Interference index prediction method and device, and electronic device

Abstract

The application discloses a kind of interference index prediction method, device and electronic equipment. Among them, the method comprises: obtaining the first parameter of satellite communication system, and obtaining the second parameter of ground international mobile communication system;First parameter and second parameter are mapped to preset parameter distribution range respectively, obtain first target parameter and second target parameter, wherein, preset parameter distribution range is the parameter distribution range corresponding to the training data used for training probability model, and probability model is the model used to determine the interference condition of ground international mobile communication system in the radiation direction of ground base station antenna;First target parameter and second target parameter are analyzed using probability model, and the target interference index predicted is obtained. The present application solves the technical problems that related technologies mainly rely on simulation means based on physical model to analyze the frequency compatibility of satellite communication system and ground international mobile communication system, which has extremely high computational complexity, long cycle and poor real-time performance.

Description

Technical Field

[0001] This application relates to the field of wireless networks, and more specifically, to a method, apparatus, and electronic device for predicting interference indicators. Background Technology

[0002] With the rapid development of satellite communication technology and terrestrial IMT (International Mobile Telecommunications) systems, the overlap and intersection of their frequency resource usage are increasing. Frequency compatibility between satellite systems and terrestrial IMT systems has become a key challenge affecting communication quality and system coexistence. Traditional methods for analyzing frequency compatibility mainly rely on physical model-based simulations, which require performing complex physical modeling processes such as micro-macro spatiotemporal transformations, antenna pattern calculations, and path loss iterations step by step. This not only results in extremely high computational complexity but also long simulation cycles and poor real-time performance, making it difficult to meet the interference analysis needs in dynamic scenarios.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method, apparatus, and electronic device for predicting interference indicators, which at least solves the technical problems of related technologies that mainly rely on simulation methods based on physical models to analyze the frequency compatibility of satellite communication systems and terrestrial international mobile communication systems, resulting in extremely high computational complexity, long cycles, and poor real-time performance.

[0005] According to one aspect of the embodiments of this application, a method for predicting interference indicators is provided, comprising: acquiring first parameters of a satellite communication system and acquiring second parameters of a terrestrial mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; mapping the first parameters and the second parameters to preset parameter distribution ranges respectively to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train a probability model, and the probability model is a model used to determine the interference situation of the terrestrial mobile communication system in the radiation direction of the ground base station antenna; and analyzing the first target parameters and the second target parameters using the probability model to obtain the predicted target interference indicators.

[0006] Optionally, the probabilistic model is trained as follows: First historical parameters of the satellite communication system and second historical parameters of the terrestrial international mobile communication system are obtained, wherein the first historical parameters include historical satellite orbit and historical space link parameters, and the second historical parameters include historical terrestrial wireless access network and historical terminal characteristic parameters; frequency sharing between the satellite communication system and the terrestrial international mobile communication system in different scenarios is simulated based on the first and second historical parameters to obtain historical interference indices; the correspondence between the first and second historical parameters and historical interference indices is determined, and labeled data is determined based on the correspondence; the original probabilistic model is trained based on the labeled data to obtain the trained probabilistic model.

[0007] Optionally, based on the first historical parameters and the second historical parameters, the frequency sharing of the satellite communication system and the terrestrial international mobile communication system under different scenarios is simulated to obtain historical interference indicators. This includes: using a preset simulation method to simulate scenarios for the first historical parameters and the second historical parameters, wherein the scenario simulation includes simulating the frequency sharing of the satellite communication system and the terrestrial international mobile communication system under different orbital positions, different geographical regions and different ground scenarios; and determining the historical interference indicators based on the results of the scenario simulation.

[0008] Optionally, before training the original probability model based on the labeled data, the method further includes: determining the amount of labeled data; if the amount of data is less than a preset amount of data, determining the original probability model as a first probability model; if the amount of data is greater than or equal to the preset amount of data, determining the original probability model as a second probability model.

[0009] Optionally, the original probabilistic model is trained based on the labeled data to obtain the trained probabilistic model, including: using the original probabilistic model to predict indicators from the labeled data to obtain historical prediction results; determining the mean square error between the historical prediction results and historical interference indicators in the labeled data; determining the total loss function of the original probabilistic model based on the mean square error and KL divergence; determining the gradient of the total loss function with respect to all trainable parameters in the original probabilistic model, and iteratively optimizing the parameters of the original probabilistic model based on the gradient until the iteration stops when a preset condition is met, thus obtaining the trained probabilistic model.

[0010] Optionally, a probability model is used to analyze the first target parameter and the second target parameter to obtain the predicted target interference index, including: determining the probability distribution of the interference index output by the probability model based on the first target parameter and the second target parameter, wherein the probability distribution includes the mean and standard deviation of the interference index; and determining the mean of the interference index as the target interference index.

[0011] Optionally, the method further includes: determining the predicted distribution of the interference power spectral density generated by the terrestrial international mobile communication system in the radiation direction of the ground base station antenna as output by the probability model; and determining the target interference index based on the predicted distribution and the physical parameters of the satellite communication system.

