loss value based on the second path parameter set and the first measurement information. Optionally, the first communication device can adjust at least one of the model parameters of the first neural network, the first path parameter set, and the first scene information based on the first loss value.

[0271] Of course, the above-mentioned first model in FIG15 is an example, and the implementation method of the first model is not actually limited.

[0272] In one possible implementation, the first communication device may obtain beam information based on the first path parameter set. The method for obtaining the beam information and the content of the beam can refer to the above-mentioned FIG. 12 for the method for obtaining the beam information and the content of the beam, respectively, and the repetitions are not listed here.

[0273] Optionally, the first communication device may calculate a gradient based on the first loss value, and then transmit the gradient back to update model parameters of some or all networks in the first model. For example, at least one of the model parameters of the first neural network, the model parameters of the second neural network, and the model parameters of the third neural network may be updated. Optionally, the first communication device may also update position vector P1 and position vector P2 based on the first loss value to obtain the second model.

[0274] Optionally, the first communication device inputs the Y1 predicted paths into the signal conversion module to obtain the predicted channel information. The content of the predicted channel information can refer to the content of the predicted channel information discussed above, and the repeated parts are not listed again.

[0275] Optionally, the first communication device may input the first path parameter set or the predicted channel information into the mobility management module to obtain mobility management (such as cell switching management) information of the communication device, such as a mobility management policy.

[0276] Optionally, the first communication device may input the first path parameter set or the predicted channel information into a positioning module to obtain the position of the communication device (e.g., positioning in an NLoS (non-line-of-sight) scenario). The position of the communication device may be, for example, the position of the first communication device and / or the second communication device.

[0277] Optionally, the first communication device may input the first path parameter set or the predicted channel information into a precoding parameter output module to obtain precoding information.

[0278] Optionally, the first communication device can input the first path parameter set or the predicted channel information into the transmission parameter output module to obtain the transmission parameters of the communication, such as at least one of the MCS indication, beam information, transmission resource indication or resource scheduling information.

[0279] Optionally, the first communication device may input the first path parameter set or the predicted channel information into a mobile route planning output module to obtain a mobile route planning of the communication device, such as suggesting that the vehicle move on a path with high throughput and less obstruction.

[0280] Optionally, S1504 is an optional step.

[0281] The following takes the first communication device as a network device and the second communication device as a UE as an example to illustrate the interaction between the first communication device and the second communication device involved in the communication method involved in FIG6 .

[0282] Please refer to Figure 16, which is a schematic diagram of a communication method provided in an embodiment of the present application. The following describes the various steps involved in Figure 16.

[0283] S1601: A network device sends third information to a UE. Correspondingly, the UE receives the third information from the network device. The third information indicates configuration information. The configuration information is used to indicate the content of measurement information fed back by the UE.

[0284] The configuration information, for example, indicates a tag data type configuration, which includes at least one of the following: the content of the feedback measurement information, the data type of the measurement information, or the format of the measurement information. The configuration information may also indicate a first threshold (also known as a weak path threshold). The first threshold indicates a power below which paths do not require feedback, thereby reducing feedback overhead. For example, a first threshold of 20 dB indicates that the UE does not need to provide feedback to the network device for paths with a power below 20 dB.

[0285] For example, the data types of measurement information are: [dod, doa, pathloss, delay] and [transmit beam index (Tx beam idx), receive beam index (Rx beam idx), power (power), time of arrival (TOF / time of flight TOA)]. The time of arrival refers to the flight time, arrival time, or delay of a signal along a path.

[0286] S1602: The UE performs channel measurement to obtain first measurement information. The content of the first measurement information and the method of obtaining the first measurement information can refer to the content of the first measurement information and the method of obtaining the first measurement information discussed above, and the repeated parts are not listed again.

[0287] S1603: The UE sends first measurement information to the network device. Correspondingly, the network device receives the first measurement information from the UE.

[0288] S1604: The network device inputs the first scenario information into the first model to obtain a first path parameter set.

[0289] The content of the first scene information, the content of the first model, and the method of obtaining the first path parameter set can all refer to the content of the first scene information, the content of the first model, and the method of obtaining the first path parameter set discussed above, and the repetitions are not listed again.

[0290] S1605: The network device obtains a first loss value according to the first path parameter set and the first measurement information, and trains a first model.

[0291] The content of the first loss value and the method for obtaining the first loss value can refer to the content of the first loss value and the method for obtaining the first loss value discussed above, and are not limited thereto. Training the first model includes updating the first model, adjusting some or all model parameters of the first model, etc.

[0292] S1606: The network device sends part or all of the first path parameter set to the UE. Correspondingly, the UE receives part or all of the first path parameter set from the network device.

[0293] S1601 , S1602 , S1603 , S1605 and S1606 are optional steps, which are indicated by dotted lines in FIG16 .

