Spectrum situation map generation method and device, computer equipment and storage medium

Abstract

The invention provides a frequency spectrum situation map generation method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a frequency spectrum situation map according to the attenuation information of a radio signal in a target region and the first receiving signal strength of a plurality of known sampling points in the target region; determining the first received signal strength of at least one corner sampling point in the target area; generating a plurality of different candidate spectrum situation maps according to the first received signal strength of the at least one angular point sampling point, the first received signal strength of the plurality of known sampling points and the random noise; for any candidate frequency spectrum situation map, calculating a signal intensity difference set and a similarity parameter set corresponding to the candidate frequency spectrum situation map; and according to the signal intensity difference set and the similarity parameter set corresponding to each candidate frequency spectrum situation map, screening out a target frequency spectrum situation map from the plurality of candidate frequency spectrum situation maps. The quality of the generated frequency spectrum situation map can be ensured under the condition that the number of sampling points is limited.

Description

Technical Field

[0001] The present application relates to the technical field of signal processing, and in particular, to a method, apparatus, computer device, and storage medium for generating a spectrum situation map. Background Art

[0002] A spectrum situation map is used to describe the communication signal parameters at various positions within the signal coverage range of a signal emission point, such as the received signal strength. The spectrum situation map will change under the influence of factors such as temperature, humidity, obstacle position, and crowd. The requirement for high real-time performance results in a single sensor being unable to obtain multiple sampling values by moving. Moreover, sensors capable of sampling radio information are expensive. Therefore, considering cost-effectiveness, the sampling rate is often limited.

[0003] In the prior art, due to the limited number of sampling points in the process of generating a spectrum situation map, the quality of the generated spectrum situation map is often low. Summary of the Invention

[0004] In view of this, the present application proposes a method, apparatus, computer device, and storage medium for generating a spectrum situation map to solve the problem in the related art that the quality of the generated spectrum situation map is low due to the limited number of sampling points.

[0005] The first aspect embodiment of the present application proposes a method for generating a spectrum situation map, the method including:

[0006] Determine the first received signal strength of at least one corner sampling point in the target area according to the attenuation information of the radio signal in the target area and the first received signal strength of a plurality of known sampling points in the target area; the corner sampling point represents the intersection of two sides in the target area; at least one emission point is included in the target area;

[0007] Generate a plurality of different candidate spectrum situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points, and random noise; the second received signal strength corresponding to the same sampling point in any two candidate spectrum situation maps is different;

[0008] For any one of the candidate spectrum situation maps, calculate the signal strength difference set and the similarity parameter set corresponding to the candidate spectrum situation map; any signal strength difference in the signal strength difference set refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectrum situation map and the corresponding preset received signal strength; any similarity parameter in the similarity parameter set refers to the similarity parameter between the received signal strength distribution map within the preset range of the corresponding emission point in the candidate spectrum situation map and the corresponding preset received signal strength distribution map;

[0009] Select a target spectrum situation map from the multiple candidate spectrum situation maps according to the set of signal strength differences and the set of similarity parameters corresponding to each candidate spectrum situation map.

[0010] In the embodiment of the present application, by determining the first received signal strength of at least one corner sampling point in the target area according to the attenuation information of the radio signal in the target area and the first received signal strength of several known sampling points in the target area, the number of sampling points can be increased in a low sampling rate scenario, thereby improving the quality of the generated spectrum situation map.

[0011] Preferably, the present application generates a plurality of different candidate spectrum situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the several known sampling points, and random noise, which can generate multiple prediction results under limited data input, that is, the above-mentioned multiple different candidate spectrum situation maps, which is beneficial to obtaining a target spectrum situation map that more conforms to the target area.

[0012] Preferably, the present application selects a target spectrum situation map from the multiple candidate spectrum situation maps according to the set of signal strength differences and the set of similarity parameters corresponding to each candidate spectrum situation map, which can screen out the target spectrum situation map that most conforms to the target area from multiple different candidate spectrum situation maps, and thus the quality of the finally generated target spectrum situation map is relatively high.

[0013] In the embodiment of the present application, generating a plurality of different candidate spectrum situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the several known sampling points, and random noise includes:

[0014] Generate a random noise matrix, and the dimension of the random noise matrix is the same as the resolution of the target spectrum situation map;

[0015] Fuse the random noise matrix, the first received signal strength of the corner sampling point, and the first received signal strength of the several known sampling points to obtain initial input data;

[0016] Input the initial input data into the denoising diffusion probability model, so that the denoising diffusion probability model gradually generates candidate spectrum situation maps through the reverse diffusion process; wherein, the reverse diffusion process includes multiple time steps, and the noise is gradually reduced at each time step;

[0017] Repeat the above steps multiple times to generate multiple different candidate spectrum situation maps.

[0018] In the embodiment of the present application, after determining the first received signal strength of at least one corner sampling point in the target area, the method further includes:

[0019] Calculate a correction factor according to the building layout diagram of the target area and the received signal strength distribution diagram within a preset range of any emission point in the candidate spectrum situation diagram;

[0020] Correct the first received signal strength of each corner sampling point in the target area through the correction factor to obtain a preset received signal strength.

