Network twinborn modeling method, channel information acquisition method and computer program product

Abstract

The invention discloses a network twinning modeling method, a channel information acquisition method and a computer program product, and the method comprises the steps: determining first channel state information according to a calculation graph of a network twinning environment and the position information of a first node of the network twinning environment; training or updating a first parameter and / or the computational graph according to the first channel state information and the second channel state information until a set convergence condition is reached; the computational graph is used for predicting channel state information according to the position information of the nodes, the computational graph is generated according to a signal propagation path of the network twin environment and a first parameter, and the first parameter is determined according to one or more of material attribute information of the network twin environment, a transmission model of a second node and an antenna mode of the second node; the second node is a node included in a signal propagation path of the network twin environment; the second channel state information represents channel state information actually measured at the first node.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a network twin modeling method, a channel information acquisition method, and a computer program product. Background Technology

[0002] Wireless signals interact with objects in the environment during propagation, causing changes in propagation direction and signal power. To ensure the feasibility of various applications, the predicted signal characteristics of the wireless channel must match the actual measured signal characteristics at various locations. Although site-specific simulation predictions can supplement actual measurement data to twinnize the wireless channel, the signal characteristics of the twinned wireless channel cannot be obtained in related technologies, and channel measurement in the real world is costly and difficult. Summary of the Invention

[0003] To address the related technical issues, embodiments of this application provide a network twin modeling method, a channel information acquisition method, and a computer program product.

[0004] The technical solution of this application embodiment is implemented as follows:

[0005] This application provides a method for network twin modeling, including:

[0006] Based on the computation graph of the network twin environment and the location information of the first node in the network twin environment, first channel state information is determined; wherein, the computation graph is used to predict the channel state information based on the location information of the nodes, and the computation graph is generated based on the signal propagation path of the network twin environment and a first parameter, the first parameter being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node;

[0007] The first parameter and / or the computation graph are trained or updated based on the first channel state information and the second channel state information until the set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

[0008] In the above scheme, training or updating the first parameter and / or the computation graph based on the first channel state information and the second channel state information includes:

[0009] The loss value is determined based on the first channel state information and the second channel state information;

[0010] The first parameter and / or the computation graph are trained or updated based on the loss value.

[0011] The method in the above scheme further includes:

[0012] The first parameter related to the second node is determined based on the location information of the second node and the first model. The first model is used to predict the first parameter based on the location information of the second node.

[0013] In the above scheme, the antenna pattern of the second node is constructed based on several of the following parameters:

[0014] The azimuth angle of the antenna;

[0015] Antenna elevation angle;

[0016] Antenna radiation efficiency;

[0017] Concentration parameters;

[0018] Weighting factors;

[0019] Amplitude factor of antenna element;

[0020] The angle of incidence of the antenna;

[0021] The antenna's emission angle.

[0022] In the above scheme, the expression for the antenna mode of the second node is:

[0023] G(θ i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented by φ, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) represents the direction θ i and φ i The unit norm in the direction of the mean, a i The amplitude factor characterizing the antenna element.

[0024] The method in the above scheme further includes:

[0025] The amplitude factor of the antenna element is determined based on the signal transmission frequency and wavelength, as well as the transmission model and antenna mode of the second node.

[0026] In the above scheme, the transmission model of the second node represents the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

[0027] In the above scheme, the first parameter includes one or more of the following information about the network twin environment:

[0028] The electrical conductivity of the material;

[0029] The relative permittivity of the material;

[0030] Scattering coefficient;

[0031] Cross-polarization coefficient;

[0032] The parameters or weights of the transmission model;

[0033] The angle of incidence at the second node;

[0034] The emission angle of the second node;

[0035] The wavelength of the signal;

[0036] The transmission frequency of the signal;

[0037] The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0038] In the above scheme, after training or updating the first parameter and / or the computation graph based on the first channel state information and the second channel state information until the set convergence condition is reached, the method further includes:

[0039] Based on the computation graph and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node, wherein the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0040] This application also provides a method for obtaining channel information, including:

[0041] Based on the computation graph of the network twin environment and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node. The computation graph is obtained based on any of the above methods, and the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0042] This application also provides a network twin modeling apparatus, including:

[0043] The first determining unit is configured to determine first channel state information based on a computational graph of a network twin environment and the location information of a first node in the network twin environment; wherein, the computational graph is used to predict the channel state information based on the location information of the nodes, and the computational graph is generated based on the signal propagation path of the network twin environment and a first parameter, the first parameter being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node;

[0044] The training unit is used to train or update the first parameter and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

[0045] This application embodiment also provides a channel information acquisition device, including:

[0046] The second determining unit is used to predict or output the third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment. The computation graph is obtained based on any of the above methods, and the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0047] This application also provides a first communication device, including: a first processor and a first communication interface; wherein,

[0048] The first processor is configured to determine first channel state information based on the computation graph of the network twin environment and the location information of the first node in the network twin environment; and to train or update the first parameters and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is reached; wherein,

[0049] The computation graph is used to predict channel state information based on the location information of nodes. The computation graph is generated based on the signal propagation path of the network twin environment and a first parameter. The first parameter is determined based on one or more of the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node. The first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment. The first channel state information represents the channel state information predicted based on the location information of the first node, and the second channel state information represents the channel state information actually measured at the first node.

[0050] This application also provides a second communication device, including: a second processor and a second communication interface; wherein,

[0051] The second processor is configured to predict or output third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment. The computation graph is obtained based on any of the above methods, and the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0052] This application also provides a communication device, including a processor and a memory for storing computer programs that can run on the processor.

[0053] When the processor runs the computer program, it executes the steps of any of the above methods.

[0054] This application also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of any of the above methods.

[0055] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above methods.

