CN116054888A

CN116054888A - Method and device for reconstructing original signal of antenna signal

Info

Publication number: CN116054888A
Application number: CN202310024809.7A
Authority: CN
Inventors: 刘婵梓; 曲春晓; 陈高; 刘新宇; 周清峰
Original assignee: Dongguan University of Technology
Current assignee: Dongguan University of Technology
Priority date: 2020-04-07
Filing date: 2020-04-07
Publication date: 2023-05-02
Also published as: WO2021203242A1; CN111630787A; CN111630787B

Abstract

The invention relates to an original signal reconstruction method and device of an antenna signal, wherein the method at least comprises the following steps: obtaining a first neural network model and a second neural network model by using a BP training algorithm of deep learning; solving a high-dimensional sparse signal theta from a low-dimensional target signal through the constructed first neural network model; the sparse signal theta is input into the constructed second neural network model to reconstruct to obtain an original signal. The invention can simultaneously transmit the same data volume on the basis of reducing the number of the required antennas, and improves the multiplexing gain and capacity of the MIMO system.

Description

Method and device for reconstructing original signal of antenna signal

The original basis of the divisional application is a patent application with application number (202080000955.0), application date of 2020, 4 months and 7 days and the name of 'deep learning-based MIMO multi-antenna signal transmission and detection technology'.

Technical Field

The present invention relates to the field of mobile communications technologies, and in particular, to a method and an apparatus for reconstructing an original signal of an antenna signal.

Background

MIMO (Multiple-Input Multiple-Output) technology refers to using Multiple transmit antennas and receive antennas at a device transmitting end and a device receiving end, respectively, so that signals are transmitted and received through the Multiple antennas at the device transmitting end and the device receiving end, thereby improving communication quality. MIMO includes a SIMO (Single-Input Multiple-Output) system and a MISO (Multiple-Input Single-Output) system, with respect to a general SISO (Single-Input Single-Output) system, according to the number of antennas at both transmitting and receiving ends. The MIMO system can fully utilize space resources, realize multiple transmission and multiple reception through a plurality of antennas, can doubly improve the system channel capacity under the condition of not increasing frequency spectrum resources and antenna transmitting power, shows obvious advantages, and is widely applied to wireless communication, namely, mobile equipment and a network commonly use a plurality of antennas to enhance connectivity and improve network speed and user experience. Massive MIMO is a key factor of 5G ultra-high data rate, can bring about larger network capacity, wider signal coverage and better user experience, and brings the potential of 5G into a brand new level.

MIMO techniques can be broadly divided into two categories according to the space-time mapping method: spatial diversity and spatial multiplexing. Space diversity refers to that signals with the same information are transmitted through different paths by utilizing a plurality of transmitting antennas, and simultaneously, a plurality of independently fading signals of the same data symbol are obtained at a receiver end, so that the receiving reliability of diversity improvement is obtained. For example, in slowIn Rayleigh fading channels, a transmitting antenna N is used _r A root receiving antenna for transmitting signals through N _r Different paths. If the fading between the individual antennas is independent, a maximum diversity gain of N can be obtained _r . For transmit diversity techniques, the gains of multiple paths are also utilized to improve the reliability of the system. In a single unit having N _t Root transmitting antenna N _r In a system with a root receive antenna, if the path gain between the antenna pairs is rayleigh fading with independent uniform distribution, the maximum diversity gain that can be achieved is N _t N _r . The unreliability of wireless communications is mainly caused by the time-varying and multipath characteristics of the wireless fading channel, and it is important how to reduce the influence of multipath fading on the base station and the mobile station without increasing power and sacrificing bandwidth. The only method is to adopt the anti-fading technology, and the effective method for overcoming the multipath fading is various diversity technology. Diversity techniques are mainly used to combat channel fading. Conversely, fading characteristics in a MIMO channel may provide additional information to increase the degrees of freedom in communication. Essentially, if the fading between each pair of transmit and receive antennas is independent, multiple parallel sub-channels may be created. If different information streams are transmitted on these parallel sub-channels, a transmission data rate may be provided, which is referred to as spatial multiplexing. However, in the case of high SNR (Signal to Noise Ratio, signal-to-noise ratio), the transmission rate is limited in degree of freedom.

MIMO relies on its two major advantages: (1) increasing the capacity of the channel. Between the MIMO access point and the MIMO client, a plurality of spatial streams can be simultaneously transmitted and received, and the channel capacity can be linearly increased along with the increase of the number of the antennas, so that the MIMO channel can be utilized to doubly improve the wireless channel capacity, and the frequency spectrum utilization rate can be doubly improved under the condition of not increasing the bandwidth and the antenna transmission power; (2) improving the reliability of the channel. With the spatial multiplexing gain and the spatial diversity gain provided by the MIMO channel, channel fading can be suppressed with multiple antennas. The application of the multi-antenna system enables parallel data streams to be transmitted simultaneously, can obviously overcome the fading of channels and reduce the error rate, and has become a core technology applied to 802.11 n. 802.11n is a brand new wireless local area network technology after 802.11a/b/g, and the speed can reach 600Mbps. Meanwhile, the MIMO technology can improve the performance of the existing 802.11a/b/g network.

As the number of antennas increases, the complexity of implementation of the MIMO technology increases greatly, so that the number of antennas used is limited, and the advantages of the MIMO technology cannot be fully exerted. At present, how to reduce the algorithm complexity and the implementation complexity of the MIMO technology on the basis of ensuring certain system performance becomes a great challenge facing the industry.

Artificial intelligence refers to the intelligence exhibited by machines manufactured by humans. Artificial intelligence generally refers to a technology for presenting human intelligence by simulating certain thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning and the like) of a person through a common computer program, and mainly comprises the principle of computer-implemented intelligence and manufacturing a computer similar to human brain intelligence, so that the computer can realize higher-level application. We trust that artificial intelligence still "drives in" for the years to create value for the traditional industry and change our daily life deeply, such as robot field, language recognition field, image recognition field, expert system, etc. Machine learning is a subset of artificial intelligence. Machine learning is currently mainly used for solving classification problems, clustering problems, regression problems and the like, and is widely applied to the fields of character recognition, machine translation, voice recognition, search engines, face recognition, unmanned driving and the like. The most critical of all machine learning algorithms is deep learning.

The concept of deep learning is derived from the study of artificial neural networks. The concept of deep learning was first proposed by the professor ge. Hinton, university of toronto, thai is a tabucket in the machine learning field, with two core views: (1) The artificial neural network with multiple hidden layers has excellent characteristic learning capability, has deeper display on the learned characteristic data, and can better classify or visualize the finally obtained network data; (2) Deep neural networks can overcome the difficulty in training their own network parameters by "layer-by-layer initialization", which can be achieved by unsupervised learning. The multi-layer sensor with multiple hidden layers is a deep learning structure. Deep learning forms more abstract high-level representation attribute categories or features by combining low-level features to discover distributed feature representations of data. Any neural network may have any number of layers, inputs or outputs. The layer between the input neuron and the last layer of output neurons is the hidden layer of the deep neural network.

