WO2021203243A1 - 基于人工智能的mimo多天线信号传输与检测技术 - Google Patents

基于人工智能的mimo多天线信号传输与检测技术 Download PDF

Info

Publication number
WO2021203243A1
WO2021203243A1 PCT/CN2020/083564 CN2020083564W WO2021203243A1 WO 2021203243 A1 WO2021203243 A1 WO 2021203243A1 CN 2020083564 W CN2020083564 W CN 2020083564W WO 2021203243 A1 WO2021203243 A1 WO 2021203243A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
neural network
module
network model
sparse
Prior art date
Application number
PCT/CN2020/083564
Other languages
English (en)
French (fr)
Inventor
周清峰
刘婵梓
陈高
曲春晓
Original Assignee
东莞理工学院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 东莞理工学院 filed Critical 东莞理工学院
Priority to PCT/CN2020/083564 priority Critical patent/WO2021203243A1/zh
Priority to CN202310024746.5A priority patent/CN116054887A/zh
Priority to CN202080000961.6A priority patent/CN111713035B/zh
Publication of WO2021203243A1 publication Critical patent/WO2021203243A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0041—Arrangements at the transmitter end
    • H04L1/0043—Realisations of complexity reduction techniques, e.g. use of look-up tables
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0045—Arrangements at the receiver end
    • H04L1/0047—Decoding adapted to other signal detection operation
    • H04L1/0048—Decoding adapted to other signal detection operation in conjunction with detection of multiuser or interfering signals, e.g. iteration between CDMA or MIMO detector and FEC decoder
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0045—Arrangements at the receiver end
    • H04L1/0052—Realisations of complexity reduction techniques, e.g. pipelining or use of look-up tables
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0224—Channel estimation using sounding signals
    • 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

