CN111614398A - Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network - Google Patents
Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network Download PDFInfo
- Publication number
- CN111614398A CN111614398A CN202010396238.6A CN202010396238A CN111614398A CN 111614398 A CN111614398 A CN 111614398A CN 202010396238 A CN202010396238 A CN 202010396238A CN 111614398 A CN111614398 A CN 111614398A
- Authority
- CN
- China
- Prior art keywords
- signal
- neural network
- noise ratio
- modulation format
- xor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
- H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
- H04B10/0795—Performance monitoring; Measurement of transmission parameters
- H04B10/07953—Monitoring or measuring OSNR, BER or Q
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0012—Modulated-carrier systems arrangements for identifying the type of modulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Electromagnetism (AREA)
- Optical Communication System (AREA)
Abstract
The embodiment of the invention provides a modulation format and signal-to-noise ratio identification method and device based on an XOR neural network, wherein the method comprises the following steps: acquiring a signal to be identified after preprocessing a digital domain signal received by a receiving terminal; generating a constellation diagram according to the signal to be identified; the constellation diagram is input into a signal recognition neural network which is trained in advance, the modulation format and the signal to noise ratio of the signal to be recognized are obtained, the signal recognition neural network is an exclusive OR neural network, the signal recognition neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal. Therefore, the modulation format and the signal-to-noise ratio of the signal in the optical fiber communication are identified by adopting the pre-trained XOR neural network, the scale of a network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be applied to a real-time system.
Description
Technical Field
The invention relates to the technical field of optical communication, in particular to a modulation format and signal-to-noise ratio identification method and device based on an XOR neural network.
Background
In the field of optical communications, Optical Performance Monitor (OPM) is an important component in network management, and is important for ensuring high-quality services of various intermediate nodes and destination nodes. Optical modulation format identification and osnr identification are two important aspects of OPM.
In recent years, coherent optical transmission and advanced modulation formats have been widely used in the field of optical communications. Because the constellation diagram can display amplitude and phase information and can comprehensively present a plurality of performance indexes of different modulation signals, the modulation format can be identified and the optical signal to noise ratio can be estimated through the constellation diagram.
However, the traditional constellation diagram analysis method has strong dependence on professional knowledge and is only suitable for experienced engineers. In addition, the conventional statistical method needs to acquire information of each constellation point, which means that all in-phase and quadrature data need to be collected. This is a rather time consuming process and is therefore not suitable for real-time testing systems. Therefore, a constellation diagram analysis method without manual intervention and human errors is a future development direction.
In recent years, a technology for performing optical modulation format recognition and optical signal-to-noise ratio recognition based on a complex neural network appears, and compared with a traditional method, the technology based on the complex neural network can achieve higher accuracy. However, the complex neural network has a large model scale, occupies a large amount of computing resources, and has a slow operation speed, and thus cannot be applied to a real-time system.
Disclosure of Invention
The embodiment of the invention aims to provide a modulation format and signal-to-noise ratio identification method and device based on an XOR neural network, which are used for solving the technical problems that the existing method for identifying a signal modulation format and a signal-to-noise ratio by adopting a complex neural network is large in model scale, occupies more computing resources, is low in operation speed and cannot be applied to a real-time system. The specific technical scheme is as follows:
in order to achieve the above object, an embodiment of the present invention provides a modulation format and signal-to-noise ratio identification method based on an xor neural network, where the method includes:
acquiring a signal to be identified after preprocessing a digital domain signal received by a receiving terminal;
generating a constellation diagram according to the signal to be identified;
inputting the constellation diagram into a signal recognition neural network which is trained in advance to obtain a modulation format and a signal to noise ratio of the signal to be recognized, wherein the signal recognition neural network is an exclusive OR neural network, the signal recognition neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
Optionally, the signal recognition neural network includes a first sub-network and a second sub-network, the first sub-network is a full-precision convolutional neural network, the second sub-network is a binarization exclusive-or convolutional neural network, and an output of the first sub-network is an input of the second sub-network.
Optionally, the first sub-network includes a full-precision convolutional layer, a first normalization layer, an activation function layer, and a first maximum pooling layer; the second sub-network comprises a second batch of normalization layers, a binary activation function layer, a binary exclusive-or convolutional layer, a second maximum pooling layer, a binary full-link layer and an output layer.
Optionally, the number of convolution kernels of the full-precision convolution layer is 20; the number of convolution kernels of the binary exclusive-or convolution layer is 500, and the number of neurons of the binary full-link layer is 500.
Optionally, the signal recognition neural network is trained as follows:
acquiring a preset XOR neural network model and the training set;
inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
determining a loss value based on the identification results of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal;
determining whether the XOR neural network model converges based on the loss value;
if not, adjusting parameter values in the XOR neural network model, and returning to the step of inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
and if so, determining the current XOR neural network model as the signal recognition neural network.