The problem of MIMO is solved by applying deep learning, and patent achievements are not invented at home and abroad.

In the prior art, as proposed in the patent document with the grant date of 20180619 being CN105610477B, a multi-transmission multi-reception system enhancement signal multiplexing method based on compressed sensing is provided, on the basis of the existing MIMO technology, a random measurement matrix in the compressed sensing technology is selected as a signal compression multiplexing matrix on the basis of the existing related MIMO system signal multiplexing technology, then the sparse characteristic of a transmission signal on an overcomplete redundant dictionary is fully utilized, and a high-dimensional transmission signal is solved from a low-dimensional reception multiplexing signal through a compressed sensing reconstruction algorithm, so that the signal multiplexing gain under the condition of the number of receiving antennas of a given MIMO system is greatly improved, the application requirement of the MIMO system on broadband transmission is better met, and the advantages of ensuring that a receiving end uses an optimized reconstruction algorithm mature in the compressed sensing field to reconstruct the multipath data stream transmitted by a transmitting end through a compression multiplexing step with high probability and small modification to the existing MIMO system are achieved.

The Chinese application number 201510473741.6 (self-complex neural network channel prediction method, western electronic technology university) discloses a complex neural network channel prediction method, which mainly solves the problem of channel fading caused by channel time variation in a MIMO system. The technical proposal is as follows: 1. the base station measures the channel to obtain a channel coefficient training sequence containing estimation errors; 2. obtaining corresponding training samples and expected output according to the obtained channel coefficient sequence; 3. inputting a training sample to perform complex wavelet neural network training to obtain a final network weight; 4. and the base station predicts the channel coefficient by using the trained complex wavelet neural network. The method is simple and easy to implement, has good effect, and is suitable for reducing the influence of channel time variation on the MIMO system channel.

The Chinese application number 201810177829.7 patent (a neural network-based wireless channel modeling method, university of southwest) discloses a neural network-based wireless channel modeling method. Firstly, processing a received signal fed back by a user to obtain estimated channel parameters; and then three-dimensional geographic information of the scatterer is obtained according to the two-dimensional image, the three-dimensional geographic information is clustered, and finally, channel parameters and the geographic information are used as input of the neural network, a received signal is used as output, and a nonlinear time-varying neural network model is obtained through training. The method can obtain a more accurate channel model within acceptable complexity, and can meet the channel modeling requirements of scenes such as large-scale MIMO technology, band expansion, high mobility and the like adopted in a future 5G communication system.

The Chinese application number 201810267976.3 patent (a deep neural network large-scale MIMO system detection method constructed based on BP algorithm, southeast university) provides a deep neural network large-scale MIMO system detection method constructed based on BP algorithm, which constructs a deep neural network for large-scale MIMO system detection by expanding and mapping a belief propagation iterative algorithm factor graph onto a neural network structure; the neurons of the deep neural network correspond to nodes in the iterative algorithm factor graph, and the number of the neurons of each layer is equal to the number of symbol nodes in the iterative algorithm factor graph; the mapping function between the hidden layers is an updating formula of the confidence information in the iterative algorithm, and the number of the hidden layers is equal to the iteration times of the iterative algorithm. Specifically, the invention also provides a MIMO detection method for respectively constructing two deep neural networks based on the damping belief propagation and the maximum and belief propagation iterative algorithm. The invention achieves lower error rate without increasing on-line operation complexity, and has robustness to various channel conditions and antenna matching.

The invention of China application number 201910063733.2 (optimized MIMO detection method based on deep learning, shanghai university) is characterized in that by constructing an MIMO end-to-end transmission model, obtaining complex time domain vector of the model according to signal y (t) received by a receiving end of an MIMO device and estimated imperfect channel state information as input of a Deep Neural Network (DNN), obtaining an estimated value of a bit stream at a transmitting end of the device by using the DNN, and compared with the estimated value of a transmitting bit stream obtained according to hard decision in the prior art, the invention can improve accuracy and detection rate under imperfect channel information, ensure detection performance of low bit error rate under low complexity algorithm, and has good robustness under the condition of containing inherent channel error.

Chinese application No. 201610327115.0 (a codebook selection method based on deep learning under massive MIMO, chongqing postal university) relates to a codebook selection method based on deep learning under massive MIMO. The method comprises the following steps: collecting pilot frequency information of a test area to construct a pilot frequency training sequence, and further obtaining a pilot frequency training sample; performing neural network iterative learning on the pilot training sample to obtain a final network weight value; and selecting the optimal code word from the complete codebook according to the channel output by the neural network after learning. And then, carrying out channel information matching on the unknown area and the test area to obtain a wireless channel of the unknown area, and further obtaining a codeword corresponding to the wireless channel. The invention can effectively, accurately and rapidly establish the wireless channel model and the codebook query, avoid the channel estimation of the unknown region and greatly reduce the complexity of the channel selection codebook of the unknown region.

The Chinese application No. 201811626005.X patent (Low complexity MIMO-NOMA system signal detection method based on improved gradient projection method, chongqing university) discloses a low complexity MIMO-NOMA system signal detection method based on improved gradient projection method, which relates to wireless communication technology. According to the sparse characteristics of active users of the system, converting the system model into a strict quadratic programming problem by utilizing a convex optimization algorithm idea; and then carrying out iterative solution on the problem, and carrying out preprocessing operation on the iterative result of each time to achieve effective detection on active users and signals thereof. The invention breaks through the problem of low algorithm convergence speed in the traditional detection method, and carries out preprocessing operation on each iteration result, so that the detection result can be converged rapidly, and the active user set can be detected, thereby having simple implementation process and wide application range.

The Chinese application number 201910014714.0 (a method for designing a MIMO system beamforming matrix based on deep learning, nanjing university of post) discloses a method for designing a MIMO system beamforming matrix based on deep learning, which comprises the following steps of firstly obtaining a training sample set required by a deep learning network by using a known algorithm; then constructing a deep learning neural network model, initializing relevant parameters of the model and training by using a training sample set; and then, the pilot frequency is used for obtaining a channel and sending the channel into a neural network to predict the beamforming matrix coefficient, and finally, the channel and the beamforming matrix coefficient are combined to form a beamforming matrix. The method utilizes the beam forming matrix obtained by the deep learning neural network to simultaneously consider the performance and the algorithm complexity, and can reduce the time delay on the premise of ensuring the performance, so that the MIMO system can provide real-time service.

The Chinese application number 201810182937.3 (a machine learning-based MIMO link adaptive transmission method, dongnan university) discloses a machine learning-based MIMO link adaptive transmission method, which uses an unsupervised learning self-coding algorithm to extract and reduce the dimension of features, introduces the idea of deep learning, and can reduce the feature dimension and the computation complexity on the premise of keeping the state information of main information. The invention utilizes the logistic regression algorithm to construct the mapping relation between the channel state information and the transmission parameters, is different from the traditional fixed parameterized model, can train based on sample data, can better establish the mapping relation between the channel state information and the transmission parameters under the condition that the quality of a data set is better and all states are covered, and can more fully utilize the channel state information compared with the traditional single equivalent signal-to-noise ratio. In addition, the invention also carries out CQI selection based on the channel matrix, and the MIMO link self-adaptive method based on machine learning is not limited by the design of a receiver through the research of the channel matrix and the noise variance, thereby having universality.