Definitions

  • the present invention relates to the field of mobile communication technology, in particular to artificial intelligence-based MIMO multi-antenna signal transmission and detection technology.
  • MIMO Multiple-Input Multiple-Output
  • MIMO also known as multiple-input multiple-output system
  • MIMO refers to a communication system that uses multiple antennas at the transmitting end and the receiving end of the device at the same time. It can double the communication system without increasing the bandwidth. Capacity and spectrum utilization.
  • the MIMO system uses multiple antennas at the transmitting end and the receiving end of the device.
  • the transmitted information stream is "space-time encoded" to form multiple information sub-streams. Multiple sub-streams are sent to the channel at the same time. Each transmitted signal occupies the same frequency band, so the system is not increased. bandwidth. If the channel responses between the transmitting and receiving antennas are independent, the multiple input multiple output system can create multiple parallel spatial channels. Through these parallel spatial channels to independently transmit information, the data rate will inevitably be improved.
  • MIMO technology is a huge breakthrough in the field of wireless communication.
  • many companies developed WIFI or WIMAX commercial systems based on MIMO technology.
  • all 4G communication system standards (such as TD.LTE, LET.A, WIMAX, etc.) selected MIMO technology as one of their key technologies.
  • MIMO systems have been widely used in wireless communications-mobile devices and networks generally use multiple antennas to enhance connectivity, improve network speed and user experience.
  • Massive MIMO is also a key factor in the ultra-high data rate of 5G, which can bring greater network capacity, wider signal coverage, and better user experience, bringing the potential of 5G to a whole new level.
  • AI is a computer program that takes reasonable actions based on its perception of the environment and obtains the most benefits
  • AI is a computer algorithm that can learn
  • Machine learning is essentially the key technology for extracting knowledge from data, and it is the driving force and engine for the development of artificial intelligence.
  • machine learning mainly solves classification problems, clustering problems, regression problems, etc., and has been widely used in character recognition, machine translation, speech recognition, search engines, face recognition, unmanned driving and other fields.
  • there are hundreds of different learning algorithms and there is no unified framework to describe the design process of machine learning algorithms. However, there are currently two strongest algorithms for machine learning-deep learning and reinforcement learning.
  • Deep learning belongs to the artificial neural network system, which is a new generation of neural networks under the development of traditional neural networks.
  • Neural network is an intelligent science that imitates the structure and function of the human nerve center-the brain. Has the ability to react quickly, facilitate real-time control and processing of things; excellent self-organization and self-learning capabilities; good at working in complex environments. It fully approximates any nonlinear system, and quickly obtains the optimal answer to the problems that meet a variety of constraints; it has superior performance such as high robustness and fault tolerance, so it has been more and more widely used in communication systems.
  • Neurosurgery is a computer system formed by a number of very simple processing units connected to each other in a certain way. The system processes information by its dynamic response to the input information.
  • the Chinese patent application number 201510473741.6 (Self-complex neural network channel prediction method, Xidian University) discloses a complex neural network channel prediction method, which mainly solves the channel fading problem caused by channel time variation in the MIMO system.
  • the technical solution is: 1.
  • the base station measures the channel to obtain a channel coefficient training sequence containing estimation errors; 2.
  • the base station uses the trained complex wavelet neural network to predict the channel coefficients.
  • the method is simple and easy to implement, has good effects, and is suitable for reducing the influence of the channel time change on the MIMO system channel.
  • the Chinese patent application number 201810177829.7 discloses a method for modeling a wireless channel based on a neural network.
  • the invention first processes the received signal fed back by the user to obtain the estimated channel parameters; then obtains the three-dimensional geographic information of the scatterers according to the two-dimensional image, and clusters them, and finally uses the channel parameters and geographic information as the input of the neural network , Receive the signal as the output, and train the nonlinear time-varying neural network model.
  • This method obtains a more accurate channel model within acceptable complexity, and can meet the channel modeling requirements of scenarios such as massive MIMO technology, frequency band expansion, and high mobility used in future 5G communication systems.
  • the Chinese patent application number 201810267976.3 (Deep neural network massive MIMO system detection method based on BP algorithm, Southeast University) provides a deep neural network massive MIMO system detection method based on BP algorithm, which is iteratively propagating beliefs
  • the algorithm factor graph is expanded and mapped to the neural network structure to construct a deep neural network for large-scale MIMO system detection; the neurons of the deep neural network correspond to the nodes in the iterative algorithm factor graph, and the number of neurons in each layer is equal to the iteration
  • the number of symbol nodes in the algorithm factor graph; the mapping function between hidden layers is the update formula of the confidence information in the iterative algorithm, and the number of hidden layers is equal to the number of iterations of the iterative algorithm.
  • the invention also provides a MIMO detection method for constructing two deep neural networks based on two information propagation iterative algorithms of damped confidence propagation and maximum and confidence propagation, respectively.
  • the invention achieves a lower bit error rate without increasing the complexity of online operations, and is robust to various channel conditions and antenna ratios.
  • the Chinese patent application number 201610327115.0 (a codebook selection method based on deep learning under massive MIMO, Chongqing University of Posts and Telecommunications) relates to a codebook selection method based on deep learning under massive MIMO.
  • the method includes: collecting pilot information in the test area to construct a pilot training sequence, and then obtaining pilot training samples; performing neural network iterative learning on the pilot training samples to obtain the final network weight value; according to the output of the learned neural network Channel, select the optimal codeword from the complete codebook. Then, the unknown area and the test area are matched with channel information to obtain the wireless channel, and then the codeword corresponding to the wireless channel is obtained.
  • the invention can effectively, accurately and quickly establish a wireless channel model and codebook query, avoids channel estimation in the unknown area and greatly reduces the complexity of channel selection codebook in the unknown area.
  • the Chinese application number 201811626005.X patent discloses a low-complexity MIMO-NOMA system signal detection method based on an improved gradient projection method, involving wireless Communication technology.
  • the idea of convex optimization algorithm is used to transform the system model into a strict quadratic programming problem; then the problem is solved iteratively, and the results of each iteration are preprocessed to achieve the goal The signal is effectively detected.
  • This invention breaks through the problem of slow algorithm convergence in traditional detection methods. Preprocessing the results of each iteration can not only make the detection results quickly converge, but also detect the set of active users.
  • the implementation process is simple and the application range is wide. .
  • the Chinese patent application number 201910014714.0 discloses a method for designing beamforming matrix of MIMO system based on deep learning.
  • the steps are as follows. First, use a known algorithm to obtain the depth Learn the training sample set required by the network; then build a deep learning neural network model, initialize the relevant parameters of the model and use the training sample set for training; then use the pilot to obtain the channel and send it to the neural network to predict the beamforming matrix coefficients, and finally the channel and The beamforming matrix coefficients are combined to form the beamforming matrix.
  • the method uses the beamforming matrix obtained by the deep learning neural network to take into account the performance and the algorithm complexity at the same time, and can reduce the delay under the premise of ensuring the performance, so that the MIMO system can provide real-time services.
  • the Chinese patent application number 201810182937.3 discloses a machine learning-based MIMO link adaptive transmission method, which uses an unsupervised learning self-encoding algorithm for feature extraction With dimensionality reduction, the introduction of the idea of deep learning can reduce the feature dimension and computational complexity on the premise of preserving the main information state information.
  • This invention uses a logistic regression algorithm to construct the mapping relationship between channel state information and transmission parameters. It is different from the previous fixed parameterized model. It can be trained based on sample data. When the quality of the data set is better, it can cover all states.
  • the mapping relationship between channel state information and transmission parameters can be established well, and compared with the traditional single equivalent signal-to-noise ratio, the channel state information can be more fully utilized.
  • the invention also selects CQI based on the channel matrix, and studies the MIMO link adaptation method based on machine learning through the channel matrix and noise variance, which is not restricted by the receiver design and has universal applicability.
  • the Chinese patent application number 201710495044.X discloses a joint precoding and antenna selection method for MIMO systems based on deep learning, which includes the following steps: First, use the existing antenna selection method to generate the training data set required for deep learning; then, establish a deep learning model, use the training data to train the deep learning model and save it; then use the saved deep learning model to complete the antenna selection; finally
  • the MIMO subsystem is designed for optimal precoding.
  • the invention uses deep learning technology to design MIMO system joint precoding and antenna selection, which can achieve lower computational complexity while obtaining a good system signal-to-noise ratio.
  • Chinese Patent Application No. 201910242525.9 proposes a deep signal detection method for high-speed rail.
  • the high-speed rail first determines its location information through GPS during driving, judges the area where it is located, and selects the corresponding neural network model, and then trains the real-time received signal input In the neural network, the signal sent by the base station is output in real time.
  • the system performance of the invention has been greatly improved, the signal detection bit error rate is reduced, and the algorithm is more robust.
  • the method used in the invention does not need to estimate the channel and saves pilot overhead.
  • Chinese application number 201810279530.2 invention patent discloses a visible light communication MIMO anti-interference and noise reduction method based on BP neural network, involving MIMO in the field of visible light communication Antenna technology.
  • This system includes the system transmitting terminal, the system device receiving terminal signal processing part and the BP neural network signal processing part which are connected in sequence.
  • This method is as follows: 1) The electrical signal is loaded on the LED array and emitted in the form of light signal; 2) The photodetector at the receiving end of the device converts the light signal into an electrical signal; 3) Multiple electrical signals are removed by a low-pass filter High-frequency interference; 4) After training, the BP neural network performs noise reduction and interference elimination processing on multi-channel signals, and finally becomes a binary serial data stream through parallel-serial conversion.
  • the invention improves the transmission performance of the existing MIMO technology; combines the neural network with the visible light MIMO technology to give full play to the advantages of the neural network in removing interference and noise in wireless communication; adopts the neural network receiving and processing technology to make the entire VLC system more stable.
  • Chinese Application No. 201710213235.2 Invention Patent discloses a joint equalization method of visible light channel based on orthogonal mapping and Probabilistic Neural Network, including device transmitter and device receiver At the end, the signal is transmitted from the transmitting end to the receiving end of the device through the visible light MIMO channel; the visible light MINO channel is a multiple-input multiple-output channel; the joint equalization is a combination of pre-equalization and post-equalization; the invention uses A joint equalization scheme that combines pre-equalization technology and post-equalization technology, that is, a visible light multiple-input multiple-output channel joint equalization method based on orthogonal mapping and probabilistic neural network, can effectively suppress the interference between visible light MIMO communication system channels and improve Data transmission reliability.
  • the embodiment of China Patent No. 201910125325.5 discloses a MIMO decoding method, device and storage medium based on deep learning.
  • the training data set includes multiple training data; then the neural network is trained based on the training data set to obtain the trained neural network model; when the MIMO signal to be decoded is received, the training data set will be decoded.
  • the MIMO signal is input to the neural network model for MIMO decoding, and then the MIMO decoding result output by the neural network model is obtained.
  • a neural network model for joint MIMO detection and channel decoding is designed based on deep learning, MIMO detection and channel decoding are regarded as a joint decoding process, and the approximation of the output of the neural network model is improved through training. , To ensure the overall performance of MIMO decoding, with higher decoding accuracy and faster decoding speed.
  • the Chinese Patent Application No. 201810757547.4 discloses a machine learning-assisted massive MIMO downlink user scheduling method, which includes the following steps: S1: the base station sends data from users The uplink detection signal obtains the characteristic mode energy coupling matrix in the characteristic direction; S2: The base station uses the characteristic mode energy coupling matrix to assist in the calculation of the sum rate under various user and beam combinations through the method of machine learning; S3: adopts the greedy algorithm to achieve the sum rate User scheduling based on the maximum rate criterion to obtain the optimal user beam pairing combination.
  • the invention obtains statistical channel information through the uplink sounding signal, and uses the sum rate maximization criterion for user scheduling.
  • the base station only has statistical channel information
  • the approximate calculation of the rate and the rate can be accurately achieved, which greatly reduces the complexity of user scheduling under large-scale antennas, and the performance is close to the best. Excellent, with good applicability and robustness.
  • China 201610353881.4 invention patent (a modulation recognition method under MIMO related channels based on machine learning algorithms, Beijing University of Posts and Telecommunications) is a modulation recognition method under MIMO related channels based on machine learning algorithms, which belongs to the field of communications; the specific steps are as follows: : Firstly, each data stream at the transmitting end of the communication adopts space-time coding, and each codeword is respectively transmitted through Nt transmitting antennas; then, the MIMO channel matrix H is calculated according to the correlation matrix of the receiving end of the device and the correlation matrix of the transmitting end; MIMO channel matrix H, calculate and correct the received signal on each receiving antenna; finally, each receiving antenna separately extracts the characteristics of the corrected signal, performs training tests on the extracted eigenvalues, and calculates the modulation that the sample belongs to Recognition mode; the advantages are: strong robustness and generalization ability for non-Gaussian channels, and modulation system recognition in more complex environments can be achieved through parameter iteration; by extracting the characteristics of higher-order moments and higher-order cumulants
  • MIMO technology has become one of the key technologies in the field of wireless communication. Through continuous development in recent years, MIMO technology has been increasingly applied to various wireless communication systems. With the increase in the number of antennas used, the complexity of the implementation of the MIMO technology is greatly increased, which limits the number of antennas used and cannot fully utilize the advantages of the MIMO technology.
  • the subject of artificial intelligence since 1956, when the subject of artificial intelligence was formally proposed, it has made considerable progress in the past 50 years and has become a broad cross-cutting and cutting-edge science. As an important branch of artificial intelligence, it can be seen from the above description that the neural network realizes the mathematical network through the computer, which is recognized as the "outlet" to solve some of the bottlenecks encountered in current communications.
  • This application combines the latest research progress of artificial intelligence technology and compressed sensing technology, and proposes a signal multiplexing transmission and detection scheme in a MIMO system based on artificial intelligence and compressed sensing technology.
  • This application reduces the dimensionality of the signal to be processed by introducing the thinning module and the compressed sensing multiplexing module at the transmitting end of the device, and can multiplex the parallel data stream exceeding the number of transmitting antennas to a given transmitting antenna for transmission, thus greatly
  • the increase in the amplitude of the signal multiplexing gain under the condition of the number of transmit and receive antennas of the MIMO system better meets the application requirements of the MIMO system for broadband transmission; on the other hand, the compressed multiplexing matrix proposed by the present invention does not need to rely on channel state information
  • the technical scheme adopted by the present invention can add sparsity and compressed sensing multiplexing steps at the transmitting end of the device without modifying the existing MIMO technical scheme, and adding mature optimized reconstruction algorithms in the compressed sensing field at the receiving
  • the MIMO device based on artificial intelligence and compressed sensing technology proposed in this application is mainly composed of a thinning module and a compression multiplexing module at the transmitting end of the device, and a demultiplexing module at the receiving end of the device.
  • a random number generator or original information bit generation module
  • a bit-level processing module or a modulation module
  • a channel estimation module is also provided in the receiving end of the device.
  • the transmitting end of the device the original data generated by the random number generator passes through the bit-level processing module and the modulation module to generate a modulated signal, and the modulated signal then passes through the thinning module, based on The first neural network model expresses it as a sparse signal.
  • the sparse signal passes through a compression multiplexing module.
  • the compression multiplexing module compresses the transmitted signal, reduces the dimension and multiplexes it, and the compressed multiplexed signal passes through the transmitting antenna.
  • the channel estimation module performs channel estimation on the received signal, the estimated channel state information is passed through the demultiplexing module, and the original transmission data stream x is reconstructed and output based on the second neural network model.
  • the above process specifically includes the following:
  • a MIMO multi-antenna signal transmission and detection device based on artificial intelligence.
  • the device includes: a sparser module that uses a first neural network model to sparsely represent the original signal; a compression multiplexing module that is used to perform sparsely represented signals Compression and dimensionality reduction processing; the receiving end of the device, used to process the received signal to achieve its reconstruction of the target signal, the device further includes a demultiplexing module, wherein the demultiplexing module is configured to The receiving end of the device uses the compressed sensing reconstruction algorithm to solve the sparse representation vector from the low-dimensional signal Finally, the second neural network model is used to reconstruct the received signal to obtain the original signal x.
  • the input and output of the second neural network model and the thinning module used at the transmitting end of the device extract the first sparse signal ⁇ from the low-dimensional signal to be transmitted to the compression multiplexing module for processing.
  • the input and output of the neural network model are reversed.
  • the receiving end of the device further includes a channel estimation module configured to perform channel estimation on the signal received by the receiving end of the device and use the channel parameter matrix obtained therefrom as the compressed sensing reconstruction Algorithm input.
  • the demultiplexing module obtains the sparse representation vector based on the compressed sensing multiplexing matrix and the channel parameter matrix that do not need to rely on channel state information of.
  • the thinning module creates a neural network through a BP neural network training method in artificial intelligence.
  • the sparse representation ⁇ is used as input and output respectively, a first training sample set is constructed, the neural network is trained, and the first neural network model is obtained.
  • the demultiplexing module composes a redundant dictionary D by using all possible combinations of the transmission signal vector x as different column vectors of the redundant dictionary to realize the sparseness of the transmission signal vector.
