CN-121530801-B - Modulation pattern recognition method and device for orthogonal frequency division multiplexing signal based on adaptive modulation

CN121530801BCN 121530801 BCN121530801 BCN 121530801BCN-121530801-B

Abstract

The present disclosure provides a modulation pattern recognition method and apparatus for an orthogonal frequency division multiplexing signal based on adaptive modulation, which may be applied to the technical field of wireless communication. The method comprises the steps of obtaining an orthogonal frequency division multiplexing signal, carrying out fast Fourier transform on the orthogonal frequency division multiplexing signal to obtain a NxK-dimensional frequency domain matrix, extracting a plurality of frequency domain symbol vectors associated with all subcarriers from the frequency domain matrix according to the identification of the K subcarriers based on the subcarrier dimensions to obtain K target data sequences, carrying out at least one operation of clustering or accumulating the target data sequences according to any one of the K target data sequences to obtain a plurality of modulation pattern identification features associated with the target data sequences, and determining a target modulation pattern according to the plurality of modulation pattern identification features and a preset modulation pattern mapping relation.

Inventors

TANG XINYI
Li Guangzuo
XU ZHONGQIU
DANG WENHUA
LIN KUAN
ZHANG YUSHU

Assignees

中国科学院空天信息创新研究院

Dates

Publication Date: 20260505
Application Date: 20251218

Claims (10)

1. A modulation pattern recognition method for an orthogonal frequency division multiplexing signal based on adaptive modulation, the method comprising: Obtaining an orthogonal frequency division multiplexing signal, wherein the orthogonal frequency division multiplexing signal comprises N symbols and K subcarriers; based on subcarrier dimension, extracting a plurality of frequency domain symbol vectors associated with each subcarrier from the frequency domain matrix according to the identification of K subcarriers to obtain K target data sequences; At least one operation of clustering or accumulating is carried out on the target data sequences according to any one of the K target data sequences, so that a plurality of modulation pattern recognition features associated with the target data sequences are obtained, wherein the modulation pattern recognition features are used for determining modulation orders; Determining a target modulation pattern according to the multiple modulation pattern recognition features and a preset modulation pattern mapping relation, wherein the target modulation pattern represents a modulation mode of an adaptive modulation orthogonal frequency division multiplexing signal, and the preset modulation pattern mapping relation represents a mapping relation between the modulation pattern recognition features and the modulation pattern.
2. The method of claim 1, wherein the extracting, based on the subcarrier dimensions, the plurality of frequency domain symbol vectors associated with each subcarrier from the frequency domain matrix according to the identification of K subcarriers, to obtain K target data sequences includes: for any one of the K subcarriers, According to the identification of the subcarrier, a plurality of frequency domain symbol vectors which are determined from the frequency domain matrix and are associated with the subcarrier are used as intermediate data sequences; and extracting a data sequence with a preset length from the intermediate data sequence as the target data sequence.
3. The method of claim 2, wherein the modulation pattern recognition feature comprises a first modulation pattern recognition feature and a second modulation pattern recognition feature, Wherein at least one of clustering or accumulating the target data sequences according to any one of the K target data sequences, determining a plurality of modulation pattern recognition features associated with the target data sequences, comprises: For any of the K target data sequences, Clustering the target data sequences, and determining a clustering center and the number of the clustering centers; Accumulating the target data sequences to obtain an accumulated quantity; determining a first modulation pattern recognition feature associated with the target data sequence according to at least one of the number of cluster centers and the cumulative amount; calculating the average value of the phase difference between each clustering center and a preset reference point to obtain the average phase shift of the clustering center; Obtaining a deviation angle according to the average phase deviation of the clustering center and the preset average phase deviation; And determining a second modulation pattern recognition characteristic associated with the target data sequence according to the offset angle and a preset offset angle range.
4. The method of claim 3, wherein the determining the target modulation pattern according to the plurality of modulation pattern recognition features and the preset modulation pattern mapping relationship comprises: Determining a gradient sequence of each modulation pattern recognition feature according to the similarity between adjacent modulation pattern recognition features, and fusing the gradient sequences of each modulation pattern recognition feature to obtain a target gradient sequence; determining at least one gradient extremum from the target gradient sequence as a candidate modulation boundary point; Determining at least one target modulation boundary point from the candidate modulation boundary points based on the preset threshold; determining at least two target modulation areas with different modulation patterns based on the at least one target modulation boundary point, and identifying features according to the modulation patterns of the at least two target modulation areas to obtain at least two feature vector average values; and determining the target modulation pattern according to the average value of the at least two feature vectors and the preset modulation pattern mapping relation.
5. The method of claim 4, wherein determining the gradient sequence of each modulation pattern recognition feature according to the similarity between adjacent modulation pattern recognition features, and fusing the gradient sequences of each modulation pattern recognition feature to obtain the target gradient sequence, comprises: Filtering each modulation pattern recognition feature to obtain an intermediate modulation pattern recognition feature associated with each modulation pattern recognition feature; Determining a gradient sequence of each intermediate modulation pattern recognition feature according to the similarity between each adjacent intermediate modulation pattern recognition feature; and fusing the gradient sequences of the identification features of each intermediate modulation pattern according to preset weights to obtain a target gradient sequence.
6. The method of claim 5, wherein determining at least two target modulation regions with different modulation patterns based on the at least one target modulation boundary point, and identifying features according to the modulation patterns of the at least two target modulation regions, and obtaining at least two feature vector averages comprises: determining at least two target modulation areas according to the at least one modulation boundary point; and respectively averaging the modulation pattern recognition features of at least two target modulation regions to obtain at least two feature vector averages.
7. The method of claim 4, wherein the feature vector average comprises a first feature vector average associated with the first modulation pattern recognition feature and a second feature vector average associated with the second modulation pattern recognition feature, wherein the determining the target modulation pattern from the at least two feature vector averages and the preset modulation pattern mapping relationship comprises: For any one of the at least two feature vector averages, Determining a modulation order associated with the first eigenvector average from the preset modulation pattern mapping sub-relationship based on the first eigenvector average, and determining at least one candidate target modulation pattern based on the modulation order; And determining the target modulation pattern from the at least one candidate target modulation pattern according to the second eigenvector average value and the modulation order.
8. The method of claim 7, wherein said determining said target modulation pattern from said at least one candidate target modulation pattern based on said second eigenvector average and said modulation order comprises: And determining a target modulation pattern from the at least one candidate target modulation pattern based on the second eigenvector average value under the condition of the modulation order determination.
9. The method of any one of claims 1-8, the method further comprising: and generating a frequency domain distribution map corresponding to the modulation pattern based on the modulation pattern.
10. A modulation pattern recognition apparatus for an orthogonal frequency division multiplexing signal based on adaptive modulation, the apparatus comprising: the acquisition module is used for acquiring an orthogonal frequency division multiplexing signal, wherein the orthogonal frequency division multiplexing signal comprises N symbols and K subcarriers; The first obtaining module is used for carrying out fast Fourier transform on the orthogonal frequency division multiplexing signal to obtain a NxK-dimension frequency domain matrix; the second obtaining module is used for extracting a plurality of frequency domain symbol vectors associated with each subcarrier from the frequency domain matrix based on subcarrier dimensions according to the identification of K subcarriers to obtain K target data sequences; a third obtaining module, configured to perform at least one of clustering or accumulating on the target data sequences according to any one of the K target data sequences, and determine a plurality of modulation pattern recognition features associated with the target data sequences, where the modulation pattern recognition features are used to determine a modulation order; The determining module is used for determining a target modulation pattern according to the multiple modulation pattern recognition features and a preset modulation pattern mapping relation, wherein the target modulation pattern represents a modulation mode of an adaptive modulation orthogonal frequency division multiplexing signal, and the preset modulation pattern mapping relation represents a mapping relation between the modulation pattern recognition features and the modulation pattern.

