Method for obtaining dynamic spectrum access frequency point and dynamic spectrum access method
A technology of dynamic spectrum access and frequency points, applied in the field of communication, can solve the problem of low efficiency of dynamic spectrum access frequency points, and achieve the effect of speed improvement
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0056] This embodiment provides a dynamic spectrum access decision-making network training method, figure 1 is a flowchart of this embodiment. Specifically include the following steps:
[0057] S101. Superimposing the primary user sample annotation map and the initial secondary user sample annotation map in the target environment to synthesize a comprehensive sample annotation map;
[0058] S102. Using the comprehensive sample annotation map as a condition, input it to the trained conditional confrontation generation network model, and generate a first simulated spectrogram set of the target environment;
[0059] S103. Input the first simulated spectrogram set and the comprehensive sample annotation map into a decision network to obtain a frequency band decision result;
[0060] S104. Adjust the secondary user sample labeling map according to the frequency band decision result, and obtain an adjusted comprehensive sample labeling map according to the adjusted secondary user ...
Embodiment 2
[0087] In the first embodiment, if Figure 4 As shown, the training method of the conditional confrontation generation network model specifically includes:
[0088] S201. Randomly acquire a variety of spectrograms and sample annotations corresponding to the spectrograms;
[0089] S202. Using the spectrogram and the sample annotation graph corresponding to the spectrogram as a training sample, train to obtain a conditional adversarial generative network model, wherein the sample annotation graph is a condition.
[0090] Below, combine Figure 4 The specific implementation steps of the method provided by the embodiment of this application are introduced in detail:
[0091] First, step S201 is executed to randomly acquire a variety of spectrograms and sample annotation maps corresponding to the spectrograms;
[0092] Specifically, the spectrum waterfall diagram is a diagram containing spectrum information generated by the spectrum environment, and,
[0093] The sample annotat...
Embodiment 3
[0105] In this example, if Figure 5 shown, including:
[0106] A synthesis module, used to superimpose the annotation graph of the primary user sample and the initial annotation graph of the secondary user sample in the target environment to synthesize a composite sample annotation graph;
[0107] The simulation module is used to use the comprehensive sample annotation map as a condition, input it to the trained conditional confrontation generation network model, and generate the first simulated spectrogram set of the target environment;
[0108] A training module, configured to input the first simulated spectrogram set and the comprehensive sample annotation map into a decision network to obtain a frequency band decision result;
[0109] A decision control module, configured to adjust the secondary user sample labeling map according to the frequency band decision result, and obtain an adjusted comprehensive sample labeling map according to the adjusted secondary user sample...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More