[0021] In an embodiment of the present application, calculating a signal strength difference set and a similarity parameter set corresponding to the candidate spectrum situation diagram includes:

[0022] For any one of the multiple corner sampling points of the candidate spectrum situation diagram, calculate the signal strength difference between the second received signal strength of the corner sampling point and the corresponding preset received signal strength;

[0023] Generate a signal strength difference set corresponding to the candidate spectrum situation diagram according to the multiple signal strength differences corresponding one by one to the multiple corner sampling points.

[0024] In an embodiment of the present application, after generating multiple candidate spectrum situation diagrams, the method further includes:

[0025] For any one of the multiple candidate spectrum situation diagrams, determine at least one emission point from the candidate spectrum situation diagram, and identify the received signal strength distribution diagram within the preset range of each emission point;

[0026] Obtain the radio propagation physical model of the target area;

[0027] Based on the radio propagation physical model, identify the preset received signal strength distribution diagram within the preset range of each emission point.

[0028] In an embodiment of the present application, calculating a signal strength difference set and a similarity parameter set corresponding to the candidate spectrum situation diagram includes:

[0029] For any one of at least one emission point of the candidate spectrum situation diagram, calculate the similarity parameter between the received signal strength distribution diagram within the preset range of the emission point and the corresponding preset received signal strength distribution diagram;

[0030] Generate a similarity parameter set corresponding to the candidate spectrum situation diagram according to at least one similarity parameter corresponding one by one to at least one emission point of the candidate spectrum situation diagram.

[0031] In an embodiment of the present application, according to the signal strength difference set and the similarity parameter set corresponding to each candidate spectrum situation diagram, screening out a target spectrum situation diagram from the multiple candidate spectrum situation diagrams includes:

[0032] For any one of the multiple candidate spectrum situation maps, calculate the absolute values of the multiple signal strength differences corresponding to the candidate spectrum situation map, and sum up the absolute values of the multiple signal strength differences to obtain a first comparison parameter;

[0033] Sum up at least one similarity parameter corresponding to the candidate spectrum situation map to obtain a second comparison parameter;

[0034] According to the first comparison parameter of the candidate spectrum situation map, the second comparison parameter, the first weight corresponding to the first comparison parameter, and the second weight corresponding to the second comparison parameter, calculate to obtain a third comparison parameter of the candidate spectrum situation map;

[0035] Use the candidate spectrum situation map with the largest third comparison parameter among the multiple candidate spectrum situation maps as the target spectrum situation map.

[0036] An embodiment of the second aspect of the present application provides a spectrum situation map generation device, including:

[0037] A received signal strength determination module, configured to determine the first received signal strength of at least one corner sampling point in the target area according to the attenuation information of the radio signal in the target area and the first received signal strength of several known sampling points in the target area; the corner sampling point represents the intersection of two sides in the target area; there is at least one emission point in the target area;

[0038] A candidate spectrum situation map generation module, configured to generate a plurality of different candidate spectrum situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the several known sampling points, and random noise; the second received signal strength corresponding to the same sampling point in any two candidate spectrum situation maps is different;

[0039] A set calculation module, configured to calculate a signal strength difference set and a similarity parameter set corresponding to any candidate spectrum situation map; any signal strength difference in the signal strength difference set refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectrum situation map and the corresponding preset received signal strength; any similarity parameter in the similarity parameter set refers to the similarity parameter between the received signal strength distribution map within a preset range of the corresponding emission point in the candidate spectrum situation map and the corresponding preset received signal strength distribution map;

[0040] A target spectrum situation map screening module, configured to screen out the target spectrum situation map from the multiple candidate spectrum situation maps according to the signal strength difference set and the similarity parameter set corresponding to each candidate spectrum situation map.

[0041] An embodiment of the third aspect of the present application provides an electronic device, which includes a memory and a processor. The memory and the processor are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the spectrum situation map generation method described in the first aspect above.

[0042] An embodiment of the fourth aspect of the present application provides a computer-readable storage medium, on which computer instructions are stored. The computer instructions are used to cause a computer to execute the spectrum situation map generation method described in the first aspect above.

[0043] Additional aspects and advantages of the present application will be given in part in the following description, will become apparent in part from the following description, or will be understood through the practice of the present application. Description of the Drawings

[0044] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present application. Moreover, throughout the drawings, the same reference numerals are used to represent the same components.

[0045] In the drawings:

[0046] Figure 1 A flowchart showing a spectrum situation map generation method provided by an embodiment of the present application is shown;

[0047] Figure 2 A flowchart showing a spectrum situation generation method based on a generative artificial intelligence method based on an enhancement module, a generation module, and a selection module provided by an embodiment of the present application is shown;

[0048] Figure 3 A schematic diagram showing the correction of the signal reception intensity of sampling points by a correction factor provided by an embodiment of the present application is shown;

[0049] Figure 4 A schematic diagram showing the simulation result of physical simulation with the emission point as the center of the ring provided by an embodiment of the present application is shown;

[0050] Figure 5 A schematic diagram showing the structure of a spectrum situation map generation device provided by an embodiment of the present application is shown;

[0051] Figure 6 A schematic diagram showing the structure of an electronic device provided by an embodiment of the present application is shown;

[0052] Figure 7 A schematic diagram showing a storage medium provided by an embodiment of the present application is shown. Detailed implementation manners

[0053] The exemplary implementation manners of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary implementation manners of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the implementation manners set forth herein. On the contrary, these implementation manners are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0054] It should be noted that unless otherwise specified, the technical terms or scientific terms used in the present application should have the ordinary meanings understood by those skilled in the art to which the present application belongs.

