CN114978381B - Compressed sensing processing method and device for broadband signal based on deep learning - Google Patents
Compressed sensing processing method and device for broadband signal based on deep learning Download PDFInfo
- Publication number
- CN114978381B CN114978381B CN202210450842.1A CN202210450842A CN114978381B CN 114978381 B CN114978381 B CN 114978381B CN 202210450842 A CN202210450842 A CN 202210450842A CN 114978381 B CN114978381 B CN 114978381B
- Authority
- CN
- China
- Prior art keywords
- signal
- sampling
- admm
- net
- frequency domain
- 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.)
- Active
Links
- 238000013135 deep learning Methods 0.000 title claims abstract description 60
- 238000003672 processing method Methods 0.000 title claims abstract description 25
- 238000005070 sampling Methods 0.000 claims abstract description 205
- 238000012549 training Methods 0.000 claims abstract description 72
- 238000012360 testing method Methods 0.000 claims abstract description 62
- 230000006870 function Effects 0.000 claims abstract description 39
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 238000005457 optimization Methods 0.000 claims abstract description 23
- 238000003062 neural network model Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 20
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000002939 conjugate gradient method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 description 16
- 238000001228 spectrum Methods 0.000 description 12
- 230000005540 biological transmission Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000006835 compression Effects 0.000 description 6
- 238000007906 compression Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000002360 explosive Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/382—Monitoring; Testing of propagation channels for resource allocation, admission control or handover
-
- 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
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
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)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Character Discrimination (AREA)
Abstract
The invention discloses a compressed sensing processing method and device of broadband signals based on deep learning, comprising the following steps: performing multiple set sampling on an input broadband signal in a time domain to obtain a discrete sampling sequence; preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set; designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net; in the model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters; and inputting the signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal. The ADMM-net combined with the deep learning in the invention has higher signal reconstruction accuracy under the conditions of lower sampling rate and lower signal-to-noise ratio after being trained by a network model, and has certain generalization capability.
Description
Technical Field
The invention relates to the technical field of broadband signal processing, in particular to a compressed sensing processing method and device of broadband signals based on deep learning, an intelligent terminal and a storage medium.
Background
With the explosive growth of electromagnetic devices and information systems, static spectrum management modes are not effective any more, a large number of frequency bands are low in utilization rate, unallocated frequency bands suitable for data transmission remain, and the contradiction that wireless spectrum resources are locally tense and overall idle is more obvious. Meanwhile, with the rapid development of mobile services and the coming of the age of internet of things, the technical route and key technology adopted by the future cellular network need at least hundreds of megahertz of transmission bandwidth, hundreds of transmission antennas, and ultra-densely deployed base stations and support massive users.
Compressed sensing (Compressed Sensing, CS) theory provides a new solution for acquisition and processing of broadband signals. In the prior art: currently, the mainstream compressed Sampling framework includes multiple-set Sampling (MCS), random demodulation Sampling (Random Demodulator Sampling, RDS), and modulation wideband converter (Modulated Wideband Converter, MWC). The multiple set sampling technology is an undersnyquist sampling technology of periodic non-uniform sampling, and compression sampling of broadband signals can be achieved through a plurality of low-speed ADCs with the same sampling rate and different sampling starting moments. In the wideband spectrum sensing scheme based on multiple set sampling in the prior art, the undersnyquist sampling of wideband signals is realized through a compressed sampling mode of a low-rate multichannel architecture, and then the occupied channel position is estimated by recovering multi-band signals through a greedy algorithm, so that the reconstruction of the input wideband signals is realized. In the scheme, when the sampling rate and the signal-to-noise ratio of the input broadband signal are low, the reconstruction accuracy of the signal is also greatly improved.
Namely, the compressed sensing processing of the broadband signal in the prior art has the defects in the aspects of reconstruction accuracy, robustness and the like, and has the problems of poor reconstruction performance and the like for broadband spectrum sensing based on undersnyquist sampling.
Accordingly, there is a need for improvement and advancement in the art.
Disclosure of Invention
The invention aims to solve the technical problems of the prior art and provides a compressed sensing processing method, a device, an intelligent terminal and a storage medium for broadband signals based on deep learning.
In order to solve the above technical problems, a first aspect of the present invention provides a compressed sensing processing method of a wideband signal based on deep learning, where the method includes:
acquiring an input broadband signal, and performing multiple set sampling on the input broadband signal in a time domain to obtain a discrete sampling sequence;
preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set, wherein the data set comprises: signal data of the training set and signal data of the test set;
designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net;
In the model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters;
and in the test stage or the application stage, inputting the signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal.
The compressed sensing processing method of the broadband signal based on the deep learning, wherein the step of obtaining the input broadband signal and performing multiple set sampling on the input broadband signal in the time domain to obtain a discrete sampling sequence comprises the following steps:
acquiring an input broadband signal x (t), and determining a sampling mode through a sequence forward selection algorithm
Sampling at t= (ml+c) using multiple sets i )T,i＝1,...,p,
Sampling at moment to obtain a discrete sampling sequence x ci [n]Where T represents the nyquist sampling time period interval of the input wideband signal, and the multiple set sampling period interval is L times as large as that of the nyquist sampling theorem, so the sampling frequency is reduced to 1/L of the nyquist sampling frequency.
The compressed sensing processing method of the broadband signal based on deep learning, wherein the steps of preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set comprise the following steps:
For the obtained discrete sampling sequence x ci [n]Performing discrete Fourier transform to obtain frequency domain representation of the sampling sequence
At the same time, the frequency domain representation of the resulting sample sequence
And the frequency domain signal X k of the corresponding input wideband signal]The data sets are composed, wherein the number of training sets and test sets is a set proportion.