[0012] According to another aspect of the embodiments of this application, an interference index prediction device is also provided, comprising: an acquisition module, configured to acquire first parameters of a satellite communication system and second parameters of a terrestrial mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; a mapping module, configured to map the first parameters and the second parameters to preset parameter distribution ranges respectively to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model, and the probability model is a model used to determine the interference situation of the terrestrial mobile communication system in the radiation direction of the ground base station antenna; and a prediction module, configured to analyze the first target parameters and the second target parameters using the probability model to obtain a predicted target interference index.

[0013] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory for storing program instructions; and a processor connected to the memory for executing program instructions to perform the following functions: acquiring first parameters of a satellite communication system and acquiring second parameters of a terrestrial international mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; mapping the first parameters and the second parameters to preset parameter distribution ranges respectively to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model, and the probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna; and analyzing the first target parameters and the second target parameters using the probability model to obtain a predicted target interference index.

[0014] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-mentioned method for predicting interference indicators by running the computer program.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the above-mentioned method for predicting interference indicators.

[0016] In this embodiment, by acquiring first parameters of the satellite communication system and second parameters of the terrestrial mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; the first and second parameters are mapped to preset parameter distribution ranges respectively to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model, and the probability model is a model used to determine the interference situation of the terrestrial mobile communication system in the radiation direction of the ground base station antenna; the first and second target parameters are analyzed using the probability model to obtain the predicted target interference index, thereby achieving the purpose of establishing a direct mapping relationship between system parameters and interference index using the probability model, thus realizing the technical effect of directly and quickly outputting the interference index through model inference, and solving the technical problems of related technologies mainly relying on simulation methods based on physical models to analyze the frequency compatibility of satellite communication systems and terrestrial mobile communication systems, which have extremely high computational complexity, long cycle and poor real-time performance. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a hardware structure block diagram of a computer terminal for implementing a method for predicting interference indicators according to an embodiment of this application.

[0019] Figure 2 This is a flowchart of a method for predicting interference indicators according to an embodiment of this application;

[0020] Figure 3 This is a flowchart illustrating the training process of a probability model according to an embodiment of this application;

[0021] Figure 4 This is a flowchart illustrating how to train an original probability model based on labeled data, according to an embodiment of this application.

[0022] Figure 5 This is a flowchart of a method for predicting interference indicators according to an embodiment of this application;

[0023] Figure 6 This is a structural diagram of a device for predicting interference indicators according to an embodiment of this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] First, some nouns or terms that appear in the explanation of the embodiments of this application shall be interpreted as follows:

[0027] Satellite communication system: refers to a system that uses satellites as relay stations to achieve communication between different locations such as the ground, air, and sea. It includes satellite orbit parameters, antenna parameters, frequency parameters, etc., and can provide wide-area coverage communication services. It is suitable for communication in remote areas, aviation and maritime communication and other scenarios.

[0028] Terrestrial IMT (International Mobile Telecommunications) system: This is a mobile communication system based on terrestrial base stations. It covers parameters such as operating frequency, bandwidth, transmission power, antenna parameters, and number of users, and includes technologies such as 4G, 5G, and future 6G. It mainly provides mobile communication services in terrestrial areas.

[0029] Frequency compatibility refers to the characteristic that when satellite communication systems and terrestrial IMT systems share or operate on adjacent frequencies, mutual interference remains within an acceptable range and does not affect their respective normal operation. Interference indicators are analyzed to determine whether the two systems are suitable for frequency sharing.

[0030] Probabilistic model: A model that makes predictions by learning the probabilistic relationships between variables in data. It can directly and quickly infer the interference results by training and learning the mapping relationship between satellite and ground IMT system parameters and interference indicators, without the need for complex physical modeling.

[0031] I / N: The ratio of interference power to noise power, reflecting the intensity of interference relative to noise.

[0032] C / I: The ratio of useful signal carrier power to interference power, reflecting the degree of interference to the useful signal.

[0033] Path loss model: A model used to calculate the power attenuation of a signal during propagation, such as the free space propagation model and the ITU-R P.452 model.

[0034] Monte Carlo simulation: a method for simulating the behavior of complex systems through random sampling.

[0035] In terms of application scenarios of related technologies, both traditional offline research and emerging real-time scenarios such as digital twins place higher demands on the efficiency and real-time performance of interference analysis. Traditional methods, due to their reliance on cumbersome calculations of intermediate physical quantities, cannot quickly respond to dynamic changes in system parameters, making it difficult to support real-time decision-making. This severely restricts the flexibility of collaborative work between satellite and ground IMT systems and the efficient utilization of spectrum resources.

[0036] Furthermore, existing technologies primarily focus on interference analysis, antenna design optimization, or anti-interference capability assessment for satellite communication systems themselves, lacking efficient analytical methods for frequency compatibility between satellite systems and terrestrial IMT service systems. In particular, there is a technological gap in combining probabilistic models to achieve rapid interference prediction, failing to meet the demands for rapid output and real-time inference of interference indicators in complex scenarios. Therefore, there is an urgent need for an interference simulation and prediction method that can reduce computational complexity, improve real-time performance, and simultaneously meet the needs of both traditional and emerging scenarios to overcome the aforementioned technical bottlenecks.

