CN-121997023-A - Quantum density matrix-based communication interference signal quantum characteristic extraction method

CN121997023ACN 121997023 ACN121997023 ACN 121997023ACN-121997023-A

Abstract

A communication interference signal quantum characteristic extraction method based on a quantum density matrix belongs to the technical field of signal processing. The method solves the problems of insufficient feature extraction precision and high calculation complexity of the existing method. According to the invention, firstly, the time-frequency characteristic matrix is used for extracting time-frequency information of an original communication interference signal, then the concept of the quantum density matrix is introduced to carry out quantization treatment on the time-frequency characteristic matrix to obtain TFQDM-1 characteristics, then the initially obtained TFQDM-1 characteristics are used for maximum likelihood estimation to reconstruct a new quantum density matrix, the new quantum density matrix is used as TFQDM-2 characteristics, the extracted TFQDM-1 characteristics and TFQDM-2 characteristics can bring computational complexity far lower than that of a classical algorithm, meanwhile, the extracted characteristics can describe the statistical characteristics of the communication interference signal more comprehensively, the accuracy of characteristic extraction on the communication interference signal is improved, and the extracted characteristics have the advantages which are not possessed by the traditional characteristics. The method can be applied to the quantum characteristic extraction of communication interference signals.

Inventors

LI ZHUOMING
LIU JIANYU
SHAN CHENGZHAO
DONG HENG

Assignees

哈尔滨工业大学

Dates

Publication Date: 20260508
Application Date: 20260126

Claims (10)

1. The method for extracting the quantum characteristics of the communication interference signals based on the quantum density matrix is characterized by comprising the following steps of: step one, collecting a signal to be detected; Step two, extracting a time-frequency characteristic matrix of the signal to be detected; Step three, each column vector of the time-frequency characteristic matrix is converted into a pure state respectively; step four, pure states corresponding to all column vectors in the time-frequency characteristic matrix are combined together to convert the time-frequency characteristic matrix into a quantum density matrix, and the quantum density matrix is used as TFQDM-1 characteristics of signals to be detected; when the interference signal is identified, TFQDM-1 characteristics of the signal to be detected are used as the input of a machine learning algorithm, and the interference type identification result of the signal to be detected is output through the machine learning algorithm.
2. The method for extracting the quantum characteristics of the communication interference signal based on the quantum density matrix according to claim 1, wherein the specific process of the third step is as follows: Wherein, the Representing the first in the time-frequency characteristic matrix The number of column vectors is a function of, Representing vectors Is used for the two norms of (2), Representing the first in the time-frequency characteristic matrix The pure state corresponding to each column vector.
3. The method for extracting quantum characteristics of communication interference signals based on a quantum density matrix according to claim 2, wherein in the fourth step, pure states corresponding to each column vector in the time-frequency characteristic matrix are combined together to convert the time-frequency characteristic matrix into the quantum density matrix, and the specific process is as follows: Wherein, the Representing a matrix of quantum densities that are to be applied, Representing a pure state Is a function of the outer product of (a), Representing the number of columns of the time-frequency characteristic matrix, Representing a pure state Is a probability of (2).
4. The method for extracting quantum characteristics of communication interference signals based on a quantum density matrix according to claim 3, wherein in the fourth step, the probability of each pure state satisfies that the sum of the probabilities of all pure states is 1, and the probabilities of all pure states are equal; the quantum density matrix Is symmetrical and semi-positive, satisfies the trace of the matrix 。
5. The method for extracting the quantum characteristics of the communication interference signals based on the quantum density matrix is characterized by comprising the following steps of: step one, collecting a signal to be detected; Step two, extracting a time-frequency characteristic matrix of the signal to be detected; Step three, each column vector of the time-frequency characteristic matrix is converted into a pure state respectively; Step four, pure states corresponding to each column vector in the time-frequency characteristic matrix are combined together to convert the time-frequency characteristic matrix into a quantum density matrix; Fifth, the characteristic value decomposition is carried out on the quantum density matrix in the fourth step, so that each characteristic vector of the quantum density matrix is obtained, and then a new quantum density matrix is constructed according to each characteristic vector Matrix the quantum density TFQDM-2 features as signals to be detected; When the interference signal is identified, TFQDM-2 characteristics of the signal to be detected are used as input of a deep learning algorithm, and an interference type identification result of the signal to be detected is output through the deep learning algorithm.
6. The method for extracting the quantum characteristics of the communication interference signal based on the quantum density matrix according to claim 5, wherein the specific process of the third step is as follows: Wherein, the Representing the first in the time-frequency characteristic matrix The number of column vectors is a function of, Representing vectors Is used for the two norms of (2), Representing the first in the time-frequency characteristic matrix The pure state corresponding to each column vector.
7. The method for extracting quantum characteristics of communication interference signals based on a quantum density matrix according to claim 6, wherein in the fourth step, pure states corresponding to column vectors in the time-frequency characteristic matrix are combined together to convert the time-frequency characteristic matrix into the quantum density matrix, and the specific process is as follows: Wherein, the Representing a matrix of quantum densities that are to be applied, Representing a pure state Is a function of the outer product of (a), Representing the number of columns of the time-frequency characteristic matrix, Representing a pure state Is a probability of (2).
8. The method for extracting quantum characteristics of communication interference signals based on a quantum density matrix according to claim 7, wherein in the fourth step, the probability of each pure state satisfies that the sum of the probabilities of all pure states is 1, and the probabilities of all pure states are equal; the quantum density matrix Is symmetrical and semi-positive, satisfies the trace of the matrix 。
9. The method for extracting quantum characteristics of communication interference signals based on a quantum density matrix as claimed in claim 8, wherein said decomposing the characteristic values of the quantum density matrix in the fourth step to obtain each characteristic vector of the quantum density matrix, and constructing a new quantum density matrix according to each characteristic vector Matrix the quantum density The TFQDM-2 characteristic of the signal to be detected is as follows: fifthly, carrying out eigenvalue decomposition on the quantum density matrix: Wherein, the Represent the first The value of the characteristic is a value of, , Represent the first The feature vector corresponding to the individual feature values, The number of feature vectors is indicated, Representing feature vectors Is the outer product of (2); Fifthly, reconstructing according to each feature vector and probability corresponding to each feature vector to obtain a new quantum density matrix : Wherein, the Represent the first Probability corresponding to each feature vector.
10. The method for extracting quantum characteristics of communication interference signals based on quantum density matrix according to claim 9, wherein the probability corresponding to the characteristic vector is obtained based on maximum likelihood estimation.