In order to achieve the above object, an embodiment of the present invention further provides a modulation format and snr identification apparatus based on an xor neural network, where the apparatus includes:
the acquisition module is used for acquiring a signal to be identified after the digital domain signal received by the receiving end is preprocessed;
the generating module is used for generating a constellation diagram according to the signal to be identified;
the identification module is used for inputting the constellation diagram into a signal identification neural network which is trained in advance to obtain a modulation format and a signal to noise ratio of a signal to be identified, the signal identification neural network is an exclusive OR neural network, the signal identification neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
Optionally, the signal identification neural network includes a first sub-network and a second sub-network, the first sub-network is a full-precision convolutional neural network, the second sub-network is a binarization exclusive-or convolutional neural network, and an output of the first sub-network is an input of the second sub-network;
the first sub-network comprises a full-precision convolution layer, a first batch of normalization layers, an activation function layer and a first maximum pooling layer; the second sub-network comprises a second batch of normalization layers, a binary activation function layer, a binary exclusive-or convolutional layer, a second maximum pooling layer, a binary full-link layer and an output layer.
Optionally, the apparatus further includes a training module, where the training module is configured to train the signal recognition neural network, and the training module is specifically configured to:
acquiring a preset XOR neural network model and the training set;
inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
determining a loss value based on the identification results of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal;
determining whether the XOR neural network model converges based on the loss value;
if not, adjusting parameter values in the XOR neural network model, and returning to the step of inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
and if so, determining the current XOR neural network model as the signal recognition neural network.
In order to achieve the above object, an embodiment of the present invention further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus; the processor, the communication interface and the memory complete mutual communication through a communication bus;
a memory for storing a computer program;
and the processor is used for realizing any method step when executing the program stored in the memory.
To achieve the above object, an embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements any of the above method steps.
The embodiment of the invention has the following beneficial effects:
by applying the method and the device for identifying the modulation format and the signal-to-noise ratio based on the XOR neural network, which are provided by the embodiment of the invention, the signal to be identified after the digital domain signal received by the receiving end is preprocessed can be obtained, the constellation diagram is generated according to the signal to be identified, the constellation diagram is input into the signal identification neural network which is trained in advance, the modulation format and the signal-to-noise ratio of the signal to be identified are obtained, the signal identification neural network is the XOR neural network, the signal identification neural network is trained in advance according to the training set, and the training set comprises the sample constellation diagrams corresponding to a plurality of sample signals and the modulation format label and the signal-to-noise ratio label of each. Therefore, the modulation format and the signal-to-noise ratio of the signal in the optical fiber communication are identified by adopting the pre-trained XOR neural network, the scale of a network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be applied to a real-time system.
Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a modulation format and snr identification method based on an xor neural network according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an optical communication system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a signal recognition neural network according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a training signal recognition neural network according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an xor neural network-based modulation format and snr identification apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to solve the technical problems that the existing complex neural network adopted for identifying the signal modulation format and the signal to noise ratio has large model scale, more occupied computing resources and low operation speed and cannot be applied to a real-time system, the embodiment of the invention provides a method, a device, electronic equipment and a computer readable storage medium for identifying the modulation format and the signal to noise ratio based on an exclusive-or neural network.
Referring to fig. 1, fig. 1 is a schematic flow chart of a modulation format and signal-to-noise ratio identification method based on an xor neural network according to an embodiment of the present invention, where the method includes the following steps:
s101: and acquiring a signal to be identified after the digital domain signal received by the receiving end is preprocessed.
The modulation format and signal-to-noise ratio identification method based on the XOR neural network can be applied to an optical communication system, and specifically, signal adjustment format and signal-to-noise ratio identification are carried out at a receiving end of the optical communication system.
For ease of understanding, the optical communication system will be briefly described below.
As shown in fig. 2, in an optical communication system, the transmitting end may include a laser and a signal driven optical modulator. The modulated optical signal enters the optical channel through the optical amplifier for transmission. The optical channel includes a plurality of lengths of optical fiber and an optical amplifier link. The signal passing through the optical channel reaches a receiving end, is subjected to frequency mixing with a local oscillator after passing through a filter, and then passes through a photoelectric detector and a digital-to-analog converter to obtain a digital domain signal.
And then preprocessing the digital domain signal at a receiving end to obtain a signal to be identified, and identifying the signal to be identified to obtain a modulation format and a signal-to-noise ratio of the received signal. The preprocessing of the digital domain signal may include down-sampling, dispersion compensation, and the like, which is not limited in the embodiment of the present invention.
S102: and generating a constellation diagram according to the signal to be identified.
In the embodiment of the invention, the constellation diagram can be generated according to the signal to be identified, and the modulation format and the signal-to-noise ratio of the signal can be identified according to the constellation diagram.
The method for generating the constellation diagram according to the signal may refer to related art, and the embodiment of the present invention does not limit this.