[0037] To address the problems existing in related technologies, embodiments of this application provide a method for predicting interference indicators, which can be run on... Figure 1 The computer terminal shown is explained below.

[0038] The method for predicting interference indicators provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a method for predicting interference indicators is shown. Figure 1As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0039] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0040] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the interference index prediction method in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned interference index prediction method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0041] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0042] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0043] It should be noted here that, in some optional embodiments, the above... Figure 1 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0044] In the above operating environment, this application provides an embodiment of a method for predicting interference indicators. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0045] Figure 2 This is a flowchart of a method for predicting interference indicators according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0046] Step S202: Obtain the first parameters of the satellite communication system and the second parameters of the terrestrial international mobile communication system. The first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters.

[0047] In step S202 above, the satellite communication system can be either the jammed party or the receiver, and the terrestrial international mobile communication system (i.e., the terrestrial IMT system) can be either the jammed party or the transmitter. The first parameter includes satellite orbit and space link parameters. Specifically, the first parameter may include orbit type (e.g., NGSO, which determines the basic characteristics of satellite operation), number of orbital planes, number of satellites per orbital plane, altitude, inclination angle, beam center frequency, maximum satellite antenna gain, receiver system noise temperature, and other link parameters. The second parameter includes terrestrial wireless access network and terminal characteristic parameters. Specifically, the second parameter may include operating center frequency, bandwidth, receiver noise temperature, transmit power (base station and terminal), antenna parameters (gain, radiation pattern, downtilt angle, number and spacing of elements), station spacing, number of access users, etc.

[0048] Step S204: Map the first parameter and the second parameter to a preset parameter distribution range to obtain the first target parameter and the second target parameter. The preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model. The probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna.

[0049] In step S204 above, after obtaining the first and second parameters, since these two types of parameters may change due to system upgrades or environmental changes, they need to be standardized and mapped to the parameter distribution range of the training data. That is, they need to be mapped to the aforementioned preset parameter distribution range so that the numerical ranges of the first and second parameters are consistent with the numerical ranges during probabilistic model training. The first parameter, after mapping, becomes the first target parameter, and the second parameter, after mapping, becomes the second target parameter. The probabilistic model is used to learn the complex nonlinear mapping between the first and second parameters and interference conditions (e.g., interference power spectral density (PSD), interference power, or interference indices).

[0050] Step S206: Analyze the first target parameter and the second target parameter using a probability model to obtain the predicted target interference index.

[0051] In step S206 above, the standardized satellite parameters (first target parameters) and ground IMT parameters (second target parameters) are used as inputs and fed into a pre-trained probability model (such as a Bayesian neural network or a Gaussian process regression model). Through mathematical operations within the model, the interference result is derived. This interference result can be the probability distribution of the interference index. The probability distribution can include the mean and standard deviation, etc. Then, the final target interference index is determined based on the probability distribution. The target interference index can be I / N and / or C / I, etc.

[0052] Through steps S202 to S206 above, the goal of establishing a direct mapping relationship between system parameters and interference indicators using a probabilistic model is achieved. This enables the technical effect of directly and quickly outputting interference indicators through model inference, thereby solving the technical problem that related technologies mainly rely on simulation methods based on physical models to analyze the frequency compatibility of satellite communication systems and terrestrial international mobile communication systems, which suffers from extremely high computational complexity, long cycles, and poor real-time performance. The following is a further explanation.

[0053] In the above-mentioned methods for predicting interference indicators, the probabilistic model is trained in the following way: Figure 3 This is a flowchart illustrating the training process of a probability model according to an embodiment of this application, as follows: Figure 3 As shown, it includes the following steps:

[0054] Step S302: Obtain the first historical parameters of the satellite communication system and the second historical parameters of the terrestrial international mobile communication system, wherein the first historical parameters include historical satellite orbit and historical space link parameters, and the second historical parameters include historical terrestrial wireless access network and historical terminal characteristic parameters;

[0055] Step S304: Simulate the frequency sharing of satellite communication system and terrestrial international mobile communication system in different scenarios based on the first historical parameters and the second historical parameters to obtain historical interference indicators;

[0056] Step S306: Determine the correspondence between the first historical parameter, the second historical parameter and the historical interference index, and determine the labeled data based on the correspondence;

[0057] Step S308: Train the original probability model based on the labeled data to obtain the trained probability model.

[0058] In some embodiments of this application, when training the probabilistic model, it is necessary to obtain first historical parameters and second historical parameters. The first historical parameters include historical satellite orbits and historical space link parameters. Specifically, the first historical parameters may include orbit type (e.g., determined to be NGSO, which determines the basic characteristics of satellite operation), number of orbital planes, number of satellites per orbital plane, altitude, tilt angle, beam center frequency, maximum satellite antenna gain, receiver system noise temperature, and other link parameters. The second historical parameters include historical terrestrial wireless access network and historical terminal characteristic parameters. Specifically, the second historical parameters may include operating center frequency, bandwidth, receiver noise temperature, transmit power (base station and terminal), antenna parameters (gain, radiation pattern, downtilt angle, number and spacing of elements), inter-station spacing, number of access users, etc.