The method for processing compressed sensing of the broadband signal based on the deep learning, wherein the step of designing the broadband signal reconstruction neural network model ADMM-net based on the deep learning and initializing the ADMM-net comprises the following steps:
an ADMM algorithm is used as a compressed sensing signal reconstruction algorithm, and an ADMM-net reconstructed broadband signal with expandable depth based on deep learning is designed;
parameters of the ADMM-net model are initialized to corresponding parameter values in the ADMM algorithm.
In the deep learning-based broadband signal compressed sensing processing method, in the model training stage, signal data of a training set is input into the ADMM-net, a loss function is continuously minimized through an optimization algorithm, and the step of obtaining optimal neural network parameters comprises the following steps:
in the model training phase, the frequency domain representation of the sampling sequence of the training set is obtained
Inputting the initialized ADMM-net to obtain the initially reconstructed frequency domain signal X ir [k]；
Constructing the initially reconstructed frequency domain signal X ir [k]Frequency domain signal X k with input wideband signal]A loss function E (θ) therebetween, where θ is a related neural network parameter;
and continuously minimizing the loss function E (theta) by using an optimization algorithm to obtain the optimal neural network parameter theta, thereby obtaining the trained ADMM-net model.
In the deep learning-based broadband signal compressed sensing processing method, in a testing stage or an application stage, the step of inputting signal data of a testing set into a trained ADMM-net model to obtain a reconstructed broadband signal comprises the following steps:
in the test phase or application phase, the frequency domain representation of the sampling sequence of the test set is obtained
Inputting the trained ADMM-net model for processing to obtain a reconstructed frequency domain signal X r [k]。
A compressed sensing processing apparatus for deep learning based wideband signals, wherein the apparatus comprises:
the multi-set sampling processing module is used for acquiring an input broadband signal, and performing multi-set sampling on the input broadband signal in the time domain to obtain a discrete sampling sequence;
the preprocessing module is used for preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set;
Setting and initializing a module, designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net;
the neural network training module is used for inputting signal data of a training set into the ADMM-net in a model training stage, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters;
and the test and application module is used for inputting the signal data of the test set into the trained ADMM-net model in the test stage or the application stage to obtain a reconstructed broadband signal.
A smart terminal comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs comprising means for performing the method of one of the claims.
A non-transitory computer readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform any one of the methods.
The beneficial effects are that: compared with the prior art, the invention provides a compressed sensing processing method of broadband signals based on deep learning. Firstly, based on multiple set sampling, compressing and sampling an input broadband signal in a time domain to obtain a discrete sampling sequence, preprocessing the obtained discrete sampling sequence and the input broadband signal to construct a data set, and simultaneously, combining deep learning, designing an expandable depth broadband signal reconstruction neural network model ADMM-net to reconstruct the input broadband signal. In the model training stage, the signal data of the training set is input into the initialized ADMM-net to obtain an initial reconstructed frequency domain signal, a loss function between the initial reconstructed frequency domain signal and the frequency domain signal of the input broadband signal is reconstructed, and the loss function is continuously minimized by utilizing an optimization algorithm, so that the optimal neural network parameters are obtained. And in the test stage or the application stage, inputting the signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal. Compared with a common reconstruction algorithm, the numerical analysis of the experimental result shows that the reconstruction accuracy of the ADMM-net combined with the deep learning under the conditions of lower sampling rate and lower signal to noise ratio is higher after the network model is trained, and the ADMM-net has certain generalization capability.
Drawings
Fig. 1 is a diagram of a compressed sensing theory framework.
Fig. 2 is a flowchart of a specific implementation of a compressed sensing processing method for a wideband signal based on deep learning according to an embodiment of the present invention.
Fig. 3 is a diagram of a multiple set sampling framework of a compressed sensing processing method of a wideband signal based on deep learning according to an embodiment of the present invention.
Fig. 4 is a schematic diagram illustrating spectral division transfer of an input wideband signal according to a compressed sensing processing method of a wideband signal based on deep learning according to an embodiment of the present invention.
Fig. 5 is a data flow diagram of ADMM-net of a compressed sensing processing method of a broadband signal based on deep learning according to an embodiment of the present invention.
Fig. 6 is a graph of the reconstruction results of three algorithms, namely SOMP, ADMM and ADMM-net, of the compressed sensing processing method of broadband signals based on deep learning according to an embodiment of the present invention, where the channel number l=64, the source channel k=5, and different signal-to-noise ratios.
Fig. 7 is a schematic block diagram of a compressed sensing processing device for a wideband signal based on deep learning according to an embodiment of the present invention.
Fig. 8 is a schematic block diagram of an internal structure of an intelligent terminal according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and more specific, the present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
With the explosive growth of electromagnetic devices and information systems, static spectrum management modes are not effective any more, a large number of frequency bands are low in utilization rate, unallocated frequency bands suitable for data transmission remain, and the contradiction that wireless spectrum resources are locally tense and overall idle is more obvious. Meanwhile, with the rapid development of mobile services and the coming of the age of internet of things, the technical route and key technology adopted by the future cellular network need at least hundreds of megahertz of transmission bandwidth, hundreds of transmission antennas, and ultra-densely deployed base stations and support massive users. The lowest undistorted sampling rate of the signal must be greater than or equal to twice the maximum bandwidth of the signal, according to the nyquist sampling theorem. As the bandwidth of signals increases, it becomes increasingly difficult to achieve high-speed sampling of broadband signals and storage, transmission and real-time processing of large volumes of data, which presents a significant challenge for conventional communication systems.