S103: the constellation diagram is input into a signal recognition neural network which is trained in advance, the modulation format and the signal to noise ratio of the signal to be recognized are obtained, the signal recognition neural network is an exclusive OR neural network, the signal recognition neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
In the embodiment of the invention, the constellation diagram corresponding to the signal to be identified can be input into the signal identification neural network, and the signal identification neural network is trained in advance according to the training set, so that the modulation format and the signal-to-noise ratio of the signal to be identified can be output.
Compared with the existing method for identifying the signal modulation format and the signal-to-noise ratio by adopting the complex neural network, the method has the important invention point that the exclusive-or neural network is selected as the identification network. In the XOR neural network, binarization operation is simultaneously performed on weight and input, so that the scale of the network model is greatly reduced, the consumption of operation resources is reduced, and the operation rate of the network model is accelerated. In the field of traditional computer identification, the influence of the extreme compression scheme on the accuracy is obvious. However, after theoretical analysis and practical verification, the inventor of the present invention finds that the problem of identifying the signal modulation format and the signal-to-noise ratio in the optical fiber communication system can be positioned as a simpler computer vision problem, and the identification function of the xor neural network is sufficient to meet the requirement.
In the embodiment of the present invention, the training set of the signal recognition neural network may include a sample constellation corresponding to a plurality of sample signals, and a modulation format label and a signal-to-noise ratio label of each sample signal.
Specifically, a plurality of Modulation formats may be set in advance, such as Quadrature Phase Shift Keying (QPSK), 8 Phase Shift Keying (8 QAM), 16-Quadrature Amplitude Modulation (16 QAM), 32-Quadrature Amplitude Modulation (32 QAM), and 64-Quadrature Amplitude Modulation (64 QAM); furthermore, a plurality of signal-to-noise ratios, for example 10db, 12db, 14db, 16db, 18db and 20db, are provided.
And generating a corresponding signal as a sample signal according to the set modulation format and the signal-to-noise ratio, and generating a sample constellation diagram as a training sample according to the sample signal. The modulation format and the signal-to-noise ratio serve as labels for the sample constellation.
For example, a simulation system generates a sample signal with a QPSK modulation format and a signal-to-noise ratio of 10db, and then generates a corresponding sample constellation, so that the label of the QPSK modulation format of the sample constellation is QPSK and the label of the signal-to-noise ratio is 10 db.
In the embodiment of the invention, a plurality of sample constellation diagrams, corresponding modulation format labels and signal-to-noise ratio labels are used as training samples for training a signal recognition neural network.
The modulation format and the signal-to-noise ratio may be set according to actual requirements, which is not limited.
The network structure of the signal recognition neural network and the training process of the signal recognition neural network can be seen below.
By applying the modulation format and signal-to-noise ratio identification method based on the XOR neural network provided by the embodiment of the invention, a signal to be identified after a digital domain signal received by a receiving end is preprocessed can be obtained, a constellation diagram is generated according to the signal to be identified, the constellation diagram is input into a signal identification neural network which is trained in advance, the modulation format and the signal-to-noise ratio of the signal to be identified are obtained, the signal identification neural network is the XOR neural network, the signal identification neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal-to-noise ratio label of each sample signal. Therefore, the modulation format and the signal-to-noise ratio of the signal in the optical fiber communication are identified by adopting the pre-trained XOR neural network, the scale of a network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be applied to a real-time system.
In one embodiment of the present invention, the signal recognition neural network may comprise a first sub-network and a second sub-network, wherein the first sub-network is a full-precision convolutional neural network and the second sub-network is a binary exclusive-or convolutional neural network.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a signal identifying neural network according to an embodiment of the present invention, as shown in fig. 3, the signal identifying neural network includes a first sub-network and a second sub-network, the first sub-network and the second sub-network are sequentially connected, an input of the first sub-network is a constellation diagram, an output of the first sub-network is an input of the second sub-network, and an output of the second sub-network is a modulation format and an snr identification result of a signal.
The first sub-network is a full-precision convolutional neural network, and in one embodiment, the first sub-network may include a full-precision convolutional layer, a first normalization layer, an activation function layer, and a first max-pooling layer, where the full-precision convolutional layer may include fewer convolution kernels, for example, 20.
The skilled person can understand that the full-precision convolution layer is used for extracting the features of the image through convolution operation, the batch normalization layer is used for normalizing feature data, the feature data are generally converted into the distribution with the mean value of 0 and the standard deviation of 0-1, and the problem that the learning speed of the neural network is low due to the fact that the feature data are dispersed is avoided. The activation function layer is used to increase the nonlinearity of the neural network model so that the neural network can arbitrarily approximate any nonlinear function. The maximum pooling layer is to increase the computation speed and also to increase the robustness of the extracted features.
After the input constellation diagram is processed by the first sub-network, the processing result is input into the second sub-network, and the second sub-network is a binary or convolution neural network.
In one embodiment, the second sub-network may comprise a second batch normalization layer, a binary activation function layer, a binary exclusive-or convolution layer, a second max-pooling layer, a binary full-link layer, and an output layer.
