MPSK signal modulation mode identification method, device, medium and product

Abstract

The present invention discloses a method, device, medium and product for identifying MPSK signal modulation modes, relating to the field of signal recognition. The method comprises obtaining a modulation signal sample and extracting a modulation signal constellation diagram based on the modulation signal sample; denoising and restoring the modulation signal constellation diagram using a trained constellation diagram pre-noising and denoising model; the trained constellation diagram pre-noising and denoising model is a one-dimensional convolutional neural network and includes an embedding layer; and performing MSE-based feature matching on the denoised modulation signal constellation diagram and the ideal constellation diagrams of different MPSK modulation modes to obtain a modulation mode recognition result. The present invention can improve the accuracy and efficiency of signal recognition and can be applied to blind demodulation scenarios in non-cooperative communications.

Description

Technical Field

[0001] The present invention relates to the field of signal recognition, and in particular to a method, device, medium and product for identifying an MPSK signal modulation mode. Background Art

[0002] In recent years, with the rapid development of digital communications, wireless signal transmission has become widely used in mobile communications, satellite communications, radar, and other fields, becoming a crucial means of transmitting information. However, during transmission, signals are susceptible to interference due to factors such as environmental noise, channel effects, and multipath effects, resulting in signal loss and distortion, which in turn affects signal reception quality and system demodulation performance. Therefore, to improve the performance of wireless communication systems and ensure signal transmission quality, signal enhancement and identification are crucial for subsequent demodulation.

[0003] Traditional signal processing typically occurs in cooperative communication scenarios, where the receiver knows information such as the received signal parameters and modulation scheme. However, in real-world applications, the receiver may not know the sender's signal parameters, requiring them to be estimated before demodulation. This type of communication, which requires pre-estimation of received signal parameters, is called non-cooperative communication.

[0004] Non-cooperative communications have important strategic significance in the military. For example, in radio reconnaissance, when an enemy target is detected in space, radio technology can be used to intercept the enemy's radio signals without affecting the transmitter. This allows for the reception and decryption of enemy signals, providing strong support for enemy situation analysis and battlefield decision-making. Therefore, non-cooperative communication technology plays an important role in military reconnaissance and is crucial to maintaining national security.

[0005] In non-cooperative communications, signal demodulation is called blind demodulation. Wireless signal modulation identification plays a crucial role between signal reception and demodulation, and is a prerequisite for correct signal demodulation. With the widespread application of deep learning technology, intelligent modulation identification technology has gained widespread adoption due to its ability to automatically learn signal characteristics, better handle complex data, and offer greater flexibility.

[0006] There are many types of modulation signals in wireless communication systems. Among them, Multiple Phase Shift Keying (MPSK) is one of the commonly used modulation types. It has strong anti-interference and good security. It is widely used in civil and military communications. As a research object, it has great practical significance.

[0007] In a non-cooperative communication environment, the receiver cannot obtain relevant parameter information of the received signal, and the transmission process is easily interfered by noise, making it difficult to accurately identify the signal modulation method. Therefore, how to improve the recognition of modulated signals has always been a challenging issue.

[0008] Modulation type recognition methods are mainly divided into two categories: recognition method based on likelihood ratio decision theory and recognition method based on statistical pattern recognition.

[0009] Identification methods based on likelihood ratio decision theory establish hypotheses based on the statistical characteristics of the signal. After determining a cost function, they calculate the data likelihood ratio and select an appropriate threshold for comparison to ultimately identify the modulation type. However, these methods have limitations. For example, the likelihood function computation is complex and requires accurate prior information on the signal's probability distribution. In contrast, statistical pattern recognition methods extract relevant features from the signal to be identified and then train a classification model using the most discriminative features. The trained classification model can then identify the signal's modulation type. These methods offer greater adaptability, relatively low complexity, and strong robustness, adapting to noise variations to a certain extent. Consequently, statistical pattern recognition methods have been widely used in the field of non-cooperative communications.