The chinese application No. 201710495044.X patent (a method for joint precoding and antenna selection of MIMO system based on deep learning, university of zhejiang technology) discloses a method for joint precoding and antenna selection of MIMO system based on deep learning, comprising the following steps: firstly, generating a training data set required by deep learning through an existing antenna selection method; then, establishing a deep learning model, training the deep learning model by using training data and storing the deep learning model; then, completing antenna selection by using the stored deep learning model; and finally, carrying out optimal precoding design on the selected MIMO subsystem. The invention designs the MIMO system joint precoding and antenna selection by using the deep learning technology, and can realize lower calculation complexity under the condition of obtaining good system signal-to-noise ratio.

The invention patent of China application number 201910242525.9 (a high-speed rail oriented depth signal detection method, shenzhen university) proposes a high-speed rail oriented depth signal detection method, which comprises the steps of firstly, collecting data, and collecting a plurality of sending signals and receiving signals in various scenes along the high-speed rail according to different environment types along the high-speed rail; secondly, dividing scenes, and further dividing each scene into a plurality of areas through data analysis to meet the compatibility of the neural network; thirdly, establishing a deep high-speed rail signal detection neural network model; then, training a high-speed rail signal detection neural network offline; and finally, carrying out online real-time signal detection, determining the position information of the high-speed rail through a GPS (global positioning system) in the running process, judging the area where the high-speed rail is positioned, selecting a corresponding neural network model, inputting the real-time received signal into a trained neural network, and outputting the signal sent by the base station end in real time. The system performance of the invention is greatly improved, the bit error rate of signal detection is reduced, and the algorithm is more robust. The method used by the invention does not need to estimate the channel, and saves pilot frequency expenditure.

The invention patent of China application No. 201810279530.2 (visible light communication MIMO anti-interference noise reduction method based on BP neural network, university of national university of middle and south) discloses a visible light communication MIMO anti-interference noise reduction method based on BP neural network, and relates to MIMO antenna technology in the field of visible light communication. The system comprises a system device transmitting end, a system device receiving end signal processing part and a BP neural network signal processing part which are sequentially communicated. The method comprises the following steps: 1) The electric signal is loaded on the LED array and emitted out in the form of an optical signal; 2) The photoelectric detector at the receiving end of the device converts the optical signal into an electric signal; 3) The multipath electric signals remove high-frequency interference through a low-pass filter; 4) After training, the BP neural network carries out noise reduction and interference elimination processing on the multipath signals, and finally the multipath signals are converted into binary serial data streams through parallel-serial conversion. The invention improves the transmission performance of the existing MIMO technology; combining the neural network with the visible light MIMO technology, and exerting the advantages of the neural network in the aspect of noise reduction and disturbance elimination in wireless communication; the adoption of the neural network receiving processing technology enables the whole VLC system to be more stable.

The invention patent of China application number 201710213235.2 (a visible light channel joint equalization method based on orthogonal mapping and a probabilistic neural network, university of Zhongshan) discloses a visible light channel joint equalization method based on the orthogonal mapping and the probabilistic neural network, which comprises a device transmitting end and a device receiving end, wherein signals are transmitted from the device transmitting end to the device receiving end through a visible light MIMO channel; the visible light MINO channel is a multiple-input multiple-output channel; the combined equalization is that the pre-equalization and the post-equalization are combined; the invention adopts a combined equalization scheme combining a front equalization technology and a rear equalization technology, namely a visible light multi-input multi-output channel combined equalization method based on orthogonal mapping and a probability neural network, which can effectively inhibit interference between channels of a visible light MIMO communication system and improve data transmission reliability.

The embodiment of China 201910125325.5 patent (a MIMO decoding method, device and storage medium based on deep learning, shenzhen Bao-chain artificial intelligence technology Co., ltd.) discloses a MIMO decoding method, device and storage medium based on deep learning, wherein a training data set of MIMO decoding is constructed, and the training data set comprises a plurality of training data; training the neural network based on the training data set to obtain a trained neural network model; when receiving the MIMO signal to be decoded, inputting the MIMO signal to be decoded into the neural network model for MIMO decoding, and then obtaining an MIMO decoding result output by the neural network model. Through the implementation of the invention, the neural network model for joint MIMO detection and channel decoding is designed based on deep learning, the MIMO detection and the channel decoding are regarded as a joint decoding process, and the approximation of the output result of the neural network model is improved through training, so that the overall performance of MIMO decoding is ensured, and the decoding accuracy and the decoding speed are higher.

The Chinese application number 201810757547.4 patent (a machine learning assisted massive MIMO downlink user scheduling method, university of southward) discloses a machine learning assisted massive MIMO downlink user scheduling method, comprising the following steps: s1: the base station acquires a characteristic mode energy coupling matrix in a characteristic direction through an uplink detection signal sent by a user; s2: the base station utilizes the characteristic mode energy coupling matrix to assist in sum rate calculation under various user and beam combinations by a machine learning method; s3: and adopting a greedy algorithm to realize user scheduling with the maximum speed criterion, and obtaining the optimal user beam pairing combination. The invention acquires statistical channel information through the uplink detection signal, and adopts the sum rate maximization criterion to carry out user scheduling. Under the condition that the base station only has statistical channel information, the approximate calculation of the rate is accurately realized through targeted feature extraction and the design of a neural network, the complexity of user scheduling under a large-scale antenna is greatly reduced, the performance is close to optimal, and the method has good applicability and robustness.

The invention patent number 201610353881.4 of China (a modulation recognition method under the MIMO related channel based on a machine learning algorithm, beijing university of post and telecommunications) is a modulation recognition method under the MIMO related channel based on the machine learning algorithm, and belongs to the field of communication; the method comprises the following specific steps: firstly, each data stream of a transmitting end of a communication device is respectively coded by space time, and each codeword is respectively transmitted through Nt transmitting antennas; then, calculating a MIMO channel matrix H according to the correlation matrix of the device receiving end and the correlation matrix of the device transmitting end; according to the MIMO channel matrix H, calculating a received signal on each receiving antenna and correcting the received signal; finally, each receiving antenna respectively performs feature extraction on the corrected signals, performs training test on the extracted feature values, and calculates a modulation recognition mode to which the sample belongs; the advantages are that: the robustness and generalization capability to the non-Gaussian channel are strong, and the modulation system identification under a more complex environment can be realized through parameter iteration; by extracting the features of the high-order moment and the high-order cumulant, the signal features have obvious difference under the condition of higher signal-to-noise ratio, and the classification of the machine learning algorithm is facilitated.

The application of MIMO technology makes space a resource that can be used to improve performance and can increase the coverage of a wireless system. MIMO technology has become one of the key technologies in the wireless communication field, and has been increasingly applied to various wireless communication systems by the continued development in recent years. As the number of antennas increases, the complexity of implementation of the MIMO technology increases greatly, so that the number of antennas used is limited, and the advantages of the MIMO technology cannot be fully exerted. Meanwhile, since the practical proposal of artificial intelligence in 1956, over 50 years, the development of the artificial intelligence is achieved, and the development of the artificial intelligence is a wide intersection and leading edge science. As an important branch of artificial intelligence, it can be seen from the above description that neural networks, which implement a mathematical network through a computer, are recognized as "out-ways" that solve some of the bottleneck problems encountered in current communications.