Compared with a full-precision network, the activation function layer, the convolution layer and the full connection layer of the second sub-network are binarized, and the operations in the second sub-network are exclusive-OR operations, so that compared with matrix operations in the full-precision network, the operation resource consumption is greatly reduced, and the operation speed is improved.
In one embodiment of the present invention, the number of convolution kernels of the binary exclusive-or convolutional layer may be 500, and the number of neurons of the binary full-link layer may be 500.
Therefore, in the embodiment of the present invention, the signal recognition neural network may include a first sub-network and a second sub-network, the first sub-network is a full-precision convolutional neural network with fewer convolutional kernels and a smaller scale, and the first sub-network is used for extracting the constellation features and performing preliminary processing. The second sub-network is a binary XOR convolutional neural network and is used for further processing through XOR operation to obtain a modulation format and a signal-to-noise ratio of the signal. Compared with the traditional complex network, the method has the advantages that the scale of the network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be suitable for a real-time system.
In one embodiment of the present invention, referring to FIG. 4, the signal recognition neural network may be trained as follows:
s401: and acquiring a preset neural network model and a preset training set.
The preset neural network model may be structured as shown in fig. 3 and the related description. The preset training set may include a sample constellation corresponding to a plurality of sample signals and a modulation format label and a signal-to-noise ratio label of each sample signal.
In the embodiment of the invention, the modulation format label and the signal-to-noise ratio label can be expressed in a vector form.
As an example, if the modulation formats include 5 { QPSK, 8QAM, 16QAM, 32QAM, 64QAM }, the modulation format labels may be represented by vectors of length 5. For example, if the modulation format label and the signal-to-noise ratio label are represented by vectors containing +1 and-1, then the label of the modulation format QPSK can be represented as (+1, -1, -1, -1, -1); the label of modulation format 8QAM may be denoted as (-1, +1, -1, -1, -1). Similarly, if the osnr includes 6 {10db, 12db, 14db, 16db, 18db, 20db }, the osnr tag may be represented by a length 6 vector, and the snr 10db tag may be (+1, -1, -1, -1, -1).
S402: and inputting the sample constellation diagram into an XOR neural network model to obtain the identification result of the modulation format and the signal-to-noise ratio of the sample signal.
In the training stage, the sample constellation diagram can be input into the XOR neural network model in batches, and after the XOR neural network model is operated, the modulation format of the sample signal corresponding to the sample constellation diagram and the recognition result of the signal-to-noise ratio are output.
In the embodiment of the present invention, the recognition result of the modulation format and the signal-to-noise ratio may also be represented by a vector. In the above example, the modulation format includes { QPSK, 8QAM, 16QAM, 32QAM, 64QAM }, and if the modulation format identification result is (-1, +1, -1, -1, -1), it indicates that the modulation format is identified as 8 QAM. The optical signal to noise ratio comprises {10db, 12db, 14db, 16db, 18db, 20db }, and if the signal to noise ratio identification result is (+1, -1, -1, -1, -1, -1), the signal to noise ratio identification is 10 db.
S403: and determining a loss value based on the identification result of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal.
When training is started, the recognition result of the xor neural network may be different from the real modulation format and the signal-to-noise ratio label, and the loss value of the current iteration may be determined based on the recognition result and the real modulation format and the signal-to-noise ratio label.
Because the identification result, the real modulation format and the signal-to-noise ratio label are represented by vectors, the loss value can be calculated according to the vectors of the identification result and the vectors of the real labels.
In the embodiment of the present invention, the loss value is obtained by using, but not limited to, a cross entropy formula, a Mean Squared Error (MSE) formula, and the like as the loss function.
S404: judging whether the XOR neural network model converges based on the loss value; if not, executing S405; if so, S406 is executed.
Specifically, a loss threshold may be preset, and if the loss value is smaller than the loss threshold, the xor neural network model is considered to have converged, otherwise, the xor neural network model is not converged. The loss threshold may be set according to actual requirements, which is not limited.
S405: and adjusting the parameter values in the XOR neural network model, and returning to the step S402.
If not, the parameter values in the xor neural network model may be adjusted, and the process returns to step S402, i.e., the next iteration is performed. Specifically, the parameter values in the neural network model may be adjusted according to a maximum gradient descent method or the like.
S406: and determining the current XOR neural network model as the signal recognition neural network.
And if the convergence is achieved, the training of the signal recognition neural network is completed.
The performance of the full-precision neural network and the XOR neural network is analyzed through simulation statistical results.
Referring to table 1, table 1 lists the accuracy of the modulation format recognition result for respectively recognizing the optical signal using the full-precision neural network and the xor neural network and the accuracy of the signal-to-noise ratio for respectively recognizing the optical signals of different modulation formats. As shown in table 1, the recognition accuracy rates of the two modulation formats are both 100%, and the recognition accuracy rates of the signal-to-noise ratios of the signals of different modulation formats are substantially the same. For example, for signal optical signal to noise ratio identification of 8QAM, the accuracy is reduced from 99.6% to 99.5%, the accuracy is reduced by 0.1%, and the accuracy can be ignored.