[0055] The technical scenarios involved in the embodiments of the present application are described below.

[0056] In the process of generating a spectrum situation map, the existing technologies mainly include: 1) using an interpolation-based method to estimate a radio map using mathematical and statistical methods, and 2) using a deep learning-based method to generate a spectrum situation map using a neural network. However, the interpolation-based method lacks physical priors and the generated results are poor; the existing deep learning-based methods model the task as a regression problem. To achieve better results at an extremely low sampling rate, the mean of all possible situations will be predicted. Such a method can relatively stably obtain a not-bad result, but there is a serious edge blurring situation, which seriously destroys the style information of the radio power during propagation and cannot truly simulate an accurate solution.

[0057] Compared with the existing methods, to solve the problem of extremely low sampling rate, the present application uses a diffusion model to solve the spectrum situation generation task. Compared with the existing methods, a key advantage of the generation model is its "divergent thinking" ability, which can generate a variety of predictions from limited inputs. This creativity makes the generation model very suitable for completing the spectrum situation generation task at an extremely low sampling rate.

[0058] According to an embodiment of the present application, an embodiment of a spectrum situation map generation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0059] Embodiment 1:

[0060] In this embodiment, a spectrum situation map generation method is provided. Figure 1 is a flowchart of the spectrum situation map generation method according to an embodiment of the present application, as Figure 1 shown, and this process includes the following steps:

[0061] Step S101: Determine the first received signal strength of at least one corner sampling point in the target area based on the attenuation information of the radio signal in the target area and the first received signal strengths of several known sampling points in the target area. Herein, the corner sampling point represents the intersection of two sides in the target area; there is at least one emission point in the target area.

[0062] In the embodiments of the present application, due to the limited number of sampling points, the quality of the spectral situation map generated based on the received signal strengths of the existing sampling points is low. Therefore, through step S101, the present application uses the attenuation information of the radio signal in the target area and the first received signal strengths of several known sampling points in the target area to determine the first received signal strength of at least one corner sampling point in the target area, thereby increasing the number of sampling points.

[0063] In some specific embodiments, the attenuation information of the radio signal in the target area is calculated based on the building layout map of the target area and the radio propagation law.

[0064] In some specific embodiments, the attenuation information of the radio signal in the target area and the first received signal strengths of several known sampling points in the target area can be input into the U-Net model with self-attention mechanism AttUnet (which can be understood as the "enhancement module" in Embodiment 2 below) to output the first received signal strength of at least one corner sampling point in the target area. The specific steps are as follows:

[0065] Step 1: Data preprocessing

[0066] Preprocess data such as the building layout, radio propagation law, and received signal strengths of known sampling points, and convert them into a format acceptable to the model. For example, normalize the received signal strength to the range [0, 1].

[0067] Step 2: Input to the model

[0068] Input the preprocessed data into the AttUnet model. The encoder part of the model downsamples the input data to extract features; the decoder part restores to the original size through upsampling and enhances the transmission of features through the attention gating module.

[0069] Step 3: Function of the attention mechanism

[0070] In each skip connection, the attention gating module determines which parts of the encoder feature map are more important based on the feature map of the decoder. In this way, the model can better focus on the key areas of signal propagation, such as the edges of buildings or areas with large signal attenuation.

[0071] Step 4: Output the prediction result

[0072] Finally, the model outputs the predicted received signal strength distribution map through the convolutional layer. This distribution map characterizes the received signal strength at different positions within the target area.

[0073] Step S102: Generate a plurality of different candidate spectral situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points, and random noise; the second received signal strength corresponding to the same sampling point in any two candidate spectral situation maps is different.

[0074] In the embodiments of the present application, a plurality of different candidate spectral situation maps are generated by a DDIM model (an implicit denoising model, a type of diffusion model, which can be understood as the "generation module" in Embodiment 2 below) according to the received signal strength of the enhanced sampling points (i.e., corner sampling points and known sampling points).

[0075] In some specific embodiments, the above step S102 further includes steps S1021 - S1024:

[0076] Step S1021: Generate a random noise matrix.

[0077] Specifically, the dimension of the random noise matrix is the same as the resolution of the target spectral situation map. For example, if the resolution of the target spectral situation map is 10×10, then the dimension of the random noise matrix is 10×10.

[0078] Step S1022: Fuse the random noise matrix, the first received signal strength of the corner sampling points, and the first received signal strength of the plurality of known sampling points to obtain initial input data.

[0079] Specifically, first, for the convenience of processing by the DDIM model, the first received signal strength of the corner sampling points and the first received signal strength of the plurality of known sampling points are normalized, and then the normalized corresponding received signal strengths are combined with the corresponding sampling points to obtain input data; finally, the input data and the random noise matrix are combined to form initial input data.