[0056] In the network twin modeling method, channel information acquisition method, apparatus, communication device, storage medium, and computer program product provided in the embodiments of this application, first channel state information is determined based on the computation graph of the network twin environment and the location information of the first node in the network twin environment; the first parameters and / or the computation graph are trained or updated based on the first channel state information and the second channel state information until a set convergence condition is reached; and third channel state information related to the third node is predicted or output based on the trained or updated computation graph and the location information of the third node in the network twin environment; wherein, the computation graph is used to predict the channel state based on the location information of the node. The computational graph is generated based on the signal propagation path of the network twin environment and a first parameter. The first parameter is determined based on one or more of the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node. The first node represents any node in the network twin environment, the second node is a node included in the signal propagation path of the network twin environment, and the third node represents any node in the network twin environment whose channel state needs to be predicted. The first channel state information represents the channel state information predicted based on the location information of the first node, and the second channel state information represents the channel state information actually measured at the first node. It can be seen that in this embodiment, the channel state information can be predicted through the computational graph of the network twin environment to obtain the signal characteristics of the wireless channel of the digital twin network, reducing the cost and difficulty of channel measurement. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of a related technology signal transmission scheme;

[0058] Figure 2 This is a schematic diagram of a network twin modeling method according to an embodiment of this application;

[0059] Figure 3 This is a schematic diagram of the computational environment of the network twin environment in an embodiment of this application;

[0060] Figure 4 This is a schematic flowchart of a channel information acquisition method according to an embodiment of this application;

[0061] Figure 5 This is a schematic diagram of the interaction process of a network twin modeling method according to an embodiment of this application;

[0062] Figure 6 This is a schematic diagram of the network twin modeling device according to an embodiment of this application;

[0063] Figure 7 This is a schematic diagram of the channel information acquisition device according to an embodiment of this application;

[0064] Figure 8This is a schematic diagram of the structure of the first communication device according to an embodiment of this application;

[0065] Figure 9 This is a schematic diagram of the structure of the second communication device according to an embodiment of this application. Detailed Implementation

[0066] At millimeter-wave and Asia-Pacific Hertz (THz) frequencies, reflection and transmission are the main propagation mechanisms observed in site surveys, while diffraction is negligible at certain high frequencies. In areas where specular reflection is obstructed in wireless network transmission, diffuse reflection and scattering are also important propagation mechanisms. Figure 1 As shown, the wireless signal emitted by the terminal is transmitted to the base station after being reflected by an object. Figure 1 In this context, line of sight (LOS) refers to the straight-line distance of a signal from the sender to the receiver.

[0067] In situations where there is no detailed understanding of the complex environment in which measurements are taken, such as insufficient knowledge of the fine details of the material being tested (e.g., the fine structure of the material), no obvious dependence between the incident angle of the wireless signal and the reflected power, and the highly nonlinear nature of currently used methods, it is impossible to accurately model and measure the reflection loss. Therefore, when the geometry of the network scene is relatively simple, precise spectral twin measurements are usually performed using channels to determine the material properties of the network scene.

[0068] The accuracy of twins of network environments depends on two key factors: the geometry of the network scene and the material properties involved, such as the conductivity σ and relative permittivity ε of the environment.

[0069] In related technologies, the geometry of a scene is usually obtained through high-precision manual measurement. However, this method is difficult to handle complex scenes and cannot automatically obtain material properties. The method of measuring the material properties of individual objects that make up objects in a network scene is not applicable to large and complex scenes, is very inefficient, and cannot adjust or compensate for the mismatch in geometric or twin environment calculations.

[0070] Related technologies also provide simplified models based on angle-independent reflection and penetration losses to measure material properties, such as measuring material properties through a reflection loss model. This method assumes that the reflection loss of a wireless signal colliding with an obstacle is independent of the incident angle and that the reflection loss is constant. It solves for a value that minimizes the difference between the actual measured path gain and the predicted path gain, thereby assessing the electrical properties of materials in a twin network scenario. Closed-loop calibration is achieved by attempting to adjust material parameters to match statistical data (path loss or delay spread) with the measurement results. However, these assumptions are not applicable in practice, and their feasibility is generally limited. In addition, the following problems exist: errors in geometric or twin environment calculations cannot be resolved; there are differences between the simplified model or reflection loss model and the actual network scenario; manual adjustment of material parameters is required, which demands significant expertise and experience, and it is difficult to guarantee parameter accuracy; and calibration typically requires a large amount of data, which may require expensive measurement equipment, significant time and human resources.

[0071] In summary, the related technologies cannot obtain the signal characteristics of a twin wireless channel, and channel measurement in the real world is costly and difficult.

[0072] Based on this, in various embodiments of this application, a first channel state information is determined according to the computation graph of the network twin environment and the location information of the first node in the network twin environment; the first parameter and / or the computation graph are trained or updated according to the first channel state information and the second channel state information until a set convergence condition is reached; the third channel state information related to the third node is predicted or output according to the trained or updated computation graph and the location information of the third node in the network twin environment; wherein, the computation graph is used to predict the channel state information based on the location information of the node, the computation graph is generated according to the signal propagation path of the network twin environment and the first parameter, the first parameter is determined according to one or more of the material property information of the network twin environment, the transmission model of the second node and the antenna mode of the second node; the first node represents any node in the network twin environment, the second node is a node included in the signal propagation path of the network twin environment, and the third node represents any node in the network twin environment whose channel state needs to be predicted; the first channel state information represents the channel state information predicted according to the location information of the first node, and the second channel state information represents the channel state information actually measured at the first node. As can be seen, in this embodiment of the application, channel state information can be predicted through the computational graph of the network twin environment, thereby obtaining the signal characteristics of the wireless channel of the digital twin network and reducing the cost and difficulty of channel measurement.

[0073] The present application will now be described in further detail with reference to the accompanying drawings and embodiments.

[0074] This application provides a network twin modeling method applied to a first communication device. The first communication device includes network devices and / or terminals and / or network functions of the core network. The network devices include base stations. Figure 2 As shown, the method includes:

[0075] Step 101: Determine the first channel state information based on the computation graph of the network twin environment and the location information of the first node of the network twin environment.

[0076] The computation graph is used to predict channel state information based on the location information of nodes. The computation graph is generated based on the signal propagation path of the network twin environment and a first parameter. The first parameter is determined based on one or more of the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node. The first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment. The first channel state information represents the channel state information predicted based on the location information of the first node.