However, the size of the XOR neural network is only 102KB, much smaller than the size of the full-precision neural network 1.7 MB.
TABLE 1
Therefore, in the embodiment of the invention, the XOR neural network is adopted to identify the modulation format and the signal-to-noise ratio of the optical signal, and compared with the full-precision complex neural network, the model is compressed by more than 90% under the condition of not changing the identification accuracy. The method has the advantages of remarkably reducing the scale of the network model, reducing the calculation resources occupied by the network, improving the operation speed and being suitable for a real-time system.
Based on the same inventive concept, according to the above embodiment of the method for identifying a modulation format and a signal-to-noise ratio based on an xor neural network, the embodiment of the present invention further provides a device for identifying a modulation format and a signal-to-noise ratio based on an xor neural network, and referring to fig. 5, the device may include the following modules:
an obtaining module 501, configured to obtain a signal to be identified after preprocessing a digital domain signal received by a receiving end;
a generating module 502, configured to generate a constellation map according to a signal to be identified;
the identification module 503 is configured to input the constellation diagram into a signal recognition neural network that is trained in advance, to obtain a modulation format and a signal-to-noise ratio of the signal to be recognized, where the signal recognition neural network is an exclusive or neural network, the signal recognition neural network is trained in advance according to a training set, and the training set includes a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal-to-noise ratio label of each sample signal.
In one embodiment of the invention, the signal recognition neural network comprises a first sub-network and a second sub-network, the first sub-network is a full-precision convolutional neural network, the second sub-network is a binary exclusive-or convolutional neural network, and the output of the first sub-network is the input of the second sub-network.
The first sub-network comprises a full-precision convolution layer, a first batch of normalization layers, an activation function layer and a first maximum pooling layer; the second sub-network comprises a second batch of normalization layers, a binary activation function layer, a binary exclusive-or convolutional layer, a second maximum pooling layer, a binary full-link layer and an output layer.
In one embodiment of the invention, the number of convolution kernels for a full-precision convolution layer is 20; the number of convolution kernels of the binary XOR convolution layer is 500, and the number of neurons of the binary full-link layer is 500.
In an embodiment of the present invention, the apparatus further includes a training module, the training module is configured to train a signal recognition neural network, and the training module is specifically configured to:
acquiring a preset XOR neural network model and a training set;
inputting the sample constellation diagram into an XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
determining a loss value based on the identification results of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal;
judging whether the XOR neural network model converges based on the loss value;
if not, adjusting parameter values in the XOR neural network model, and returning to the step of inputting the sample constellation diagram into the XOR neural network model to obtain the modulation format of the sample signal and the recognition result of the signal-to-noise ratio;
and if so, determining the current XOR neural network model as the signal recognition neural network.
By applying the modulation format and signal-to-noise ratio identification device based on the XOR neural network provided by the embodiment of the invention, a signal to be identified after a digital domain signal received by a receiving end is preprocessed can be obtained, a constellation diagram is generated according to the signal to be identified, the constellation diagram is input into a signal identification neural network which is trained in advance, the modulation format and the signal-to-noise ratio of the signal to be identified are obtained, the signal identification neural network is the XOR neural network, the signal identification neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal-to-noise ratio label of each sample signal. Therefore, the modulation format and the signal-to-noise ratio of the signal in the optical fiber communication are identified by adopting the pre-trained XOR neural network, the scale of a network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be applied to a real-time system.
Based on the same inventive concept, according to the above embodiments of the modulation format and the snr identification method based on the xor neural network, the embodiments of the present invention further provide an electronic device, as shown in fig. 6, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604, wherein the processor 601, the communication interface 602, and the memory 603 complete mutual communication through the communication bus 604,
a memory 603 for storing a computer program;
the processor 601 is configured to implement the following steps when executing the program stored in the memory 603:
acquiring a signal to be identified after preprocessing a digital domain signal received by a receiving terminal;
generating a constellation diagram according to the signal to be identified;
the constellation diagram is input into a signal recognition neural network which is trained in advance, the modulation format and the signal to noise ratio of the signal to be recognized are obtained, the signal recognition neural network is an exclusive OR neural network, the signal recognition neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
The electronic equipment provided by the embodiment of the invention can acquire the signal to be identified after the digital domain signal received by the receiving end is preprocessed, a constellation diagram is generated according to the signal to be identified, the constellation diagram is input into the signal identification neural network which is trained in advance, the modulation format and the signal to noise ratio of the signal to be identified are obtained, the signal identification neural network is an exclusive OR neural network, the signal identification neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal. Therefore, the modulation format and the signal-to-noise ratio of the signal in the optical fiber communication are identified by adopting the pre-trained XOR neural network, the scale of a network model is obviously reduced, the calculation resources occupied by the network are reduced, the operation speed is improved, and the method can be applied to a real-time system.
In another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above methods for identifying modulation format and signal-to-noise ratio based on an xor neural network.