Compressed sensing (Compressed Sensing, CS) theory provides a new solution for acquisition and processing of broadband signals. Compared with a traditional Nyquist sampling system which is used for sampling at a high rate and then compressing data, compressed sensing directly performs compressed sampling on signals with sparsity or compressibility at a rate lower than the Nyquist sampling rate, and then a reconstruction algorithm is combined to reconstruct original signals, wherein a theoretical framework is shown in fig. 1, and the original signals are subjected to sparse transformation, compressed sampling and signal reconstruction. Compressed sensing obviously reduces the requirements of signal sampling on hardware equipment such as storage, transmission, analysis and processing and the like in a communication system, thereby bringing new opportunities for solving the challenges of high cost, low efficiency, information redundancy and the like which are difficult to break through by the traditional sampling method.
In wideband spectrum sensing based on undersnyquist sampling, the main roles are the selection of the compressed sampling framework and the design of sparse signal reconstruction algorithms. The compressive sampling framework determines the manner of compressive sampling of the broadband signal, and is generally related to the design of a random observation matrix, so that the subsequent signal reconstruction process is affected. The reconstruction algorithm has the greatest influence on the whole broadband spectrum sensing, and the reconstruction algorithm is related to the reconstruction complexity and accuracy of the signals and determines the whole broadband spectrum sensing performance. Therefore, the research design of a combination of a stable and low-reconstruction-complexity compressed sampling frame and a reconstruction algorithm is very important for improving the broadband spectrum sensing performance.
In the prior art: the currently prevailing compressed sampling framework has multiple set sampling (Multi-CosetSampling, MCS), random demodulation sampling (Random Demodulator Sampling, RDS), modulated wideband converter (Modulated Wideband Converter, MWC). The multiple set sampling technology is an undersnyquist sampling technology of non-uniform period sampling, compression sampling of broadband signals can be achieved through a plurality of low-speed ADCs with the same sampling rate and different sampling starting moments, and compared with other time-based ADC interleaving sampling schemes, the multiple set sampling technology only needs to select p < L channels in L channels for sampling, the sampling matrix dimension is low, the sampling rate of a compression sampling system is reduced, a front-end circuit is simple, the cost of hardware implementation is lower, and the main difficulty is the accurate design of sampling time delay. The reconstruction algorithm essentially reconstructs the original signal by solving an underdetermined linear equation, and the conventional reconstruction algorithm comprises a greedy algorithm, such as a synchronization orthogonal matching pursuit algorithm trace (Simultaneous Orthogonal Matching Pursuit, SOMP) and the like; convex relaxation algorithms, such as the alternate direction multiplier algorithm (Alternating Direction Method of Multipliers, ADMM), etc. The greedy algorithm has lower computational complexity, but has poorer reconstruction performance when the observed value is smaller, and the operand is gradually increased along with the increase of the observed value. While the convex relaxation algorithm can utilize a small number of samples to achieve approximate reconstruction of the input wideband signal, its reconstruction is more complex. In the prior art, a wideband spectrum sensing scheme based on multiple set sampling is provided, the underscore sampling of wideband signals is realized through a compressed sampling mode of a low-rate multichannel architecture, and then the occupied channel position is estimated by recovering multiband signals through a greedy algorithm, so that the reconstruction of input wideband signals is realized. In the scheme, when the sampling rate and the signal-to-noise ratio of the input signal are low, the reconstruction accuracy of the signal is also greatly improved.
In order to solve the problems in the prior art, the embodiment provides a compressed sensing processing method of a broadband signal based on deep learning. Firstly, based on multiple set sampling, compressing and sampling an input broadband signal in a time domain to obtain a discrete sampling sequence, preprocessing the obtained discrete sampling sequence and the input broadband signal to construct a data set, and simultaneously, combining deep learning, designing an expandable depth broadband signal reconstruction neural network model ADMM-net to reconstruct the input broadband signal. In the model training stage, the signal data of the training set is input into the initialized ADMM-net to obtain an initial reconstructed frequency domain signal, a loss function between the initial reconstructed frequency domain signal and the frequency domain signal of the input broadband signal is reconstructed, and an optimization algorithm is utilized to minimize the loss function, so that the optimal neural network parameters are obtained. And in the test stage or the application stage, inputting the signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal.
Compared with a common reconstruction algorithm, the numerical analysis of the experimental result shows that the reconstruction accuracy of the broadband signal reconstruction neural network model ADMM-net combined with the deep learning is higher under the conditions of lower sampling rate and lower signal to noise ratio after the network model is trained, and the reconstruction accuracy has a certain generalization capability.
Exemplary method
The method of the embodiment of the invention can be applied to an intelligent terminal, and when the method is concretely implemented, as shown in fig. 2, the compressed sensing processing method of the broadband signal based on deep learning provided by the embodiment of the invention specifically comprises the following steps:
step S100, acquiring an input broadband signal, and performing multiple set sampling on the input broadband signal in a time domain to obtain a discrete sampling sequence;
when the invention is implemented, an input broadband signal is firstly obtained, a sampling mode is determined through a sequence forward selection algorithm, and multiple set sampling is carried out on a time domain to obtain a discrete sampling sequence.
Step 200, preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set, wherein the data set comprises: signal data of the training set and signal data of the test set;
in the step, the frequency domain representation of the sampling sequence is obtained through discrete Fourier transform (DFT transform) on the obtained discrete sampling sequence;
Meanwhile, the obtained frequency domain representation of the sampling sequence and the corresponding frequency domain signals of the input broadband signals form a data set, wherein the quantity of the training set and the testing set is set proportion.
Step S300, designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net;
in the step, an ADMM algorithm is used as a compressed sensing signal reconstruction algorithm, and a depth-learning-based extended depth broadband signal reconstruction neural network model ADMM-net is designed to reconstruct broadband signals;
parameters of the ADMM-net model are initialized to corresponding parameter values in the ADMM algorithm.