In summary, the research of the key technologies surrounding the neural network and the communication problem at home and abroad at present has a lot of research results, and the results have provided feasible solutions from the angles of channel estimation, signal detection and the like of MIMO. At the same time, however, how to increase the system capacity, reduce the algorithm complexity and implementation complexity of the MIMO technology on the basis of ensuring a certain system performance becomes a great challenge facing the industry. The related research under the condition of surrounding MIMO does not research on a scheme for transmitting and detecting signals from the point of improving the multiplexing gain of a system by combining the neural network technology under the condition of given transmitting antenna and receiving antenna number.

Furthermore, there are differences in one aspect due to understanding to those skilled in the art; on the other hand, as the inventors studied numerous documents and patents while the present invention was made, the text is not limited to details and contents of all that are listed, but it is by no means the present invention does not have these prior art features, the present invention has all the prior art features, and the applicant remains in the background art to which the rights of the related prior art are added.

Disclosure of Invention

Aiming at the problems of improving the system capacity and reducing the algorithm complexity and the implementation complexity of the MIMO technology on the basis of ensuring certain system performance, the prior related research on the MIMO condition does not research the scheme of signal transmission and detection from the angle of improving the system multiplexing gain by combining the neural network technology under the condition of giving the number of transmitting antennas and receiving antennas. According to the method, the latest research progress of a compressed sensing technology and a neural network technology are combined, and a multiplexing transmission and detection scheme of signals in a MIMO system based on deep learning and compressed sensing is provided; on the other hand, compared with the existing MIMO space multiplexing technical scheme which only focuses on eliminating the interference of adjacent data, the method and the device have the advantages that the method and the device focus on eliminating the interference of the adjacent data, and focus on how to multiplex and transmit more data streams to the receiving end of the device on the basis of guaranteeing the detection performance of the receiving end of the device under the condition of given number of transmitting antennas, so that multiplexing gain and transmission capacity which exceed inherent multiplexing gain and transmission capacity of a MIMO system are obtained.

The MIMO device based on deep learning and compressed sensing provided in the present application mainly comprises a compression multiplexing module 104 of a device transmitting end 1, and a first neural network signal processing module 202 and a second neural network signal processing module 203 of a device receiving end 2. In addition, preferably, the device transmitting end 1 is further provided with a random number generator 101 (or called an original information bit generation module), a bit level processing module 102 and a modulation module 103, and the device receiving end 2 is further provided with a channel estimation module 201.

Preferably, as shown in fig. 5, the device transmitting end 1: the original data generated by the random number generator 101 sequentially passes through the bit level processing module 102 and the modulation module 103 to generate a modulated signal, the modulated signal then passes through the compression multiplexing module 104, the compression multiplexing module 104 performs compression reduction and multiplexing on the transmission signal, and the compressed and multiplexed signal is transmitted through the transmitting antenna; device receiving end 2: the channel estimation module 201 performs channel estimation on the received signal, obtains the input of the first neural network signal processing module 202 based on the received signal and the estimated channel state information, obtains the input of the second neural network signal processing module 203 based on the output of the first neural network signal processing module 202, and the second neural network model 203 reconstructs and outputs the original transmission data stream x.

The process specifically comprises the following steps:

MIMO multi-antenna signal transmission and detection device based on deep learning, the device includes: the compression multiplexing module is used for carrying out compression dimension reduction processing on the modulated signals; the device transmitting end is used for transmitting the target signal processed by the compression multiplexing module through the transmitting antenna under the condition that the number of the transmitting antennas is given; the device receiving end is used for processing the received signals to reconstruct the target signals, and comprises a first neural network signal processing module and a second neural network signal processing module, wherein the second neural network signal processing module is used for reconstructing the original signals x by inputting high-dimensional sparse signals theta which are solved from the low-dimensional target signals by the first neural network signal processing module through a first neural network model constructed by the first neural network signal processing module into a second neural network model constructed by the second neural network signal processing module.

According to a preferred embodiment, the apparatus receiving end further includes a channel estimation module configured to perform channel estimation according to the low-dimensional target signal subjected to the compressed dimensionality reduction process received by the apparatus receiving end, and take the obtained channel parameter matrix as an input of the first neural network model.

According to a preferred embodiment, the first neural network signal processing module creates a neural network through a BP algorithm of deep learning, takes a received signal vector y of the receiving end of the device and a sparse representation θ determined based on a transmitted signal vector x of the transmitting end of the device as samples, constructs a first training sample set, trains the neural network, and obtains the first neural network model.

According to a preferred embodiment, the sparse representation θ of the transmit signal vector is achieved by composing the redundant dictionary D in such a way that all possible combinations of the transmit signal vector x are respectively different column vectors of the redundant dictionary.

According to a preferred embodiment, the second neural network signal processing module creates a neural network through a BP algorithm of deep learning, takes a transmission signal vector x of the device transmitting end and the sparse representation θ determined based on the transmission signal vector x of the device transmitting end as samples, constructs a second training sample set, trains the neural network, and obtains the second neural network model.

According to a preferred embodiment, the transmitting end of the device adopts a random number generator to generate a group of random 0, 1 binary bit sequences to form the original data; the original data is BPSK modulated to produce modulated signal x.

According to a preferred embodiment, the modulated signal x is compressed into a signal ρ1 by the compression multiplexing module, and the device transmits the compressed multiplexed signal z via the transmitting antenna.

According to a preferred embodiment, the compression multiplexed ρ1-way signal z is obtained by the calculation formula z=ax, wherein a is N _t A compressed dimension-reducing matrix of rows and columns,

representing the compression ratio.

The system obtains the input of the neural network signal processing according to the target signal subjected to compression dimension reduction processing and the estimated channel state information received by the receiving end of the device by constructing a MIMO end-to-end transmission model, and reconstructs an original signal by utilizing the neural network signal processing.

A deep learning based MIMO multi-antenna signal transmission and detection method, the method comprising at least one of the following steps: transmitting the target signal subjected to the compression multiplexing processing through a transmitting antenna under the condition that the number of the transmitting antennas is given; acquiring a first neural network model and a second neural network model by utilizing a BP training algorithm of deep learning in advance; solving a high-dimensional sparse signal theta from a low-dimensional target signal through the constructed first neural network model; the original signal x is reconstructed by inputting the sparse signal theta into the constructed second neural network model.

Drawings

Fig. 1 is a schematic block diagram of a signal processing flow of a deep learning-based MIMO multi-antenna signal transmission and detection system provided by the present invention;

FIG. 2 is a schematic block diagram of a preferred signal compression multiplexing and detection process provided by the present invention;

fig. 3 is a graph of the error rate performance of the classical detection algorithm ZF and the MIMO system multi-antenna signal transmission and detection technique of the present invention under different transceiver antenna configurations;

fig. 4 is a graph of the error rate performance of the classical detection algorithm ZF and the MIMO system multi-antenna signal transmission and detection technique of the present invention under different transceiver antenna configurations; and

fig. 5 is a schematic diagram of module connection of a signal transmission and detection system of a MIMO multi-antenna system based on artificial intelligence and compressed sensing technology provided by the present invention.