[0059] Using traditional physical simulation methods (such as Monte Carlo simulation), the collected first and second historical parameters are used as inputs to simulate frequency sharing between satellite communication systems and terrestrial IMT systems under different scenarios. This allows for the acquisition of historical interference indices for various scenarios. For example, it can simulate the communication conditions of satellites at different orbital positions and terrestrial IMT systems in different geographical areas, covering various orbital phases of satellite operation and diverse ground communication environments. It considers different ground scenarios, such as densely populated urban areas, urban areas, and suburbs, where the network topology and frequency configuration of ground systems differ significantly. The simulation generates a large amount of labeled data containing system parameters and corresponding interference indices (such as I / N and C / I). Specifically, the labeled data includes the first historical parameter, the second historical parameter, historical interference indices, and the correspondence between the two types of historical parameters and the historical interference indices. The original probabilistic model is then trained based on the labeled data to obtain the trained probabilistic model.

[0060] In the above steps, the frequency sharing of the satellite communication system and the terrestrial international mobile communication system under different scenarios is simulated based on the first historical parameters and the second historical parameters to obtain historical interference indicators. This includes: using a preset simulation method to simulate scenarios based on the first historical parameters and the second historical parameters, wherein the scenario simulation includes simulating the frequency sharing of the satellite communication system and the terrestrial international mobile communication system under different orbital positions, different geographical regions and different ground scenarios; and determining the historical interference indicators based on the results of the scenario simulation.

[0061] In some embodiments of this application, the preset simulation method may be, for example, the Monte Carlo simulation method. The Monte Carlo simulation method performs complex physical processes, including geometric calculations (calculating the instantaneous distance, elevation angle, and azimuth angle between the satellite and each ground base station / terminal), path loss calculations (calculating free space loss), antenna pattern calculations (calculating the gain of the satellite antenna and the ground base station antenna at a specific angle), and interference integration (e.g., summing up the contributions of all ground interference sources to the satellite receiver).

[0062] To ensure the generalization ability of the probabilistic model, the training data must cover the widest possible spatial and environmental dimensions. Monte Carlo simulations are used to simulate different scenarios, including different orbital positions (i.e., satellites passing over ground base stations at different phases, altitudes, and inclinations), different geographical regions (i.e., satellites passing over oceans, deserts, city centers, etc.), and different ground scenarios (i.e., densely populated urban areas, suburbs, and city centers). Based on these simulation scenarios, the frequency sharing between the satellite communication system and the terrestrial international mobile communication system is determined, thereby identifying historical interference indicators, such as I / N and C / I ratios.

[0063] In the above steps, before training the original probability model based on the labeled data, the method further includes: determining the amount of labeled data; if the amount of data is less than a preset amount of data, determining the original probability model as the first probability model; if the amount of data is greater than or equal to the preset amount of data, determining the original probability model as the second probability model.

[0064] In some embodiments of this application, the architecture or hyperparameters of the probabilistic model can be dynamically selected according to the data scale. For example, a lightweight / small-capacity probabilistic model can be used when the data volume is small, and a heavyweight / large-capacity probabilistic model can be used when the data volume is large, in order to balance computing resources, training time and prediction accuracy.

[0065] Specifically, the amount of labeled data used to train the original probabilistic model is first determined. The original probabilistic model to be trained is then determined based on the relationship between the amount of labeled data and the preset data amount. For example, if the amount of data is less than the preset data amount (e.g., 10,000), a first probabilistic model is selected as the original probabilistic model. This first probabilistic model could be, for example, a Gaussian process regression model, a shallow Bayesian neural network model, or a linear regression model. If the amount of data is greater than or equal to the preset data amount, a second probabilistic model is selected as the original probabilistic model to capture complex nonlinear relationships. This second probabilistic model could be, for example, a deep Bayesian neural network model. Bayesian neural networks can handle high-dimensional inputs and quantify prediction uncertainty through weighted probability distributions. Gaussian process regression can effectively capture the nonlinear relationship between input and output and is suitable for small to medium sample scenarios (e.g., scenarios where the amount of labeled data is less than the preset data amount).

[0066] Furthermore, since the first probability model and the second probability model are suitable for different amounts of data, in terms of model structure, the number of hidden layer nodes in the first probability model is less than that in the second probability model, or the training regularization strength of the first probability model is greater than that of the second probability model, etc.

[0067] In the above steps, the original probability model is trained based on the labeled data to obtain the trained probability model, such as... Figure 4 As shown, the specific steps include the following:

[0068] Step S402: Use the original probability model to predict the indicators of the labeled data and obtain historical prediction results;

[0069] Step S404: Determine the mean square error between the historical prediction results and the historical interference indicators in the labeled data;

[0070] Step S406: Determine the total loss function of the original probability model based on the mean square error and KL divergence;

[0071] Step S408: Determine the gradient of the total loss function with respect to all trainable parameters in the original probability model, and iteratively optimize the parameters of the original probability model based on the gradient until the preset conditions are met, and stop the iteration to obtain the trained probability model.