  • the thinned sparse representation ⁇ is compressed into a ⁇ 1 signal through the compression multiplexing module, and then the device transmits the compressed and multiplexed signal Z through the transmitting antenna.
  • a MIMO multi-antenna signal transmission and detection system based on artificial intelligence.
  • the system constructs a MIMO end-to-end transmission model. Compress the sparse signal before processing by the multiplexing module, and the device receiver uses the second neural network model constructed by it which is opposite to the input and output of the first neural network model of the sparse module to reconstruct the received signal to obtain the original signal .
  • the input of the second neural network model is a sparse representation vector obtained by using a compressed sensing reconstruction algorithm
  • FIG. 1 is a schematic block diagram of the signal processing flow of the signal transmission and detection system of the MIMO multi-antenna system based on artificial intelligence and compressed sensing technology provided by the present invention
  • FIG. 2 is a schematic block diagram of a preferred signal compression multiplexing and detection processing process provided by the present invention
  • Figure 3 is a bit error rate performance curve of the classic detection algorithm ZF and the multi-antenna signal transmission and detection technology of the MIMO system of the present invention under different transceiver antenna configurations;
  • Figure 4 is a bit error rate performance curve of the classic detection algorithm ZF and the multi-antenna signal transmission and detection technology of the MIMO system of the present invention under different transceiver antenna configurations;
  • FIG. 5 is a schematic diagram of the module connection of the signal transmission and detection system of the MIMO multi-antenna system based on artificial intelligence and compressed sensing technology provided by the present invention.
  • Device transmitter 101 Random number generator
  • Bit-level processing module 103 Modulation module
  • Thinning module 105 Compression multiplexing module
  • Device receiving end 201 Channel estimation module
  • the MIMO device based on artificial intelligence and compressed sensing technology proposed in this application is mainly composed of the thinning module 104 and the compression multiplexing module 105 of the transmitting end 1 of the device, and the demultiplexing module 202 of the receiving end 2 of the device.
  • a random number generator 101 or original information bit generation module
  • a bit-level processing module 102 or a modulation module 103 are further provided in the transmitting terminal 1 of the device
  • a channel estimation module 201 is also provided in the receiving terminal 2 of the device.
  • the device transmitting terminal 1 the original data generated by the random number generator 101 passes through the bit-level processing module 102 and the modulation module 103 successively to generate a modulated signal, and the modulated signal is then sparsely processed.
  • the transformation module 104 expresses it as a sparse signal based on the first neural network model.
  • the sparse signal passes through the compression multiplexing module 105.
  • the compression multiplexing module 105 compresses the transmitted signal and performs dimensionality reduction and multiplexing processing.
  • the used signal is transmitted through the transmitting antenna; the receiving end of the device 2:
  • the channel estimation module 201 performs channel estimation on the received signal, and the estimated channel state information is passed through the demultiplexing module 202, and the original output is reconstructed based on the second neural network model.
  • Send data stream is transmitted through the transmitting antenna; the receiving end of the device 2:
  • the channel estimation module 201 performs channel estimation on the received signal, and the estimated channel state information is passed through the demultiplexing module 202, and the original output is reconstructed based on the second neural network model.
  • the technical solution adopted by the present invention is a signal sparse design, compressed multiplexing matrix design, and signal detection method based on compressed sensing in a MIMO system.
  • the technical solutions are as follows:
  • the modulated l-channel signal x is the first neural network model generated by pre-training and is mapped to the sparse signal ⁇ m ⁇ 1 .
  • the first neural network model training algorithm uses the BP algorithm.
  • the sparse signal ⁇ m ⁇ 1 obtained after thinning is compressed into a ⁇ 1 signal through the compression multiplexing module 105, and then the compressed and multiplexed signal z is transmitted through the transmitting antenna.
  • the compression ratio ⁇ is determined by the size of the measurement matrix A in the compressed sensing technology.
  • a Gaussian random matrix is selected as the compressed sensing multiplexing matrix/compressed dimensionality reduction matrix A in the transmitter 1 of the device.
  • y is the received signal vector of Nr ⁇ 1, that is , the ⁇ l modulation symbol after compression and multiplexing received by the receiving end 2 of the device receiving N t transmitting antennas
  • n is the Gaussian white noise vector of Nr ⁇ 1, which The element is an independent and identically distributed complex Gaussian variable with a mean of 0 and a variance of 1.
  • H is a channel propagation matrix of Nr ⁇ Nt, and it is a deterministic matrix that remains constant within a coherent time interval.
  • the receiving end 2 of the apparatus may estimate the channel propagation matrix H according to the pilot signal inserted in the transmission data.
  • the decompression multiplexing process is realized by solving the following optimization problem, and the sent sparse vector is calculated and determined: min
  • 0 sty H ⁇ A ⁇ .
  • BCS Bayesian Compressive Sensing
  • the biggest advantage of the technical solution adopted by the present invention is to pass the thinning module 104 and the compressed sensing multiplexing module on the basis of the existing MIMO system.
  • the traditional signal transmission process is that the codeword is modulated and sent and then the receiving end is installed.
  • this application reduces the dimensionality of the signal to be processed, and can multiplex parallel data streams exceeding the number of transmitting antennas to a given transmitting antenna for transmission, thereby greatly improving
  • Given the signal multiplexing gain under the condition of the number of transmit and receive antennas of the MIMO system it can better meet the application requirements of the MIMO system for broadband transmission.
  • the BP neural network training method in artificial intelligence is combined to realize the sparse representation and reconstruction of the signal.
  • the compressed multiplexing matrix proposed by the present invention does not need to rely on channel state information.
  • the technical solution adopted by the present invention can add sparseness and compressed sensing complex at the transmitting end 1 of the device without modifying the existing MIMO technical solution. With steps, the receiving end 2 of the device adds a mature optimized reconstruction algorithm in the compressed sensing field to reconstruct a compressed multiplexed signal, which has the advantages of small modification to the existing MIMO system and convenient implementation.
  • the artificial intelligence method used in the present invention sparsely represents the signal, and reconstructs the original signal according to the sparse signal, ensuring the feasibility of compressing and transmitting multiple data streams, and taking into account the bit error rate.
  • Compressed sensing also known as compressed sampling or sparse sampling, is a method of finding sparse solutions for underdetermined linear systems. Compressed sensing is one of the most dazzling achievements in the field of signal processing since the beginning of the 21st century, and it has achieved effective applications in fields such as magnetic resonance imaging, image processing, and wireless communication systems.
  • the compressed sensing theory is different from the traditional Nyquist sampling theorem. As long as the signal x is compressible or sparse in a certain transform domain D, then an observation matrix A that is not related to the transform base D can be used to transform the high The one-dimensional sparse signal is projected onto a low-dimensional space, and then the original signal is reconstructed with high probability from these few projections by solving an optimization problem. Under this theoretical framework, the sampling rate is not determined by the bandwidth of the signal, but by the structure and content of the information in the signal. Compressed sensing theory mainly includes three aspects: signal sparse representation, coding sampling and reconstruction algorithm.
  • Neural network is a concept proposed by scientists led by Rumelhart and McClelland in 1986. It is a multi-layer feedforward neural network trained according to the error back propagation algorithm. At present, in the actual application of artificial neural networks, most of the neural network models adopt BP networks and their variations. It is also the core part of the forward network and embodies the essence of the artificial neural network.
  • BP algorithm to train the first neural network model (Neural Network Model, NN1) to realize the sparse representation of the signal.
  • observation matrix in the compressed sampling system Next, in the compressed sensing theory, it is necessary to design the observation matrix A of the compressed sampling system, how to sample a small amount of observations and ensure that the original signal can be reconstructed from it. Obviously, if the observation process destroys the information in the original signal, the reconstruction quality cannot be guaranteed. In order to ensure that the linear projection of the signal can maintain the original structure of the signal, the projection matrix must meet the Restricted Isometry Property (RIP) condition, and then the linear projection measurement of the original signal is obtained by the product of the original signal and the measurement matrix.
  • RIP Restricted Isometry Property
  • the RIP condition is defined as follows: If there is a constant ⁇ K ⁇ (0,1] for all signals ⁇ with a sparsity of K, the matrix A satisfies the following formula: It is said that the matrix A satisfies the constraint equidistant property of order K, where the sparsity K refers to the number of non-zero elements of the signal ⁇ .
  • the reconstruction algorithm of compressed sensing can still completely recover the original signal from the measurement number much smaller than the signal dimension.
  • ⁇ determines the reduction in the number of transmitting and receiving antennas and the performance of the reconstruction of the receiving end 2 of the device.
  • Greedy tracking algorithm This type of method gradually approximates the original signal by selecting a local optimal solution at each iteration.
  • These algorithms include "Sparse solution of underdo-termined linear equations by stagewise orthogonal matching pursuit proposed by Donoho," Technical Report, 2006 segmented OMP algorithm, etc.;
  • Bayesian compressed sensing reconstruction BCS algorithm This kind of method uses Bayesian prior to give a reasonable prior distribution of the solution signal, and then derive the original signal, such as "Bayesian compressive sensing using laplace priorors," IEEE Trans. Image Process, vol. 19, no. 1, pp. 53-63, 20 10 proposed BCS (Bayesian Compressive Sensing) algorithm, etc.;
  • each algorithm has its inherent shortcomings.
  • the convex relaxation method requires the least number of observations to reconstruct the signal, but it often has a heavy computational burden.
  • Greedy tracking algorithm is between these types of algorithms in terms of running time and sampling efficiency, and its anti-noise performance is unstable.
  • the appropriate reconstruction algorithm can be selected according to different environments. Once the sparse representation vector is obtained, the original signal can be recovered.
  • the codeword is modulated and sent, and the receiving end 2 of the device performs signal detection and reconstruction of the original signal.
  • the transmitting terminal 1 adds a thinning module 104 and a compression multiplexing module 105.
  • the modulated signal is expressed as a sparse signal through deep learning to realize the nonlinear expression of the signal; then the sparse signal is compressed and reduced in dimensionality.
  • the selection of the compressed dimensionality reduction matrix does not require channel state information, and the compressed sensing technology is selected.
  • the measurement matrix can be used as a signal compression matrix to complete the compression and dimensionality reduction and multiplexing processing of the transmitted signal.
  • the receiver 2 of the device is divided into two steps to reconstruct the signal: (1) Through the compressed sensing reconstruction algorithm, the high-dimensional sparse signal is solved from the low-dimensional received signal (2)
  • the second neural network model (Neural Network Model, NN2) trained by the deep learning algorithm, reconstructs the original signal
  • the present invention can be based on the existing related MIMO system signal multiplexing technology, by introducing the thinning module 104 and the compression multiplexing module 105, and the demultiplexing module 202 of the receiving end 2 of the device, compared with the traditional MIMO scheme , On the basis of reducing the number of required antennas, the same amount of data is transmitted at the same time, which improves the multiplexing gain and capacity of the MIMO system.
  • the existing MIMO spatial multiplexing technical solutions we no longer only focus on eliminating the interference of adjacent data, but more on how to ensure the detection performance of the receiving end 2 of the device under the condition of a given number of transmitting antennas. On this basis, more data streams are multiplexed and transmitted to the receiving end 2 of the device to obtain multiplexing gain and transmission capacity that exceed the inherent MIMO system.
  • This embodiment integrates the MIMO multi-antenna signal transmission and detection technology based on artificial intelligence and compressed sensing proposed in the present application, and gives a detailed description of the specific implementation steps of the present invention with examples.
  • the information source is generated by using the random number generator 101 to generate a 0,1 bit sequence.
  • Modulation is to modulate bit data, including BPSK, QPSK, 16QAM and 64QAM.
  • This embodiment uses BPSK modulation as an example for description.
  • S1 Signal processing at the transmitter 1 of the device.
  • the random number generator 101 is used to generate a 0,1 bit sequence to form the original data.
  • S12 Generate signal x after BPSK modulation.
  • the sending data of each group can be different.
  • the neural network is created by the BP neural network training method in artificial intelligence, and the transmission signal vector x of the device transmitter 1 and the sparse representation ⁇ determined based on the transmission signal vector x of the device transmitter 1 are respectively used as input and output, Construct a first training sample set, train the neural network, and obtain the first neural network model.
  • S2 Signal detection at the receiving end 2 of the device.
  • the signal matrix H is estimated.
  • the new matrix HA still obeys the Gaussian distribution and meets the conditions that the compressed sensing measurement matrix needs to meet.
  • the under-determined problem of compressed sensing can be solved by the reconstruction algorithm in compressed sensing.
  • the Bayesian compressed sensing BCS reconstruction algorithm is used, and the sparse representation vector is obtained according to the multiplexing matrix A and the channel matrix H known by the receiving end 2 of the device.
  • S23 Training the neural network model.
  • the input and output of the first training sample set are exchanged in order to form the second training sample.
  • the neural network is created by the BP neural network training method in artificial intelligence to be based on the sparse representation ⁇ and the transmission signal vector As input and output respectively, construct a second set of training samples, train the neural network, and obtain the second neural network model.
  • the neural network model is composed of three parts: the input layer (layer1), the middle layer (layer2, .., L-1), and the output layer (layerL).
  • the input layer plays the role of signal transmission and is responsible for receiving
  • each unit of the input layer represents a feature
  • the middle layer can be a single middle layer or multiple middle layers, which play the role of internal information processing and are responsible for information transformation
  • the output layer plays the role of outputting information to the outside, and each of the output layers
  • the unit represents a category.
  • a BP neural network is used to simulate a mapping function, which can map the input space data to the output space; the BP neural network will try to fit an input and output as the transmission signal vector x and the transmitting end 1 of the device.
  • a function between the sparse representation ⁇ determined based on the transmission signal vector x of the transmitting end 1 of the device, and the input and output are respectively the sparse representation ⁇ and the original transmission data stream.
  • the function between; based on the mapping function model generated by the trained BP neural network, the sparse representation ⁇ of the sparse original signal x on the over-complete redundancy dictionary D can be obtained according to the modulated transmission signal vector x of the transmitter 1 of the device, And according to the sparse representation ⁇ calculated by the Bayesian reconstruction algorithm, the original transmission data x is reconstructed.
  • a cost function is used to measure the difference between the output of the BP neural network and the real output, and the BP neural network is trained so that the output of the network input (the signal received by the device receiving end 2) through the neural network can be as close as possible to the theoretical output.
  • the gradient descent method is used in this application to solve the neural network parameters.
  • the optimal neural network weight is solved, the first neural network model or the second neural network model is established. Create a BP neural network, collect a large amount of sample data, and artificially label the correct classification results, and then use these labeled data to train the created neural network.
  • each layer in the neural network is constantly adjusting its weight and bias until it can accurately output the target value.
  • the weight parameter matrix between the layers of the neural network Indicates, where the superscript of the weight parameter W represents the number of layers, and the number of nodes in each of the two adjacent layers of the subscript.
  • the superscript of the weight parameter W represents the number of layers, and the number of nodes in each of the two adjacent layers of the subscript.
  • E.g Represents the weight of the line segment between the first node of Layer1 and the second node of Layer2 in the input layer. These weights determine the role of the model, and the goal of the neural network is to calculate the weights through samples.
  • E.g Represents the input value of the first node of Layer2 and brings it into the Logistic function to get the output
  • the input of the neural network is known as:
  • Y [y 1 y 2 ...y n ] T
  • the output of the neural network is: Introduce a nonlinear operator: It can be derived:
  • Training the neural network model The process of training the neural network model is mainly divided into two steps, one is to calculate the cost function J( ⁇ ), and the other is to adjust the parameter ⁇ to make the cost function value J( ⁇ ) as small as possible.
  • the forward propagation algorithm is used as follows to calculate the output of the sample under the current neural network model for each sample, find the cost function, and then update the weight parameter according to the output. Define the cost function J( ⁇ ), where m is the number of samples. Since the neural network has K outputs, its cost function also calculates the cost of K outputs accordingly.
  • the calculation formula is:
  • the back propagation algorithm is used to adjust the parameter ⁇ as follows, so that the cost function value J( ⁇ ) is as small as possible.
  • the backpropagation algorithm updates each weight coefficient by finding the partial derivative of the cost function with respect to each weight coefficient. For example, first, calculate the gradient of the last layer: (1) calculate the gradient of the cost function value to the nonlinear operator, (2) calculate the gradient of the output of the neural network to the bias and the weight between adjacent layers.
  • Figure 3 shows the bit error rate performance of different transceiver antenna configurations of the MIMO system after using compressed sensing and neural network signal transmission and detection technologies.
  • a flat fading channel is assumed here.
  • 4 transmitting antennas can only transmit 4 data symbols at the same time.
  • the BPSK modulation signal is first used to obtain the original signal x 4 ⁇ 1 , and the sparse signal ⁇ 16 ⁇ 1 is obtained through the first neural network model.
  • Device receiving end 2 also through 2 or 3 receiving antennas, using the BCS reconstruction algorithm and the second neural network model trained by the neural network to obtain the reconstructed transmission signal
  • the bit error rate performance of this scheme is shown in the figure (2 ⁇ 2)-4.
  • the first number in the brackets indicates the number of transmitting antennas
  • the second number indicates the number of receiving antennas
  • the last number indicates the length of the original data.
  • the number of receiving antennas has increased, and the scheme is recorded as (3 ⁇ 3)-4.
  • the solution proposed in this application can ensure a bit error rate while reducing the number of required transmitting and receiving antennas under the condition of high SNR.
  • Figure 4 shows the bit error rate performance of different transceiver antenna configurations of the MIMO system after using compressed sensing and neural network signal transmission and detection technologies.
  • a flat fading channel is assumed here.
  • 20 transmitting antennas can only transmit 20 data symbols at the same time.
  • the random Gaussian matrix A 4 ⁇ 16 is selected as the compressed dimension reduction matrix, thereby obtaining z i .
  • the solution proposed in this application can ensure the bit error rate while reducing the number of required transmitting and receiving antennas under the condition of high SNR. It can be seen that the proposed scheme can reduce the number of transceiver antennas required while ensuring the bit error rate.
  • the enhanced spatial multiplexing method proposed by the present invention can be combined with artificial intelligence on the basis of the existing MIMO system, and on the basis of reducing the number of antennas, the same amount of data can be transmitted and the number of antennas required can be reduced. , Improve multiplexing gain and system capacity.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Power Engineering (AREA)
  • Radio Transmission System (AREA)