[0080] Step S1023: Input the initial input data into the denoising diffusion probability model so that the denoising diffusion probability model gradually generates candidate spectral situation maps through the reverse diffusion process.

[0081] Specifically, the reverse diffusion process includes a plurality of time steps, and the noise is gradually reduced at each time step;

[0082] Step S1024, repeat the above steps multiple times to generate multiple different candidate spectrum situation maps.

[0083] Specifically, each candidate spectrum situation map characterizes the distribution of the received signal strengths at different sampling points within the target area.

[0084] Step S103, for any one of the candidate spectrum situation maps, calculate the set of signal strength differences and the set of similarity parameters corresponding to the candidate spectrum situation map.

[0085] Specifically, any signal strength difference in the set of signal strength differences refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectrum situation map and the corresponding preset received signal strength; any similarity parameter in the set of similarity parameters refers to the similarity parameter between the received signal strength distribution map within the preset range of the corresponding emission point in the candidate spectrum situation map and the corresponding preset received signal strength distribution map.

[0086] In some specific embodiments, after the above step S101, the method further includes:

[0087] Step a1, calculate a correction factor according to the building layout map of the target area and the received signal strength distribution map within the preset range of any emission point in the candidate spectrum situation map.

[0088] Specifically, this correction factor res is used to overcome the offset of the first received signal strength of the corner sampling point detected at a low sampling rate relative to the true received signal strength of this corner sampling point.

[0089] Step a2, correct the first received signal strength of each corner sampling point within the target area through the correction factor to obtain the preset received signal strength.

[0090] In some specific embodiments, the above step S103 includes steps b1 - b2:

[0091] Step b1, for any one of the multiple corner sampling points of the candidate spectrum situation map, calculate the signal strength difference between the second received signal strength of the corner sampling point and the corresponding preset received signal strength;

[0092] Step b2, generate the set of signal strength differences corresponding to the candidate spectrum situation map according to the multiple signal strength differences corresponding one by one to the multiple corner sampling points.

[0093] Specifically, for example: The candidate spectrum situation diagram includes corner sampling point 1, corner sampling point 2, and corner sampling point 3. Among them, corner sampling point 1 - signal intensity difference 1, corner sampling point 2 - signal intensity difference 2, corner sampling point 3 - signal intensity difference 3. The set of signal intensity differences corresponding to the candidate spectrum situation diagram is {signal intensity difference 1, signal intensity difference 2, signal intensity difference 3}.

[0094] In some specific embodiments, after the above step S102, the method further includes steps c1 - c3:

[0095] Step c1, for any one of the multiple candidate spectrum situation diagrams, determine at least one emission point from the candidate spectrum situation diagram, and identify the received signal intensity distribution diagram within the preset range of each emission point.

[0096] Step c2, obtain the radio propagation physical model of the target area.

[0097] Specifically, the radio propagation physical model refers to evaluating the performance of the spectrum situation diagram in the environment around the emission point through physical laws, that is, performing physical simulation on a circle centered on the emission station to obtain the preset received signal intensity distribution diagram within the preset range of the emission point, where the preset range is predefined.

[0098] Step c3, based on the radio propagation physical model, identify the preset received signal intensity distribution diagram within the preset range of each emission point.

[0099] In some specific embodiments, the above step S103 includes steps d1 - d2:

[0100] Step d1, for any one of the at least one emission point of the candidate spectrum situation diagram, calculate the similarity parameter between the received signal intensity distribution diagram within the preset range of the emission point and the corresponding preset received signal intensity distribution diagram.

[0101] Specifically, the similarity parameter can be understood as the similarity between the received signal intensity distribution diagram and the corresponding preset received signal intensity distribution diagram. The specific calculation method of the similarity is not specifically limited here.

[0102] Step d2, generate the set of similarity parameters corresponding to the candidate spectrum situation diagram according to the at least one similarity parameter corresponding to the at least one emission point of the candidate spectrum situation diagram.

[0103] Specifically, each emission point has a corresponding received signal strength distribution map. The number of received signal strength distribution maps of its candidate spectrum situation map can be determined according to the number of emission points in the candidate spectrum situation map, and then the number of similarity parameters can be determined, so as to obtain the set of similarity parameters corresponding to the candidate spectrum situation map.

[0104] Step S104, according to the set of signal strength differences and the set of similarity parameters corresponding to each candidate spectrum situation map, screen out the target spectrum situation map from the multiple candidate spectrum situation maps.

[0105] Specifically, this step S104 can be understood as the "selection module" in Embodiment 2 below.

[0106] In some specific embodiments, the above step S104 includes steps S1041 - S1044:

[0107] Step S1041, for any one of the multiple candidate spectrum situation maps, calculate the absolute values of the multiple signal strength differences corresponding to the candidate spectrum situation map, and sum the absolute values of the multiple signal strength differences to obtain a first comparison parameter.