Description

Quantum density matrix-based communication interference signal quantum characteristic extraction method Technical Field The invention belongs to the technical field of signal processing, and particularly relates to a communication interference signal quantum characteristic extraction method. Background The rapid development of wireless communication technology has led to its widespread use in various fields over the last decades, but has also faced the serious challenges of the ever-increasing complexity of the electromagnetic environment. The current communication countermeasure scene shows the characteristics of dense distribution of radiation sources, wide frequency range, improved interference power intensity, diversified modulation modes, popularization of low interception probability signals and the like, so that the traditional signal detection and processing means are difficult to deal with. The interference source includes natural thermal noise and other non-human factors, and also includes artificial interference, especially intentional interference for destroying communication system, and its form includes deceptive and pressed, single and multitone, narrow band and wide band, and forms omnibearing threat from air, land to underwater. The diversity and intelligent evolution of the interference make it difficult for the communication system to decode useful information only by receiving signals and further difficult to guarantee the timeliness of information transmission. Therefore, the accurate and rapid intelligent recognition of the interference signals is a primary premise and a core link for implementing an effective countermeasure strategy. The recognition of the interference signals is essentially dependent on the quality of feature extraction, and the accuracy of the recognition directly determines the reliability of subsequent classification and situation understanding. The traditional method generally extracts manual design characteristics from multiple dimensions such as time domain, frequency domain, space domain and the like, wherein the time domain focuses on transient characteristics and statistics (such as high-order cumulant and peak-to-average ratio), the frequency domain analyzes power spectral density distribution, spectrum symmetry and spectrum peak position, and the space domain utilizes array signal processing to extract arrival direction and polarization characteristics. While the prior researches attempt to automatically mine features through deep learning, the problems of high model complexity, large-scale labeling data dependence, catastrophic forgetting and the like still exist, the problems are difficult to deploy on communication equipment with limited resources, and a new paradigm capable of capturing high-dimensional nonlinear associated features under low complexity is needed. In summary, the existing method still has the problems of insufficient feature extraction precision and high computation complexity, and researchers are turning to quantum machine learning to seek breakthrough in face of inherent limitations of classical features in dimension and information quantity. The quantum neural network theoretically has the advantages of exponential storage capacity, simplified network structure, stronger stability, avoidance of catastrophic forgetting and the like by virtue of the Hilbert space of the quantum system, and provides brand new possibility for high-dimensional signal processing. Disclosure of Invention The invention aims to solve the problems of insufficient feature extraction precision and high calculation complexity of the existing method, and provides a communication interference signal quantum feature extraction method based on a quantum density matrix. The technical scheme adopted by the invention for solving the technical problems is as follows: According to one aspect of the invention, a method for extracting quantum characteristics of communication interference signals based on a quantum density matrix specifically comprises the following steps: step one, collecting a signal to be detected; Step two, extracting a time-frequency characteristic matrix of the signal to be detected; Step three, each column vector of the time-frequency characteristic matrix is converted into a pure state respectively; step four, pure states corresponding to all column vectors in the time-frequency characteristic matrix are combined together to convert the time-frequency characteristic matrix into a quantum density matrix, and the quantum density matrix is used as TFQDM-1 characteristics of signals to be detected; when the interference signal is identified, TFQDM-1 characteristics of the signal to be detected are used as the input of a machine learning algorithm, and the interference type identification result of the signal to be detected is output through the machine learning algorithm. Further, the specific process of the third step is as follows: Wherein, the Repre