In another embodiment of the present invention, there is also provided a computer program product containing instructions, which when run on a computer, causes the computer to execute any one of the above-mentioned methods for identifying modulation format and signal-to-noise ratio based on an xor neural network.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The procedures or functions according to the embodiments of the invention are brought about in whole or in part when the computer program instructions are loaded and executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wirelessly (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the modulation format and snr identification apparatus, the electronic device, the computer readable storage medium and the computer program product based on the xor neural network, since they are substantially similar to the embodiments of the modulation format and snr identification method based on the xor neural network, the description is simple, and the relevant points can be found in the partial description of the embodiments of the modulation format and snr identification method based on the xor neural network.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A modulation format and signal-to-noise ratio identification method based on an XOR neural network is characterized by comprising the following steps:
acquiring a signal to be identified after preprocessing a digital domain signal received by a receiving terminal;
generating a constellation diagram according to the signal to be identified;
inputting the constellation diagram into a signal recognition neural network which is trained in advance to obtain a modulation format and a signal to noise ratio of the signal to be recognized, wherein the signal recognition neural network is an exclusive OR neural network, the signal recognition neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
2. The method of claim 1, wherein the signal-identifying neural network comprises a first sub-network and a second sub-network, wherein the first sub-network is a full-precision convolutional neural network, the second sub-network is a binary exclusive-or convolutional neural network, and wherein an output of the first sub-network is an input of the second sub-network.
3. The method of claim 2, wherein the first sub-network comprises a full-precision convolutional layer, a first normalization layer, an activation function layer, and a first max-pooling layer; the second sub-network comprises a second batch of normalization layers, a binary activation function layer, a binary exclusive-or convolutional layer, a second maximum pooling layer, a binary full-link layer and an output layer.
4. The method of claim 3, wherein the number of convolution kernels of the full-precision convolution layer is 20; the number of convolution kernels of the binary exclusive-or convolution layer is 500, and the number of neurons of the binary full-link layer is 500.
5. The method of claim 1, wherein the signal-recognizing neural network is trained as follows:
acquiring a preset XOR neural network model and the training set;
inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
determining a loss value based on the identification results of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal;
determining whether the XOR neural network model converges based on the loss value;
if not, adjusting parameter values in the XOR neural network model, and returning to the step of inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
and if so, determining the current XOR neural network model as the signal recognition neural network.
6. An XOR neural network-based modulation format and signal-to-noise ratio identification device, the device comprising:
the acquisition module is used for acquiring a signal to be identified after the digital domain signal received by the receiving end is preprocessed;
the generating module is used for generating a constellation diagram according to the signal to be identified;
the identification module is used for inputting the constellation diagram into a signal identification neural network which is trained in advance to obtain a modulation format and a signal to noise ratio of a signal to be identified, the signal identification neural network is an exclusive OR neural network, the signal identification neural network is trained in advance according to a training set, and the training set comprises a sample constellation diagram corresponding to a plurality of sample signals and a modulation format label and a signal to noise ratio label of each sample signal.
7. The apparatus of claim 6, wherein the signal recognition neural network comprises a first sub-network and a second sub-network, the first sub-network is a full-precision convolutional neural network, the second sub-network is a binary exclusive-or convolutional neural network, and an output of the first sub-network is an input of the second sub-network;
the first sub-network comprises a full-precision convolution layer, a first batch of normalization layers, an activation function layer and a first maximum pooling layer; the second sub-network comprises a second batch of normalization layers, a binary activation function layer, a binary exclusive-or convolutional layer, a second maximum pooling layer, a binary full-link layer and an output layer.
8. The apparatus according to claim 6, further comprising a training module, the training module being configured to train the signal recognition neural network, the training module being specifically configured to:
acquiring a preset XOR neural network model and the training set;
inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
determining a loss value based on the identification results of the modulation format and the signal-to-noise ratio of the sample signal and the real modulation format label and the signal-to-noise ratio label of the sample signal;
determining whether the XOR neural network model converges based on the loss value;
if not, adjusting parameter values in the XOR neural network model, and returning to the step of inputting the sample constellation diagram into the XOR neural network model to obtain the identification results of the modulation format and the signal-to-noise ratio of the sample signal;
and if so, determining the current XOR neural network model as the signal recognition neural network.
9. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any one of claims 1 to 5 when executing a program stored in the memory.