Step S400, in the model training stage, signal data of a training set are input into the ADMM-net, and the loss function is continuously minimized through an optimization algorithm, so that the optimal neural network parameters are obtained;
in the step, in the model training stage, the frequency domain representation of the sampling sequence of the training set is input into the initialized ADMM-net to obtain an initial reconstructed frequency domain signal;
constructing a loss function between the initially reconstructed frequency domain signal and the frequency domain signal of the input wideband signal;
and continuously minimizing a loss function by using a conjugate gradient method (Conjugate Gradient, CG) to obtain optimal neural network parameters, thereby obtaining the trained ADMM-net model.
And step S500, in a test stage or an application stage, inputting signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal.
In the step, in a testing stage or an application stage, the frequency domain representation of the sampling sequence of the testing set is input into the trained ADMM-net model for processing, and a reconstructed frequency domain signal is obtained.
The process according to the invention is described in further detail below by way of more detailed examples:
in the embodiment of the present invention, regarding the step S100 of obtaining an input wideband signal, performing multiple set sampling on the input wideband signal in the time domain to obtain a discrete sampling sequence, which is specifically implemented as follows:
the input broadband signal x (t) is acquired, and multiple set sampling is adopted to be t= (mL+c) i )T,i＝1,2,…,p,
The discrete sampling sequence is obtained by sampling at the moment, wherein T represents the Nyquist sampling time period interval of the input broadband signal, compared with the Nyquist sampling theorem, the multiple set sampling time period interval is L times of the Nyquist sampling time period interval, so that the sampling frequency is reduced to 1/L of the Nyquist sampling frequency, and FIG. 3 is a multiple set sampling frame diagram of the embodiment of the invention. Set->
Represents p selected from {0,1,2, …, L-1}, p<L different integers are used to represent the number of different integers,in the multiple set sampling referred to as the (L, p) sampling mode. Then the ith discrete sample sequence x ci [n]Can be defined as equation (1).
Wherein the method comprises the steps of
i＝1,2,…,p，
Represents the sampling delay of the i-th channel, +.>
Representing the nyquist sampling frequency.
In the embodiment of the invention, a sequence forward selection algorithm is adopted to determine the sampling mode.
Further, in the embodiment of the present invention, regarding step S200, preprocessing the obtained discrete sampling sequence and the input wideband signal and constructing a data set is specifically implemented as follows:
in the present invention, as described in step S200, the sampled sequence x ci [n]Performing discrete Fourier transform (DFT transform) to obtain frequency domain representation of the ith sample sequence
I.e.
Wherein k is c =0, 1,2, …, M-1, i=1, 2, …, p, M represents the number of sampling points obtained per sampling channel.
Set at Nyquist sampling frequency f s Uniformly sampling the input broadband signal x (T) with the time interval of T to obtain a discrete sequence x [ n ] with the total length of K]. Then pair sequence x [ n ]]Performing discrete Fourier transform to obtain inputFrequency domain signal X k of the wideband signal of (2)]I.e.
Where k=0, 1,2, …, K-1, K represents the number of points of the discrete sequence x [ n ].
Frequency domain representation of the resulting sample sequence
i=1, 2, …, p and the corresponding frequency domain signal X [ k ] of the input wideband signal]The data sets are composed, wherein the number of training sets and test sets is a set ratio (4:1). The training set is used for training the network model, and the testing set is used for testing the reconstruction performance of the network model and evaluating the generalization capability of the network model.
Frequency domain representation of a sampling sequence based on formulas (2), (3)
Frequency domain signal X k with input wideband signal]The relation between the matrix expressions of the input and output of the multiple set sampling can be as follows
Y＝AX(4)
Wherein Y is C p×M A frequency domain representation matrix representing a sample sequence, the representation of which is:
t represents the nyquist sampling time period interval of the input wideband signal; x epsilon C L×M Frequency domain representation matrix representing an input wideband signal, frequency domain signal X k of the input wideband signal]Dividing the signal into L channels in 1/LT unit, and moving L-1 channels outside the first channel to left in 1/LT unit, wherein the spectrum division transfer diagram is shown in figure 4, so as to obtain frequency domain signal X [ k ] of X and input broadband signal]Relational expressions of (2), i.e
Where L represents the number of spectrally divided channels, M represents the number of signal points per channel, and k=l·m represents the number of discrete signal points of the input wideband signal based on the nyquist sampling frequency.
Further, the present invention relates to step S300, designing a deep learning-based broadband signal reconstruction neural network model ADMM-net, and initializing the ADMM-net specifically as follows:
Specifically, the present invention reconstructs an input wideband signal using a convex relaxation method, and then its regularized unconstrained model can be expressed as:
wherein X represents the frequency domain representation matrix of the input broadband signal to be reconstructed, A is the measurement matrix of the multiple set sampling, Y represents the frequency domain representation matrix of the sampling sequence, mu is the regularization parameter, and the auxiliary variables Z= [ Z1, Z2, …, Z p ].
The extended Lagrangian expansion L of the above formula (8) P (X, Z, alpha) is:
where α represents the Lagrangian multiplier for the dual variable and ρ represents the Lagrangian multiplier coefficient.
By using the ADMM algorithm as a compressed sensing signal reconstruction algorithm, the { X, Z, alpha } can be alternately and optimally solved through three sub-problems separated by the above formula, so as to obtain:
where k represents the kth iteration, X (k) A frequency domain representation matrix representing a kth iterative reconstruction, Z (k) Auxiliary variable Z, M representing the kth iteration update (k) The dual variables α, ρ, η and representing the kth iteration update
Representing the relevant parameters of the reconstruction signal operation, the auxiliary variable updating operation and the dual variable updating operation respectively.