Description

Modulation pattern recognition method and device for orthogonal frequency division multiplexing signal based on adaptive modulation Technical Field The present disclosure relates to the field of wireless communications technologies, and in particular, to a method and apparatus for identifying a modulation pattern of an orthogonal frequency division multiplexing signal based on adaptive modulation. Background With the wide application of software radio technology, the development of the signal processing field is promoted, so that various new protocols and new types of signals are continuously emerging. Based on the software radio technology, the modulation of signals with different systems can be completed by carrying out different modulation, spread spectrum, coding, encryption and other processing means on the baseband signals. The recognition of the signal modulation pattern is an important problem in the detection of communication signals, particularly the automatic recognition of the signal modulation pattern in non-cooperative communication, and a communication information receiver can correctly demodulate data by effectively recognizing the signal modulation pattern, so as to complete the reception of information. The modulation pattern recognition method in the related art has difficulty in performing high-precision recognition of the modulation pattern on the adaptively modulated orthogonal frequency division multiplexing signal. Disclosure of Invention In view of the foregoing, the present disclosure provides a modulation pattern recognition method, apparatus, device, medium, and program product for an orthogonal frequency division multiplexing signal based on adaptive modulation. According to a first aspect of the present disclosure, a modulation pattern recognition method for an orthogonal frequency division multiplexing signal based on adaptive modulation is provided, which includes obtaining an orthogonal frequency division multiplexing signal, where the orthogonal frequency division multiplexing signal includes N symbols and K subcarriers, performing fast fourier transform on the orthogonal frequency division multiplexing signal to obtain a frequency domain matrix of NxK dimensions, extracting a plurality of frequency domain symbol vectors associated with each subcarrier from the frequency domain matrix according to the identification of the K subcarriers based on the subcarrier dimensions to obtain K target data sequences, performing at least one of clustering or accumulating on the target data sequences according to any one of the K target data sequences to obtain a plurality of modulation pattern recognition features associated with the target data sequences, where the modulation pattern recognition features are used to determine a modulation order, determining a target modulation pattern according to the modulation pattern recognition features and a preset modulation pattern mapping relation, where the target modulation pattern represents a modulation pattern of the orthogonal frequency division multiplexing signal of adaptive modulation, and the preset modulation pattern mapping relation represents a mapping relation between the modulation pattern recognition feature and the modulation pattern. According to the embodiment of the disclosure, a plurality of frequency domain symbol vectors associated with each subcarrier are extracted from a frequency domain matrix according to the identification of the K subcarriers to obtain K target data sequences, wherein the K target data sequences comprise a plurality of frequency domain symbol vectors associated with the subcarriers, which are determined from the frequency domain matrix according to the identification of the subcarriers, as intermediate data sequences, for any one of the K subcarriers, and a data sequence with a preset length is extracted from the intermediate data sequences as the target data sequences. According to the embodiment of the disclosure, the modulation pattern recognition features comprise first modulation pattern recognition features and second modulation pattern recognition features, wherein the method comprises the steps of carrying out at least one operation of clustering or accumulating on target data sequences according to any one of K target data sequences, determining a plurality of modulation pattern recognition features associated with the target data sequences, wherein the method comprises the steps of carrying out clustering on the target data sequences according to any one of the K target data sequences, determining the number of clustering centers and the number of the clustering centers, accumulating the target data sequences to obtain the accumulated amount, determining the first modulation pattern recognition features associated with the target data sequences according to at least one of the number of the clustering centers and the accumulated amount, averaging the phase difference between each clusterin