List of reference numerals

1: device transmitting end 101: random number generator

102: bit level processing module 103: modulation module

104: compression multiplexing module 2: device receiving end

201: channel estimation module 202: first neural network signal processing module

203: second neural network signal processing module

Detailed Description

The present invention will be described in detail with reference to the accompanying drawings.

As shown in fig. 2, the technical scheme adopted by the invention is a signal transmission and detection technology and a detection method of a MIMO multi-antenna system based on artificial intelligence and compressed sensing technology in a MIMO system, and the technical scheme is as follows:

1. Device transmitting end 1 signal processing: for a pair of N _t Root transmit antenna and N _r A MIMO communication system with a root receive antenna. The l paths of signals x after channel coding and signal modulation at the system transmitting end are compressed into ρl paths of signals by the compression multiplexing module 104, and then the compressed and multiplexed signals z are transmitted by the transmitting antenna. Wherein the compression multiplexing module 104 performs compression processing on the input signal as follows: z=ax. Wherein x= [ x ] ₁ ，x ₂ ，...，x _l ] ^T Represents the l-path modulation symbol after coded modulation, A is N _t A compressed dimension-reducing matrix of rows and columns,

representing the compression ratio. Preferably, the gaussian random matrix is selected as the compressed sensing multiplexing matrix/compressed dimension-reducing matrix a in this embodiment. The compression ratio ρ is determined by the size of the compressed dimension-reduction matrix a in the compressed sensing technique.

2. Signal detection at the device receiving end 2: the device receiving end 2 link is approximately the reverse process of the device transmitting end 1 link, and the signal received by the device receiving end 2 is: y=hz+n= HAx +n. Where y is the received signal vector of nr×1, i.e. N received by the device receiving end 2 _t L paths of modulation symbols transmitted by a root transmitting antenna; z is a transmission signal vector of nt×1; n is a gaussian white noise vector of nr×1; h is the nr×nt channel propagation matrix and is a deterministic matrix that remains unchanged for a coherence time interval. The apparatus receiving end 2 may estimate the channel propagation matrix H from pilot signals inserted in the transmission data.

The above-described channel estimation, i.e., the device receiving end 2 determines the state (uncertainty) of the wireless transmission channel by data processing to the device transmitting end 1. A common approach is non-blind channel estimation based on pilot symbols, i.e. the device transmitting end 1 sends known pilot information, which is processed by the device receiving end 2 to derive the channel state.

3. Reconstructing an original transmission data stream according to the first neural network model and the second neural network model which are obtained through training

The input signal of the training set of the first neural network model is y, and the output signal is sparse representation theta of the original signal x on the overcomplete redundant dictionary D. A second neural network model, the input signal of the training set is theta, and the output signal is the original transmission data stream +.>

I.e. device sender N _t The compressed and multiplexed l-path modulation symbols transmitted by the transmitting antenna. θ is the encoded l-path modulation symbol x= [ x ] ₁ ，x ₂ ，...，x _l ] ^T Sparse representation on an overcomplete redundant dictionary D. />

The specific implementation steps of the above process are described as follows:

for "construction of compressed multiplexing matrix": compressed sensing (Compressed sensing), also known as compressed sampling or sparse sampling, is a method of finding sparse solutions for underdetermined linear systems. This method has been in existence for at least forty years, and has been a long-standing development in this field recently due to the work of David Donoho, emmanel candles and Tao Zhexuan. In recent years, compressed sensing technology has been largely applied to the fifth generation mobile communication system, and has gained a lot of attention and research.

Compressed sensing arises from acquiring and reconstructing sparse or compressible signals. CandGs and Donoho are described In the literature "Compressed sensing," IEEE Transactions on Information Theory, vol.52, no.4, pp.1289-1306, 2006 and "Compressive sampling," In: proceedings of International Congress of Mathematicians, switzerland: european Mathematical Society Publishing House, pp.1433-1452, 2006 formally propose the concept of compressed sensing, which exploits the sparsity of the original signal to recover the original whole high-dimensional signal from fewer measured values than the nyquist theory. The core idea is to combine compression and sampling, first to collect a non-adaptive linear projection (measurement) of the signal, and then to reconstruct the original signal from the measurement according to the corresponding reconstruction algorithm. The conventional signal acquisition and processing process mainly comprises four parts of sampling, compression, transmission and decompression. The sampling process must meet shannon's sampling theorem that the sampling frequency must not be less than 2 times the highest frequency in the analog signal spectrum.

Unlike the traditional nyquist sampling theorem, as long as the signal x is compressible or sparse in some transform domain D, the high-dimensional sparse signal obtained by the transform can be projected onto a low-dimensional space using an observation matrix a that is uncorrelated with the transform domain D, and then the original signal can be reconstructed from these small projections with high probability by solving an optimization problem. Within this theoretical framework, the sampling rate is not dependent on the bandwidth of the signal, but on the structure and content of the information in the signal. The compressed sensing theory mainly comprises three aspects of sparse representation of signals, coded sampling and a reconstruction algorithm.

Sparse representation of a signal refers to representing the original signal as a sparse linear combination over a suitably selected set of overcomplete bases (dictionary d= [ D1, D2 … dp ], or transform domain), where D1, D2 … dp are atoms in the dictionary. By "overcomplete base" is meant a base in which the number of atoms greatly exceeds the dimension of the original signal. Since signals commonly existing in nature are generally not sparse, when signals are projected to a certain transform domain D, only a few elements are non-zero, the obtained transform vector is said to be sparse or approximately sparse, i.e., x=dθ, θ is a concise expression of the original signal x, which is a priori condition of compressed sensing, i.e., the signals must be sparsely expressed under a certain transform. Theoretically, a transform domain D can always be found, so as to realize sparse representation of signals. If the original signal x itself is sparse, x=θ. The multiple atoms with the best linear combination are found from the redundant dictionary to represent a signal, known as a sparse approximation or a highly nonlinear approximation of the signal.