By combining the ideas of deep learning and data flow diagram, the iterative solution of the three sub-problems of the ADMM algorithm can be designed into a data flow diagram, so that the iterative solution is expanded into an alternate direction multiplication sub-network with any layer number, the data flow diagram is shown in fig. 5, and fig. 5 is the data flow diagram of the ADMM-net model in the embodiment of the invention. The dataflow graph consists of nodes in the ADMM algorithm corresponding to different operations, the directed edges corresponding to the data flow between the operations. In the kth stage of the graph, the three operation types in the ADMM algorithm map three types of nodes, namely defined as signal reconstruction layers (X (k) ) Nonlinear conversion layer (Z) (k) ) And a multiplication update layer (M (k) ). The entire dataflow graph is a multiple repetition of the above phases, corresponding to successive iterations in the ADMM algorithm.
Parameters for layers in ADMM-net model
Which is initialized to the corresponding parameter values in the ADMM algorithm.
In a further embodiment, in the step S400 of the present invention, in a model training stage, signal data of a training set is input into the ADMM-net, and the loss function is continuously minimized through an optimization algorithm, so that the optimal neural network parameters are specifically obtained as follows:
in the model training phase, the frequency domain representation of the sampling sequence of the training set is obtained
Inputting the initialized ADMM-net to obtain the initially reconstructed frequency domain signal X ir [k]。
The embodiment of the invention constructs the initially reconstructed frequency domain signal X of the training set ir [k]And the frequency domain signal X [ k ] of the input broadband signal]Is defined as:
wherein the method comprises the steps of
An initial reconstructed frequency domain signal representing an ith signal data of the training set N signal data, (Y, θ) represents a frequency domain representation matrix Y of the sampling sequence and a network parameter θ, respectively, where θ comprises a signal reconstruction layer (X (k) ) Nonlinear conversion layer (Z) (k) ) And a multiplication update layer (M (k) ) Parameter of->
X (i) A frequency domain signal representing the input wideband signal of the training set ith signal data.
In model training, the optimization algorithm of the network parameters is a conjugate gradient method (Conjugate Gradient, CG), and the gradient descending direction is determined by an Armijo criterion of an inaccurate one-dimensional search method. And continuously minimizing a loss function E (theta) through a conjugate gradient method to obtain an optimal neural network parameter theta, thereby obtaining the trained ADMM-net model.
Further, in step S500 of the embodiment of the present invention, in a testing stage or an application stage, signal data of a testing set is input into a trained ADMM-net model, and a reconstructed wideband signal is obtained specifically as follows:
in the embodiment of the invention, the reconstruction effect of the broadband signal under different signal-to-noise ratios is tested by using the trained ADMM-net model in the test stage or the application stage. Frequency domain representation of a sample sequence of a test set
Inputting the trained ADMM-net model for processing to obtain a reconstructed frequency domain signal X r [k]。
In the test stage of the invention, the MSE represents the final reconstruction error of the frequency domain signal:
wherein X is r [k]Frequency domain signal, X k, representing a reconstructed wideband signal]A frequency domain signal representing the input wideband signal.
From the above, the compressed sensing processing method of the wideband signal based on deep learning in the embodiment of the invention is mainly divided into five steps, firstly, the input wideband signal is obtained, and the input wideband signal is sampled in multiple sets in the time domain to obtain a discrete sampling sequence; secondly, preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set; thirdly, designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net; fourthly, in a model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters; and fifthly, in a testing stage or an application stage, inputting signal data of a testing set into the trained ADMM-net model to obtain a reconstructed broadband signal.
The implementation method of the invention comprises the following steps: the 300 broadband signals are sampled in multiple sets at two different compression sampling rates of p/L=17% and p/L=42%, and are preprocessed to form a data set, wherein the number of data pairs of a training set and a testing set is set to be a set ratio (4:1), the signal to noise ratio of the broadband signals of the training set is uniformly set in a range of [5,30] dB, and the signal to noise ratio of the broadband signals of the testing set is set to be 5dB,10dB,15dB,20dB,25dB and 30dB.
In the model training stage, signal data of training sets obtained by p/L=17% and p/L=42% of two different compression sampling rates are respectively input into an ADMM-net model, and the loss function is continuously minimized through an optimization algorithm to obtain optimal neural network parameters, so that the trained ADMM-net model is obtained.
And in the test stage or the application stage, inputting the signal data of the test set into the trained ADMM-net model to obtain the reconstruction result of the broadband signal under different signal-to-noise ratios.
The channel l=64 of the wideband signal x (t) in the present invention includes 5 source channels with different bandwidths, and for the signal data with two different compression sampling rates of p/l=17% and p/l=42% in the test set, the signal reconstruction results of three reconstruction algorithms of SOMP, ADMM and ADMM-net are shown in fig. 6 under different signal-to-noise ratio conditions, and fig. 6 is a schematic diagram of the reconstruction results of three reconstruction algorithms of SOMP, ADMM and ADMM-net under different signal-to-noise ratio conditions in the embodiment of the present invention with channel number l=64 and source channel k=5.
As shown by the results of the test set, compared with the conventional reconstruction algorithm, the ADMM-net combined with the deep learning in the invention has higher reconstruction accuracy under the conditions of lower sampling rate and lower signal-to-noise ratio after being trained by a network model and has certain generalization capability.
The invention has the following advantages:
1) Combining with the idea of deep learning, the frequency domain reconstruction is carried out on the broadband signal by multiplying the sub-network in the alternative direction of the expandable depth for the first time, and the result proves that the reconstruction performance on the broadband signal can be improved.
2) The loss function between the network reconstruction result and the true value is constructed, so that the loss function is continuously minimized by using an Armijo criterion as a conjugate gradient method in the gradient descent direction, the neural network parameters of the sub-network in the alternate direction are trained, and the reconstruction performance of the broadband signal and the generalization of the network model are improved.