[0077] Here, a first parameter is determined based on one or more of the following: the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node. A computational graph of the network twin environment is generated based on the signal propagation path and the first parameter. First channel state information is determined based on the computational graph and the location information of the first node in the network twin environment. The node's location information serves as the input to the computational graph of the network twin environment, and the channel state information serves as the output. Determining the first channel state information can be described as predicting the first channel state information, which can be understood as the channel state information related to the first node.

[0078] It should be noted that a network twin environment can be described as a digital twin network, the environment of a digital twin network, the environment of a twin network, a twin network environment, or the network environment of a digital twin network. The computational graph of a network twin environment can be understood as a computational graph constructed for a digital twin network or network twin environment. That is, the parameter calculation of the network twin environment is regarded as a large computational graph, and the signal propagation path of the network twin environment is abstracted as similar to a neural network. The first parameter is similar to the weights or trainable parameters of a neural network, the node location information serves as the input of the neural network, and the channel state information serves as the output of the neural network, such as... Figure 3As shown, a neural network can include input cells, hidden cells, and output cells. A first node can be understood as any node or any second node in a network twin environment; the number of first nodes can be one or more, and each first node corresponds one-to-one with the first channel state information. First and second nodes can be understood as intersections, scattering points, and diffraction points in a network twin environment; an intersection can be understood as the intersection between different signal propagation paths. In a network twin environment, any signal propagation path between the sender and receiver can include one or more intersections and / or scattering points and / or diffraction points. Any signal propagation path can consist of multiple transmission events, which can include scattering events and / or reflection events and / or diffraction events. Reflection events include specular reflection events and / or diffuse reflection events. For example, a network propagation path consists of a series of Q scattering events (such as specular reflection, diffuse reflection, and diffraction). Scattering events can be described as scattering type, scattering process, or scattering mode. For each scattering event, the current scattering point S needs to be calculated. j The incident field at point S and the next scattering point S j+1 The relationship between the far field and the S j Let S be a real number R, j is greater than or equal to 1 and less than or equal to Q, where Q is a positive integer or a positive integer greater than zero, and S is a real number. Q and S Q+1 Let p and q represent the transmitting and receiving ends of the antenna, respectively. A network twin environment under specific local conditions can be represented as a lateral environment, and therefore can be represented by two orthogonal components p and q. For each scattering event in any network propagation path within the network twin environment, the electric field can be represented as a complex vector containing p and q components. Let Eout represent the complex vector containing p and q of the outgoing field at the (j+1)th scattering point of the j-th scattering event. In the (j+1)-th scattering event, Eout can also be represented as Ein of the incident field at the (j+1)-th scattering point. The p component represents the component of the electric field in a reference direction, which is usually orthogonal to the propagation direction, such as the horizontal direction. The q component represents the component of the electric field in a direction orthogonal to both the p component and the propagation direction, such as the vertical direction. A reflection may be followed by a diffuse reflection. An effective scattering point Eout = Fj(Ein) can be found by first determining a set of candidate reflection paths or reflection points and then using the shortest distance principle, where Fj represents the j-th scattering event. Therefore, the sequence of Q scattering events used to form the network propagation path can be applied to the computation graph of the neural network's forward feedback, and the results of each layer in the neural network can be computed and stored sequentially from the input layer to the output layer.

[0079] First-level channel state information (CLS) can be described as channel estimation data or a twin of the channel state environment. CLS characterizes information related to the signal characteristics of the uplink and / or downlink channels, including the uplink channel frequency response (CFR) and / or channel impulse response (CIR), and / or the downlink channel CFR and / or CIR. CFR refers to the signal response across different frequency ranges, generally including both amplitude / frequency and phase / frequency responses. CIR is the signal energy value reaching the receiver after different times (different propagation paths result in different propagation times). CIR and CFR satisfy the Fourier transform of each other. CFR can be obtained by measuring the received channel power through channel information at the receiver, or it can be acquired at the receiver from specific reflected signals injected at the transmitter.

[0080] It's important to note that in cyber twin modeling, it's crucial to determine how to set parameters for each material in the cyber twin environment to accurately simulate the interaction between light and materials. One simple approach is to define all objects as being made of the same material, or to categorize objects in the cyber twin environment into different categories, defining each category using different materials, such as the floor, walls, and ceiling in the cyber twin environment. However, in some extreme cases, the material of each mesh in the cyber twin environment might be different. If each mesh requires a separate material, the number of materials needed could be enormous, resulting in a very high computational cost.

[0081] The material properties of a network twin environment include, but are not limited to, one or more of the following: conductivity σ, relative permittivity ε, scattering coefficient S, and cross-polarization coefficient. Specifically, conductivity is greater than or equal to zero, relative permittivity is greater than or equal to 1, and both the scattering coefficient and cross-polarization coefficient are greater than or equal to zero and less than or equal to 1. These material properties determine how much energy is transferred to the orthogonal polarization direction after diffuse reflection. Parameterizing one or more of the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node can reduce the computational complexity of subsequent steps.

[0082] In one embodiment, the first parameter includes one or more of the following information about the network twin environment:

[0083] The electrical conductivity of the material;

[0084] The relative permittivity of the material;

[0085] Scattering coefficient;

[0086] Cross-polarization coefficient;

[0087] The parameters or weights of the transmission model;

[0088] The angle of incidence at the second node;

[0089] The emission angle of the second node;

[0090] The wavelength of the signal;

[0091] The transmission frequency of the signal;

[0092] The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0093] It should be noted that the first value can also be understood as the total number of signal propagation paths between the starting point and the ending point of the signal propagation path in the network twin environment.

[0094] In one embodiment, the antenna pattern of the second node is constructed based on several of the following parameters:

[0095] The azimuth angle of the antenna;

[0096] Antenna elevation angle;

[0097] Antenna radiation efficiency;

[0098] Concentration parameters;

[0099] Weighting factors;

[0100] Amplitude factor of antenna element;

[0101] The angle of incidence of the antenna;

[0102] The antenna's emission angle.