Abstract

本发明涉及基于人工智能的MIMO多天线信号传输与检测装置，稀疏化模块，其利用第一神经网络模型对原始信号进行稀疏化表示；压缩复用模块，用于对稀疏表示后的信号进行压缩降维处理；装置接收端，用于对接收到的信号进行处理以实现其对目标信号的重构，所述装置还包括解复用模块，其中，所述解复用模块被配置为在装置接收端，利用压缩感知重构算法，从低维的信号中求解得到稀疏表示向量aa，最后，利用第二神经网络模型对其接收到的信号重构得到原始信号x。

Description

基于人工智能的MIMO多天线信号传输与检测技术 技术领域
本发明涉及移动通讯技术领域，尤其涉及基于人工智能的MIMO多天线信号传输与检测技术。
背景技术
通信技术研发可以认为是全人类共同的财富，但其带来的全球性的通信标准不仅仅是一项技术标准，而是关系到产业发展和国家战略。中国，经历2G时代的一无所有、3G时代初登舞台、4G时代陪跑，在5G时代，我们的目标是重量级参与者，这将对中国通信业发展和整个国民经济发展起到巨大推动作用。
MIMO(Multiple-Input Multiple-Output)又称为多入多出系统，指在发射端和装置接收端同时使用多个天线的通信系统，在不增加带宽的情况下可以成倍地提高通信系统的容量和频谱利用率。MIMO系统在发射端和装置接收端均采用多天线，传输信息流经过“空时编码”形成多个信息子流，多个子流同时发送到信道，各发射信号占用同一频带，因而并未增加系统带宽。若各发射接收天线间的信道响应独立，则多入多出系统可以创造多个并行空间信道，通过这些并行空间信道独立的传输信息，数据率必然会得到提高。
MIMO技术是无线通信领域的一个巨大突破，2011年，多家公司开发了基于MIMO技术的WIFI或WIMAX商用系统。2012年，所有的4G通信系统的标准(例如TD.LTE，LET.A，WIMAX等)都选用MIMO技术作为其关键技术之一。MIMO系统已经广泛应用于无线通信中——移动设备和网络普遍都会使用多根天线来增强连接性、提升网络速度和用户体验。大规模MIMO也是5G超高数据速率的关键因素，可以带来更大网络容量，更广信号覆盖和更好的用户体验，将5G的潜力发挥到一个全新的水平。
从方法论上来说，从4G的目标到5G的目标升级是巨大的，4G考虑的是连接人与人强调提高传输速率，而5G已经跳出之前的思维局限进而考虑人与物的互联，不但强调传输速率的提高，还考虑超大连接和超低时延。从社会学范畴来说，4G改变的是我们的生活，而5G改变的是我们的社会结构。针对5G的三大KPI，即超高带宽、超大连接和超低时延，为了实现以上目的就需要对现有的网络架构进行升级，并根据需求进行智能化的调整，因而人工智能(Artificial Intelligence，AI)可以给5G赋能。反过来，5G技术可以带来更多的应用需求，反过来加速AI的发展。
迄今，人工智能仍然没有一个明确且统一的定义。一种教科书式的定义是“AI就是根据对环境的感知，做出合理的行动，并获得最大收益的计算机程序”，另一种更技术导向的定义是“AI就是会学习的计算机算法”。从技术应用的角度，特别是人工智能在通信技术研发的角度，学界更倾向于第二种定义，那么可以从技术上将人工智能等同于机器学习。机器学习本质上是从数据中提取知识的关键技术，是人工智能发展的动力和引擎。机器学习目前主要解决分类问题、聚类问题、回归问题等，已广泛应用于字符识别、机器翻译、语音识别、搜索引擎、人脸识别、无人驾驶等领域。目前学习算法大大小小有上百种之多，尚不存在一个统一的框架来描述机器学习算法的设计过程，但目前机器学习有两大最强算法——深度学习和强化学习。
深度学习隶属于人工神经网络体系，是在传统神经网络发展下的新一代神经网络。神经网络是一门模仿人类神经中枢——大脑构造与功能的智能科学。具有快速反应能力，便于对事物进行实时控制与处理；卓越的自组织、自学习能力；善于在复杂的环境中的。充分逼近任意非线性系统，快速获得满足多种约束条件问题的最优化答案；具有高度的鲁棒性和容错能力等优越的性能，因此在通信系统中获得了越来越广泛的应用。神经外科是由多个非常简单的处理单元彼此按某种方式相互连接而形成的计算机系统，该系统靠其状态对不外输入信息的动态响应来处理信息。
围绕应用神经网络解决MIMO的问题，国内外也有不少发明专利成果。中国申请号201510473741.6专利(自复数神经网络信道预测方法，西安电子科技大学)公开了一种复数神经网络信道预测方法，主要解决MIMO系统中由于信道时变而导致的信道衰落问题。其技术方案是，1.基站对信道进行测量得到含有估计误差的信道系数训练序列；2.根据得到的信道系数序列得到相应的训练样本和期望输出；3.输入训练样本进行复数小波神经网络训练，得到最终的网络权值；4.基站利用训练后的复数小波神经网络进行信道系数预测。本方法简便易行，效果良好，适用于减少由信道时变对MIMO系统信道的影响。
中国申请号201810177829.7专利(一种基于神经网络的无线信道建模方法，东南大学)公开了一种基于神经网络的无线信道建模方法。该发明首先对用户反馈的接收信号进行处理，得到估计的信道参数；然后根据二维图像得到散射体的三维地理信息，并对它们进行聚类，最后将信道参数、地理信息作为神经网络的输入，接收信号作为输出，训练得到非线性时变的神经网络模型。该方法在可接受的复杂度内得到更加准确的信道模型，能够满足未来5G通信系统中采用的大规模MIMO技术、频带扩展、高移动性等场景的信道建模需求。
中国申请号201810267976.3专利(基于BP算法构建的深度神经网络大规模MIMO系统检测方法，东南大学)提供了一种基于BP算法构建的深度神经网络大规模MIMO系统检测方法，该方法通过将置信传播迭代算法因子图展开并映射到神经网络结构上来构建用于大规模MIMO系统检测的深度神经网络；深度神经网络的神经元对应于迭代算法因子图中的结点，各层神经元个数均等于迭代算法因子图中符号结点的个数；隐藏层之间的映射函数为迭代算法中置信信息的更新公式，隐藏层的数量等于迭代算法的迭代次数。具体地，该发明还提供了基于阻尼置信传播和最大和置信传播这两种信息传播迭代算法分别构建两种深度神经网络的MIMO检测方法。该发明在不增加在线运算复杂度的情况下，达到更低的误码率，并且对各种信道状况和天线配比都具有稳健性。
中国申请号201910063733.2专利(基于深度学习的优化MIMO检测方法，上海大学)通过构建MIMO端到端传输模型，根据MIMO装置接收端收到的信号y(t)和估计的不完美信道状态信息得到模型的复时域向量作为深度神经网络(DNN)的输入，利用DNN得到装置发射端比特流的估计值与现有技术根据做硬判决得到发送比特流的估计值相比，该发明能够在不完美的信道信息下，提高准确度和检测速率，保证在低复杂度算法下实现低误比特率的检测性能，同时在含有固有信道误差的情况下具有很好的鲁棒性。
中国申请号201610327115.0专利(一种大规模MIMO下基于深度学习的码本选择方法，重庆邮电大学)涉及一种大规模MIMO下基于深度学习的码本选择方法。该方法包括：采集测试区的导频信息构建导频训练序列，进而得到导频训练样本；对导频训练样本进行神经网络迭代学习，得到最终的网络权重值；根据学习后的神经网络输出的信道，从完备码本中选出最优码字。之后将未知区与测试区进行信道信息匹配，得到其无线信道，进而得到与无线信道对应的码字。该发明能有效、准确、快速地建立无线信道模型与码本查询，避免了未知区的信道估计且大大降低了未知区信道选择码本的复杂 度。
中国申请号201811626005.X专利(基于改进梯度投影法的低复杂度MIMO-NOMA系统信号检测方法，重庆邮电大学)公开一种基于改进梯度投影法的低复杂度MIMO NOMA系统信号检测方法，涉及无线通信技术。根据系统活跃用户的稀疏特性，利用凸优化算法思想，将系统模型转化为严格的二次规划问题；然后对该问题进行迭代求解，并对每次迭代结果进行预处理操作，达到对活跃用户及其信号有效的检测。该发明突破了传统检测方法中算法收敛速度慢的问题，对每次迭代结果进行预处理操作，不仅可使检测结果快速收敛，而且还能检测出活跃用户集合，其实现过程简单，应用范围广泛。
中国申请号201910014714.0专利(一种基于深度学习的MIMO系统波束成形矩阵设计方法，南京邮电大学)公开了一种基于深度学习的MIMO系统波束成形矩阵设计方法，步骤如下，首先利用已知算法获得深度学习网络所需要的训练样本集；接着构建深度学习神经网络模型，初始化模型相关参数并利用训练样本集进行训练；然后利用导频获取信道送入神经网络预测出波束成形矩阵系数，最后将信道与波束成形矩阵系数组合，构成波束成形矩阵。该方法利用深度学习神经网络获得的波束成形矩阵能够同时兼顾性能与算法复杂度，可以在保证性能的前提下降低时延，使得MIMO系统能提供实时服务。
中国申请号201810182937.3专利(一种基于机器学习的MIMO链路自适应传输方法，东南大学)公开了一种基于机器学习的MIMO链路自适应传输方法，使用非监督学习自编码算法进行特征的提取与降维，引入深度学习的思想，能够在保留主要信息状态信息的前提下降低特征维度与计算复杂度。该发明利用逻辑回归算法构建信道状态信息与传输参数间的映射关系，不同于以往固定的参数化模型，能够基于样本数据进行训练，在数据集质量较优，覆盖所有状态的情况下，能够更好的建立信道状态信息与传输参数间的映射关系，并相较于传统的单一等效信噪比，能够更充分地利用信道状态信息。此外，该发明还基于信道矩阵进行CQI选择，通过信道矩阵和噪声方差研究基于机器学习的MIMO链路自适应方法不受接收机设计的约束，具有普适性。
中国申请号201710495044.X专利(一种基于深度学习的MIMO系统联合预编码和天线选择方法，浙江理工大学)公开了一种基于深度学习的MIMO系统联合预编码和天线选择方法，包括以下步骤：首先通过已有的天线选择方法产生深度学习所需要的训练数据集；接着，建立深度学习模型，利用训练数据训练深度学习模型并保存；然后利用保存的深度学习模型完成天线选择；最后对所选择的MIMO子系统进行最优预编码设计。该发明利用深度学习技术设计MIMO系统联合预编码和天线选择，能够在获得良好系统信噪比的情况下实现较低的计算复杂度。
中国申请号201910242525.9号发明专利(一种面向高铁的深度信号检测方法，深圳大学)提出了一种面向高铁的深度信号检测方法，首先，收集数据，依据高铁沿线不同环境类型，在高铁沿线各个场景内收集若干发送信号和接收信号；其次，划分场景，通过数据分析将每个场景进一步划分为多个区域以满足神经网络的兼容性；再次，建立深度高铁信号检测神经网络模型；接着，离线训练高铁信号检测神经网络；最后，进行在线实时检测信号，高铁在行驶过程中先通过GPS确定其位置信息，判断其所在的区域，并选择相应的神经网络模型，接着将实时接收到信号输入训练好的神经网络中，实时输出基站端发送信号。该发明的系统性能有了很大的提高，信号检测误比特率降低，算法更具鲁棒性。该发明所使用的方法无需对信道进行估计，节省了导频开销。
中国申请号201810279530.2号发明专利(基于BP神经网络的可见光通信MIMO抗扰降噪方法，中南民族大学)公开了一种基于BP神经网络的可见光通信MIMO抗扰 降噪方法，涉及可见光通信领域的MIMO天线技术。本系统包括依次连通的系统发射端、系统装置接收端信号处理部分和BP神经网络信号处理部分。本方法是：1)电信号加载到LED阵列上，以光信号的形式发射出去；2)装置接收端光电探测器将光信号转为电信号；3)多路电信号经低通滤波器去除高频干扰；4)BP神经网络经过训练后对多路信号进行降噪以及消除干扰的处理，最后通过并串转换成为二进制串行数据流。该发明改进了现有MIMO技术的传输性能；将神经网络与可见光MIMO技术结合，发挥神经网络在无线通信中去扰降噪方面的优势；采用神经网络接收处理技术使整个VLC系统更加稳定。
中国申请号201710213235.2发明专利(基于正交映射与概率神经网络的可见光信道联合均衡方法，中山大学)公开一种基于正交映射与概率神经网络的可见光信道联合均衡方法，包括装置发射端和装置接收端，信号通过可见光MIMO信道从发射端传送到装置接收端；所述的可见光MINO信道为多输入多输出信道；所述的联合均衡为前置均衡与后置均衡相结合；该发明采用的是前置均衡技术与后置均衡技术相结合的联合均衡方案，即基于正交映射与概率神经网络的可见光多输入多输出信道联合均衡方法，能有效抑制可见光MIMO通信系统信道之间的干扰，提高数据传输可靠性。