10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of the claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010396238.6A CN111614398B (en) | 2020-05-12 | 2020-05-12 | Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010396238.6A CN111614398B (en) | 2020-05-12 | 2020-05-12 | Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111614398A true CN111614398A (en) | 2020-09-01 |
| CN111614398B CN111614398B (en) | 2021-06-11 |
Family
ID=72204447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010396238.6A Expired - Fee Related CN111614398B (en) | 2020-05-12 | 2020-05-12 | Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111614398B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112187266A (en) * | 2020-09-29 | 2021-01-05 | 中国电子科技集团公司第三十六研究所 | Nonlinear correction method and device for analog-to-digital converter and electronic equipment |
| CN112270263A (en) * | 2020-10-28 | 2021-01-26 | 电子科技大学 | Modulation identification method based on differential density constellation diagram |
| CN112953871A (en) * | 2021-02-23 | 2021-06-11 | 北京邮电大学 | New signal modulation format identification method based on neural network |
| CN113052081A (en) * | 2021-03-26 | 2021-06-29 | 江苏师范大学 | Modulation mode identification method based on Cauchy Score constellation diagram |
| CN113285762A (en) * | 2021-02-25 | 2021-08-20 | 广西师范大学 | Modulation format identification method based on relative entropy calculation |
| CN113626960A (en) * | 2021-07-29 | 2021-11-09 | 南京信息工程大学 | Anti-mode coupling few-mode signal complex format analysis method and device |
| CN113919401A (en) * | 2021-11-05 | 2022-01-11 | 深圳市唯特视科技有限公司 | Modulation type identification method and device based on constellation diagram characteristics and computer equipment |
| CN114553650A (en) * | 2022-04-27 | 2022-05-27 | 南京信息工程大学 | Multi-level neural network-based anti-mode coupling signal complex format analysis method |
| WO2022151069A1 (en) * | 2021-01-13 | 2022-07-21 | Oppo广东移动通信有限公司 | Method and apparatus for processing received information, computer device, and storage medium |
| CN115173936A (en) * | 2022-06-30 | 2022-10-11 | 烽火通信科技股份有限公司 | Optical module identification marking method and device |
| CN115276789A (en) * | 2022-05-16 | 2022-11-01 | 聊城大学 | Optical performance monitoring method, device, equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6601049B1 (en) * | 1996-05-02 | 2003-07-29 | David L. Cooper | Self-adjusting multi-layer neural network architectures and methods therefor |
| US8468244B2 (en) * | 2007-01-05 | 2013-06-18 | Digital Doors, Inc. | Digital information infrastructure and method for security designated data and with granular data stores |
| CN108234370A (en) * | 2017-12-22 | 2018-06-29 | 西安电子科技大学 | Modulation mode of communication signal recognition methods based on convolutional neural networks |
| US10051475B2 (en) * | 2015-09-28 | 2018-08-14 | Department 13, Inc. | Unmanned aerial vehicle intrusion detection and countermeasures |
| CN109785828A (en) * | 2017-11-13 | 2019-05-21 | 通用汽车环球科技运作有限责任公司 | Spatial term based on user speech style |
| CN110324080A (en) * | 2019-06-28 | 2019-10-11 | 北京邮电大学 | A kind of method, apparatus of optical information networks, electronic equipment and medium |
| CN111082919A (en) * | 2019-12-18 | 2020-04-28 | 南京信息工程大学 | Dynamic probability 16cap mapping encryption method |
| CN111126579A (en) * | 2019-11-05 | 2020-05-08 | 复旦大学 | Memory computing device suitable for binary convolution neural network computing |
-
2020
- 2020-05-12 CN CN202010396238.6A patent/CN111614398B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6601049B1 (en) * | 1996-05-02 | 2003-07-29 | David L. Cooper | Self-adjusting multi-layer neural network architectures and methods therefor |
| US8468244B2 (en) * | 2007-01-05 | 2013-06-18 | Digital Doors, Inc. | Digital information infrastructure and method for security designated data and with granular data stores |
| US10051475B2 (en) * | 2015-09-28 | 2018-08-14 | Department 13, Inc. | Unmanned aerial vehicle intrusion detection and countermeasures |
| CN109785828A (en) * | 2017-11-13 | 2019-05-21 | 通用汽车环球科技运作有限责任公司 | Spatial term based on user speech style |
| CN108234370A (en) * | 2017-12-22 | 2018-06-29 | 西安电子科技大学 | Modulation mode of communication signal recognition methods based on convolutional neural networks |
| CN110324080A (en) * | 2019-06-28 | 2019-10-11 | 北京邮电大学 | A kind of method, apparatus of optical information networks, electronic equipment and medium |
| CN111126579A (en) * | 2019-11-05 | 2020-05-08 | 复旦大学 | Memory computing device suitable for binary convolution neural network computing |
| CN111082919A (en) * | 2019-12-18 | 2020-04-28 | 南京信息工程大学 | Dynamic probability 16cap mapping encryption method |
Non-Patent Citations (2)
| Title |
|---|
| FAQUAN YANG: "Method of neural network modulation recognition based on clustering and Polak-Ribiere algorithm", 《JOURNAL OF SYSTEMS ENGINEERING AND ELECTRONICS 》 * |
| 马晓丽: "运用遗传规划解决异或问题的新方法", 《计算机应用研究》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112187266B (en) * | 2020-09-29 | 2024-03-08 | 中国电子科技集团公司第三十六研究所 | Nonlinear correction method and device of analog-to-digital converter and electronic equipment |