[0294] Optionally, after obtaining part or all of the first path parameter set, the UE can select a beam.

[0295] For example, in a millimeter wave communication scenario, the predicted angles in the first path parameter set can be used to assist UE beam selection. For example, the first path parameter set indicates X paths, where X is 100. The UE can filter the path probabilities according to a second threshold. For example, if the second threshold is 0.9, the UE can filter out all paths from the 100 paths with a probability greater than 0.9. For example, after filtering, two paths are obtained: one path (e.g., path 1) with a probability of 0.95 and one path (e.g., path 2) with a probability of 0.99. The path parameters of these two paths are as follows: [0.99, azimuth of arrival angle: 130 degrees, elevation of arrival angle: 85 degrees, path loss 80 dB, delay 100 nanoseconds] and [0.95, azimuth of arrival angle: 100 degrees, elevation of arrival angle: 89 degrees, path loss 100 dB, delay 150 nanoseconds].

[0296] To obtain better signal strength, the UE can select the beam corresponding to path 1. If the beam corresponding to path 1 is blocked by a moving object in the environment, the UE can quickly select the beam corresponding to path 2 to complete a fast switching.

[0297] In the above optional implementation, the BS can directly inform the UE of which direction the beam will be stronger, and provide the probability of the existence of multiple paths, so that the UE can make decisions autonomously and quickly.

[0298] It is understood that, in order to implement the functions in the above embodiments, the base station and the terminal include hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily appreciate that, in conjunction with the units and method steps of the various examples described in the embodiments disclosed in this application, this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in hardware or in a computer software-driven hardware manner depends on the specific application scenario and design constraints of the technical solution.

[0299] Please refer to Figure 17, which is a schematic diagram of the structure of a communication device provided in an embodiment of the present application. The communication device can be used to implement the functions of the first communication device or the second communication device in the above method embodiment, and thus can also achieve the beneficial effects possessed by the above method embodiment. In an embodiment of the present application, the communication device can be the first communication device involved in Figure 1, the terminal device involved in Figure 4 or Figure 5, the network device involved in Figure 4, the network device involved in Figure 4, the terminal device involved in Figure 4 or Figure 5, or a satellite or ground station, etc., and can also be a module (such as a chip) in these devices or equipment.

[0300] As shown in Figure 17, the communication device 1700 includes a processing module 1710 and a transceiver module 1720. The communication device 1700 is used to implement the functions of the first communication device or the second communication device in the method embodiments shown in Figures 6, 12, 15 or 16 above.

[0301] For example, the communication device 1700 is used to implement the function of the first communication device in the method embodiment shown in FIG. 6 .

[0302] Exemplarily, the transceiver module 1720 is used to obtain the first scenario information, and the processing module 1710 is used to obtain the first path parameter set, etc. The communication device 1700 may also perform other steps performed by the first communication device involved in FIG6 , which are not listed here one by one.

[0303] For another example, the communication device 1700 is used to implement the function of the first communication device in the method embodiment shown in FIG12 .

[0304] Exemplarily, the transceiver module 1720 is used to obtain the first scenario information, and the processing module 1710 is used to obtain the first path parameter set, etc. The communication device 1700 may also perform other steps performed by the first communication device involved in FIG12 , which are not listed here one by one.

[0305] For another example, the communication device 1700 is used to implement the function of the first communication device in the method embodiment shown in FIG15 .

[0306] Exemplarily, the transceiver module 1720 is used to obtain the first scenario information, and the processing module 1710 is used to obtain the first path parameter set, etc. The communication device 1700 may also perform other steps performed by the first communication device involved in FIG15 , which are not listed here one by one.

[0307] For another example, the communication device 1700 is used to implement the function of the first communication device in the method embodiment shown in FIG16 .

[0308] Exemplarily, the processing module 1710 is used to obtain the first scene information; the processing module 1710 is used to obtain the first path parameter set, etc. The communication device 1700 may also perform other steps performed by the first communication device involved in FIG16 , which are not listed here one by one.

[0309] A more detailed description of the processing module 1710 and the transceiver module 1720 can be directly obtained by referring to the relevant description in the method embodiment shown in Figure 6, Figure 12, Figure 15 or Figure 16, and will not be repeated here.

[0310] Please refer to Figure 18, which is a structural diagram of a communication device provided in an embodiment of the present application. As shown in Figure 18, the communication device 1800 includes a processor 1810 and an interface circuit 1820. The processor 1810 and the interface circuit 1820 are coupled to each other. It will be understood that the interface circuit 1820 can be a transceiver or an input / output (I / O) interface. Optionally, the communication device 1800 may further include a memory 1830 for storing instructions executed by the processor 1810 or storing input data required for the processor 1810 to run instructions or storing data generated after the processor 1810 runs instructions. Optionally, the processor 1810 may include one or more modules, such as the first model mentioned above. For example, the processor includes an AI chip, which is deployed with one or more models.