中国201910125325.5号发明专利(一种基于深度学习的MIMO解码方法、装置及存储介质，深圳市宝链人工智能科技有限公司)实施例公开了一种基于深度学习的MIMO解码方法、装置及存储介质，通过构建MIMO解码的训练数据集合，训练数据集合中包括多个训练数据；然后基于训练数据集合对神经网络进行训练，得到训练完成的神经网络模型；在接收到待解码MIMO信号时，将待解码MIMO信号输入至神经网络模型进行MIMO解码，然后得到神经网络模型所输出的MIMO解码结果。通过该发明的实施，基于深度学习来设计用于联合MIMO检测和信道解码的神经网络模型，将MIMO检测和信道解码视为联合解码过程，并且通过训练来改善神经网络模型所输出结果的近似性，保证了MIMO解码的整体性能，具有更高的解码准确性以及更快的解码速度。
中国申请号201810757547.4专利(一种机器学习辅助的大规模MIMO下行用户调度方法，东南大学)公开了一种机器学习辅助的大规模MIMO下行用户调度方法，包括以下步骤：S1：基站通过用户发送的上行探测信号获取特征方向上的特征模式能量耦合矩阵；S2：基站利用特征模式能量耦合矩阵，通过机器学习的方法辅助进行各种用户和波束组合下的和速率计算；S3：采用贪婪算法实现和速率最大准则的用户调度，获取最优用户波束配对组合。该发明通过上行探测信号获取统计信道信息，采用和速率最大化准则进行用户调度。在基站仅有统计信道信息的情况下，通过有针对性的特征提取以及神经网络的设计，精确地实现和速率的近似计算，极大地降低大规模天线下用户调度的复杂度，并且性能接近最优，具有较好的适用性和鲁棒性。
中国201610353881.4号发明专利(一种基于机器学习算法的MIMO相关信道下的调制识别方法，北京邮电大学)是一种基于机器学习算法的MIMO相关信道下的调制识别方法，属于通信领域；具体步骤为：首先，将通信发射端的每个数据流分别采用空时编码，每个码字分别通过Nt根发射天线发射出去；然后，根据装置接收端的相关矩阵和发射端的相关矩阵计算MIMO信道矩阵H；根据MIMO信道矩阵H，计算每个接收天线上的接收信号并进行修正；最后，每根接收天线分别对修正后的信号进行特征提取，针对提取的特征值进行训练测试，计算出该样本属于的调制识别模式；优点在于：对非高斯信道的鲁棒性和泛化能力较强，通过参数迭代可实现更加复杂环境下的调制体制识别；通过提取高阶矩和高阶累积量的特征，在较高信噪比下，信号特征差异明显，便于机器学习算法的分类。
MIMO技术的应用，使空间成为一种可以用于提高性能的资源，并能够增加无线系 统的覆盖范围。MIMO技术已经成为无线通信领域的关键技术之一，通过近几年的持续发展，MIMO技术已经越来越多地应用于各种无线通信系统。随着使用天线数目的增加，MIMO技术实现的复杂度大幅度增高，从而限制了天线的使用数目，不能充分发挥MIMO技术的优势。同时，从1956年正式提出人工智能学科算起，50多年来，取得长足的发展，成为一门广泛的交叉和前沿科学。作为人工智能的一个重要分支，从上面的描述可以看出，神经网络通过计算机来实现数学上的网络，是公认解决目前通信遇到的部分瓶颈问题的“出路”。
综合目前国内外围绕神经网络和通信问题关键技术的研究，目前已有不少的研究成果，这些成果已经从MIMO的信道估计、检测信号等角度提出了可行的解决方案。如何在保证一定的系统性能的基础上提高系统容量、降低MIMO技术的算法复杂度和实现复杂度，成为业界面对的巨大挑战。然而现有围绕MIMO条件下的相关研究还没有从给定发送天线和接收天线数的条件下，结合人工智能技术提高系统复用增益这个角度来研究信号传输与检测的方案。
此外，一方面由于对本领域技术人员的理解存在差异；另一方面由于发明人做出本发明时研究了大量文献和专利，但篇幅所限并未详细罗列所有的细节与内容，然而这绝非本发明不具备这些现有技术的特征，相反本发明已经具备现有技术的所有特征，而且申请人保留在背景技术中增加相关现有技术之权利。
发明内容
针对如何在保证一定的系统性能的基础上提高系统容量、降低MIMO技术的算法复杂度和实现复杂度的问题，现有围绕MIMO条件下的相关研究还没有从给定发送天线和接收天线数的条件下，结合人工智能技术提高系统复用增益这个角度来研究信号传输与检测的方案。
本申请结合人工智能技术与压缩感知技术的最新研究进展，提出了一种基于人工智能和压缩感知技术的MIMO系统中信号的复用传输与检测方案，一方面，与传统的MIMO技术方案相比，本申请通过引入装置发射端的稀疏化模块和压缩感知复用模块，降低了待处理信号的维度，可将超过发送天线数目的并行数据流复用到给定的发射天线上发送出去，从而大幅度的提高给定MIMO系统收发天线数条件下的信号复用增益，更好地满足MIMO系统对宽带传输的应用要求；另一方面，本发明所提出的压缩复用矩阵不需要依赖信道状态信息，本发明所采用的技术方案可以在不修改现有MIMO技术方案的基础上，在装置发射端增加稀疏化和压缩感知复用步骤，在装置接收端增加压缩感知领域成熟的优化重构算法，即可重构出压缩复用信号，对现有MIMO系统修改小，具有实现方便的优点。
本申请所提出的基于人工智能和压缩感知技术的MIMO装置主要由装置发射端的稀疏化模块和压缩复用模块、以及装置接收端的解复用模块所构成。此外优选地，装置发射端中还设置有随机数发生器(或称原始信息比特生成模块)、比特级处理模块、调制模块，装置接收端中还设置有信道估计模块。
优选地，如图5所示，装置发射端：由随机数发生器产生的原始数据先后经过比特级处理模块和调制模块，生成调制后的信号，经调制后的信号继而经过稀疏化模块，基于第一神经网络模型将其表达为稀疏信号，经过稀疏化后的信号再经过压缩复用模块，压缩复用模块对发送信号进行压缩降维和复用处理，经压缩复用后的信号经发射天线发射；装置接收端：由信道估计模块对接收到的信号进行信道估计，估计的信道状态信息 经过解复用模块，基于第二神经网络模型重构输出原始的发送数据流x。
上述过程具体包括以下内容：
基于人工智能的MIMO多天线信号传输与检测装置，该装置包括：稀疏化模块，其利用第一神经网络模型对原始信号进行稀疏化表示；压缩复用模块，用于对稀疏表示后的信号进行压缩降维处理；装置接收端，用于对接收到的信号进行处理以实现其对目标信号的重构，所述装置还包括解复用模块，其中，所述解复用模块被配置为在装置接收端，利用压缩感知重构算法，从低维的信号中求解得到稀疏表示向量
Figure PCTCN2020083564-appb-000001
最后，利用第二神经网络模型对其接收到的信号重构得到原始信号x。
根据一种优选实施方式，所述第二神经网络模型的输入输出与用于装置发射端的稀疏化模块从低维的信号中解出待传输至压缩复用模块进行处理的稀疏信号θ的第一神经网络模型的输入输出相反。
根据一种优选实施方式，所述装置接收端还包括信道估计模块，其被配置为对所述装置接收端收到的信号进行信道估计并将其得到的信道参数矩阵作为所述压缩感知重构算法的输入。
根据一种优选实施方式，所述解复用模块是根据无需依赖信道状态信息的压缩感知复用矩阵和信道参数矩阵来求解得到稀疏表示向量
Figure PCTCN2020083564-appb-000002
的。
根据一种优选实施方式，所述稀疏化模块通过人工智能中的BP神经网络训练方法创建神经网络，以所述装置发射端的发送信号向量x和基于所述装置发射端的发送信号向量x所确定的稀疏表示θ分别作为输入输出，构造第一组训练样本集，训练所述神经网络，得到所述第一神经网络模型。
根据一种优选实施方式，所述解复用模块通过将所述发送信号向量x的所有可能组合分别作为冗余字典的不同列向量的方式组成冗余字典D，实现所述发送信号向量的稀疏表示θ，即x＝Dθ。
根据一种优选实施方式，经稀疏化后的稀疏表示θ经过所述压缩复用模块，压缩成ρl路信号，然后所述装置将压缩复用后的信号Z经发射天线发射。
根据一种优选实施方式，压缩复用后的信号Z是通过计算式z＝Aθ来得到的，其中，A是N t行m列的压缩降维矩阵，ρ＝N t/l∈(0,1]代表压缩比例。
基于人工智能的MIMO多天线信号传输与检测系统，所述系统通过构建MIMO端到端传输模型，装置发射端通过其所构建的第一神经网络模型来从低维的信号中解出在传输至压缩复用模块进行处理前的稀疏信号，装置接收端利用其所构建的与稀疏化模块的第一神经网络模型的输入输出相反的第二神经网络模型对其接收到的信号重构得到原始信号。
根据一种优选实施方式，所述第二神经网络模型的输入是利用压缩感知重构算法所求解得到的稀疏表示向量
Figure PCTCN2020083564-appb-000003
附图说明
图1是本发明提供的基于人工智能和压缩感知技术的MIMO多天线系统的信号传输与检测系统的信号处理流程的示意框图；
图2是本发明提供的优选的信号压缩复用与检测处理过程示意框图；
图3是不同收发天线配置下经典检测算法ZF和本发明MIMO系统多天线信号传输与检测技术的误码率性能曲线；
图4是不同收发天线配置下经典检测算法ZF和本发明MIMO系统多天线信号传输 与检测技术的误码率性能曲线；和
图5是本发明提供的基于人工智能和压缩感知技术的MIMO多天线系统的信号传输与检测系统的模块连接示意图。
附图标记列表
1：装置发射端                     101：随机数发生器
102：比特级处理模块               103：调制模块
104：稀疏化模块                   105：压缩复用模块
2：装置接收端                     201：信道估计模块
202：解复用模块
具体实施方式
下面结合附图对本发明进行详细说明。
本申请所提出的基于人工智能和压缩感知技术的MIMO装置主要由装置发射端1的稀疏化模块104和压缩复用模块105、以及装置接收端2的解复用模块202所构成。此外优选地，装置发射端1中还设置有随机数发生器101(或称原始信息比特生成模块)、比特级处理模块102、调制模块103，装置接收端2中还设置有信道估计模块201。
优选地，如图5所示，装置发射端1：由随机数发生器101产生的原始数据先后经过比特级处理模块102和调制模块103，生成调制后的信号，经调制后的信号继而经过稀疏化模块104，基于第一神经网络模型将其表达为稀疏信号，经过稀疏化后的信号再经过压缩复用模块105，压缩复用模块105对发送信号进行压缩降维和复用处理，经压缩复用后的信号经发射天线发射；装置接收端2：由信道估计模块201对接收到的信号进行信道估计，估计的信道状态信息经过解复用模块202，基于第二神经网络模型重构输出原始的发送数据流
Figure PCTCN2020083564-appb-000004
如图2所示，本发明采用的技术方案是一种MIMO系统中基于压缩感知的信号稀疏化设计、压缩复用矩阵设计和信号检测方法，其技术方案如下所述：
1.装置发射端1信号处理：
对于一个配有N t根发送天线和N r根接收天线的MIMO通信系统。系统发射端经过信道编码、信号调制后得到l路信号x。经调制后的l路信号x经过提前训练产生的第一神经网络模型，映射为稀疏信号θ m×1。其中，第一神经网络模型训练算法采用BP算法。
2.装置发射端1的压缩复用处理：
经过稀疏化后得到的稀疏信号θ m×1，经过压缩复用模块105，压缩成ρl路信号，然后将压缩复用后的信号z经发射天线发射。其中压缩复用模块105对输入信号的压缩处理可以表示为：z＝Aθ，其中，A是N t行m列的压缩降维矩阵，ρ＝N t/l∈(0,1]代表压缩比例。压缩比例ρ由压缩感知技术中的测量矩阵A的尺寸决定。优选地，装置发射端1中选择高斯随机矩阵作为压缩感知复用矩阵/压缩降维矩阵A。
3.装置接收端2的信号检测：
装置接收端2链路大致是装置发射端1链路的逆过程，装置接收端2接收到的信号为：y＝Hz+n＝HAθ+n。其中，y是Nr×1的接收信号向量，即为装置接收端2收到N t根发射天线发射的经压缩复用后的ρl路调制符号；n是Nr×1的高斯白噪声向量，其元素是均值为0、方差为1的独立同分布复高斯变量；H是Nr×Nt的信道传播矩阵，并且是一个确定性的、在一个相干时间间隔内都保持不变的矩阵。装置接收端2可以是根据发 送数据中所插入的导频信号估计出信道传播矩阵H。
假设信道矩阵H服从高斯分布，则可以证明新的矩阵HA依然服从高斯分布，满足压缩感知测量矩阵需要满足的条件。根据压缩感知复用矩阵HA，通过求解下述优化问题实现解压缩复用过程，计算和确定发送的稀疏向量：min||θ|| 0 s.t.y＝H·A·θ。这里可以采用压缩感知技术中基于贝叶斯的压缩感知(Bayesian Compressive Sensing,BCS)重构算法求解得到稀疏信号
Figure PCTCN2020083564-appb-000005
4.根据深度学习的训练方法，训练产生第二神经网络模型，稀疏信号