3) The compressed sensing performance of the broadband signal is improved under the conditions of lower sampling rate and lower signal-to-noise ratio by taking the multiple set sampling and the alternative direction multiplier sub-network based on deep learning as the compressed sensing framework of the broadband signal.
Exemplary apparatus
As shown in fig. 7, an embodiment of the present invention provides a compressed sensing processing apparatus for a wideband signal based on deep learning, the apparatus comprising:
The multiple set sampling processing module 410 is configured to obtain an input wideband signal, and perform multiple set sampling on the input wideband signal in a time domain to obtain a discrete sampling sequence;
a preprocessing module 420, configured to preprocess the obtained discrete sampling sequence and the input wideband signal and construct a data set;
a setting and initializing module 430, configured to design a deep learning-based broadband signal reconstruction neural network model ADMM-net, and initialize the ADMM-net;
the neural network training module 440 is configured to input signal data of a training set into the ADMM-net in a model training stage, and continuously minimize a loss function through an optimization algorithm to obtain an optimal neural network parameter;
the test and application module 450 is configured to input the signal data of the test set into the trained ADMM-net model during the test phase or the application phase, to obtain a reconstructed wideband signal, as described above.
Based on the above embodiment, the present invention further provides an intelligent terminal, and a functional block diagram thereof may be shown in fig. 8. The intelligent terminal of the embodiment of the invention can be an intelligent television, and comprises a processor, a memory, a network interface and a display screen which are connected through a system bus. The processor of the intelligent terminal is used for providing computing and control capabilities. The memory of the intelligent terminal comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the intelligent terminal is used for communicating with an external terminal through network connection. The computer program, when executed by a processor, implements a compressed sensing processing method for a wideband signal based on deep learning. The display screen of the intelligent terminal can be a liquid crystal display screen or an electronic ink display screen.
It will be appreciated by those skilled in the art that the schematic block diagram shown in fig. 8 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and is not limiting of the smart terminal to which the present inventive arrangements are applied, and that a particular smart terminal may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a smart terminal is provided that includes a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs comprising instructions for:
acquiring an input broadband signal, and performing multiple set sampling on the input broadband signal in a time domain to obtain a discrete sampling sequence;
preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set;
designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net;
in the model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters;
And in a testing stage or an application stage, inputting signal data of a testing set into the trained ADMM-net model to obtain a reconstructed broadband signal.
The step of obtaining the input broadband signal and performing multiple set sampling on the input broadband signal in the time domain to obtain a discrete sampling sequence comprises the following steps:
acquiring an input broadband signal x (t), and determining a sampling mode through a sequence forward selection algorithm
Sampling at moment to obtain a discrete sampling sequence x ci [n]Where T represents the nyquist sampling time period interval of the input wideband signal, and the multiple set sampling period interval is L times as large as that of the nyquist sampling theorem, so the sampling frequency is reduced to 1/L of the nyquist sampling frequency. />
The step of preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set comprises the following steps:
for the obtained discrete sampling sequence x ci [n]Performing discrete Fourier transform to obtain frequency domain representation of the sampling sequence
At the same time, the frequency domain representation of the resulting sample sequence
And the frequency domain signal X k of the corresponding input wideband signal]The data sets are composed, wherein the number of training sets and test sets is a set proportion.
The method for processing compressed sensing of the broadband signal based on the deep learning, wherein the step of designing the broadband signal reconstruction neural network model ADMM-net based on the deep learning and initializing the ADMM-net comprises the following steps:
An ADMM algorithm is used as a compressed sensing signal reconstruction algorithm, and an ADMM-net reconstructed broadband signal with expandable depth based on deep learning is designed;
parameters of the ADMM-net model are initialized to corresponding parameter values in the ADMM algorithm.
In the model training stage, signal data of a training set is input into the ADMM-net, a loss function is continuously minimized through an optimization algorithm, and the step of obtaining optimal neural network parameters comprises the following steps:
in the model training phase, the frequency domain representation of the sampling sequence of the training set is obtained
Inputting the initialized ADMM-net to obtain the initially reconstructed frequency domain signal X ir [k]；
Constructing the initially reconstructed frequency domain signal X ir [k]Frequency domain signal X k with input wideband signal]A loss function E (θ) therebetween, where θ is a related neural network parameter;
and continuously minimizing the loss function E (theta) by using an optimization algorithm to obtain the optimal neural network parameter theta, thereby obtaining the trained ADMM-net model.
The step of inputting the signal data of the test set into the trained ADMM-net model to obtain the reconstructed broadband signal in the test stage or the application stage comprises the following steps:
in the test phase or application phase, the frequency domain representation of the sampling sequence of the test set is obtained 
Inputting the trained ADMM-net model for processing to obtain a reconstructed frequency domain signal X r [k]。
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
In summary, the invention discloses a compressed sensing processing method, a device, an intelligent terminal and a storage medium of broadband signals based on deep learning, wherein the method comprises the following steps: acquiring an input broadband signal, and performing multiple set sampling on the input broadband signal in a time domain to obtain a discrete sampling sequence; preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set; designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net; in the model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters; and in the test stage or the application stage, inputting the signal data of the test set into the trained ADMM-net model to obtain a reconstructed broadband signal. Compared with a common reconstruction algorithm, the numerical analysis of the experimental result shows that the reconstruction accuracy of the ADMM-net combined with the deep learning under the conditions of lower sampling rate and lower signal to noise ratio is higher after the network model is trained, and the ADMM-net has certain generalization capability.