Next, in the compressed sensing theory, an observation matrix a of the compressed sampling system needs to be designed, how to sample to obtain a small amount of observation values, and the original signal can be reconstructed from the small amount of observation values. Obviously, if The observation process destroys the information in the original signal and the quality of the reconstruction cannot be guaranteed. To ensure that the linear projection of the signal can preserve the original structure of the signal, the projection matrix must satisfy the constraint equidistance (Restricted Isometry Property, RIP) condition, and then a linear projection measurement of the original signal is obtained by the product of the original signal and the measurement matrix. RIP conditions are defined as follows: if a constant delta exists _K ∈(0，1]For all signals theta with sparsity K, the matrix A satisfies the following formula

The matrix a is said to satisfy the constrained equidistant property of order K, where sparsity K refers to the number of non-zero elements of the signal θ. A is N _t A compressed dimension-reducing matrix of rows and columns. The advantage of compressed sensing techniques is that even N _t > l, (l refers to the length of the signal), can still be derived from N _r (N _r ＝N _t ) The primary data of length l is recovered from the secondary measurements. Order the

Representing the compression ratio. According to the compressed sensing principle, as long as the measurement matrix A meets the RIP condition, even if the A is a matrix with the number of rows being far smaller than the number of columns, the signal theta is projected to a space with one dimension reduced, and the original signal can still be completely recovered from the measurement number far smaller than the signal dimension through a compressed sensing reconstruction algorithm. The compression ratio ρ determines the number of transmit and receive antennas that can be reduced and the performance of the device receiver 2 reconstruction. Davenport in its doctor paper "Random observation on random observations: sparse signal acquisition and processing "theorem 3.5 states that: a is the sum of 2K-order RIP constants delta _2K ∈(0，1]As long as->

C is a constant approximately equal to 0.28, the original signal can be recovered. Meanwhile, donoho is in document "Extensions of compressed sensing," SigThree conditions necessary for observing the matrix are given in nal Processing, vol.86, no.3, pp.533-548, 2006, and it is pointed out that most uniformly distributed random matrices have these three conditions, which can be used as the observation matrix, for example: a partial Fourier set, a partial Hadamard set, a uniformly distributed random projection (uniform Random Projection) set, etc. Documents "ecoding by linear programming," IEEE Transactions on Information Theory, vol.51, no.12, pp.4201-4215, 2005 and "Stable signal recovery from incomplete and inaccurate measurements," Communications on Pure and Applied Mathematics, vol.59, no.8, pp.1207-1223, 2006 demonstrate that a can meet RIP conditions with a high probability when the measurement matrix a is a gaussian random matrix. The gaussian random matrix is chosen as the compressed sensing multiplexing matrix a in this application.

For "signal reconstruction at device receiving end 2": the underdetermined problem can be solved by a reconstruction algorithm in compressed sensing. Solving this problem requires an exhaustive list of all permutations of non-zero values in the sparse vector θ and is therefore difficult to solve. In view of this, researchers have proposed a series of algorithms for finding sub-optimal solutions, mainly including: greedy tracking algorithms, convex relaxation methods, bayesian algorithms, combinatorial algorithms, and the like. Each algorithm has its advantages and disadvantages. The convex relaxation method requires a minimum number of observations to reconstruct the signal, but tends to be computationally burdensome. The greedy tracking algorithm is located between these several classes of algorithms in terms of both run time and sampling efficiency, and noise immunity is unstable. An appropriate reconstruction algorithm can be selected according to different environments, and once the sparse representation vector is obtained, the original signal can be recovered.

Traditional MIMO signal transmission flow: the transmitting data stream s is processed by a space-time coding, digital-to-analog conversion and analog module, separated into Nt sub-data streams, and simultaneously transmitted by Nt transmitting antennas at the same frequency. The transmitted signals propagate through reflection, scattering and the like of the wireless channel, and the parallel sub-signals reach the device receiving end 2 at different moments through different paths and are received by Nr antennas. The device receiving end 2 adopts a signal processing technology to perform joint processing on signals received by all antennas, so as to recover the original data stream. The code word is modulated and then transmitted, and the device receiving end 2 performs signal detection to reconstruct the original signal.

Compared with the traditional MIMO signal transmission flow, a signal transmission and detection scheme based on the combination of deep learning and compressed sensing technology is proposed. As shown in fig. 1, the application adds a compression multiplexing module 104 at the device transmitting end 1. Firstly, the modulated signals are subjected to compression dimension reduction processing, channel state information is not needed in the selection of a compression dimension reduction matrix, a measurement matrix in a compression sensing technology is selected to be used as a signal compression matrix, the compression dimension reduction and multiplexing processing of the transmitted signals are completed, and the data volume is reduced. The receiving end 2 of the device of the application is divided into the following two steps to reconstruct signals: (1) A first neural network model (Neural Network Model, NN 1) is trained by the BP (back propagation) algorithm in deep learning to solve a high-dimensional sparse signal θ from a low-dimensional received signal. (2) The second neural network model (Neural Network Model, NN 2) is trained by the deep-learning BP algorithm, reconstructing the original signal x.

The Back Propagation (BP) neural network is a concept proposed by scientists, including Rumelhart and McClelland, in 1986, and is a multi-layer feedforward neural network trained according to an error back propagation algorithm. The BP algorithm uses the square of the network error as an objective function and adopts a gradient descent method to calculate the minimum value of the objective function. The BP network is to add several layers (one or several layers) of neurons between the input layer and the output layer, these neurons are called hidden units, they have no direct connection with the outside, but the change of state can affect the relationship between the input and the output, and each layer can have several nodes. The calculation process of the BP neural network consists of a forward calculation process and a reverse calculation process. In the forward propagation process, the input mode is processed layer by layer from the input layer through the hidden unit layer and is transferred to the output layer, and the state of each layer of neurons only affects the state of the next layer of neurons. If the expected output cannot be obtained at the output layer, the reverse propagation is carried out, the error signal is returned along the original connecting path, and the weight of each neuron is modified to minimize the error signal. The above "back propagation" is an algorithm used in neural networks to efficiently calculate gradients, or more generally, a feed-forward computational graph (feedforward computational graph). It can be attributed to the application of a differentiated chain law starting from the network output, followed by a backward propagation gradient. The first application of back propagation can be traced back to Vapnik et al in 1960, but papers Learning representations by back-propagating errors are often used as a reference source. At present, in the practical application of the artificial neural network, most neural network models adopt BP networks and variation forms thereof. The artificial neural network is also a core part of the forward network, and the essence of the artificial neural network is embodied.

The BP network is mainly used for the following four aspects: 1) Function approximation: training a network to approximate a function using the input vector and the corresponding output vector; 2) Pattern recognition: associating an input vector with a pending output vector; 3) Classification: classifying the appropriate manner defined by the input vector; 4) Data compression: the output vector dimension is reduced for transmission or storage. Here, the application adopts the BP algorithm to train the neural network, and realizes the approximation of the function.

Compared with the traditional MIMO scheme, the method can simultaneously transmit the same data volume on the basis of reducing the number of required antennas by introducing the compression multiplexing module 104 and the demultiplexing module of the device receiving end 2 on the basis of the existing related MIMO system signal multiplexing technology, and improves the multiplexing gain and capacity of the MIMO system. Compared with the existing MIMO space multiplexing technical scheme, the method and the device have the advantages that interference of adjacent data is not only eliminated, but also more data streams are multiplexed and transmitted to the device receiving end 2 on the basis of guaranteeing the detection performance of the device receiving end 2 under the condition of given sending antenna number, and the multiplexing gain and the transmission capacity which exceed the inherent multiplexing gain and the transmission capacity of the MIMO system are obtained.

Examples

The embodiment fuses the MIMO multi-antenna signal transmission and detection technology based on deep learning and compressed sensing, and the specific implementation steps of the invention are illustrated in detail.

First, the information source is generated by generating a 0,1 bit sequence using a random number generator 2.