Figure PCTCN2020083564-appb-000006
被映射为
Figure PCTCN2020083564-appb-000007
最后，对
Figure PCTCN2020083564-appb-000008
进行解调，还原出原始的发送数据流。
与现有技术相比，本发明的有益效果是：
一、本发明所采用的技术方案的最大优点是在现有MIMO系统的基础上通过稀疏化模块104和压缩感知复用模块，传统的信号传输流程是码字经过调制后发送然后装置接收端2进行信号检测重构原始信号，与之相比，本申请降低了待处理信号的维度，可将超过发送天线数目的并行数据流复用到给定的发射天线上发送出去，从而大幅度的提高给定MIMO系统收发天线数条件下的信号复用增益，更好地满足MIMO系统对宽带传输的应用要求。同时，结合人工智能中的BP神经网络训练方法，实现信号的稀疏表示和重构。
二、本发明所提出的压缩复用矩阵不需要依赖信道状态信息，本发明所采用的技术方案可以在不修改现有MIMO技术方案的基础上，在装置发射端1增加稀疏化和压缩感知复用步骤，装置接收端2增加压缩感知领域成熟的优化重构算法的基础上，即可重构出压缩复用信号，对现有MIMO系统修改小，具有实现方便的优点。
三、本发明所采用的人工智能的方法稀疏化表示信号，且根据稀疏信号重构原始信号，确保了压缩传输多路数据流的可行性，兼顾了误码率。
如下对上述过程的具体实现步骤进行说明：
针对“信号的稀疏表示”：压缩感知(Compressed sensing)，也被称为压缩采样或稀疏采样，是一种寻找欠定线性系统的稀疏解的方法。压缩感知是信号处理领域进入21世纪以来取得的最耀眼的成果之一，并在磁共振成像、图像处理、无线通信系统等领域取得了有效应用。
将模拟信号转换为计算机能够处理的数字信号，必然要经过采样的过程。为了保证信号的完整性，奈奎斯特给出了答案——采样频率应该是信号最高频率的两倍。一直以来，奈奎斯特采样定律被视为数字信号处理领域的金科玉律。Candès最早意识到了突破的可能，并在陶哲轩以及Donoho的协助下，提出了压缩感知理论，该理论认为：如果信号是稀疏的，那么它可以由远低于采样定理要求的采样点重建恢复。
压缩感知理论与传统奈奎斯特采样定理不同，只要信号x是可压缩的或在某个变换域D是稀疏的，那么就可以用一个与变换基D不相关的观测矩阵A将变换所得高维的稀疏信号投影到一个低维空间上，然后通过求解一个优化问题从这些少量的投影中以高概率重构出原信号。在该理论框架下，采样速率不决定于信号的带宽，而决定于信息在信号中的结构和内容。压缩感知理论主要包括信号的稀疏表示、编码采样和重构算法三个方面。由于自然界中普遍存在的信号一般都不是稀疏的，信号的稀疏表示就是将信号投影到某个变换域D时，只有少数元素是非零的，则称所得到的变换向量是稀疏或者近似稀疏的，即x＝Dθ，θ是原始信号x的一种简洁表达，这是压缩感知的先验条件。
如何找到信号最佳的稀疏域是压缩感知理论应用的基础和前提，只有选择合适的基表示信号才能保证信号的稀疏度，从而保证信号的恢复精度。在研究信号的稀疏表示时，可以通过变换系数衰减速度来衡量变换基的稀疏表示能力。Candès等在文献”Near o ptimal signal recovery from random projections:Universal encoding strategies，”IEEE Trans.Information Theory，vol.52，no.12，pp.5406-5425，2006中指出表明，满足具有幂次速度衰减的信号，可利用压缩感知理论得到恢复。最近几年，对稀疏表示研究的热点是信号在冗余字典下的稀疏分解。这是一种全新的信号表示理论：用超完备的冗余函数库取代基函数，称之为冗余字典D，字典中的元素被称为原子。Temlyakov在文献”Nonlinear Methods of Approximation，IMI Research Reports，Dept.of Mathematics，University of South Carolina，2001中指出字典D的选择应尽可能好地符合被逼近信号的结构，其构成可以没有任何限制。从冗余字典中找到具有最佳线性组合的少量原子来表示一个信号x＝Dθ，称作信号的稀疏逼近或高度非线性逼近，其中，θ只有K项元素是非零的。过完备冗余字典D的构成应尽量符合信号本身所固有的特性，对信号的稀疏表示非常重要。过完备冗余字典的结构越逼近信号的特性，所需要的原子越少，θ越稀疏，所需的测量数越少，重构性能越精确。从理论上来说，总是可以找到一个变换域D，实现信号的稀疏表示。
深度学习的概念源于人工神经网络的研究。神经网络是1986年由Rumelhart和Mc Clelland为首的科学家提出的概念，是一种按照误差逆向传播算法训练的多层前馈神经网络。目前，在人工神经网络的实际应用中，绝大部分的神经网络模型都采用BP网络及其变化形式。它也是前向网络的核心部分，体现了人工神经网络的精华。这里，我们采用BP算法训练出第一神经网络模型(Neural Network Model，NN1)，实现信号的稀疏表示。
针对“压缩采样系统中的观测矩阵”：接下来在压缩感知理论中，需要设计压缩采样系统的观测矩阵A，如何采样得到少量的观测值，并保证从中能重构出原始的信号。显然，如果观测过程破坏了原始信号中的信息，重构质量是不可能得到保证的。为了确保信号的线性投影能够保持信号的原始结构，投影矩阵必须满足约束等距性(Restricted Isometry Property，RIP)条件，然后通过原始信号与测量矩阵的乘积获得原始信号的线性投影测量。RIP条件定义如下：如果存在常数δ K∈(0,1]对所有稀疏度为K的信号θ，矩阵A满足下式：
Figure PCTCN2020083564-appb-000009
则称矩阵A满足阶数为K的约束等距性质，其中稀疏度K是指信号θ的非零元素的个数。A是N t行l列的压缩降维矩阵。压缩感知技术的优势在于即使N t<l(l是指信号的长度)，我们依然可以从N r(N r＝N t)次测量值中恢复出长度为l的原始数据。令ρ＝N t/l∈(0,1]代表压缩比例。根据压缩感知原理，只要测量A满足RIP，即使A是行数远远小于列数的矩阵，将信号θ投影到了一个维度减少的空间上，依然可以通过压缩感知的重构算法从远小于信号维度的测量数中完整地恢复出原始信号。ρ决定了能够减少的发送和接收天线数和装置接收端2重构的性能。M.Davenport在其博士论文“Random observation on random observations:Sparse signal acquisition and processing”的定理3.5中指出：A是满足2K阶RIP常数δ 2K∈(0,1]的矩阵，只要
Figure PCTCN2020083564-appb-000010
C是约等于0.28的常数，则可恢复出原始信号。同时，Donoho在文献“Extensions of compressed sensing，”Signal Processing，vol.86，no.3，pp.533-548，2006中给出了观测矩阵所必需具备的三个条件，并指出大部分一致分布 的随机矩阵都具备这三个条件，均可作为观测矩阵，如：部分Fourier集、部分Hadamard集、一致分布的随机投影(uniform Random Projection)集等。文献“decoding by linear programming，”IEEE Transactions on Information Theory，vol.51，no.12，pp.4201-4215，2005和”Stable signal recovery from incomplete and inaccurate mea surements，”Communications on Pure and Applied Mathematics，vol.59，no.8，pp.1207-1223，2006证明当测量矩阵A是高斯随机矩阵时，A能以较大概率满足RIP。所以在本申请中选择高斯随机矩阵作为压缩感知复用矩阵A。
针对“装置接收端2的信号重构”：最后通过压缩感知中的重构算法解决求解该欠定问题。Donoho在文献”For most large underdetermined systems of linear equations，the minimal l 0-norm solution is also the sparsest solution，”Communications on Pure and Applied Mathematics，vol.59，no.6，pp.797-829，2006中指出，最小l 0范数问题是一个NP-hard问题，需要穷举θ中非零值的所有排列可能，因而无法求解。鉴于此，本领域研究人员提出了一系列求得次最优解的算法，主要包括以下四大类：
(1)贪婪追踪算法：这类方法是通过每次迭代时选择一个局部最优解来逐步逼近原始信号。这些算法包括文献Donoho提出的”Sparse solution of underdo-termined linear equations by stagewise orthogonal matching pursuit，”Technical Report，2006分段OMP算法等；
(2)凸松弛法：这类方法通过将求解l 0范数的非凸问题转化为凸问题求解找到信号的逼近，如文献”A method for large scale regularized least squares，”IEEE Journal on Selected Topics in Signal Processing，vol.4，no.1，pp.606-617，2007提出的内点法、文献”Gradient projection for sparse reconstruction:Application to compressed sensing and other inverse problems，”Journal of Selected Topics in Signal Processing:Special Issue on Convex Optimization Methods for Signal Processing，vol.1，no.4，pp.586-598，2007提出的梯度投影方法、Daubechies在”Aniterative thresholding algorithm for linear inverse problems with a sparsity constraint，”Comm.Pure Appl.Math.，vol.57，no.11，pp.1413-1457，2004一文中提出的迭代阈值法等；
(3)贝叶斯压缩感知重构BCS算法：这类方法运用贝叶斯先验，给求解信号一个合理的先验分布，然后推导出原信号，如”Bayesian compressive sensing using laplace priors，”IEEE Trans.Image Process，vol.19，no.1，pp.53-63，20 10提出的BCS(Bayesian Compressive Sensing)算法等；
(4)组合算法：这类方法要求信号的采样支持通过分组测试快速重建，如文献“Improved time bounds for near optimal sparse Fourier representation，”Proceedings of SPIE，Wavelets XI，Bellingham WA:International Society for Optical Engineering，2005提出的傅立叶采样、文献”One sketch for all:Fast algorithms for compressed sensing，”Proceedings of the 39th Annual ACM Symposium on Theory of Computing，New York:Association for Computing Machiner，pp.237-246，2007提出的HHS(Heavg Hitters on Steroids)追踪等。
如上可以看出，每种算法都有其固有的缺点。凸松弛法重构信号所需的观测次数最少，但往往计算负担很重。贪婪追踪算法在运行时间和采样效率上都位于这几类算法之间，并且抗噪性能不稳定。可以根据不同的环境选择合适的重构算法，一旦得到稀疏表示向量，就可以恢复出原始的信号。