[0311] Communication device 1800 can be used to implement the method shown in Figure 6, Figure 12, Figure 15, or Figure 16. Optionally, processor 1810 can execute any of the method embodiments shown in Figure 6, Figure 12, Figure 15, or Figure 16 based on the model. Optionally, processor 1810 is used to implement the functions of processing module 1710, and interface circuit 1820 is used to implement the functions of transceiver module 1720.

[0312] When the communication device is a chip used in a terminal device, the terminal device chip implements the functions of the terminal device in the above method embodiments. The terminal device chip receives information from other modules in the terminal device (such as a radio frequency module or antenna), and the information is sent by the network device to the terminal device; or the terminal device chip sends information to other modules in the terminal device (such as a radio frequency module or antenna), and the information is sent by the terminal device to the network device.

[0313] When the above-mentioned communication device is a module applied to a network device, the network device module implements the functions of the network device in the above-mentioned method embodiment. The network device module receives information from other modules in the network device (such as a radio frequency module or an antenna), and the information is sent by the terminal device to the network device; or the network device module sends information to other modules in the network device (such as a radio frequency module or an antenna), and the information is sent by the network device to the terminal device. The network device module here can be a baseband chip of the network device, or it can be a DU or other module. The DU here can be a DU under the open radio access network (O-RAN) architecture.

[0314] It is understood that the processor involved in the various embodiments of the present application may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor. In addition, the memory involved in the various embodiments of the present application may include volatile memory, such as random access memory (RAM). The memory may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD) or solid state drive (SSD).

[0315] An embodiment of the present application provides another example of a communication device, which includes at least one processor and at least one memory, the at least one processor and the at least one memory being coupled, the at least one memory being used to store instructions, and when the instructions are executed by the at least one processor, the communication device executes the method in the above embodiment. Taking the communication device including a processor and a memory as an example, as shown in Figure 19, the communication device 1900 includes a processor 1910 and a memory 1920. The processor 1910 and the memory 1920 are coupled, and instructions are stored in the memory 1920. When the instructions stored in the memory 1920 are executed by the processor 1910, the communication device 1900 executes any of the method embodiments involved in Figures 6, 12, 15 or 16 above. Optionally, the communication device can also implement the functions of any of the first communication devices described above, or the functions of any of the second communication devices described above.

[0316] Optionally, the processor 1910 may include one or more modules, such as the first model mentioned above. For example, the processor includes an AI chip that is deployed with one or more models. Optionally, the processor 1910 may execute the method shown in Figures 6, 12, 15, or 16 above based on the model.

[0317] An embodiment of the present application provides a communication system, comprising a first communication device and a second communication device. The first communication device may, for example, implement the functions of the first communication device in any of the method embodiments described in FIG. 6 , FIG. 12 , FIG. 15 , or FIG. 16 . The second communication device may, for example, implement the functions of the second communication device in any of the method embodiments described in FIG. 6 , FIG. 12 , FIG. 15 , or FIG. 16 .

[0318] An embodiment of the present application provides a chip system, comprising: a processor and an interface, wherein the processor is configured to call and execute instructions from the interface, and when the processor executes the instructions, implements any of the method embodiments described in FIG. 6 , FIG. 12 , FIG. 15 , or FIG. 16 .

[0319] An embodiment of the present application provides a computer-readable storage medium for storing computer programs or instructions, which, when executed, implements any of the method embodiments described in FIG. 6 , FIG. 12 , FIG. 15 , or FIG. 16 .

[0320] An embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, implements any of the method embodiments described in FIG. 6 , FIG. 12 , FIG. 15 , or FIG. 16 .

[0321] The method steps in each embodiment of the present application can be implemented in hardware or in software instructions that can be executed by a processor. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, mobile hard disk, CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. The storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC. In addition, the ASIC can be located in a base station or a terminal. The processor and storage medium can also exist in a base station or a terminal as discrete components.

[0322] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are performed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, or other programmable device. The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video disk; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or nonvolatile storage medium, or may include both volatile and nonvolatile types of storage media.

[0323] In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0324] It should be understood that the various numbers used in the various embodiments of this application are merely for ease of description and are not intended to limit the scope of the embodiments of this application. The order of the sequence numbers of the above-mentioned processes does not necessarily imply a specific order of execution; the order of execution of the processes should be determined by their functions and inherent logic.

Claims

1. A communication method, characterized in that: include: Acquiring first scene information, where the first scene information is used to determine a scene in which the first communication device is located and / or a scene in which the second communication device is located; The first scenario information is input into a first model to obtain a first path parameter set, where the first path parameter set is used to determine at least one path for signal transmission between the first communication device and the second communication device.