[0072] In some embodiments of this application, during the forward propagation process of training the original probabilistic model, labeled data is input into the original probabilistic model, and historical prediction results are output. These historical prediction results refer to the distribution parameters, including the mean and standard deviation, currently predicted by the original probabilistic model based on random initial weights. The mean squared error (MSE) is calculated based on the historical prediction results and historical interference indicators in the labeled data. In the original probabilistic model, such as a Bayesian neural network model, the weights are not fixed values ​​but follow a certain probability distribution (e.g., a normal distribution), which serves as the prior distribution. During training, the original probabilistic model learns the posterior distribution, and the KL divergence is used to measure the difference between the posterior distribution and the initial prior distribution. The KL divergence, as a regularization term, is used to penalize weight distributions that deviate excessively from the prior assumptions. The total loss function of the original probabilistic model can be determined based on the MSE and KL divergence, and then gradient descent is performed based on the total loss function.

[0073] In the original probabilistic model, such as a Bayesian neural network model, the trainable parameters include the mean and standard deviation (or variance) of the distribution. By calculating the gradient of the total loss function with respect to the mean and standard deviation (or variance) of the trainable parameters, an optimizer (such as Adam) updates the parameters of the original probabilistic model based on the gradient. This process is executed iteratively until preset conditions are met (e.g., reaching the maximum number of iterations, the change in the total loss function being less than a threshold, or the validation set error no longer decreasing), at which point the iteration stops, resulting in the trained probabilistic model.

[0074] Taking the original probability model as a Bayesian neural network model as an example, the model architecture can include an input layer, a hidden layer, and an output layer. The dimension of the input layer matches the number of input parameters. The hidden layer has 3-5 fully connected layers to perform deep feature extraction and nonlinear transformation on the input data. The output layer outputs the mean and standard deviation (or variance) of I / N and C / I, providing the probability distribution of interference indicators.

[0075] In step S206 above, a probability model is used to analyze the first target parameter and the second target parameter to obtain the predicted target interference index, including: determining the probability distribution of the interference index output by the probability model based on the first target parameter and the second target parameter, wherein the probability distribution includes the mean and standard deviation of the interference index; and determining the mean of the interference index as the target interference index.

[0076] In some embodiments of this application, after inputting standardized satellite parameters (first target parameters) and ground IMT parameters (second target parameters), the probabilistic model (such as a Bayesian neural network model) outputs the statistical distribution characteristics of the interference index (i.e., the probability distribution of the aforementioned interference index), including the mean and standard deviation of the interference index. The mean represents the most likely value of the interference index (i.e., the expected value) given the input parameters, reflecting the deterministic mapping relationship between system parameters and interference. The standard deviation represents the degree of uncertainty or variance in the prediction, reflecting the ambiguity of the prediction caused by data noise, model approximation errors, or the input parameters being at the edge of the training distribution. After obtaining the probability distribution of the interference index, both the interference estimate (reflected by the mean, i.e., the aforementioned target interference index) and reliability information (reflected by the standard deviation) can be provided simultaneously, meeting the decision-making needs of multiple scenarios.

[0077] In the inference phase of interference indicators, in practical applications, the parameters of the new satellite communication system and the ground IMT system are input, namely the first target parameter and the second target parameter. The probabilistic model directly infers interference indicators such as I / N and C / I between the two systems based on the mapping relationship obtained through training. The parameters inside the probabilistic model and the rules learned through training transform the input parameters into predicted values ​​of interference indicators, skipping the complex steps of antenna pattern calculation, antenna pointing calculation, and spatiotemporal conversion between micro and macro systems in traditional methods. The probability distribution of interference indicators (such as mean and standard deviation) is output, and the most likely interference indicator (i.e., the target interference indicator) and confidence interval (such as the I / N range at 90% confidence) are output based on the probability distribution.

[0078] The method for predicting the aforementioned interference indicators further includes: determining the predicted distribution of the interference power spectral density generated by the terrestrial international mobile communication system in the radiation direction of the ground base station antenna, as output by the probability model; and determining the target interference indicators based on the predicted distribution and the physical parameters of the satellite communication system.

[0079] In some embodiments of this application, the probabilistic model can output a predicted distribution of the power spectral density (PSD) of interference generated by a terrestrial IMT system in the radiation direction of a terrestrial base station antenna, or a predicted distribution of the total interference power. Here, PSD represents the interference power intensity of the terrestrial IMT system within a specific direction and frequency unit bandwidth, and the predicted distribution includes the mean and standard deviation (or variance). This indicates that the probabilistic model has learned how to handle complex geometric summations, antenna pattern integration, and path loss calculations. Combining the predicted distribution of PSD or total interference power with the physical parameters of the satellite communication system (which may include satellite antenna gain, path loss, receiver noise temperature, channel bandwidth, etc.), the predicted distribution of PSD or total interference power is transformed into a probability distribution of an interference index (such as I / N) using physical formulas, thereby determining the target interference index (e.g., using the mean of the probability distribution of the interference index as the target interference index).