[0010] Modulated signals possess a variety of characteristics. In addition to traditional signal spectra, high-order cumulants, and transient features, constellation diagrams are widely used in digital signal modulation type identification due to their intuitiveness, ease of implementation, and strong anti-interference properties. In 2020, Doan et al. proposed a modulation type identification method that combines deep learning with constellation diagrams. This method achieves a classification accuracy of approximately 87% under low signal-to-noise ratios. Since signals are susceptible to noise during transmission, the constellation diagram's position shifts. To eliminate redundancy in the original constellation diagram and extract key information to synthesize an enhanced constellation diagram, Song et al. proposed an enhanced constellation diagram modulation recognition method based on a convolutional neural network. This method improves the constellation diagram by capturing correlation information between adjacent points in the signal constellation and incorporating amplitude information into pixel values. This method effectively improves recognition accuracy and robustness. However, these methods treat constellation diagrams as images and directly incorporate image classification and recognition methods from deep learning, resulting in slow processing and high computational complexity.

[0011] Therefore, in order to improve the accuracy and efficiency of signal recognition, it is urgent to provide a new signal recognition method or system. Summary of the Invention

[0012] The purpose of the present invention is to provide a method, device, medium and product for identifying the MPSK signal modulation mode, which can improve the accuracy and efficiency of signal recognition and can be applied in blind demodulation scenarios of non-cooperative communications.

[0013] To achieve the above object, the present invention provides the following solutions:

[0014] A method for identifying an MPSK signal modulation mode, the method comprising:

[0015] Obtain a certain number of modulated signal samples and extract a modulated signal constellation diagram based on the modulated signal;

[0016] Denoising and restoring the constellation of the modulated signal using a trained constellation pre-addition and denoising model (ADMC); the trained ADMC is a one-dimensional convolutional neural network including an embedding layer;

[0017] The denoised modulation signal constellation diagram is matched with the ideal constellation diagrams of different MPSK modulation modes based on MSE features to obtain the modulation mode recognition result.

[0018] Optionally, acquiring a modulation signal and extracting a modulation signal constellation diagram according to the modulation signal specifically includes:

[0019] Carrier frequency estimation of modulated signal based on Welch power spectrum method;

[0020] Symbol rate estimation based on complex envelope detection method;

[0021] Symbol synchronization is achieved by using a self-synchronization method based on the Gardner algorithm;

[0022] Carrier synchronization is achieved based on the Costas in-phase orthogonal loop to obtain the modulation signal constellation diagram.

[0023] Optionally, the process of determining the trained denoising diffusion probability model is:

[0024] Construct a training data set based on the ideal BPSK modulated signal constellation sample;

[0025] Constructing an ADMC convolutional neural network; the input layer of the ADMC convolutional neural network is used to receive a modulation signal constellation sample, and then the modulation signal constellation sample passes through a convolutional layer, a BN layer, a ReLU activation function layer, and a fully connected layer, and finally outputs predicted noise data;

[0026] Noise addition to training dataset samples based on ADMC pre-noising process;

[0027] The ADMC convolutional neural network is trained using the noisy training dataset samples to learn the ADMC denoising process and obtain the trained ADMC applied to the BPSK modulated signal.

[0028] The trained ADMC applied to the QPSK modulated signal and the 8PSK modulated signal is determined.

[0029] Optionally, performing MSE-based feature matching on the denoised modulation signal constellation diagram with ideal constellations of different MPSK modulation modes to obtain a modulation mode recognition result specifically includes:

[0030] The ideal constellation sample points of BPSK, QPSK and 8PSK modulation signals are used as matching templates;

[0031] The denoised modulation signal constellation is matched and compared with the matching template based on MSE;

[0032] The similarity between the denoised modulation signal constellation and the matching template is determined based on the comparison result;

[0033] According to the minimum MSE rule, the most matching modulation mode is selected as the modulation mode of the modulated signal.

[0034] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the MPSK signal modulation mode identification method.

[0035] A computer-readable storage medium stores a computer program, which implements the MPSK signal modulation mode identification method when executed by a processor.

[0036] A computer program product includes a computer program, which implements the MPSK signal modulation mode identification method when executed by a processor.