| CN112187266A (en) * | 2020-09-29 | 2021-01-05 | 中国电子科技集团公司第三十六研究所 | Nonlinear correction method and device for analog-to-digital converter and electronic equipment |
| CN112270263A (en) * | 2020-10-28 | 2021-01-26 | 电子科技大学 | Modulation identification method based on differential density constellation diagram |
| CN112270263B (en) * | 2020-10-28 | 2022-03-25 | 电子科技大学 | Modulation identification method based on differential density constellation diagram |
| WO2022151069A1 (en) * | 2021-01-13 | 2022-07-21 | Oppo广东移动通信有限公司 | Method and apparatus for processing received information, computer device, and storage medium |
| CN112953871A (en) * | 2021-02-23 | 2021-06-11 | 北京邮电大学 | New signal modulation format identification method based on neural network |
| CN113285762A (en) * | 2021-02-25 | 2021-08-20 | 广西师范大学 | Modulation format identification method based on relative entropy calculation |
| CN113285762B (en) * | 2021-02-25 | 2022-08-05 | 广西师范大学 | Modulation format identification method based on relative entropy calculation |
| CN113052081A (en) * | 2021-03-26 | 2021-06-29 | 江苏师范大学 | Modulation mode identification method based on Cauchy Score constellation diagram |
| CN113052081B (en) * | 2021-03-26 | 2024-03-15 | 江苏师范大学 | Modulation mode identification method based on Cauchy Score constellation diagram |
| CN113626960A (en) * | 2021-07-29 | 2021-11-09 | 南京信息工程大学 | Anti-mode coupling few-mode signal complex format analysis method and device |
| CN113919401A (en) * | 2021-11-05 | 2022-01-11 | 深圳市唯特视科技有限公司 | Modulation type identification method and device based on constellation diagram characteristics and computer equipment |
| CN114553650A (en) * | 2022-04-27 | 2022-05-27 | 南京信息工程大学 | Multi-level neural network-based anti-mode coupling signal complex format analysis method |
| CN114553650B (en) * | 2022-04-27 | 2022-07-29 | 南京信息工程大学 | Multi-level neural network-based anti-mode coupling signal complex format analysis method |
| CN115276789A (en) * | 2022-05-16 | 2022-11-01 | 聊城大学 | Optical performance monitoring method, device, equipment and storage medium |
| CN115276789B (en) * | 2022-05-16 | 2024-06-11 | 聊城大学 | Optical performance monitoring method, device, equipment and storage medium |
| CN115173936A (en) * | 2022-06-30 | 2022-10-11 | 烽火通信科技股份有限公司 | Optical module identification marking method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111614398B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111614398B (en) | Method and device for identifying modulation format and signal-to-noise ratio based on XOR neural network | |
| CN110798417B (en) | Signal modulation identification method and device based on cyclic residual error network | |
| CN107124381B (en) | Automatic identification method for digital communication signal modulation mode | |
| US11962344B2 (en) | Method and apparatus for obtaining transmitter test parameter, and storage medium | |
| CN109361471B (en) | Optical communication signal modulation format identification method based on amplitude deviation analysis | |
| CN111510408B (en) | Signal modulation mode identification method and device, electronic equipment and storage medium | |
| WO2023123688A1 (en) | Signal transmission method, signal transmission apparatus, and electronic device | |
| Alharbi et al. | Automatic modulation classification of digital modulations in presence of HF noise | |
| CN113141325A (en) | Training method, identification method and device for optical OFDM signal subcarrier modulation format identification model | |
| WO2021143126A1 (en) | Modulation mode determining method and apparatus | |
| CN113869227B (en) | Signal modulation mode identification method, device, equipment and readable storage medium | |
| CN113033457B (en) | Soft fault recognition model training and soft fault recognition method and device | |
| CN113726711A (en) | OFDM receiving method and device, and channel estimation model training method and device | |
| CN111541482A (en) | Digital signal preprocessing method and device in direct detection optical communication system | |
| CN116760491A (en) | Signal-to-noise ratio estimation method based on deep learning | |
| CN114006798B (en) | Signal processing method and device, electronic equipment and storage medium | |
| CN116506273A (en) | Novel MPSK modulation signal identification and classification method | |
| CN113591787B (en) | Method, device, equipment and storage medium for identifying optical fiber link component | |
| CN113518049B (en) | Modulation identification method based on fractional low-order polar coordinates and deep learning | |
| WO2020215911A1 (en) | Method and apparatus for classification of mixed signals, and electronic device | |
| CN117544272A (en) | Signal type and azimuth joint reasoning method | |
| CN114598581A (en) | Training method, recognition method and device for two-stage detection model of probability shaping signal | |
| CN111988252A (en) | Signal modulation mode identification method based on deep learning | |
| CN114936580A (en) | Modulation type identification method and related device, electronic equipment and storage medium | |
| CN114285701B (en) | Method, system, equipment and terminal for identifying transmitting power of main user |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210611 |