[0080] Specifically, if the probabilistic model outputs a predicted distribution of the PSD (Power Segment Distribution), the total interference power can be determined based on the PSD and channel bandwidth. The total noise power can be determined based on the known Boltzmann constant, noise temperature, and channel bandwidth. The I / N (Target Interference Index) is then determined based on the ratio of the total interference power to the total noise power. This calculation method improves the versatility of the probabilistic model. If the probabilistic model outputs a predicted distribution of the total interference power, then after calculating the total noise power, the I / N (Target Interference Index) is determined based on the ratio of the total interference power to the total noise power.

[0081] In some other embodiments of this application, the interference power spectral density generated by the terrestrial IMT system in various directions can also be calculated based on a probabilistic model. The interference power distribution in different directions can be directly derived from the relationship between system parameters and interference learned by the probabilistic model. Combined with path loss models (such as the free-space propagation model, ITU-R P.452 model, etc.), the interference value to a satellite or satellite terminal at a specified location can be directly obtained based on factors such as the interference power spectral density and signal propagation distance. The free-space propagation model calculates the loss based on the signal frequency and propagation distance, and combines this with the output of the probabilistic model to calculate the interference impact. The entire calculation process utilizes a probabilistic model to skip complex intermediate physical quantity calculations, avoiding tedious intermediate steps, reducing computational complexity, and improving computational efficiency.

[0082] The aforementioned method for predicting interference indicators also includes a step of periodically correcting the probabilistic model through feedback. By comparing the inference results with traditional simulation results, the model is periodically corrected to ensure its accuracy and reliability. This allows for the rapid and accurate determination of interference conditions based on simplified simulation results and model parameter inference when inputting new satellite communication system parameters or new terrestrial IMT system parameters.

[0083] The interference index prediction method provided in this application is a probability model-based simulation and rapid prediction method for interference between satellite communication systems and terrestrial IMT systems. It can solve the problems of high computational complexity and poor real-time performance of traditional simulation methods when analyzing the frequency compatibility of satellite communication systems and terrestrial IMT systems.

[0084] Specifically, by training a probabilistic model to learn the mapping relationship between system parameters and interference indicators (or other interference parameters), the interference power spectral density of the terrestrial IMT system in each direction can be directly calculated based on the probabilistic model. Combined with path loss models or the physical parameters of the satellite communication system, the interference to the satellite or satellite terminal can be derived. This skips the complex intermediate steps of antenna pattern calculation and micro-macro spatiotemporal conversion in traditional methods, thereby reducing computational complexity and improving computational efficiency. In other embodiments, a probabilistic model based on the interference situation of the terrestrial IMT system in each direction is established to achieve real-time inference of interference to the satellite system. When new system parameters are input, interference indicators can be quickly output, meeting the need for immediate feedback on interference change trends in real-time scenarios.

[0085] This method can also optimize the frequency compatibility analysis process between satellite communication systems and terrestrial IMT systems, providing an efficient tool for frequency sharing planning. By quickly outputting interference indicators, it helps determine whether the two systems can coexist, supporting traditional offline research and dynamic spectrum resource management (such as adjusting frequency allocation strategies and beam avoidance). It simplifies the compatibility analysis process under multiple frequency bands and scenarios, breaks through the limitations of traditional offline simulation, extends interference analysis capabilities to real-time scenarios, provides quantitative basis for network operation and maintenance and dynamic spectrum optimization, and improves the reliability and resource utilization of satellite and terrestrial communication systems working together.

[0086] The interference index prediction method provided in this application has the following effects: 1. Based on the mapping relationship between system parameters and interference indexes learned by probabilistic model learning, the interference index can be directly output without complex physical modeling, reducing computational complexity; 2. By adopting probabilistic models such as Bayesian neural networks or Gaussian process regression, the uncertainty of prediction can be quantified and confidence intervals can be provided, improving decision reliability; 3. By combining the path loss model to calculate the interference value (or interference index), the interference power spectral density can be quickly converted into the actual interference value (or interference index), simplifying the calculation process; 4. Multi-scenario training data is generated through Monte Carlo simulation, covering scenarios such as dense urban areas and suburbs, ensuring the model's generalization ability; 5. The input parameters are standardized to adapt to different parameter distribution ranges, improving the model's inference ability for new parameters; 6. The probability distribution (mean, standard deviation) of the interference index is output, and the interference estimate (target interference index) and reliability information are provided to meet the decision-making needs of multiple scenarios; 7. Periodic feedback and correction of the probabilistic model (compared with traditional simulation results) can dynamically optimize the model accuracy and ensure long-term prediction accuracy; 8. In the structural design of the prediction model, 3-5 fully connected layers can be set in the hidden layer to extract features in depth and improve the prediction accuracy of interference indicators.