[0103] Here, to reduce the computational complexity of the various parameters in the computation graph of the network twin environment, an antenna pattern of the second node that satisfies normalization constraints can be introduced. The weighting factor includes mixed weights, and the amplitude factor of the antenna element includes the amplitude weights of the antenna array elements. The parameters used to construct the antenna pattern of the second node can serve as the input layer of the forward feedback of the computation graph of the network twin environment. A scattering pattern z = W(i)x can also be added to the input layer of the forward feedback of the computation graph of the network twin environment to construct the hidden layer or concealed layer of the forward feedback of the computation graph of the network twin environment. W(i) represents the mixed weighting factor of the network twin environment and can include multiple parameters, such as two or more parameters from the antenna incident angle, antenna exit angle, direction of incident light, direction of exit light, and surface normal. x represents the i-th scattering pattern and can be represented as Fi(Ein). Since different features or parameters related to the computation graph in a network twin environment can have different dimensions and units, this can affect the data analysis results and gradient convergence efficiency. Therefore, to improve the accuracy of the data analysis results and gradient convergence efficiency, W(i) can be normalized. The incident angle can be understood as the angle of arrival.

[0104] The antenna mode of the second node is an integrable function on a sphere. In one embodiment, the expression for the antenna mode of the second node is: Wherein G(θ) i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented by φ, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) represents the direction θ i and φ i The unit norm in the direction of the mean, a i The amplitude factor characterizing the antenna element.

[0105] It should be noted that the azimuth angle of an antenna can be understood as the direction angle of the antenna, where η is greater than or equal to zero and less than or equal to 1; w i This can be called a mixed weighting; In this context, e represents the natural constant, and λ i Greater than zero; a i The characterization can also be referred to as the amplitude weighting of the antenna element.

[0106] In one embodiment, the method further includes:

[0107] Based on the signal transmission frequency and wavelength, and based on the transmission model and antenna mode of the second node, the amplitude factor of the antenna element is determined. The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0108] Here, the antenna modes of the second node include transmit antenna mode and / or receive antenna mode, and the antenna mode can be described as an antenna model; the amplitude factor of the antenna element can be determined based on the wavelength of the signal, the transmission model of the second node, the transmit antenna mode of the second node, the receive antenna mode of the second node, and the wavelength of the signal.

[0109] In order to accurately calculate the amplitude factor of the antenna element and thus obtain an accurate first parameter, thereby improving the accuracy of the trained computation graph, in one embodiment, determining the amplitude factor of the antenna element based on the signal transmission frequency and wavelength, and based on the transmission model and antenna mode of the second node, includes:

[0110] The amplitude factor of the antenna element is determined using the following formula:

[0111]

[0112] Among them, a i λ represents the amplitude factor of the antenna element in the i-th signal propagation path; λ represents the wavelength of the signal. Characterization The conjugate transpose of; The receiving antenna mode of the second node representing the i-th signal propagation path; The incident angle of the antenna at the second node in the i-th signal propagation path; The emission angle of the antenna at the second node representing the i-th signal propagation path; T i The transmission model representing the second node of the i-th signal propagation path; The transmitting antenna mode of the second node representing the i-th signal propagation path. The incident angle of the antenna at the second node of the i-th signal propagation path. f represents the emission angle of the antenna at the second node of the i-th signal propagation path; f represents the transmission frequency of the signal.

[0113] It should be noted that the transmission model for the sum of M signal propagation paths can be expressed as: H(f) represents the transmission model of the sum of M signal propagation paths between the start and end points of a signal propagation path in a network twin environment; M represents the number of signal propagation paths, or the number of second nodes contained between the start (sender) and end (receiver) points of a signal propagation path in a network twin environment. For example, M is the number of second nodes contained in the signal transmission path from the base station to the terminal.

[0114] To facilitate the analysis of the characteristics or properties of materials in a network twin environment, in one embodiment, the transmission model of the second node characterizes the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

[0115] Here, the transmission model of the second node can be represented as Eout_i+1=Fi(Ein); Fi(Ein) is the incident field of the i-th second node, and Eout_i+1 is the outgoing field of the (i+1)-th second node.

[0116] Step 102: Train or update the first parameter and / or the computation graph based on the first channel state information and the second channel state information until the set convergence condition is reached.

[0117] The second channel state information represents the channel state information actually measured at the first node.

[0118] Here, to obtain the second channel information, the first parameters and / or computation graph can be trained or updated multiple times based on the first and second channel state information. After reaching the set convergence condition, the trained computation graph and the first parameters are obtained. The second channel state information can be measured by the terminal and / or network device. For example, the network device listens to the sounding reference signal (SRS) sent by the terminal located at the first node and measures the second channel state information based on the SRS.

[0119] When the first channel state information includes a first CFR and the second channel state information includes a second CFR, the first parameters and / or the computational graph of the network twin environment can be trained or updated based on the first CFR and the second CFR. When the first channel state information includes a first CIR and the second channel state information includes a second CIR, the first parameters and / or the computational graph of the network twin environment can be trained or updated based on the first CIR and the second CIR.

[0120] It should be noted that stochastic gradient descent can be used to calibrate the loss function of the first parameter, and the gradient of the average loss of the batch can be used to update the first parameter until the set convergence condition is met. Specifically, all trainable parameters (including the first parameter) of the computational graph of the network twin environment can be randomly initialized, and stochastic gradient descent (SGD) with a learning rate β can be defined to iteratively calculate the first parameter. The partial derivatives of the loss function with respect to the global optimal solution are calculated step by step, and the gradient of the average loss of the batch is used to update the first parameter until the set convergence condition is met; this allows for accurate digital twin modeling of the network environment.

[0121] To improve the accuracy of the computation graph of the first parameter and / or the network twin environment, in one embodiment, training or updating the first parameter and / or the computation graph based on the first channel state information and the second channel state information includes:

[0122] The loss value is determined based on the first channel state information and the second channel state information;

[0123] The first parameter and / or the computation graph are trained or updated based on the loss value.