[0108] Specifically, among the multiple signal strength differences corresponding to the candidate spectrum situation map, there are positive numbers and negative numbers. By taking the absolute values of the multiple signal strength differences first and then performing the summation process to obtain the first comparison parameter, it is to reflect the degree of change between the received signal strength of each sampling point in the candidate spectrum situation map and the corresponding preset received signal strength, so as to facilitate screening out the candidate spectrum situation maps in which the received signal strength of all sampling points is closer to the preset received signal strength. This is to screen the multiple candidate spectrum situation maps from the perspective of the received signal strength of the sampling points.

[0109] Step S1042, sum at least one similarity parameter corresponding to the candidate spectrum situation map to obtain a second comparison parameter.

[0110] Specifically, first, calculate the similarity parameters between the received signal strength distribution map of each emission point in the candidate spectrum situation map and the corresponding preset received signal strength distribution map, in order to ensure that the received signal strength distribution map of each emission point in the finally screened target spectrum situation map is relatively similar to the corresponding preset received signal strength distribution map; secondly, sum at least one similarity parameter corresponding to the candidate spectrum situation map to obtain a second comparison parameter, and screen the multiple candidate spectrum situation maps through the second comparison parameter, in order to ensure that the received signal strength distribution maps of all emission points in the finally screened target spectrum situation map are relatively similar to the corresponding preset received signal strength distribution maps as a whole.

[0111] Step S1043: Calculate the third comparison parameter of the candidate spectrum situation map based on the first comparison parameter, the second comparison parameter, the first weight corresponding to the first comparison parameter, and the second weight corresponding to the second comparison parameter of the candidate spectrum situation map.

[0112] Specifically, the first weight and the second weight can be artificially set according to the actual situation, and no specific limitation is made here. The specific calculation method of the third comparison parameter is as follows:

[0113] a1*A1 + b1*B1 = C1

[0114] where a1 represents the first weight, A1 represents the first comparison parameter, b1 represents the second weight, B1 represents the second comparison parameter, and C1 represents the third comparison parameter.

[0115] Step S1044: Take the candidate spectrum situation map with the largest third comparison parameter among the multiple candidate spectrum situation maps as the target spectrum situation map.

[0116] Embodiment 2:

[0117] Based on the above Embodiment 1, the present invention also provides a spectrum situation generation method based on a generative artificial intelligence method using an enhancement module, a generation module, and a selection module. For example, Figure 2 as shown, the main idea is to first use a diffusion model to generate a set of candidate fine-grained spectrum situation maps, and then find the best result from the candidate set as the final output. Specifically as follows:

[0118] Enhancement module

[0119] Although the invention uses a creative generation model such as a diffusion model, due to the extremely limited number of sampling points, the candidate set often consists mainly or entirely of low-quality maps. Therefore, it is very important to design an enhancement module before the generation stage to improve the quality of the maps in the candidate set. The diffusion model is good at focusing on important features and performing information processing, but it lacks a deep understanding of the basic physical laws. Therefore, in the enhancement module of the invention, by adopting the output of a U-Net model with a self-attention mechanism, sampling points are supplemented to the DDIM model according to the physical laws of radio propagation, thereby improving the performance of the architecture.

[0120] In this stage, the invention designed an AttUnet, a U-Net model with a self-attention mechanism, which predicts the received power (received signal strength) of these critical points based on prior information such as building layouts, radio propagation laws, and the reception of known radio samples. DDIM can more easily predict the propagation of high-power radios in a building-free environment. On the contrary, predicting the propagation of low-power radios in a complex building environment is a greater challenge for DDIM. The invention regards the complex building environment as an area containing "key points" for DDIM. Then these prediction results are provided to DDIM. In practice, the invention regards corner points as "key points".

[0121] Generation module

[0122] In the generation module, the invention utilizes an existing advanced DDIM model (an implicit denoising model, a type of diffusion model) to generate a set of candidate radio maps based on the enhanced sampling values.

[0123] DDIM is an improved method of DDPM (Denoising Diffusion Probability Model). DDPM is an image generation model based on Markov chains, which generates images with good quality, strong diversity, and more stable training. The biggest drawback of DDPM is that it requires setting a relatively large number of diffusion steps to achieve good results, which leads to a slow speed of generating samples. For example, if the number of diffusion steps is 1000, then the model needs to perform 1000 inferences to generate a sample. The DDIM method is a method to accelerate the image generation of DDPM with little degradation in image quality. Specifically, the method uses DDPM for training and DDIM for inference.

[0124] Selection module

[0125] Even with the enhancement module, at extremely low sampling rates, many of the spectral situation maps in the candidate set generated by DDIM have poor quality. Therefore, it is necessary to design a selection module after the generation stage to find the best results in the candidate set. In the selection module, the invention is guided by mathematics and radio propagation models (a method of integrating information processing into physical laws) to find the map in the candidate set that best conforms to the propagation model and use it as the final output.

[0126] The selection module is realized through the cooperation of a method based on information processing and a method based on physical laws.