The invention has the following advantages:
1) Combining with the idea of deep learning, the frequency domain reconstruction is carried out on the broadband signal by the sub-network in the alternative direction of the expandable depth for the first time, and the result proves that the reconstruction performance of the broadband signal can be improved.
2) The loss function between the network reconstruction result and the true value is constructed, so that the loss function is continuously minimized by using an Armijo criterion as a conjugate gradient method in the gradient descent direction, the neural network parameters of the sub-network in the alternate direction are trained, and the reconstruction performance of the broadband signal and the generalization of the network model are improved.
3) The compressed sensing performance of the broadband signal is improved under the conditions of lower sampling rate and lower signal-to-noise ratio by taking the multiple set sampling and the alternative direction multiplier sub-network based on deep learning as the compressed sensing framework of the broadband signal.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. A compressed sensing processing method of a broadband signal based on deep learning, the method comprising:
acquiring an input broadband signal, and performing multiple set sampling on the input broadband signal in a time domain to obtain a discrete sampling sequence;
preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set, wherein the data set comprises: signal data of the training set and signal data of the test set;
designing a broadband signal reconstruction neural network model ADMM-net based on deep learning, and initializing the ADMM-net;
in the model training stage, inputting signal data of a training set into the ADMM-net, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters;
in a test stage or an application stage, inputting signal data of a test set into a trained ADMM-net model to obtain a reconstructed broadband signal;
the step of obtaining the input broadband signal, and performing multiple set sampling on the input broadband signal in the time domain to obtain a discrete sampling sequence further comprises the following steps:
acquiring an input broadband signal x (t), and determining a sampling mode through a sequence forward selection algorithm
Sampling at +. >
Sampling at moment to obtain a discrete sampling sequence x ci [n]Wherein T represents a nyquist sampling time period interval of an input broadband signal, and compared with the nyquist sampling theorem, the multiple set sampling time period interval is L times of the nyquist sampling time period interval, so that the sampling frequency is reduced to 1/L of the nyquist sampling frequency;
minimizing the loss function with a conjugate gradient method using Armijo's criterion;
in the model training stage, the step of inputting the signal data of the training set into the ADMM-net and continuously minimizing the loss function through an optimization algorithm to obtain the optimal neural network parameters comprises the following steps:
in the model training phase, the frequency domain representation of the sampling sequence of the training set is obtained
Inputting the initialized ADMM-net to obtain the initially reconstructed frequency domain signal X ir [k]；
Constructing the initially reconstructed frequency domain signal X ir [k]Frequency domain signal X k with input wideband signal]A loss function E (θ) therebetween, where θ is a related neural network parameter;
continuously minimizing a loss function E (theta) by using an optimization algorithm to obtain an optimal neural network parameter theta, thereby obtaining a trained ADMM-net model;
the loss function is defined as:
wherein the method comprises the steps of
An initial reconstructed frequency domain signal representing an ith signal data of the training set N signal data, (Y, θ) represents a frequency domain representation matrix Y of the sampling sequence and a network parameter θ, respectively, where X () A frequency domain signal representing the input wideband signal of the training set ith signal data.
2. The compressed sensing processing method of a deep learning based wideband signal of claim 1, wherein the step of preprocessing the obtained discrete sample sequence and the inputted wideband signal and constructing a data set comprises:
for the obtained discrete sampling sequence x ci [n]Performing discrete Fourier transform to obtain frequency domain representation of the sampling sequence
At the same time, the frequency domain representation of the resulting sample sequence
And the frequency domain signal X k of the corresponding input wideband signal]Forming a data set, wherein the quantity of the training set and the testing set is a set proportion;
at the Nyquist sampling frequency f s Uniformly sampling the input broadband signal x (T) with the time interval of T to obtain a discrete sequence x [ n ] with the total length of K]Then to the sequence x [ n ]]Performing discrete Fourier transform to obtain frequency domain signal X [ k ] of input broadband signal]I.e.
Where k=0, 1,2, …, K-1, K represents the number of points of the discrete sequence x [ n ].
3. The compressed sensing processing method for deep learning based broadband signals according to claim 1, wherein the step of designing the deep learning based signal reconstruction neural network model ADMM-net and initializing the ADMM-net comprises:
Using an ADMM algorithm as a compressed sensing signal reconstruction algorithm, and designing an extensible depth signal reconstruction neural network model ADMM-net based on deep learning;
parameters of the ADMM-net model are initialized to corresponding parameter values in the ADMM algorithm.