[0124] Here, a predefined loss function can be used to determine the loss value based on the first channel state information and the second channel state information, and the first parameters and / or computation graph can be trained or updated based on the loss value. When there are multiple first nodes, a loss value can be determined based on one first channel state information and the corresponding second channel state information; the total loss value is determined based on all loss values, and the first parameters and / or computation graph can be trained or updated based on the total loss value.

[0125] Loss values ​​include, but are not limited to, one or more of the following: differences in relative path power delay distribution, differences in absolute power, power angle delay, root mean square (RMS) delay, and absolute power. In practical applications, the loss function can be set as the root mean square (RMS) delay breadth, with the total channel gain as the objective function. The purpose of the loss function is to quantify the difference between the predicted and actual values, while the objective function refers to the function or metric to be optimized.

[0126] It should be noted that when there are multiple first nodes, the total loss value is for a batch of prediction results (first channel state information) at different locations in the network twin environment. Normalizing each loss value ensures that all loss values ​​are assigned the same weight, thereby reducing the situation where the network twin optimization process is concentrated in areas with strong channel coverage, resulting in insufficient network twin optimization in areas with weak signal coverage. When the loss value includes power parameters (differences in relative path power delay distribution, differences in absolute power, power angular delay, and absolute power), the power parameters in the network twin environment do not have absolute values. A common scaling factor α can be used to scale the loss value; α is a calibration scaling factor, and α is a real number.

[0127] To reduce the difficulty and improve the efficiency of obtaining the first parameter, the first parameter at different locations in the network twin environment can be predicted using an AI model. Based on this, in one embodiment, the method further includes:

[0128] The first parameter related to the second node is determined based on the location information of the second node and the first model. The first model is used to predict the first parameter based on the location information of the second node.

[0129] Here, the position information of the second node can be input into the first model to obtain the first parameters related to the second node output by the first model. There can be one or more second nodes, and the position information of the second nodes can be position coordinates, which can be UV coordinates or coordinates applied to UV space. UV coordinates are two-dimensional coordinates, where U represents the position in the horizontal direction and V represents the position in the vertical direction. The first model can be a neural network model or a model composed of neural networks, including Artificial Neural Networks (ANNS) and / or Neural Networks (NNS).

[0130] The first parameter includes, but is not limited to, one or more of the following: conductivity σ, relative permittivity ε, scattering coefficient S, cross-polarization coefficient, parameters or weights of the transmission model, and weights or parameters of the scattering pattern or scattering mode.

[0131] It should be noted that, in order to make the location of the network twin environment unique, the location information of the second node can be converted into a matrix, normalized and encoded, and then input into the first model.

[0132] Once the computational graph in the network twin environment has been trained, the trained computational graph can be used to predict channel state information at different locations in the network twin environment. Based on this, in one embodiment, after training or updating the first parameters and / or the computational graph according to the first channel state information and the second channel state information until a set convergence condition is met, the method further includes:

[0133] Based on the computation graph and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node, wherein the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0134] Here, when it is necessary to predict the channel state information related to the third node, the third channel state information related to the third node is predicted or output based on the computation graph of the network twin environment, the first parameter, and the location information of the third node in the network twin environment. The location information of the third node serves as the input to the computation graph, and the third channel state information serves as the output of the computation graph. Outputting the third channel state information can be described as displaying the third channel state information, while predicting the third channel state information can be described as determining the third channel state information.

[0135] This application also provides a channel information acquisition method applied to a second communication device. The second communication device includes network equipment and / or terminals and / or network functions of a core network, and the network equipment includes a base station. The second communication device may be the same as or different from the first communication device; the second communication device and the first communication device may be the same device or different devices. Figure 4 As shown, the above includes:

[0136] Step 401: Based on the computation graph of the network twin environment and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node, wherein the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0137] The computation graph is based on Figure 3 The corresponding embodiment was obtained.

[0138] The following section provides a more detailed description of this application with reference to application examples.

[0139] like Figure 5 As shown, the network twin modeling method includes the following steps:

[0140] Step 1: The base station sends a reference signal configuration to the terminal.

[0141] Here, the base station can send reference signal configuration to the terminal via broadcast, unicast, or multicast. Reference signal configuration includes related configurations for the reference signals, which include Channel State Information-Reference Signal (CSI-RS) and / or SRS. Related configurations for the reference signals may include resource configurations and reporting configurations.

[0142] For example, the base station sends CSI-RS to the terminal according to the CSI-RS resource configuration. Based on information such as the complexity and performance of channel estimation, the base station configures the time-frequency resources required for the terminal to send CSI-RS feedback information to the base station. The CSI-RS reporting configuration is then sent to the terminal via the Physical Downlink Control Channel (PDCCH) and / or the Physical Downlink Shared Channel (PDSCH). The base station can select the time-frequency resources for CSI feedback based on performance parameters such as the number of downlink subbands.

[0143] Step 2: The terminal receives the reference signal configuration and performs channel estimation based on the reference signal configuration.

[0144] Here, the terminal configures itself to receive the CSI-RS transmitted by the base station based on the reference signal; it performs channel estimation based on the CSI-RS to obtain the second channel state information, and then sends the second channel state information to the base station. The base station receives the second channel state information. The terminal is located at the position of the first node in the network twin environment, or the terminal can act as the first node in the network twin environment. The first node can be any node in the network twin environment.

[0145] Step 3: The base station determines the first channel state information based on the computation graph of the network twin environment and the location information of the first node in the network twin environment.

[0146] Here, the base station can determine the signal propagation path in the network twin environment and also determine the first parameter. The first reference can be randomly initialized parameters or obtained through model prediction. The base station can determine the signal propagation path and the first parameter separately. The first channel state information represents the channel state information predicted based on the location information of the first node. The specific implementation process for determining the first channel state information is described above and will not be repeated here.

[0147] It should be noted that the base station can listen to the SRS sent by the terminal, perform channel estimation based on the SRS, and obtain the second channel state information.