[0127] Method based on information processing: For example Figure 3 As shown, the DDIM method can be regarded as establishing a mapping from the noise distribution N(0, I) to the conditional distribution X of the real data in a 256×256-dimensional space under the condition c. c However, the spectral situation generation task requires us to be at a specific point x of the conditional distribution. cAs close as possible, rather than simply conforming to the distribution. Naturally, since the value of x c is unknown, we use the known sparse sampling s c in the condition to estimate x c . However, due to the extremely low dimensionality of the sparse sampling s c at low sampling rates, the most optimal point x z cannot be effectively selected. Therefore, since the output x a of AttUnet in the enhancement module has all-dimensional information and is relatively close to x c , it can guide the selection of the most optimal point. Also, because x a will also produce an offset relative to x c at low sampling rates, we can introduce a correction factor res for correction. Among them, the value of the correction factor res can be obtained through various information analyses.

[0128] Method based on information processing: The method based on information processing only processes based on the information flow and does not reflect the physical information of the spectrum situation characteristics. Therefore, in the case of the presence of a transmitting station in the area, we can use the method based on physical laws for selection. However, in complex situations such as low-power propagation, electromagnetic wave interaction, and building reflection of electromagnetic waves, it is very difficult for the method based on physical laws to outperform the results of deep learning methods. Only in scenarios where the physical laws are relatively clear, such as in high-power (around the emission point) environments, can we simulate results very close to the real situation through physical laws. Therefore, in the case of high power, this application can evaluate the performance of the spectrum situation map in the environment around the emission point through physical laws, and then judge the overall performance of the spectrum situation map and make a selection, such as Figure 4 shown. Therefore, this application conducts physical simulations on the ring centered on the emission point and compares them with the results output by the model. Those closer to the simulation results are considered to be more effective, and the degree of closeness is measured by the mean square error. At the same time, this application also emits detection rays outward centered on the transmitting station, and then studies the change in the received power on the line segment from the emission to the building contact of the rays, and judges the credibility of the results by observing whether it conforms to the smooth decline of physical laws. The rays extend outward up to 300m at most.

[0129] This embodiment has the following technical effects:

[0130] 1. Using the generative method based on the diffusion model to solve the spectrum situation generation task based on sampling, it can achieve the same effect at a sampling rate one-fifth of the existing methods in the case of extremely low sampling rates.

[0131] 2. The model has strong robustness and can obtain good results in generating spectral situation maps for complex wireless electromagnetic wave scenarios in space, such as scenarios where multiple transmitting stations interfere with each other and scenarios where there is no transmitting station in the area.

[0132] Corresponding to the implementation manner of the above spectral situation map generation method, an embodiment of the present application further provides a spectral situation map generation device for executing the spectral situation map generation method described in any one of the above Figures 1 to 4 illustrated embodiments. As Figure 5 shown, the spectral situation map generation device includes:

[0133] A received signal strength determination module, configured to determine the first received signal strength of at least one corner sampling point in the target area according to the attenuation information of the radio signal in the target area and the first received signal strength of a plurality of known sampling points in the target area; the corner sampling point represents the intersection of two sides in the target area; the target area contains at least one emission point;

[0134] A candidate spectral situation map generation module, configured to generate a plurality of different candidate spectral situation maps according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points, and random noise; the second received signal strength corresponding to the same sampling point in any two candidate spectral situation maps is different;

[0135] A set calculation module, configured to calculate a signal strength difference set and a similarity parameter set corresponding to any one candidate spectral situation map; any signal strength difference in the signal strength difference set refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectral situation map and the corresponding preset received signal strength; any similarity parameter in the similarity parameter set refers to the similarity parameter between the received signal strength distribution map within the preset range of the corresponding emission point in the candidate spectral situation map and the corresponding preset received signal strength distribution map;

[0136] A target spectral situation map screening module, configured to screen out a target spectral situation map from the plurality of candidate spectral situation maps according to the signal strength difference set and the similarity parameter set corresponding to each candidate spectral situation map.

[0137] Optionally, the candidate spectrum situation map generation module is further configured to generate a random noise matrix, where the dimension of the random noise matrix is the same as the resolution of the target spectrum situation map; fuse the random noise matrix, the first received signal strength of the corner sampling points, and the first received signal strength of the several known sampling points to obtain initial input data; input the initial input data into the denoising diffusion probability model, so that the denoising diffusion probability model gradually generates a candidate spectrum situation map through the reverse diffusion process; where the reverse diffusion process includes multiple time steps, and the noise is gradually reduced in each time step; repeat the above steps multiple times to generate multiple different candidate spectrum situation maps.

[0138] Optionally, the apparatus further includes a signal reception strength correction module, configured to, after determining the first received signal strength of at least one corner sampling point in the target area, calculate a correction factor according to the building layout map of the target area and the received signal strength distribution map within a preset range of any emission point in the candidate spectrum situation map; correct the first received signal strength of each corner sampling point in the target area through the correction factor to obtain a preset received signal strength.

[0139] Optionally, the set calculation module is further configured to, for any one of the multiple corner sampling points of the candidate spectrum situation map, calculate the signal strength difference between the second received signal strength of the corner sampling point and the corresponding preset received signal strength; generate a signal strength difference set corresponding to the candidate spectrum situation map according to the multiple signal strength differences corresponding to the multiple corner sampling points one by one.