[0037] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0038] The present invention provides a method, device, medium, and product for identifying MPSK signal modulation modes. This method uses a constellation pre-addition and denoising model-mean square error (ADMC-MSE) to identify modulation signals. The method uses the sample point coordinates of the modulation signal constellation as input data, uses a trained ADMC to denoise and restore the modulation signal constellation, and then performs MSE-based feature matching on the denoised modulation signal constellation with the ideal constellation for different MPSK modulation modes to achieve intelligent recognition of the modulation mode. The present invention can improve the accuracy and efficiency of signal recognition and can be applied to blind demodulation scenarios in non-cooperative communications. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 A schematic flow chart of a method for identifying an MPSK signal modulation mode provided in Example 1 of the present invention;

[0041] Figure 2 A schematic diagram of the ADMC convolutional neural network structure provided in Example 1 of the present invention;

[0042] Figure 3 This is a signal modulation identification framework diagram provided in Example 1 of the present invention;

[0043] Figure 4 Schematic diagram of the recognition results of three modulation signals under different signal-to-noise ratios; DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] The purpose of the present invention is to provide a method, device, medium and product for identifying the MPSK signal modulation mode, which can improve the accuracy and efficiency of signal recognition and can be applied in blind demodulation scenarios of non-cooperative communications.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] Example 1

[0048] like Figure 1 As shown, the present invention provides a method for identifying an MPSK signal modulation mode, comprising:

[0049] S101, obtaining a modulation signal, and extracting a modulation signal constellation diagram according to the modulation signal;

[0050] Specifically include:

[0051] Carrier frequency estimation of modulated signal based on Welch power spectrum method;

[0052] Symbol rate estimation based on complex envelope detection method;

[0053] Symbol synchronization is achieved by using a self-synchronization method based on the Gardner algorithm;

[0054] Carrier synchronization is achieved based on the Costas in-phase orthogonal loop to obtain the modulation signal constellation diagram.

[0055] S102, using the trained ADMC to perform denoising and restoration on the constellation diagram of the modulated signal; the trained ADMC is a one-dimensional convolutional neural network and includes an embedding layer;

[0056] The process of determining the trained ADMC is:

[0057] A training dataset is constructed based on ideal BPSK modulated signal constellation samples. Each sample represents a point in the constellation using two-dimensional coordinates. All samples have the same modulation signal type.

[0058] Construct an ADMC convolutional neural network, such as Figure 2 As shown; the input layer of the ADMC convolutional neural network is used to receive the modulation signal constellation sample, and then the modulation signal constellation sample passes through the convolution layer, BN (BatchNormalization) layer, ReLU (Rectified LinearUnit) activation function layer and the fully connected layer, and finally outputs the predicted noise data; the ADMC convolutional neural network contains a total of three convolution layers, of which the convolution kernel size of the first layer is set to 2, and the second and third layers are both set to 1. A BN layer and a ReLU activation function layer are used after each convolution layer, where the role of the BN layer is to normalize the input data, increase the training speed, prevent gradient explosion, and effectively improve the stability of the model. The role of the ReLU layer is to enhance the nonlinear relationship between the various network layers of the model by introducing an activation function, thereby improving the sparsity of the model. Since the entire training process of ADMC includes time parameters, an embedding layer is added to the model framework so that the model considers the influence of the time step during the generation and prediction process, thereby better processing the generated data.

[0059] The ADMC-based constellation pre-noising process adds noise to the training dataset samples until the training constellation sample points are scattered within the circumscribed circles of the three MPSK constellation points in the form of Gaussian noise. The specific noise addition principle is as follows:

[0060] For MPSK modulated signals, it can be expressed as:

[0061] S N =A m m(t)cos(2πf c t+φ c)+n(t) (1)

[0062] Among them, A m represents the signal amplitude; m(t) represents the low-pass pulse signal; cos(2πf c t+φ c ) represents the high-frequency carrier signal, f c Indicates the carrier frequency, φ c represents the initial phase of the carrier signal, and n(t) represents the noise at the receiving end. Expanding the above formula yields:

[0063]

[0064] n i (t), n q (t) represents the noise component after the noise is removed from the carrier frequency, and still satisfies the same distribution as n(t).

[0065] According to the MPSK modulation principle, each modulated code element can be expressed in the form of a vector:

[0066]

[0067] Among them, δ g is the signal energy, This vector corresponds to a constellation point on the constellation plane.

[0068] The constellation diagram pre-noising method based on ADMC is:

[0069] (a) Select L original MPSK modulated signal samples (noise-free), and all samples are taken as the original element S0 according to the constellation point set determined by (5).