[0148] In one embodiment, a first parameter related to the second node is determined based on the location information of the second node and a first model, wherein the first model is used to predict the first parameter based on the location information of the second node.

[0149] In one embodiment, the antenna pattern of the second node is constructed based on several of the following parameters:

[0150] The azimuth angle of the antenna;

[0151] Antenna elevation angle;

[0152] Antenna radiation efficiency;

[0153] Concentration parameters;

[0154] Weighting factors;

[0155] Amplitude factor of antenna element;

[0156] The angle of incidence of the antenna;

[0157] The antenna's emission angle.

[0158] The amplitude factor of the antenna element can be determined using the following formula:

[0159]

[0160] The expression for the antenna mode of the second node is:

[0161] G(θ i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented by φ, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) characterizes θ i and φ i The direction of the mean of the unit norm at point a i The amplitude factor characterizing the antenna element.

[0162] The transmission model of the second node represents the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

[0163] The first parameter includes one or more of the following information about the network twin environment:

[0164] The electrical conductivity of the material;

[0165] The relative permittivity of the material;

[0166] Scattering coefficient;

[0167] Cross-polarization coefficient;

[0168] The parameters or weights of the transmission model;

[0169] The angle of incidence at the second node;

[0170] The emission angle of the second node;

[0171] The wavelength of the signal;

[0172] The transmission frequency of the signal;

[0173] The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0174] Step 4: The base station trains or updates the first parameter and / or the computation graph based on the first channel state information and the second channel state information until the set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

[0175] Step 5: Based on the computation graph of the network twin environment and the location information of the third node in the network twin environment, the base station predicts or outputs the third channel state information related to the third node. The third node represents any node in the network twin environment whose channel state needs to be predicted.

[0176] Here, the number of third nodes can be one or more. The base station can construct a digital twin network based on the computational graph of the trained network twin environment and the first parameters.

[0177] It should be noted that the base station can listen to the SRS sent by the terminal, perform channel estimation based on the SRS, and obtain the second channel state information.

[0178] To implement the network twin modeling method of this application embodiment, this application embodiment also provides a network twin modeling apparatus, which is installed on a first communication device, such as... Figure 6 As shown, the device includes:

[0179] The first determining unit 601 is configured to determine first channel state information based on a computational graph of the network twin environment and the location information of a first node in the network twin environment; wherein, the computational graph is used to predict the channel state information based on the location information of the nodes, and the computational graph is generated based on the signal propagation path of the network twin environment and a first parameter, the first parameter being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node;

[0180] The training unit 602 is used to train or update the first parameter and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

[0181] In one embodiment, the training unit 602 is specifically configured to determine a loss value based on first channel state information and second channel state information; and to train or update the first parameter and / or the computation graph based on the loss value.

[0182] In one embodiment, the device further includes:

[0183] The third determining unit is used to determine the first parameters related to the second node based on the location information of the second node and the first model. The first model is used to predict the first parameters based on the location information of the second node.

[0184] In one embodiment, the antenna pattern of the second node is constructed based on several of the following parameters:

[0185] The azimuth angle of the antenna;

[0186] Antenna elevation angle;

[0187] Antenna radiation efficiency;

[0188] Concentration parameters;

[0189] Weighting factors;

[0190] Amplitude factor of antenna element;

[0191] The angle of incidence of the antenna;

[0192] The antenna's emission angle.

[0193] In one embodiment, the expression for the antenna pattern of the second node is:

[0194] G(θ i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented by φ, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) represents the direction θ i and φ i The unit norm in the direction of the mean, a i The amplitude factor characterizing the antenna element.

[0195] In one embodiment, the device further includes:

[0196] The fourth determining unit is used to determine the amplitude factor of the antenna element based on the transmission frequency and wavelength of the signal, as well as the transmission model and antenna mode of the second node.

[0197] In one embodiment, the transmission model of the second node characterizes the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

[0198] In one embodiment, the first parameter includes one or more of the following information about the network twin environment:

[0199] The electrical conductivity of the material;

[0200] The relative permittivity of the material;

[0201] Scattering coefficient;

[0202] Cross-polarization coefficient;

[0203] The parameters or weights of the transmission model;

[0204] The angle of incidence at the second node;

[0205] The emission angle of the second node;

[0206] The wavelength of the signal;

[0207] The transmission frequency of the signal;

[0208] The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0209] In one embodiment, the device further includes:

[0210] The fifth determining unit is used to predict or output the third channel state information related to the third node based on the computation graph and the location information of the third node in the network twin environment. The third node represents any node in the network twin environment whose channel state needs to be predicted.

[0211] In practical applications, the first determining unit 601, the training unit 602, the third determining unit, the fourth determining unit, and the fifth determining unit can be implemented by the processor in the network twin modeling device.

[0212] It should be noted that the network twin modeling device provided in the above embodiments is only illustrated by the division of the above program modules when performing network twin modeling. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the network twin modeling device and the network twin modeling method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0213] To implement the channel information acquisition method of this application embodiment, this application embodiment also provides a channel information acquisition device, which is disposed on a second communication device, such as... Figure 7 As shown, the device includes:

[0214] The second determining unit 701 is configured to predict or output third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment; wherein,

[0215] The computation graph is obtained based on any of the above methods, and the third node represents any node in the network twin environment that needs to predict the channel state.

[0216] In practical applications, the second determining unit 701 can be implemented by the processor in the network twin modeling device.

[0217] It should be noted that the channel information acquisition device provided in the above embodiments is only illustrated by the division of the above-described program modules. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the channel information acquisition device and the channel information acquisition method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0218] Based on the hardware implementation of the above program modules, and in order to implement the network twin modeling method of this application embodiment, this application embodiment also provides a first communication device, such as... Figure 8 As shown, the first communication device 800 includes:

[0219] The first communication interface 801 is capable of exchanging information with other network nodes;

[0220] The first processor 802 is connected to the first communication interface 801 to enable information interaction with other network nodes. When running a computer program, it executes the methods provided by one or more technical solutions on the first communication device side. The computer program is stored in the first memory 803.