[0140] Optionally, the preset received signal strength distribution map generation module is configured to, for any one of the multiple candidate spectrum situation maps, determine at least one emission point from the candidate spectrum situation map, and identify the received signal strength distribution map within a preset range of each emission point; obtain the radio propagation physical model of the target area; based on the radio propagation physical model, identify the preset received signal strength distribution map within a preset range of each emission point.

[0141] Optionally, the set calculation module is further configured to, for any one of the at least one emission point of the candidate spectrum situation map, calculate the similarity parameter between the received signal strength distribution map within the preset range of the emission point and the corresponding preset received signal strength distribution map; generate a similarity parameter set corresponding to the candidate spectrum situation map according to the at least one similarity parameter corresponding to the at least one emission point of the candidate spectrum situation map one by one.

[0142] Optionally, the target spectrum situation map screening module is further configured to, for any one of the multiple candidate spectrum situation maps, calculate the absolute values of the multiple signal intensity differences corresponding to the candidate spectrum situation map, sum the absolute values of the multiple signal intensity differences to obtain a first comparison parameter; sum at least one similarity parameter corresponding to the candidate spectrum situation map to obtain a second comparison parameter; calculate a third comparison parameter of the candidate spectrum situation map according to the first comparison parameter, the second comparison parameter, the first weight corresponding to the first comparison parameter, and the second weight corresponding to the second comparison parameter; and use the candidate spectrum situation map with the largest third comparison parameter among the multiple candidate spectrum situation maps as the target spectrum situation map.

[0143] The spectrum situation map generation device provided in the above embodiments of the present application and the spectrum situation map generation method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0144] The embodiments of the present application also provide an electronic device to execute the above spectrum situation map generation method. Please refer to Figure 6 , which shows a schematic diagram of an electronic device provided in some embodiments of the present application. As Figure 6 shown, the electronic device 6 includes: a processor 600, a memory 601, a bus 602, and a communication interface 603. The processor 600, the communication interface 603, and the memory 601 are connected through the bus 602; a computer program that can run on the processor 600 is stored in the memory 601, and when the processor 600 runs the computer program, it executes the spectrum situation map generation method provided in any one of the foregoing Figures 1 to 4 schematic embodiments of the present application.

[0145] Among them, the memory 601 may include a high-speed random access memory (Random Access Memory, RAM), and may also include a non-volatile memory, such as at least one disk memory. Through at least one communication interface 603 (which can be wired or wireless), a communication connection between the system network element and at least one other network element is realized, and the Internet, a wide area network, a local area network, a metropolitan area network, etc. can be used.

[0146] The bus 602 may be an ISA bus, a PCI bus, an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 601 is used to store a program, and after receiving an execution instruction, the processor 600 executes the program, and the foregoing Figures 1 to 4The method for generating a spectral situation diagram disclosed in any of the schematic embodiments can be applied to or implemented by a processor 600.

[0147] The processor 600 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 in the processor 600 or the instructions in the form of software. The above-mentioned processor 600 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. This storage medium is located in the memory 601, and the processor 600 reads the information in the memory 601 and combines its hardware to complete the steps of the above method.

[0148] The electronic device provided in the embodiments of the present application and the method for generating a spectral situation diagram provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the method adopted, run, or implemented by it.

[0149] The embodiments of the present application also provide a computer-readable storage medium corresponding to the method for generating a spectral situation diagram provided in the foregoing embodiments. Please refer to Figure 7 which shows that the computer-readable storage medium is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method for generating a spectral situation diagram provided in any of the foregoing embodiments.

[0150] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other optical or magnetic storage media, which will not be elaborated here one by one.

[0151] The computer-readable storage medium provided by the above embodiments of the present application and the method for generating a spectrum situation map provided by the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0152] It should be noted that:

[0153] In the specification provided here, a large number of specific details are described. However, it can be understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this specification.

[0154] Similarly, it should be understood that, in order to streamline the present application and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting the following schematic: that the claimed present application requires more features than those expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim itself serves as a separate embodiment of the present application.

[0155] In addition, those skilled in the art can understand that, although some of the embodiments described herein include certain features included in other embodiments but not other features, the combination of the features of different embodiments means that it is within the scope of the present application and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0156] As described above, only the preferred specific embodiments of the present application are provided, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claimed rights.

Claims

1. A method for generating a spectrum situation diagram, characterized in that: The method comprises: Determine the first received signal strength of at least one corner sampling point in the target area according to the attenuation information of the radio signal in the target area and the first received signal strength of a plurality of known sampling points in the target area; the corner sampling point represents the intersection of two edges in the target area; the target area contains at least one transmitting point; Generate a plurality of different candidate spectrum situation diagrams according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points, and random noise; the second received signal strengths corresponding to the same sampling point in any two candidate spectrum situation diagrams are different; For any candidate spectrum situation diagram, calculate the signal strength difference set and similarity parameter set corresponding to the candidate spectrum situation diagram; any signal strength difference in the signal strength difference set refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectrum situation diagram and the corresponding preset received signal strength; any similarity parameter in the similarity parameter set refers to the similarity parameter between the received signal strength distribution diagram within the preset range of the corresponding transmitting point in the candidate spectrum situation diagram and the corresponding preset received signal strength distribution diagram; According to the signal strength difference set and the similarity parameter set corresponding to each candidate spectrum situation map, a target spectrum situation map is screened out from the multiple candidate spectrum situation maps.