[0087] The following combination Figure 5 The method flow for predicting interference indicators in embodiments of this application will be briefly described. Figure 5 This is a flowchart of a method for predicting interference indicators according to an embodiment of this application, such as... Figure 5 As shown, during the training of the probabilistic model, satellite system data and ground IMT system data (the first and second historical parameters mentioned above, respectively) are collected. Simulations based on the traditional Monte Carlo method are performed using the collected system data to obtain interference-related index results (i.e., the historical interference indicators mentioned above). The data is then labeled based on these results to obtain labeled data (containing the correspondence between system data or parameter data and historical interference indicators). The probabilistic model is trained using this labeled data, and model inference and parameter tuning are performed during training. For example, the parameters of the probabilistic model can be adjusted by comparing the inference results with traditional simulation results. After the probabilistic model training is complete, new satellite system parameters and new ground IMT system parameters (the first and second parameters mentioned above, respectively) are obtained. The model parameters from the trained probabilistic model are used to infer the new satellite system parameters and new ground IMT system parameters to obtain the target interference indicators. Furthermore, the probabilistic model can be periodically fed back and corrected based on the simplified simulation results to improve model accuracy.

[0088] Figure 6 This is a structural diagram of a device for predicting interference indicators according to an embodiment of this application, as shown below. Figure 6 As shown, the device includes:

[0089] The acquisition module 60 is used to acquire first parameters of the satellite communication system and second parameters of the terrestrial international mobile communication system. The first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters.

[0090] The mapping module 62 is used to map the first parameter and the second parameter to a preset parameter distribution range respectively to obtain the first target parameter and the second target parameter. The preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model. The probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna.

[0091] The prediction module 64 is used to analyze the first target parameter and the second target parameter using a probability model to obtain the predicted target interference index.

[0092] Through the acquisition module 60, mapping module 62 and prediction module 64 in the above-mentioned interference index prediction device, the purpose of establishing a direct mapping relationship between system parameters and interference indexes using a probability model is achieved. This realizes the technical effect of directly and quickly outputting interference indexes through model inference, thereby solving the technical problem that related technologies mainly rely on simulation methods based on physical models to analyze the frequency compatibility of satellite communication systems and terrestrial international mobile communication systems, which has extremely high computational complexity, long cycle and poor real-time performance.

[0093] The aforementioned interference index prediction device also includes a training module 66, which is used to train a probability model. Specifically, the probability model is trained in the following way: acquiring first historical parameters of the satellite communication system and second historical parameters of the terrestrial international mobile communication system, wherein the first historical parameters include historical satellite orbit and historical space link parameters, and the second historical parameters include historical terrestrial wireless access network and historical terminal characteristic parameters; simulating frequency sharing between the satellite communication system and the terrestrial international mobile communication system in different scenarios based on the first and second historical parameters to obtain historical interference indices; determining the correspondence between the first historical parameters, the second historical parameters, and the historical interference indices, and determining the labeled data based on the correspondence; and training the original probability model based on the labeled data to obtain the trained probability model.

[0094] In the aforementioned training module, the training module is also used to perform scenario simulations on the first historical parameters and the second historical parameters using a preset simulation method. The scenario simulation includes simulating the frequency sharing between the satellite communication system and the terrestrial international mobile communication system under different orbital positions, different geographical regions, and different ground scenarios. Based on the results of the scenario simulation, historical interference indicators are determined.

[0095] In the training module described above, the module is also used to determine the amount of labeled data; when the amount of data is less than the preset amount of data, the original probability model is determined to be the first probability model; when the amount of data is greater than or equal to the preset amount of data, the original probability model is determined to be the second probability model.

[0096] In the aforementioned training module, the module is also used to predict indicators from labeled data using the original probabilistic model to obtain historical prediction results; determine the mean square error between the historical prediction results and historical interference indicators in the labeled data; determine the total loss function of the original probabilistic model based on the mean square error and KL divergence; determine the gradient of the total loss function with respect to all trainable parameters in the original probabilistic model, and iteratively optimize the parameters of the original probabilistic model based on the gradient until the iteration stops when the preset conditions are met, thus obtaining the trained probabilistic model.

[0097] In the aforementioned prediction module, the prediction module is also used to determine the probability distribution of the interference index output by the probability model based on the first target parameter and the second target parameter, wherein the probability distribution includes the mean and standard deviation of the interference index; the mean of the interference index is determined as the target interference index.

[0098] In the aforementioned prediction module, the prediction module is also used to determine the predicted distribution of the interference power spectral density generated by the terrestrial international mobile communication system in the radiation direction of the ground base station antenna, as output by the probability model; and to determine the target interference index based on the predicted distribution and the physical parameters of the satellite communication system.

[0099] It should be noted that, Figure 6 The interference index prediction device shown is used to perform... Figure 2 The method for predicting interference indicators shown above is also applicable to the device for predicting interference indicators, and will not be repeated here.

[0100] This application also provides an electronic device, which includes a memory and a processor. The memory stores program instructions, and the processor is connected to the memory to execute program instructions that perform the following functions: acquiring first parameters of a satellite communication system and acquiring second parameters of a terrestrial international mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; mapping the first and second parameters to preset parameter distribution ranges to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model, and the probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna; and analyzing the first and second target parameters using the probability model to obtain predicted target interference indicators.