[0221] Specifically, the first processor 802 is configured to determine first channel state information based on a computational graph of the network twin environment and the location information of a first node in the network twin environment; and to train or update the first parameters and / or the computational graph based on the first channel state information and the second channel state information until a set convergence condition is reached; wherein, the computational graph is used to predict channel state information based on the location information of the nodes, the computational graph is generated based on the signal propagation path of the network twin environment and the first parameters, the first parameters being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node; and the second channel state information represents the channel state information actually measured at the first node.

[0222] In one embodiment, the first processor 802 is specifically configured to determine a loss value based on first channel state information and second channel state information; and to train or update the first parameter and / or the computation graph based on the loss value.

[0223] In one embodiment, the first processor 802 is further configured to determine a first parameter related to the second node based on the location information of the second node and the first model, wherein the first model is configured to predict the first parameter based on the location information of the second node.

[0224] In one embodiment, the antenna pattern of the second node is constructed based on several of the following parameters:

[0225] The azimuth angle of the antenna;

[0226] Antenna elevation angle;

[0227] Antenna radiation efficiency;

[0228] Concentration parameters;

[0229] Weighting factors;

[0230] Amplitude factor of antenna element;

[0231] The angle of incidence of the antenna;

[0232] The antenna's emission angle.

[0233] In one embodiment, the expression for the antenna pattern of the second node is:

[0234] G(θ i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented by φ, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) represents the direction θ i and φ i The unit norm in the direction of the mean, a i The amplitude factor characterizing the antenna element.

[0235] In one embodiment, the first processor 802 is further configured to determine the amplitude factor of the antenna element based on the transmission frequency and wavelength of the signal, as well as the transmission model and antenna mode of the second node.

[0236] In one embodiment, the transmission model of the second node characterizes the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

[0237] In one embodiment, the first parameter includes one or more of the following information about the network twin environment:

[0238] The electrical conductivity of the material;

[0239] The relative permittivity of the material;

[0240] Scattering coefficient;

[0241] Cross-polarization coefficient;

[0242] The parameters or weights of the transmission model;

[0243] The angle of incidence at the second node;

[0244] The emission angle of the second node;

[0245] The wavelength of the signal;

[0246] The transmission frequency of the signal;

[0247] The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

[0248] In one embodiment, the first processor 802 is further configured to predict or output third channel state information related to the third node based on the computation graph and the location information of the third node in the network twin environment, wherein the third node represents any node in the network twin environment whose channel state needs to be predicted.

[0249] It should be noted that the specific processing procedures of the first processor 802 and the first communication interface 801 can be understood by referring to the above method.

[0250] Of course, in practical applications, the various components in the first communication device 800 are coupled together through the bus system 804. It can be understood that the bus system 804 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 804 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in... Figure 8 The general labeled all buses as Bus System 804.

[0251] The first memory 803 in this embodiment is used to store various types of data to support the operation of the first communication device 800. Examples of such data include any computer program used to operate on the first communication device 800.

[0252] The methods disclosed in the above embodiments of this application can be applied to the first processor 802, or implemented by the first processor 802. The first processor 802 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware or by instructions in the form of software in the first processor 802. The first processor 802 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The first processor 802 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in the first memory 803. The first processor 802 reads the information in the first memory 803 and completes the steps of the aforementioned method in combination with its hardware.

[0253] In an exemplary embodiment, the first communication device 800 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0254] Based on the hardware implementation of the above program modules, and in order to implement the method on the second communication device side of the embodiments of this application, the embodiments of this application also provide a second communication device. For example... Figure 9 As shown, the second communication device 900 includes:

[0255] The second communication interface 901 is capable of exchanging information with other network nodes;

[0256] The second processor 902 is connected to the second communication interface 901 to enable information interaction with other network nodes. When running a computer program, it executes the methods provided by one or more technical solutions on the second communication device side. The computer program is stored in the second memory 903.

[0257] Specifically, the second processor 902 is configured to predict or output third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment; wherein,

[0258] The computation graph is obtained based on any of the above methods, and the third node represents any node in the network twin environment that needs to predict the channel state.

[0259] It should be noted that the specific processing procedures of the second processor 902 and the second communication interface 901 can be understood by referring to the above method.

[0260] Of course, in practical applications, the various components in the second communication device 900 are coupled together through the bus system 904. It can be understood that the bus system 904 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 904 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in... Figure 9 The general labeled all buses as Bus System 904.

[0261] The second memory 903 in this embodiment is used to store various types of data to support the operation of the second communication device 900. Examples of such data include any computer program used to operate on the second communication device 900.

[0262] The methods disclosed in the embodiments of this application can be applied to, or implemented by, the second processor 902. The second processor 902 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by the integrated logic circuitry of the hardware or by instructions in the form of software within the second processor 902. The second processor 902 may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The second processor 902 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, specifically a second memory 903. The second processor 902 reads information from the second memory 903 and, in conjunction with its hardware, completes the steps of the aforementioned method.

[0263] In an exemplary embodiment, the second communication device 900 may be implemented by one or more ASICs, DSPs, PLDs, CPLDs, FPGAs, general-purpose processors, controllers, MCUs, microprocessors, or other electronic components to perform the aforementioned method.

[0264] It is understood that the memories (first memory 803 and second memory 903) in the embodiments of this application can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memories.

[0265] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a first memory 803 storing a computer program, which can be executed by the first processor 802 of the first communication device 800 to complete the steps described in the aforementioned network twin modeling method. Another example is a second memory 903 storing a computer program, which can be executed by the second processor 902 of the second communication device 900 to complete the steps described in the aforementioned channel information acquisition method. The computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM.

[0266] For example, this application also provides a computer program product, including a computer program that can be executed by a first processor 802 of a first communication device 800 to complete the steps of the aforementioned network twin modeling method. The computer program can also be executed by a second processor 902 of a second communication device 900 to complete the steps of the aforementioned channel information acquisition method.