4. The compressed sensing processing method for wideband signals based on deep learning according to claim 1, wherein the step of inputting the signal data of the test set into the trained ADMM-net model to obtain the reconstructed wideband signals in the test stage or the application stage comprises:
in the test phase or application phase, the frequency domain representation of the sampling sequence of the test set is obtained
Inputting the trained ADMM-net model for processing to obtain a reconstructed frequency domain signal X r [k]。
5. A compressed sensing processing apparatus for deep learning based wideband signals, the apparatus comprising:
the multi-set sampling processing module is used for acquiring an input broadband signal, and performing multi-set sampling on the input broadband signal in the time domain to obtain a discrete sampling sequence;
the preprocessing module is used for preprocessing the obtained discrete sampling sequence and the input broadband signal and constructing a data set, wherein the data set comprises: signal data of the training set and signal data of the test set;
The setting and initializing module is used for designing a deep learning-based signal reconstruction neural network model ADMM-net and initializing the ADMM-net;
the neural network training module is used for inputting signal data of a training set into the ADMM-net in a model training stage, and continuously minimizing a loss function through an optimization algorithm to obtain optimal neural network parameters;
the test and application module is used for inputting the signal data of the test set into the trained ADMM-net model in the test stage or the application stage to obtain a reconstructed broadband signal;
acquiring an input broadband signal x (t), and determining a sampling mode through a sequence forward selection algorithm
Sampling at +.>
Sampling at moment to obtain a discrete sampling sequence x ci [n]Wherein T represents a nyquist sampling time period interval of an input broadband signal, and compared with the nyquist sampling theorem, the multiple set sampling time period interval is L times of the nyquist sampling time period interval, so that the sampling frequency is reduced to 1/L of the nyquist sampling frequency;
minimizing the loss function with a conjugate gradient method using Armijo's criterion;
in the model training stage, the step of inputting the signal data of the training set into the ADMM-net and continuously minimizing the loss function through an optimization algorithm to obtain the optimal neural network parameters comprises the following steps:
In the model training phase, the frequency domain representation of the sampling sequence of the training set is obtained
Inputting the initialized ADMM-net to obtain the initially reconstructed frequency domain signal X ir [k]；
Constructing the initially reconstructed frequency domain signal X ir [k]Frequency domain signal X k with input wideband signal]A loss function E (θ) therebetween, where θ is a related neural network parameter;
continuously minimizing a loss function E (theta) by using an optimization algorithm to obtain an optimal neural network parameter theta, thereby obtaining a trained ADMM-net model;
the loss function is defined as:
wherein the method comprises the steps of
An initial reconstructed frequency domain signal representing an ith signal data of the training set N signal data, (Y, θ) represents a frequency domain representation matrix Y of the sampling sequence and a network parameter θ, respectively, where X () A frequency domain signal representing the input wideband signal of the training set ith signal data.
6. An intelligent terminal comprising a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs comprising instructions for performing the method of any of claims 1-4.
7. A non-transitory computer readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210450842.1A CN114978381B (en) | 2022-04-27 | 2022-04-27 | Compressed sensing processing method and device for broadband signal based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210450842.1A CN114978381B (en) | 2022-04-27 | 2022-04-27 | Compressed sensing processing method and device for broadband signal based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114978381A CN114978381A (en) | 2022-08-30 |
| CN114978381B true CN114978381B (en) | 2023-06-23 |
Family
ID=82978551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210450842.1A Active CN114978381B (en) | 2022-04-27 | 2022-04-27 | Compressed sensing processing method and device for broadband signal based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114978381B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111680666A (en) * | 2020-06-30 | 2020-09-18 | 西安电子科技大学 | Under-sampling frequency hopping communication signal deep learning recovery method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9734601B2 (en) * | 2014-04-04 | 2017-08-15 | The Board Of Trustees Of The University Of Illinois | Highly accelerated imaging and image reconstruction using adaptive sparsifying transforms |
| EP3618287B1 (en) * | 2018-08-29 | 2023-09-27 | Université de Genève | Signal sampling with joint training of learnable priors for sampling operator and decoder |
| CN111224671A (en) * | 2020-01-15 | 2020-06-02 | 高跃 | Signal processing apparatus |
| CN112615801B (en) * | 2020-12-16 | 2021-11-19 | 西安交通大学 | Channel estimation method, medium, and apparatus based on compressed sensing and deep learning |
-
2022
- 2022-04-27 CN CN202210450842.1A patent/CN114978381B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111680666A (en) * | 2020-06-30 | 2020-09-18 | 西安电子科技大学 | Under-sampling frequency hopping communication signal deep learning recovery method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114978381A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111680666B (en) | Under-sampling frequency hopping communication signal deep learning recovery method | |
| CN111915007B (en) | Magnetic resonance spectrum noise reduction method based on neural network | |
| CN110365437B (en) | Fast power spectrum estimation method based on sub-Nyquist sampling | |
| He et al. | Maximum correntropy adaptation approach for robust compressive sensing reconstruction | |
| CN109688074A (en) | A kind of channel estimation methods of compressed sensing based ofdm system | |
| CN114978381B (en) | Compressed sensing processing method and device for broadband signal based on deep learning | |
| Kato et al. | Uniform confidence bands for nonparametric errors-in-variables regression | |
| CN107426737B (en) | Broadband spectrum sensing method based on single-channel structure modulation broadband converter | |
| CN112688746A (en) | Spectrum prediction method based on space-time data | |
| Geng et al. | HFIST-Net: High-throughput fast iterative shrinkage thresholding network for accelerating MR image reconstruction | |
| Wang et al. | Parallel non-Cartesian spatial-temporal dictionary learning neural networks (stDLNN) for accelerating 4D-MRI | |
| CN103929256B (en) | A kind of multiframe compressed sensing signal spectrum detection method | |
| CN111478706B (en) | Compressed sensing-oriented sparse multi-band signal blind reconstruction method | |
| Huang et al. | High-speed signal reconstruction for compressive sensing applications | |
| Fu et al. | A parameter estimation algorithm for multiple frequency-hopping signals based on compressed sensing | |
| Lyons et al. | A deep compound Gaussian regularized unfolded imaging network | |
| Pelissier et al. | Hardware platform of Analog-to-Information converter using Non Uniform Wavelet Bandpass Sampling for RF signal activity detection | |
| CN113095215B (en) | Solar radio filtering method and system based on improved LSTM network | |
| Liu et al. | HTR-CTO algorithm for wireless data recovery | |
| Zhu et al. | ROAST: Rapid orthogonal approximate Slepian transform | |
| CN111982856B (en) | Substance marker-free detection and identification method based on terahertz waves | |
| Hu et al. | Adaptive and blind wideband spectrum sensing scheme using singular value decomposition | |
| Wang et al. | WDU-Net: Wavelet-guided deep unfolding network for image compressed sensing reconstruction | |
| Rateb et al. | A generic top-level mechanism for accelerating signal recovery in compressed sensing | |
| CN108092668A (en) | A kind of linear FM signal restructing algorithm based on minimum sparse ruler sampling |
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 |