2. The method according to claim 1, characterized in that The first model includes at least one first neural network, a second neural network, and a third neural network, and the first scene information includes at least one type of information; Inputting the first scenario information into a first model to obtain a first path parameter set includes: Inputting each type of information in the at least one type of information into one of the at least one first neural network to obtain a type of feature, thereby obtaining at least one type of feature in total; Inputting the at least one type of feature into the second neural network to obtain a fusion feature; The fused features are input into a third neural network to obtain the first path parameter set.

3. The method according to claim 2, characterized in that The first scene information includes at least one type of information, wherein the at least one type of information includes: Environmental information, the environmental information indicating an environment in which the first communication device and / or the second communication device is located; and / or, Communication information, where the communication information indicates communication-related parameters of the first communication device and / or the second communication device.

4. The method according to claim 3, characterized in that The environmental information indicates at least one of the following: the layout of objects in the environment; The materials of objects in the environment; The location of objects in the environment; the layout of objects in the environment; The size of objects in the environment; The velocity of objects in the environment.

5. The method according to claim 3 or 4, characterized in that The communication information indicates at least one of the following: the location of the first communication device; configuration information of the first communication device; the location of the second communication device; Configuration information of the second communication device.

6. The method according to any one of claims 2 to 5, characterized in that: The second neural network includes a deformation network based on an attention mechanism, wherein the attention mechanism is used to perform weighted fusion on the at least one type of features to obtain the fused features.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: Obtaining a first loss value based on the first path parameter set and first measurement information, wherein the first measurement information includes information about the second path parameter set and / or a measured channel, and the first measurement information is obtained by measuring the channel by the first communication device or the second communication device, where the channel is a channel between the first communication device and the second communication device; Determine at least one of the second scene information, the third path parameter set and the second model, wherein the second scene information is obtained by updating the first scene information based on the first loss value, the third path parameter set is obtained by updating the first path parameter set based on the first loss value, and the second model is obtained by updating the first model based on the first loss value.

8. The method according to claim 7, characterized in that The second model is obtained by updating the first model based on the first loss value, including: The second model is obtained by updating at least one model parameter of the first neural network, the model parameter of the second neural network, or the model parameter of the third neural network in the first model based on the first loss value.

9. The method according to claim 7 or 8, characterized in that Before obtaining the first loss value based on the first path parameter set and the first measurement information, the method further includes: performing screening processing on the second path parameter set; The filtered second path parameter set is normalized to obtain a preprocessed second path parameter set.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: Acquire a first image, where the first image represents a scene where the first communication device is located and / or a scene where the second communication device is located; Preprocessing the first image to obtain a second image; calibrating a position of the first communication device and / or a position of the second communication device based on the second image; The second image is updated based on the calibrated position of the first communication device and / or the calibrated position of the second communication device, wherein the first scene information is carried on the updated second image.

11. The method according to any one of claims 1 to 10, characterized in that The method further comprises: Performing screening and / or normalization on the first path parameter set to obtain a fourth path parameter set; Based on the fourth path parameter set, channel information is predicted and / or the fourth path parameter set is sent to the second communication device, where the channel is the channel between the first communication device and the second communication device, and the fourth path parameter set is used by the second communication device to determine at least one path for signal transmission between the second communication device and the first communication device.

12. The method according to any one of claims 1 to 11, characterized in that The first path parameter set information includes a path parameter of each path in the at least one path.

13. The method according to claim 12, characterized in that Path parameters include at least one of the following: Probability of existence of a path; The departure angle information of the path; Angle of arrival information of the path; Pitch angle information of the path; Azimuth information of the path; Path loss information; Path delay information; The channel impulse response of the path; or, Phase information of the path.

14. A communication device, characterized in that: include: A module for executing the method according to any one of claims 1 to 13.

15. A communication device, characterized in that: The method comprises a processor and an interface circuit, wherein the interface circuit is used to receive signals from other communication devices outside the communication device and transmit them to the processor or send signals from the processor to other communication devices outside the communication device, and the processor is used to implement the method according to any one of claims 1 to 13 through a logic circuit or executing code instructions.

16. A communication device, characterized in that: include: A processor is coupled to a memory, wherein the memory is used to store instructions, and when the instructions are executed by the processor, the communication device performs the method according to any one of claims 1 to 13.

17. A computer program product comprising instructions, characterized in that When the instruction is executed by a communication device, the communication device is caused to perform the method according to any one of claims 1 to 13.

18. A computer-readable storage medium, characterized in that The storage medium stores a computer program or instruction. When the computer program or instruction is executed by the communication device, the method according to any one of claims 1 to 13 is implemented.

Images