针对“装置接收端2的信号重构”：如图1所示，与传统的信号传输流程：码字经过调制后发送，装置接收端2进行信号检测重构原始信号相比，本申请在装置发射端1增加了稀疏化模块104和压缩复用模块105。首先，通过深度学习将调制后的信号表达为稀疏信号，实现信号的非线性表达；然后将稀疏信号进行压缩降维处理，压缩降维矩阵的选择不需要信道状态信息，选择压缩感知技术中的测量测量矩阵即可作为信号压缩矩阵，完成对发送信号的压缩降维和复用处理。装置接收端2分为两步重构信号：(1)通过压缩感知重构算法，从低维的接收信号中解出高维的稀疏信号
Figure PCTCN2020083564-appb-000011
(2)通过深度学习算法训练出来的第二神经网络模型(Neural Network Model，NN2)，重构出原始信号
Figure PCTCN2020083564-appb-000012
本发明可以在现有的相关MIMO系统信号复用技术的基础上，通过引入稀疏化模块104和压缩复用模块105，以及装置接收端2的解复用模块202，和传统的MIMO方案相比，在减少所需天线数的基础上，同时传输相同的数据量，提高MIMO系统的复用增益和容量。与现有的MIMO空间复用技术方案相比，我们不再仅仅关注于消除相邻数据的干扰，而更关注于如何在给定发送天线数的条件下，在保证装置接收端2检测性能的基础上，将更多的数据流复用传输到装置接收端2，获得超过MIMO系统固有的复用增益和传输容量。
实施例
本实施例融合本申请提出的基于人工智能和压缩感知的MIMO多天线信号传输与检测技术，对本发明的具体实施步骤进行举例详细说明。
首先，信息源的产生是采用随机数发生器101产生0,1比特序列。
调制是对比特数据进行调制，包括BPSK、QPSK、16QAM和64QAM等。
本实施例采用BPSK调制作为例子进行说明。
根据如图2所示出的信号在装置发射端1和装置接收端2的处理流程，具体步骤如下：
S1：装置发射端1的信号处理。
S11：采用随机数发生器101产生0,1比特序列，构成原始数据。
S12：经过BPSK调制，产生信号x。每一组的发送数据都可以不相同。
S13：训练神经网络模型。通过人工智能中的BP神经网络训练方法创建神经网络，以所述装置发射端1的发送信号向量x和基于所述装置发射端1的发送信号向量x所确定的稀疏表示θ分别作为输入输出，构造第一组训练样本集，训练所述神经网络，得到所述第一神经网络模型。
当发送信号为x时，装置接收端2信号为y，同时，从理论上来说，我们总可以找到一个合适的基，实现信号的稀疏表示。Temlyakov在文献”Nonlinear Methods of Approximation,IMI Research Reports,Dept.of Mathematics,University of South Carolina,2001中指出字典D的选择应尽可能好地符合被逼近信号的结构，其构成可以没有任何限制。这里将x的所有可能组合，分别作为冗余字典D的不同列向量，组成冗余字典D，找到x对应的稀疏表示：除了相应索引位置为1，其他位置均为0的稀疏向量θ。
S14：信号的稀疏表示。经由人工智能中的BP神经网络训练方法，得到第一神经网络模型，当输入为信号x时，输出对应的稀疏信号θ。
S15：压缩复用处理，稀疏信号θ发送数据乘以压缩复用矩阵A，得到数据向量z，即z＝Aθ。
这里所提及的压缩复用矩阵A选择为高斯矩阵。其中，ρ＝N t/l代表压缩比，表示天 线减少数量的比例。
S16：将数据经信道发射出去。
S2：装置接收端2的信号检测。
S21：接收到的信号为y＝Hz+n＝HAx+n，其中，n表示噪声。根据信道估计模块201，估计信号矩阵H。
S22：根据已知的复用矩阵A和信道矩阵H，求解下述优化问题，计算稀疏向量θ：
min||θ|| 0 s.t.y＝HA·θ
上式中，||·|| 0为向量的l 0范数，表示稀疏向量θ中非零元素的个数。
假设信道矩阵服从高斯分布，可以证明新的矩阵HA，依然服从高斯分布，满足压缩感知测量矩阵需要满足的条件。压缩感知的欠定问题可以通过压缩感知中的重构算法解决。本实施例中采用贝叶斯压缩感知BCS重构算法，根据装置接收端2已知的复用矩阵A和信道矩阵H，求解得到稀疏表示向量
Figure PCTCN2020083564-appb-000013
S23：训练神经网络模型。将第一组训练样本集的输入和输出交换顺序，构成第二组训练样本。通过人工智能中的BP神经网络训练方法创建神经网络，以基于所述稀疏表示θ和所述发送信号向量
Figure PCTCN2020083564-appb-000014
分别作为输入输出，构造第二组训练样本集，训练所述神经网络，得到所述第二神经网络模型。
S24：根据第二神经网络模型，当输入为信号
Figure PCTCN2020083564-appb-000015
时，得到重构的发送数据
Figure PCTCN2020083564-appb-000016
如下对步骤S13与S23进一步说明：神经网络模型由输入层(layer1)、中间层(layer2,..,L-1)、输出层(layerL)三部分组成，输入层起信号传输作用，负责接收外部输入信息，输入层每个单元代表一个特征；中间层可以是单中间层或多中间层，其起内部信息处理作用，负责信息变换；输出层起向外部输出信息作用，输出层的每个单元代表一个类别。本申请中利用BP神经网络模拟一个映射函数，此函数可以把输入空间数据映射到输出空间；BP神经网络会尽量的去拟合一个输入输出分别为所述装置发射端1的发送信号向量x和基于所述装置发射端1的发送信号向量x所确定的稀疏表示θ之间的函数，以及输入输出分别为稀疏表示θ与原始的发送数据流
Figure PCTCN2020083564-appb-000017
之间的函数；基于训练的BP神经网络产生的映射函数模型，可以根据装置发射端1经调制后的发送信号向量x得到稀疏化原始信号x在过完备冗余字典D上的稀疏表示θ，以及根据贝叶斯重构算法推算出的稀疏表示θ重构出原始的发送数据x。
本申请中使用代价函数来衡量BP神经网络输出与真实输出间的差异，训练BP神经网络使得网络的输入(装置接收端2接收信号)经过神经网络后的输出能够尽可能的接近理论输出。为了使代价函数最小，本申请中使用梯度下降法来求解神经网络参数，当求解到最优的神经网络权重，建立第一神经网络模型或第二神经网络模型。创建BP神经网络，收集大量的样本数据，并且人为的标记正确的分类结果，然后用这些标记好的数据来训练所创建的神经网络。在这个过程中，根据当前的输出值以及被标记的正确的目标值之间的差异，神经网络中的每一层都在不断的调整自身的权重和偏置，直到能够准确输出目标值。
针对训练神经网络时需用到的两个参数——权重和偏置进一步说明：本申请中，神经网络各层间的权重参数矩阵，用
Figure PCTCN2020083564-appb-000018
表示，其中，权重参数W的上标表 示层数，下标的相邻两层各自的第几个节点。例如，
Figure PCTCN2020083564-appb-000019
表示输入层Layer1的第1个节点与Layer2的第2个节点的线段的权重。这些权重决定了模型的作用，神经网络的目标就是通过样本来计算权重。每一个中间层和输出层的节点都是一个Logistic函数g(z)＝a。例如
Figure PCTCN2020083564-appb-000020
表示Layer2的第1个节点的输入值，带入Logistic函数得到输出
Figure PCTCN2020083564-appb-000021
神经网络各层间的偏置参数矩阵为：B＝[b 1b 2...b n] T，已知神经网络的输入为：Y＝[y 1y 2...y n] T,神经网络的输出为：
Figure PCTCN2020083564-appb-000022
引入一个非线性算子：
Figure PCTCN2020083564-appb-000023
则可以推导出：
Figure PCTCN2020083564-appb-000024
Figure PCTCN2020083564-appb-000025
初始化权重参数：将权重参数W随机初始化为[-ε,ε]之间，ε是预设的一个足够小的值。
训练神经网络模型：训练神经网络模型的过程主要分为两步，一是计算代价函数J(θ)，二是调整参数θ，使得代价函数值J(θ)尽量小。如下采用正向传播算法，对每一个样本计算当前神经网络模型下该样本的输出，求出代价函数，再根据输出来更新权重参数。定义代价函数J(θ)，m是样本的个数，由于神经网络有K个输出，其代价函数也相应地计算了K个输出的代价，其计算公式为：
Figure PCTCN2020083564-appb-000026
如下采用反向传播算法调整参数θ，使得代价函数值J(θ)尽量小。反向传播算法通过求代价函数关于各个权重系数的偏导数，以此来更新各个权重系数。例如，首先，计算最后一层的梯度：(1)计算代价函数值对非线性算子的梯度，(2)计算神经网络的输出对偏置及相邻层间权重的梯度。并按照梯度的负方向更新梯度；其次，计算倒数第二层的梯度：(1)计算上层回传的误差对非线性算子的梯度，(2)计算H n-1(H为每层经过激活函数后的输出)对偏置及相邻层间权重的梯度，并按照梯度的负方向更新梯度；最后，经过一层一层回传之后，最后计算第一层的梯度：(1)计算第二层回传的误差对非线性算子的梯度；(2)计算H 1对偏置及相邻层间权重的梯度。并按照梯度的负方向更新梯度。如此，在第一次反向传播过程循环结束后，继续进行上面的正向传播得到输出，后向传播更新参数这两步，直到均方差最小，即完成神经网络模型的训练过程。
图3给出了MIMO系统不同收发天线配置在采用了压缩感知和神经网络的信号传输与检测技术后的误码率性能。这里假定平坦衰落信道。在传统的方案下，4根发射天线只能同时发送4个数据符号。应用本申请所提出的方案，先采用BPSK调制信号，得到原始信号x 4×1，经过第一神经网络模型，得到稀疏信号θ 16×1。随机高斯矩阵A 4ρ×16作为压缩降维矩阵，从而得到z。如ρ＝0.5，则只需要2根发射天线，就可以实现原始数据的发送。如ρ＝0.75，则只需要3根发射天线，就可以实现原始数据的发送。装置接收端2，同样 通过2或者3根接收天线，采用BCS重构算法和神经网络训练出的第二神经网络模型，得到重构的发送信号
Figure PCTCN2020083564-appb-000027
该方案的误码率性能如图中(2×2)-4所示，括号内第一个数字发送天线数，第二个数字表示接收天线数，最后的数字表示原始数据长度。接收天线数有所增加，该方案记为(3×3)-4。和经典检测算法ZF(Zero Forcing)比起来，本申请所提出的方案可以在高SNR的条件下，兼顾保证误码率的同时，减少需要的收发天线数。
图4给出了MIMO系统不同收发天线配置在采用了压缩感知和神经网络的信号传输与检测技术后的误码率性能。这里假定平坦衰落信道。在传统的方案下，20根发射天线只能同时发送20个数据符号。应用本申请的方案，先采用BPSK调制信号，然后将x 20×1分为5组，则每组的向量x i(i＝1,2,3,4,5)的长度等于4。每组信号经过第一神经网络模型，得到稀疏信号θ i 16×1(i＝1,2,3,4,5)。选择随机高斯矩阵A 4ρ×16作为压缩降维矩阵，从而得到z i。将z i级联起来，得到待发送向量
Figure PCTCN2020083564-appb-000028
如ρ＝0.5，则只需要10根发射天线，就可以实现原始数据的发送。如ρ＝0.75，则需要15根发射天线，就可以实现原始数据的发送。装置接收端2，同样通过10根接收天线，该方案的误码率性能如图中(10×10)-20所示，括号内第一个数字发送天线数，第二个数字表示接收天线数，最后的数字表示原始数据长度。同样的数据，分组数也相同，ρ＝0.75，压缩降维矩阵为A 3×4时，接收天线数有所增加，该方案记为(15×15)-20。和经典检测算法ZF(Zero Forcing)比起来，本申请提出的方案可以在高SNR的条件下，兼顾保证误码率的同时，减少需要的收发天线数。可见，我们提出的方案可以在兼顾保证误码率的同时，减少需要的收发天线数。
如上所述，采用本发明所提出的增强空间复用方法，可以在已有的MIMO系统的基础上，结合人工智能，在减少天线数的基础上，传输相同的数据量，减少需要的天线数，提高复用增益和系统容量。
需要注意的是，上述具体实施例是示例性的，本领域技术人员可以在本发明公开内容的启发下想出各种解决方案，而这些解决方案也都属于本发明的公开范围并落入本发明的保护范围之内。本领域技术人员应该明白，本发明说明书及其附图均为说明性而并非构成对权利要求的限制。本发明的保护范围由权利要求及其等同物限定。