[0267] It should be noted that terms such as "first" and "second" are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. "Multiple" can refer to two or more items, and "multiple" can refer to two or more items. The term "and / or" in this document merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, the term "one or more" in this document refers to any combination of at least two of the multiple elements. For example, including one or more of A, B, and C can represent including any one or at least two or more elements selected from the set consisting of A, B, and C.

[0268] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0269] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application.

Claims

1. A method for modeling network twins, characterized in that, include: Based on the computation graph of the network twin environment and the location information of the first node in the network twin environment, first channel state information is determined; wherein, the computation graph is used to predict the channel state information based on the location information of the nodes, and the computation graph is generated based on the signal propagation path of the network twin environment and a first parameter, the first parameter being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node; The first parameter and / or the computation graph are trained or updated based on the first channel state information and the second channel state information until the set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

2. The method according to claim 1, characterized in that, The step of training or updating the first parameter and / or the computation graph based on the first channel state information and the second channel state information includes: The loss value is determined based on the first channel state information and the second channel state information; The first parameter and / or the computation graph are trained or updated based on the loss value.

3. The method according to claim 1, characterized in that, The method further includes: The first parameter related to the second node is determined based on the location information of the second node and the first model. The first model is used to predict the first parameter based on the location information of the second node.

4. The method according to claim 1, characterized in that, The antenna pattern of the second node is constructed based on several of the following parameters: The azimuth angle of the antenna; Antenna elevation angle; Antenna radiation efficiency; Concentration parameters; Weighting factors; Amplitude factor of antenna element; The angle of incidence of the antenna; The antenna's emission angle.

5. The method according to claim 4, characterized in that, The expression for the antenna mode of the second node is: G(θ i ,φ i θ represents the antenna gain. i The azimuth angle of the antenna of the i-th second node is represented, where i is a positive integer and i+1 is less than or equal to the total number of second nodes. φ i Let η represent the elevation angle of the antenna of the i-th second node, and let w represent the radiation efficiency of the antenna of the i-th second node. i Characteristic weighting factor, λ i Characterizing concentration parameter, μ i (θ i ,φ i ) represents the direction θ i and φ i The unit norm in the direction of the mean, a i The amplitude factor characterizing the antenna element.

6. The method according to claim 4 or 5, characterized in that, The method further includes: The amplitude factor of the antenna element is determined based on the signal transmission frequency and wavelength, as well as the transmission model and antenna mode of the second node.

7. The method according to claim 1, characterized in that, The transmission model of the second node represents the relationship between the incident field of the i-th second node and the outgoing field of the (i+1)-th second node, where i is a positive integer and i+1 is less than or equal to the total number of second nodes.

8. The method according to claim 1, characterized in that, The first parameter includes one or more of the following information about the network twin environment: The electrical conductivity of the material; The relative permittivity of the material; Scattering coefficient; Cross-polarization coefficient; The parameters or weights of the transmission model; The angle of incidence at the second node; The emission angle of the second node; The wavelength of the signal; The transmission frequency of the signal; The first value represents the number of second nodes contained between the start and end points of the signal propagation path in the network twin environment.

9. The method according to any one of claims 1 to 5, 7 to 8, characterized in that, After training or updating the first parameter and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is met, the method further includes: Based on the computation graph and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node, wherein the third node represents any node in the network twin environment whose channel state needs to be predicted.

10. A method for acquiring channel information, characterized in that, include: Based on the computation graph of the network twin environment and the location information of the third node in the network twin environment, predict or output the third channel state information related to the third node. The computation graph is obtained based on the method described in any one of claims 1 to 9. The third node represents any node in the network twin environment whose channel state needs to be predicted.

11. A network twin modeling device, characterized in that, include: The first determining unit is configured to determine first channel state information based on a computational graph of a network twin environment and the location information of a first node in the network twin environment; wherein, the computational graph is used to predict the channel state information based on the location information of the nodes, and the computational graph is generated based on the signal propagation path of the network twin environment and a first parameter, the first parameter being determined based on one or more of the material property information of the network twin environment, the transmission model of the second node, and the antenna mode of the second node; the first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment; the first channel state information represents the channel state information predicted based on the location information of the first node; The training unit is used to train or update the first parameter and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is reached. The second channel state information represents the channel state information actually measured at the first node.

12. A channel information acquisition device, characterized in that, include: The second determining unit is configured to predict or output third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment. The computation graph is obtained based on the method described in any one of claims 1 to 9, and the third node represents any node in the network twin environment whose channel state needs to be predicted.

13. A first communication device, characterized in that, include: A first processor and a first communication interface; wherein... The first processor is configured to determine first channel state information based on the computation graph of the network twin environment and the location information of the first node in the network twin environment; and to train or update the first parameters and / or the computation graph based on the first channel state information and the second channel state information until a set convergence condition is reached; wherein, The computation graph is used to predict channel state information based on the location information of nodes. The computation graph is generated based on the signal propagation path of the network twin environment and a first parameter. The first parameter is determined based on one or more of the material properties of the network twin environment, the transmission model of the second node, and the antenna mode of the second node. The first node represents any node in the network twin environment, and the second node is a node included in the signal propagation path of the network twin environment. The first channel state information represents the channel state information predicted based on the location information of the first node, and the second channel state information represents the channel state information actually measured at the first node.

14. A second communication device, characterized in that, include: A second processor and a second communication interface; wherein... The second processor is configured to predict or output third channel state information related to the third node based on the computation graph of the network twin environment and the location information of the third node in the network twin environment, wherein the computation graph is obtained based on the method described in any one of claims 1 to 9, and the third node represents any node in the network twin environment whose channel state needs to be predicted.

15. A communication device, characterized in that, This includes a processor and memory for storing computer programs that can run on the processor. When the processor is used to run the computer program, it performs the steps of the method according to any one of claims 1 to 9, or the steps of the method according to claim 10.

16. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 9, or the steps of the method according to claim 10.

17. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 9, or the steps of the method according to claim 10.

Images