The modulation is to modulate bit data including BPSK, QPSK, 16QAM, 64QAM, and the like.

This embodiment will be described using BPSK modulation as an example.

According to the processing flow of the signal shown in fig. 2 at the device transmitting end 1 and the device receiving end 2, the specific steps are as follows:

s1: signal processing at the transmitting end 1 of the device.

S11: the 0,1 bit sequence is generated by a random number generator 101 to constitute the original data.

S12: the signal x is generated by BPSK modulation. The transmission data of each group may be different.

S13: and (3) performing compression multiplexing processing, namely multiplying the sparse vector transmission data by the compression multiplexing matrix A to obtain a data vector z. A represents a measurement matrix within compressed sensing, here chosen as N _t X l gaussian matrix. Wherein, the liquid crystal display device comprises a liquid crystal display device,

representing the compression ratio, represents the ratio of the number of antenna reductions. Finally, the data is transmitted via the channel.

S2: signal detection at the receiving end 2 of the device.

S21: the received signal is y=hz+n= HAx +n, where n represents noise. The signal matrix H is estimated according to the channel estimation module 201.

S22: and obtaining a first neural network model and a second neural network model by utilizing a BP training algorithm of deep learning in advance.

Specifically, when the transmitted signal is x, the signal of the device receiving end 2 is y, and meanwhile, in theory, we can always find a proper base to realize sparse representation of the signal. Temlyakov in documents Nonlinear Methods of Approximation, IMI Research Reports, dept. Of materials, university of South Carolina,2001, states that the dictionary D should be selected to fit the structure of the approximated signal as well as possible, and its construction may be without any limitation. All possible combinations of x are herein taken,and respectively serving as different column vectors of the redundant dictionary D to form the redundant dictionary D, so as to realize sparse representation of x: the sparse vector θ with all positions 0 except for the corresponding index position 1. Using y and θ, a first set of training samples is constructed; θ and x, constitute a second set of training samples. Through a BP training algorithm of deep learning, a first group of training samples obtain a first neural network model, a second group of training samples obtain a second neural network model, and when the input is a signal y, reconstructed transmission data is obtained

S23: the first neural network signal processing module 202 solves the high-dimensional sparse signal θ from the low-dimensional target signal by the first neural network model it builds. The input signal of the first neural network model is y, and the output signal is the sparse representation theta of the original signal x on the overcomplete redundant dictionary D.

S24: the second neural network signal processing module reconstructs an original signal x by inputting the sparse signal theta into a second neural network model constructed by the second neural network signal processing module. The input signal of the second neural network model is sparse representation theta of the original signal x on the overcomplete redundant dictionary D, and the output signal is the original transmission data stream

Steps S22 to S24 are further described as follows: the neural network model consists of three parts, namely an input layer (layer 1), a middle layer (layer 2, the layer of the first layer, L-1) and an output layer (layerL), wherein the input layer plays a role in signal transmission and is responsible for receiving external input information, and each unit of the input layer represents a characteristic; the middle layer can be a single middle layer or multiple middle layers, plays a role in processing internal information and is responsible for information transformation; the output layer functions to output information to the outside, and each unit of the output layer represents a category. In the application, a BP neural network is utilized to simulate a mapping function, and the mapping function can map input space data to output space; the BP neural network can try to fit an original device receiving end 2 signal y and an original signal x as much as possible when the original device receiving end 2 signal y and the original signal x are over-run Functions between sparse representation θ on redundant dictionary D, and sparse representation θ and original transmit data stream

Function between: based on a mapping function model generated by a trained BP neural network, the sparse representation theta of the original signal x calculated theoretically on the overcomplete redundant dictionary D can be restored according to the signal y received by the device receiving end 2, and the original transmission data x can be reconstructed according to the sparse representation theta calculated theoretically.

In the method, the cost function is used for measuring the difference between the BP neural network output and the real output, and the BP neural network is trained so that the output of the input (the signal received by the device receiving end 2) of the network after passing through the neural network can be as close to the theoretical output as possible. In order to minimize the cost function, the gradient descent method is used for solving the neural network parameters, and when the optimal neural network weight is solved, the first neural network model or the second neural network model is built. A BP neural network is created, a large amount of sample data is collected, and the correct classification results are artificially labeled, and then the created neural network is trained with the labeled data. In this process, each layer in the neural network continuously adjusts its own weight and bias based on the difference between the current output value and the labeled correct target value until the target value can be accurately output.

For two parameters that are needed in training the neural network-weight and bias further description: in the present application, a weight parameter matrix between layers of a neural network is used

And (3) representing the number of layers by the superscript of the weight parameter w, and the number of nodes of each of two adjacent layers of the subscript. For example, a->

The weights of the line segments representing the 1 st node of Layer1 and the 2 nd node of Layer2 are input. These weights determine the model's behaviorWith neural networks, the goal is to compute weights from samples. The nodes of each intermediate layer and output layer are a Logistic function g (z) =a. For example

The input value representing node 1 of Layer2 is brought into the Logistic function to obtain the output +.>

The bias parameter matrix between each layer of the neural network is as follows: b= [ B ] ₁ b ₂ ...b _n ] ^T Inputs to a neural network are known as: y= [ Y ] ₁ y ₂ ...y _n ] ^T The output of the neural network is:

introducing a nonlinear operator:

it can be deduced that:

initializing weight parameters: the weight parameter w is randomly initialized to lie between [ -epsilon, epsilon ] which is a preset, sufficiently small value.

Training a neural network model: the process of training the neural network model is mainly divided into two steps, namely, calculating a cost function J (theta), and adjusting a parameter theta to enable the cost function J (theta) to be as small as possible. The forward propagation algorithm is adopted as follows, the output of each sample under the current neural network model is calculated for the sample, the cost function is obtained, and the weight parameters are updated according to the output. Defining a cost function J (theta), wherein m is the number of samples, and since the neural network has K outputs, the cost function correspondingly calculates the cost of the K outputs, and the calculation formula is as follows:

The back propagation algorithm is used to adjust the parameter θ such that the cost function value J (θ) is as small as possible as follows. The back propagation algorithm updates each weight coefficient by taking the partial derivative of the cost function with respect to each weight coefficient. For example, first, the gradient of the last layer is calculated: (1) Calculating the gradient of the cost function value to the nonlinear operator, (2) calculating the gradient of the output of the neural network to the bias and the weight between adjacent layers. Updating the gradient according to the negative direction of the gradient; next, the gradient of the penultimate layer is calculated: (1) Calculating the gradient of the error returned by the upper layer to the nonlinear operator, (2) calculating the gradient of Hn-1 (H is the output of each layer after the activation function) to the bias and the weight between adjacent layers, and updating the gradient according to the negative direction of the gradient; finally, after passing back layer by layer, the gradient of the first layer is finally calculated: (1) Calculating the gradient of the error returned by the second layer to the nonlinear operator; (2) calculating the gradient of H1 to bias and adjacent inter-layer weight. And updates the gradient in the negative direction of the gradient. And after the cycle of the first back propagation process is finished, continuing the forward propagation to obtain output, and backward propagation to update parameters until the mean square error is minimum, thereby completing the training process of the neural network model.