Claims (10)

  1. 基于人工智能的MIMO多天线信号传输与检测装置，该装置包括：
    稀疏化模块(104)，其利用第一神经网络模型对原始信号进行稀疏化表示；
    压缩复用模块(105)，用于对稀疏表示后的信号进行压缩降维处理；
    装置接收端(2)，用于对接收到的信号进行处理以实现其对目标信号的重构，
    其特征是，所述装置还包括解复用模块(202)，其中，
    所述解复用模块(202)被配置为在装置接收端(2)，利用压缩感知重构算法，从低维的信号中求解得到稀疏表示向量
    Figure PCTCN2020083564-appb-100001
    最后，利用第二神经网络模型对其接收到的信号重构得到原始信号x。
  2. 根据权利要求1所述的装置，其特征是，所述第二神经网络模型的输入输出与用于装置发射端(1)的所述稀疏化模块(104)从低维的信号中解出待传输至压缩复用模块(105)进行处理的稀疏信号θ的第一神经网络模型的输入输出相反。
  3. 根据权利要求2所述的装置，其特征是，所述装置接收端(2)还包括信道估计模块(201)，其被配置为对所述装置接收端(2)收到的信号进行信道估计并将其得到的信道参数矩阵作为所述压缩感知重构算法的输入。
  4. 根据权利要求3所述的装置，其特征是，所述解复用模块(202)是根据无需依赖信道状态信息的压缩感知复用矩阵和信道参数矩阵来求解得到稀疏表示向量
    Figure PCTCN2020083564-appb-100002
    的。
  5. 根据权利要求4所述的装置，其特征是，所述稀疏化模块(104)通过人工智能中的BP神经网络训练方法创建神经网络，以所述装置发射端(1)的发送信号向量x和基于所述装置发射端(1)的发送信号向量x所确定的稀疏表示θ分别作为输入输出，构造第一组训练样本集，训练所述神经网络，得到所述第一神经网络模型。
  6. 根据权利要求5所述的装置，其特征是，所述解复用模块(202)通过将所述发送信号向量x的所有可能组合分别作为冗余字典的不同列向量的方式组成冗余字典D，实现所述发送信号向量的稀疏表示θ，即x＝Dθ。
  7. 根据权利要求6所述的装置，其特征是，经稀疏化后的稀疏表示θ经过所述压缩复用模块(105)，压缩成ρl路信号，然后所述装置将压缩复用后的信号Z经发射天线发射。
  8. 根据权利要求7所述的装置，其特征是，压缩复用后的信号Z是通过计算式z＝Aθ来得到的，其中，A是N t行m列的压缩降维矩阵，ρ＝N t/l∈(0,1]代表压缩比例。
  9. 基于人工智能的MIMO多天线信号传输与检测系统，
    其特征是，
    所述系统通过构建MIMO端到端传输模型，装置发射端(1)通过其所构建的第一 神经网络模型来从低维的信号中解出在传输至压缩复用模块(105)进行处理前的稀疏信号，装置接收端(2)利用其所构建的与稀疏化模块(104)的第一神经网络模型的输入输出相反的第二神经网络模型对其接收到的信号重构得到原始信号。
  10. 根据权利要求9所述的系统，其特征是，所述第二神经网络模型的输入是利用压缩感知重构算法所求解得到的稀疏表示向量
    Figure PCTCN2020083564-appb-100003
PCT/CN2020/083564 2020-04-07 2020-04-07 基于人工智能的mimo多天线信号传输与检测技术 WO2021203243A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/CN2020/083564 WO2021203243A1 (zh) 2020-04-07 2020-04-07 基于人工智能的mimo多天线信号传输与检测技术
CN202310024746.5A CN116054887A (zh) 2020-04-07 2020-04-07 一种基于神经网络模型的天线信号调制方法
CN202080000961.6A CN111713035B (zh) 2020-04-07 2020-04-07 基于人工智能的mimo多天线信号传输与检测技术

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/083564 WO2021203243A1 (zh) 2020-04-07 2020-04-07 基于人工智能的mimo多天线信号传输与检测技术

Publications (1)

Publication Number Publication Date
WO2021203243A1 true WO2021203243A1 (zh) 2021-10-14

Family

ID=72547481

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/083564 WO2021203243A1 (zh) 2020-04-07 2020-04-07 基于人工智能的mimo多天线信号传输与检测技术

Country Status (2)

Country Link
CN (2) CN116054887A (zh)
WO (1) WO2021203243A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114423033A (zh) * 2022-01-17 2022-04-29 南京工业大学 基于vssamp的抗干扰多用户检测算法及其设计方法
CN115242271A (zh) * 2022-07-27 2022-10-25 南京邮电大学 一种强化学习辅助的大规模MIMO的Damped-BP检测方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112073156B (zh) * 2020-11-11 2021-03-26 电子科技大学 一种高维非正交传输方法
CN115842601A (zh) * 2021-08-04 2023-03-24 华为技术有限公司 数据处理方法及装置
CN113777578B (zh) * 2021-11-12 2022-02-08 上海闪马智能科技有限公司 一种信号处理方法、装置、存储介质及电子装置
CN114205001B (zh) * 2021-11-29 2023-07-11 武汉邮电科学研究院有限公司 一种低复杂度的波长标签生成与检测方法及系统
CN115085808B (zh) * 2022-06-09 2023-10-17 重庆邮电大学 一种基于小波神经网络的vlc系统时频联合后均衡方法
CN115695105B (zh) * 2023-01-03 2023-03-17 南昌大学 基于深度迭代智能超表面辅助通信的信道估计方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103188058A (zh) * 2011-12-27 2013-07-03 财团法人工业技术研究院 通道信息反馈方法及其无线通信装置
US20140293904A1 (en) * 2013-03-28 2014-10-02 Futurewei Technologies, Inc. Systems and Methods for Sparse Beamforming Design
CN106301384A (zh) * 2016-08-26 2017-01-04 辽宁工程技术大学 一种基于分块压缩感知的信号重构方法
CN107368887A (zh) * 2017-07-25 2017-11-21 江西理工大学 一种深度记忆卷积神经网络的结构及其构建方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018107414A1 (zh) * 2016-12-15 2018-06-21 上海寒武纪信息科技有限公司 压缩/解压缩神经网络模型的装置、设备和方法
US10771488B2 (en) * 2018-04-10 2020-09-08 Cisco Technology, Inc. Spatio-temporal anomaly detection in computer networks using graph convolutional recurrent neural networks (GCRNNs)
CN108768585B (zh) * 2018-04-27 2021-03-16 南京邮电大学 基于深度学习的上行免信令非正交多址接入noma系统多用户检测方法
CN109741256B (zh) * 2018-12-13 2023-01-06 西安电子科技大学 基于稀疏表示和深度学习的图像超分辨率重建方法
CN110099016B (zh) * 2019-05-14 2022-05-31 东南大学 一种基于深度学习网络的毫米波稀疏阵面信道估计方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103188058A (zh) * 2011-12-27 2013-07-03 财团法人工业技术研究院 通道信息反馈方法及其无线通信装置
US20140293904A1 (en) * 2013-03-28 2014-10-02 Futurewei Technologies, Inc. Systems and Methods for Sparse Beamforming Design
CN106301384A (zh) * 2016-08-26 2017-01-04 辽宁工程技术大学 一种基于分块压缩感知的信号重构方法
CN107368887A (zh) * 2017-07-25 2017-11-21 江西理工大学 一种深度记忆卷积神经网络的结构及其构建方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
CHEN MUHAN, GUO JIAJIA, LI XIAOJINSHI: "An overview of the CSI feedback based on deep learning for massive MIMO systems", CHINESE JOURNAL ON INTERNET OF THINGS, vol. 4, no. 1, 31 March 2020 (2020-03-31), pages 33 - 44, XP055856457, DOI: 10.11959/j.issn.2096−3750.2020.00157 *
LIU CHANZI; ZHOU QINGFENG; WANG XINDI; CHEN KAIPING: "MIMO Signal Multiplexing and Detection Based on Compressive Sensing and Deep Learning", IEEE ACCESS, IEEE, USA, vol. 7, 1 January 1900 (1900-01-01), USA , pages 127362 - 127372, XP011746358, DOI: 10.1109/ACCESS.2019.2937490 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114423033A (zh) * 2022-01-17 2022-04-29 南京工业大学 基于vssamp的抗干扰多用户检测算法及其设计方法
CN114423033B (zh) * 2022-01-17 2023-08-08 南京工业大学 基于vssamp的抗干扰多用户检测算法及其设计方法
CN115242271A (zh) * 2022-07-27 2022-10-25 南京邮电大学 一种强化学习辅助的大规模MIMO的Damped-BP检测方法
CN115242271B (zh) * 2022-07-27 2023-06-16 南京邮电大学 一种强化学习辅助的大规模MIMO的Damped-BP检测方法

Also Published As

Publication number Publication date
CN116054887A (zh) 2023-05-02
CN111713035A (zh) 2020-09-25
CN111713035B (zh) 2022-12-20

Similar Documents

Publication Publication Date Title
WO2021203242A1 (zh) 基于深度学习的mimo多天线信号传输与检测技术
WO2021203243A1 (zh) 基于人工智能的mimo多天线信号传输与检测技术
CN109672464B (zh) 基于fcfnn的大规模mimo信道状态信息反馈方法
CN110912598B (zh) 基于长短时注意力机制的大规模mimo系统csi反馈方法
Liao et al. CSI feedback based on deep learning for massive MIMO systems
Liu et al. A Markovian model-driven deep learning framework for massive MIMO CSI feedback
Chen et al. Deep learning-based implicit CSI feedback in massive MIMO
CN109257309A (zh) 一种高性能的大规模mimo下行链路传输信道估计方法
CN113972939B (zh) 基于双时间尺度和深度学习的天线系统预编码方法和装置
Xu et al. Deep joint source-channel coding for CSI feedback: An end-to-end approach
CN116248156A (zh) 基于深度学习的大规模mimo信道状态信息反馈及重建方法
CN113381790B (zh) 一种基于ai的环境知识辅助的无线信道反馈方法
CN105162556A (zh) 基于空时相关性的大规模mimo系统信道反馈方法
Zhang et al. Pilot-assisted MIMO-V-OFDM systems: Compressed sensing and deep learning approaches
CN112929304A (zh) 一种环境知识库辅助的无线信道获取方法
CN113660020A (zh) 一种无线通信信道信息传输方法、系统和解码器
Liu et al. CSI feedback based on complex neural network for massive MIMO systems
CN114844541B (zh) 一种基于深度学习的混合预编码和反馈方法
CN116436551A (zh) 信道信息传输方法及装置
Liao et al. Structured neural network with low complexity for MIMO detection
Liao et al. M-SAMP: A low-complexity modified SAMP algorithm for massive MIMO CSI feedback
Yue et al. Deep learning-based CSI feedback for IoT-oriented massive MIMO systems
Liang et al. The Application of Deep Learning in Wireless Communications of Maglev System
He et al. 5 Channel Estimation, Feedback, and Signal Detection
Cao et al. Information Bottleneck Based Joint Feedback and Channel Learning in FDD Massive MIMO Systems

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20930015

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 07/03/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20930015

Country of ref document: EP

Kind code of ref document: A1