[0101] It should be noted that the aforementioned electronic equipment is used to perform Figure 2 The method for predicting interference indicators shown above is also applicable to this electronic device, and will not be repeated here.

[0102] This application also provides a non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the above-mentioned method for predicting interference indicators by running the computer program.

[0103] This application also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the interference index prediction method in various embodiments of this application.

[0104] This application also provides a computer program that, when executed by a processor, implements the steps of the interference index prediction method in various embodiments of this application.

[0105] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0106] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0107] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0108] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0110] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0111] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for predicting interference indicators, characterized in that, include: The system acquires first parameters of a satellite communication system and second parameters of a terrestrial international mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters. The first parameter and the second parameter are respectively mapped to a preset parameter distribution range to obtain a first target parameter and a second target parameter. The preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model. The probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna. The probability model is used to analyze the first target parameter and the second target parameter to obtain the predicted target interference index.

2. The method according to claim 1, characterized in that, The probability model is trained in the following way: The system acquires first historical parameters of the satellite communication system and second historical parameters of the terrestrial international mobile communication system, wherein the first historical parameters include historical satellite orbit and historical space link parameters, and the second historical parameters include historical terrestrial wireless access network and historical terminal characteristic parameters. Based on the first historical parameters and the second historical parameters, the frequency sharing between the satellite communication system and the terrestrial international mobile communication system under different scenarios is simulated to obtain historical interference indicators; Determine the correspondence between the first historical parameter, the second historical parameter and the historical interference index, and determine the labeled data based on the correspondence; The original probability model is trained based on the labeled data to obtain the trained probability model.

3. The method according to claim 2, characterized in that, Based on the first historical parameters and the second historical parameters, the frequency sharing between the satellite communication system and the terrestrial international mobile communication system under different scenarios is simulated to obtain historical interference indicators, including: A preset simulation method is used to simulate scenarios for the first historical parameters and the second historical parameters. The scenario simulation includes simulating the frequency sharing between the satellite communication system and the terrestrial international mobile communication system under different orbital positions, different geographical regions and different ground scenarios. Based on the results of the scenario simulation, the historical interference indicators are determined.

4. The method according to claim 2, characterized in that, Before training the original probability model based on the labeled data, the method further includes: Determine the amount of labeled data; If the amount of data is less than the preset amount of data, the original probability model is determined to be the first probability model; If the amount of data is greater than or equal to the preset amount of data, the original probability model is determined to be the second probability model.

5. The method according to claim 2, characterized in that, The original probability model is trained based on the labeled data to obtain the trained probability model, including: The original probability model is used to predict indicators from the labeled data to obtain historical prediction results. Determine the mean square error between the historical prediction results and the historical interference indicators in the labeled data; The total loss function of the original probability model is determined based on the mean square error and KL divergence. The gradient of the total loss function with respect to all trainable parameters in the original probability model is determined, and the parameters of the original probability model are iteratively optimized based on the gradient until a preset condition is met, at which point the iteration stops, and the trained probability model is obtained.

6. The method according to claim 1, characterized in that, The probability model is used to analyze the first target parameter and the second target parameter to obtain the predicted target interference index, including: Determine the probability distribution of the interference index output by the probability model based on the first target parameter and the second target parameter, wherein the probability distribution includes the mean and standard deviation of the interference index; The mean value of the interference index is determined as the target interference index.

7. The method according to claim 1, characterized in that, The method further includes: Determine the predicted distribution of the interference power spectral density generated by the terrestrial international mobile communication system in the radiation direction of the ground base station antenna, as output by the probability model; The target interference index is determined based on the predicted distribution and the physical parameters of the satellite communication system.

8. A device for predicting interference indicators, characterized in that, include: The acquisition module is used to acquire first parameters of the satellite communication system and second parameters of the terrestrial international mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters. A mapping module is used to map the first parameter and the second parameter to a preset parameter distribution range respectively to obtain a first target parameter and a second target parameter. The preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model. The probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna. The prediction module is used to analyze the first target parameter and the second target parameter using the probability model to obtain the predicted target interference index.

9. An electronic device, characterized in that, include: Memory, used to store program instructions; A processor, connected to the memory, is configured to execute program instructions to perform the following functions: acquiring first parameters of a satellite communication system and second parameters of a terrestrial international mobile communication system, wherein the first parameters include satellite orbit and space link parameters, and the second parameters include terrestrial wireless access network and terminal characteristic parameters; mapping the first parameters and the second parameters to preset parameter distribution ranges to obtain first target parameters and second target parameters, wherein the preset parameter distribution range is the parameter distribution range corresponding to the training data used to train the probability model, and the probability model is a model used to determine the interference situation of the terrestrial international mobile communication system in the radiation direction of the ground base station antenna; and analyzing the first target parameters and the second target parameters using the probability model to obtain predicted target interference indicators.

10. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the method for predicting interference indicators according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the method for predicting interference indicators as described in any one of claims 1 to 7.

Images