[0070] (b) Generate a random Gaussian noise sequence that conforms to N(0,I), with a mean of 0 and a variance of I, where I is an L-dimensional vector with an element of 1. The generated noise sequence is used as the label for pre-noise training.

[0071] (c) Randomly select an integer N in the range of 0-10000 as the step selection for a training epoch.

[0072] (d) Starting from S0, noise is added to S0 in N steps. The adding method is:

[0073]

[0074] η t The value ranges from 0 to 1, and as t increases, η t Getting smaller and smaller, S t represents the modulated signal sample obtained after adding noise t times, ε trepresents the noise added each time, and obeys the same distribution for all t, that is, it conforms to N(0,I), with a mean of 0 and a variance of I, where I is an L-dimensional vector with 1 elements.

[0075] Use the noisy training dataset samples (and the labels generated in step b) to train the ADMC convolutional neural network, learn the ADMC inverse denoising process, and obtain the trained ADMC applied to the BPSK modulated signal;

[0076] Generate ideal QPSK and 8PSK modulated signal constellation samples to determine the trained ADMC applied to QPSK modulated signals and 8PSK modulated signals.

[0077] In S102, the trained ADMC model for denoising BPSK, QPSK, and 8PSK modulated signals is fed into the three channels respectively to perform the inverse denoising process. By gradually predicting and removing the noise, the model will generate the coordinates of the denoised constellation sample points.

[0078] S103 , performing feature matching based on MSE on the denoised modulation signal constellation diagram and the ideal constellation diagrams of different MPSK modulation modes to obtain a modulation mode recognition result.

[0079] like Figure 3 As shown, S103 specifically includes:

[0080] The ideal constellation sample points of BPSK, QPSK and 8PSK modulation signals are used as matching templates;

[0081] The denoised modulation signal constellation is matched and compared with the matching template based on MSE;

[0082] According to the comparison result, the similarity between the constellation diagram of the modulation signal after denoising and restoration and the matching template is judged;

[0083] According to the minimum MSE rule, the most matching modulation mode is selected as the modulation mode of the modulated signal.

[0084] For the three modulation signals of BPSK, QPSK and 8PSK, the experimental parameters are set as follows: the number of code elements of the signal data is set to 1000, the sampling frequency f s =10kHz, carrier frequency f c =1.5kHz, the channel is a Gaussian white noise channel. The signal-to-noise ratio range is set to -5 to 10dB, with a step of 1dB. 300 Monte Carlo simulation experiments are performed on the signal at each signal-to-noise ratio to obtain the recognition accuracy of each modulation signal. The results are as follows: Figure 4 shown.

[0085] like Figure 4As shown in the figure, as the signal-to-noise ratio increases, the recognition accuracy of the three modulation signals shows a steady upward trend. When the signal-to-noise ratio of the BPSK signal is 0dB, when the signal-to-noise ratio of the QPSK signal is 4dB, and when the signal-to-noise ratio of the 8PSK signal is 8dB, the recognition accuracy of the present invention reaches 100.0%.

[0086] To verify the effective improvement in recognition performance achieved by the present invention, we compared it with three methods based on high-order cumulants, a convolutional neural network, and ISPP-ResNet18 for signal-to-noise ratios of 0dB, 5dB, and 10dB. The recognition results are summarized in Tables 1, 2, and 3, respectively, to visually demonstrate the performance of different methods under low signal-to-noise ratio conditions.

[0087] Table 1 Comparison of recognition performance of three modulation signals under different methods (0db)

[0088]

[0089] Table 2 Comparison of recognition performance of three modulation signals under different methods (5dB)

[0090]

[0091] Table 3 Comparison of recognition performance of three modulation signals under different methods (10db)

[0092]

[0093] According to the data in Table 1, when the signal-to-noise ratio is 0dB, the method proposed in this paper is better in identifying BPSK signals, surpassing the other three methods. Although it is slightly inferior to ISPP-ResNet18 in the recognition of QPSK signals, and the recognition effect of 8PSK signals is also slightly lower than that of convolutional neural networks, the overall average recognition rate is still significantly better than the other three methods. Observing the data in Table 2, when the signal-to-noise ratio is increased to 5dB, the present invention is better than the other three methods in the recognition effect of the three modulation signals, and the recognition rate of BPSK signals and QPSK signals reaches 100%. As the signal-to-noise ratio is further increased to 10dB, as shown in Table 3, it can be seen that under high signal-to-noise ratio conditions, all methods show excellent recognition effects, and the recognition rate of the present invention for the three modulation signals reaches 100%. It can be seen that the present invention achieves good recognition effects at a signal-to-noise ratio of 0dB and obtains the best recognition effect at a signal-to-noise ratio of 5dB.