2. The method according to claim 1, characterized in that Generating a plurality of different candidate spectrum situation diagrams according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points, and random noise, including: Generate a random noise matrix, wherein the dimension of the random noise matrix is ​​the same as the resolution of the target spectrum situation map; fusing the random noise matrix, the first received signal strength of the corner sampling point, and the first received signal strength of the plurality of known sampling points to obtain initial input data; Inputting the initial input data into a denoising diffusion probability model so that the denoising diffusion probability model gradually generates a candidate spectrum situation map through a reverse diffusion process; wherein the reverse diffusion process includes a plurality of time steps, and each time step gradually reduces noise; Repeat the above steps multiple times to generate multiple different candidate spectrum situation diagrams.

3. The method according to claim 1, characterized in that After determining a first received signal strength of at least one corner sampling point within the target area, the method further includes: Calculating a correction factor based on a building layout diagram of the target area and a received signal strength distribution diagram within a preset range of any transmitting point in the candidate spectrum situation diagram; The first received signal strength of each corner sampling point in the target area is corrected by the correction factor to obtain a preset received signal strength.

4. The method according to claim 3, characterized in that Calculating a signal strength difference set and a similarity parameter set corresponding to the candidate spectrum situation diagram, including: For any one of the multiple corner sampling points of the candidate spectrum situation diagram, calculating a signal strength difference between a second received signal strength of the corner sampling point and a corresponding preset received signal strength; According to the multiple signal strength differences corresponding one-to-one to the multiple corner sampling points, a signal strength difference set corresponding to the candidate spectrum situation diagram is generated.

5. The method according to claim 1, characterized in that After generating a plurality of candidate spectrum situation diagrams, the method further includes: For any one of the multiple candidate spectrum situation maps, determine at least one transmission point from the candidate spectrum situation map, and identify a received signal strength distribution map within a preset range of each transmission point; Acquiring a radio propagation physical model of the target area; Based on the radio propagation physical model, a preset received signal strength distribution diagram within a preset range of each transmitting point is identified.

6. The method according to claim 5, characterized in that Calculating a signal strength difference set and a similarity parameter set corresponding to the candidate spectrum situation diagram, including: For any one of the at least one transmitting point of the candidate spectrum situation diagram, calculating a similarity parameter between a received signal strength distribution diagram within a preset range of the transmitting point and a corresponding preset received signal strength distribution diagram; According to at least one similarity parameter corresponding to at least one emission point of the candidate spectrum situation diagram, a similarity parameter set corresponding to the candidate spectrum situation diagram is generated.

7. The method according to claim 1, characterized in that According to the signal strength difference set and the similarity parameter set corresponding to each candidate spectrum situation map, a target spectrum situation map is screened out from the plurality of candidate spectrum situation maps, including: For any one of the multiple candidate spectrum situation graphs, calculate the absolute values ​​of multiple signal strength differences corresponding to the candidate spectrum situation graph, and sum the absolute values ​​of the multiple signal strength differences to obtain a first comparison parameter; Summing at least one similarity parameter corresponding to the candidate spectrum situation diagram to obtain a second comparison parameter; Calculate a third comparison parameter of the candidate spectrum situation diagram according to the first comparison parameter of the candidate spectrum situation diagram, the second comparison parameter, the first weight corresponding to the first comparison parameter, and the second weight corresponding to the second comparison parameter; The candidate spectrum situation map with the largest third comparison parameter among the multiple candidate spectrum situation maps is used as the target spectrum situation map.

8. A spectrum situation diagram generating device, characterized in that: The device comprises: A received signal strength determination module, configured to determine a first received signal strength of at least one corner sampling point in the target area according to attenuation information of the radio signal in the target area and first received signal strengths of a plurality of known sampling points in the target area; the corner sampling point represents an intersection of two edges in the target area; and the target area contains at least one transmitting point; A candidate spectrum situation diagram generation module is used to generate a plurality of different candidate spectrum situation diagrams according to the first received signal strength of the at least one corner sampling point, the first received signal strength of the plurality of known sampling points and random noise; the second received signal strength corresponding to the same sampling point in any two candidate spectrum situation diagrams is different; A set calculation module is used to calculate, for any candidate spectrum situation diagram, a signal strength difference set and a similarity parameter set corresponding to the candidate spectrum situation diagram; any signal strength difference in the signal strength difference set refers to the difference between the second received signal strength of the corresponding corner sampling point in the candidate spectrum situation diagram and the corresponding preset received signal strength; any similarity parameter in the similarity parameter set refers to the similarity parameter between the received signal strength distribution diagram within the preset range of the corresponding transmitting point in the candidate spectrum situation diagram and the corresponding preset received signal strength distribution diagram; The target spectrum situation diagram screening module is used to screen out a target spectrum situation diagram from the multiple candidate spectrum situation diagrams according to the signal strength difference set and the similarity parameter set corresponding to each candidate spectrum situation diagram.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the spectrum situation diagram generating method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the spectrum situation diagram generating method according to any one of claims 1 to 7.

Images