[0094] In addition, since the present invention uses the signal constellation sample points as the data to be processed, the calculation complexity is low.

[0095] Example 2

[0096] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement an MPSK signal modulation mode identification method in embodiment 1.

[0097] Example 3

[0098] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements an MPSK signal modulation mode identification method in embodiment 1.

[0099] Example 4

[0100] A computer program product includes a computer program. When the computer program is executed by a processor, the method for identifying an MPSK signal modulation mode in embodiment 1 is implemented.

[0101] Example 5

[0102] A computer device, which may be a database. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The I / O interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements an MPSK signal modulation method identification method in Example 1.

[0103] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0104] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but are not limited to these.

[0105] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0106] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for identifying MPSK signal modulation mode, characterized in that: The method comprises: Obtain a certain number of modulation signal samples and extract the modulation signal constellation diagram according to the modulation signal; Denoising the modulated signal constellation using a trained constellation pre-noising and denoising model ADMC; the trained ADMC is a one-dimensional convolutional neural network and includes an embedding layer; the ADMC is a constellation pre-noising and denoising model; the embedding layer is an embedding layer; The denoised modulation signal constellation is matched with the ideal constellation of different MPSK modulation modes based on MSE to obtain the modulation mode recognition result; MPSK modulation mode is multi-phase shift keying; MSE is mean square error; The obtaining of a certain number of modulation signal samples and extracting a modulation signal constellation diagram according to the modulation signal specifically includes: Carrier frequency estimation of modulated signal is performed based on Welch power spectrum method, which is a modern spectrum analysis method. Symbol rate estimation based on complex envelope detection method; Symbol synchronization is achieved using a self-synchronization method based on the Gardner algorithm; the Gardner algorithm is a non-data-assisted timing error detection algorithm; Carrier synchronization is achieved based on the Costas in-phase orthogonal loop to obtain the modulation signal constellation diagram; Costas is a Costas loop, a feedback control system used for carrier synchronization; The process of determining the trained ADMC is: Construct a training data set based on the ideal BPSK modulated signal constellation sample; Construct an ADMC convolutional neural network; the input layer of the ADMC convolutional neural network is used to receive the modulation signal constellation sample, and then the modulation signal constellation sample passes through the convolution layer, BN layer, ReLU activation function layer and fully connected layer respectively, and finally outputs the predicted noise data; the BN layer is a normalization layer; Noise addition to training dataset samples based on ADMC constellation diagram noising process; The ADMC convolutional neural network is trained using the noisy training dataset samples to learn the ADMC denoising process and obtain the trained ADMC applied to the BPSK modulated signal. Based on the ideal QPSK modulated signal constellation diagram samples and the ideal 8PSK modulated signal constellation diagram samples, the ADMC convolutional neural network is trained respectively to learn the ADMC denoising process and determine the trained ADMC applied to the QPSK modulated signal and the 8PSK modulated signal; The modulation signal constellation after denoising is matched with the ideal constellation of different MPSK modulation modes based on MSE features to obtain the recognition result of the modulation mode, specifically including: The ideal constellation sample points of BPSK, QPSK and 8PSK modulation signals are used as matching templates; The denoised modulation signal constellation is matched and compared with the matching template based on MSE; The similarity between the denoised modulation signal constellation and the matching template is determined based on the comparison result; According to the minimum MSE rule, the most matching modulation mode is selected as the modulation mode of the modulated signal.

2. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the MPSK signal modulation mode identification method described in claim 1.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying the MPSK signal modulation mode described in claim 1 is implemented.

4. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for identifying the MPSK signal modulation mode described in claim 1 is implemented.

Images