Fig. 3 shows bit error rate performance of different transceiver antenna configurations of the MIMO system after the signal transmission and detection techniques of compressed sensing and neural networks are adopted. Here a flat fading channel is assumed. Under the conventional scheme, 4 transmit antennas can only transmit 4 data symbols at the same time. By applying the scheme of the application, BPSK modulation signals are firstly adopted to obtain original signals x ^4×1 Random Gaussian matrix A ^4ρ×4 As a compressed dimension-reducing matrix, z is obtained. If ρ=0.5, only 2 transmitting antennas are needed to realize the transmission of the original data. If ρ=0.75, only 3 transmitting antennas are needed to realize the transmission of the original data. The receiving end 2 of the device obtains a reconstructed transmitting signal by adopting the models 1 and 2 trained by the neural network through 2 or 3 receiving antennas

. The error rate performance of the scheme is shown as (2×2) -4 in the figure, the first number in the brackets represents the number of transmitting antennas, the second number represents the number of receiving antennas, and the last number represents the original data length. The number of receiving antennas is increased and this scheme is denoted as (3 x 3) -4. Compared with the traditional signal detection algorithm-Zero Forcing detection (ZF) of the MIMO system, the scheme provided by the application can reduce the number of required receiving and transmitting antennas while ensuring the error rate under the condition of high SNR.

The zero forcing detection algorithm eliminates interference between each transmission signal by multiplying the received signal y by a filter matrix WZF, so as to estimate each transmission symbol, where the filter matrix is: w (W) _ZF ＝H ^-1 ＝(H ^H H) ^-1 H ^H The estimated signal vector is therefore:

after obtaining the estimated signal vector, mapping the estimated signal vector to a constellation point with the nearest Euclidean distance in a constellation diagram, wherein the constellation point is used as an optimal solution so as to recover a final symbol vector XZF.

Fig. 4 shows bit error rate performance of different transceiver antenna configurations of a MIMO system after signal transmission and detection techniques using compressed sensing and neural networks. Here a flat fading channel is assumed. Under the conventional scheme, only 20 data symbols can be simultaneously transmitted by 20 transmit antennas. By applying the scheme of the application, BPSK modulation signals are adopted first, and then x is calculated ^20×1 Divided into 5 groups, the vector x of each group _i (i=1, 2,3,4, 5) is equal to 4. Random Gaussian matrix A ^4ρ×4 As a compressed dimension-reducing matrix, thereby obtaining z _i . Will z _i Cascading to obtain the vector to be sent

. If ρ=0.5, only 10 transmit antennas are neededThe transmission of the original data can be achieved. If ρ=0.75, 15 transmitting antennas are needed to realize the transmission of the original data. The device receiving end 2 also passes through 10 receiving antennas, the error rate performance of the scheme is shown as (10×10) -20 in the figure, the first number of transmitting antennas in brackets, the second number represents the number of receiving antennas, and the last number represents the original data length. The same data, the same packet number, ρ=0.75, the compressed dimension-reduction matrix is a ^3×4 The number of receiving antennas increases, and this scheme is denoted as (15×15) -20. Compared with the classical detection algorithm ZF (Zero Forcing), the scheme provided by the invention can reduce the number of the required receiving and transmitting antennas while guaranteeing the error rate under the condition of high SNR. Therefore, the scheme provided by the application can reduce the number of the required receiving and transmitting antennas while guaranteeing the error rate under the condition of high SNR.

As described above, the enhanced spatial multiplexing method provided by the present invention can combine the neural network technology on the basis of the existing MIMO system, transmit the same data amount on the basis of reducing the number of antennas, reduce the number of antennas required, and improve the multiplexing gain and the system capacity.

It should be noted that the above-described embodiments are exemplary, and that a person skilled in the art, in light of the present disclosure, may devise various solutions that fall within the scope of the present disclosure and fall within the scope of the present disclosure. It should be understood by those skilled in the art that the present description and drawings are illustrative and not limiting to the claims. The scope of the invention is defined by the claims and their equivalents.

Claims

1. A method for reconstructing an original signal of an antenna signal, the method comprising at least:

obtaining a first neural network model and a second neural network model by using a BP training algorithm of deep learning;

solving a high-dimensional sparse signal theta from a low-dimensional target signal through the constructed first neural network model;

the sparse signal theta is input into the constructed second neural network model to reconstruct to obtain an original signal.

2. The method of claim 1, wherein the step of obtaining the first neural network model and the second neural network model using a BP training algorithm for deep learning comprises:

using y and θ, a first set of training samples is constructed;

using θ and x, a second set of training samples is constructed;

through a deep learning BP training algorithm, a first set of training samples results in a first neural network model, a second set of training samples results in a second neural network model,

when the input signal is the signal y, the reconstructed transmission data is obtained

y represents an input signal of the first neural network model, and θ represents a sparse vector; x represents the vector of the transmitted signal,

representing the output original transmit data stream.

3. The method for original signal reconstruction of an antenna signal according to claim 1 or 2, further comprising:

and carrying out channel estimation on the low-dimensional target signal and taking the obtained channel parameter matrix as the input of the first neural network model.

4. A method of raw signal reconstruction of an antenna signal according to any one of claims 1 to 3, characterized in that the method further comprises:

and forming a redundant dictionary D by taking all possible combinations of the transmitted signal vectors x as different column vectors of the redundant dictionary respectively, so as to realize sparse signals theta of the transmitted signal vectors.

5. The method for reconstructing an original signal of an antenna signal according to any one of claims 1 to 4, wherein the process of training a neural network model mainly comprises:

the cost function J (theta) is calculated,

the parameter theta is adjusted so that the cost function value J (theta) is as small as possible,

the cost function J (theta) is obtained by calculating the output of each sample under the current neural network model by adopting a forward propagation algorithm.

6. The method for reconstructing an original signal of an antenna signal according to claim 5, wherein,

adopting a back propagation algorithm to adjust the parameter theta so as to enable the cost function value J (theta) to be as small as possible;

The back propagation algorithm updates each weight coefficient by taking the partial derivative of the cost function J (θ) with respect to each weight coefficient.

7. An original signal reconstruction device of an antenna signal is characterized by comprising at least a first neural network signal processing module (202) and a second neural network signal processing module (203),

the second neural network signal processing module (203) reconstructs an original signal by inputting a high-dimensional sparse signal θ, which is solved from a low-dimensional target signal by the first neural network signal processing module (202) by using the first neural network model constructed by the first neural network signal processing module, into the second neural network model constructed by the first neural network signal processing module.

8. The apparatus for reconstructing an original signal of an antenna signal according to claim 7, further comprising a channel estimation module (201) configured to perform channel estimation based on the received compressed reduced-dimension processed low-dimension target signal and to take the obtained channel parameter matrix as an input of the first neural network model.

9. The apparatus for raw signal reconstruction of an antenna signal according to claim 7 or 8, wherein the first neural network signal processing module (202) utilizes y and θ to construct a first set of training samples;

The second neural network signal processing module (203) utilizes θ and x to form a second set of training samples;

representing the output original transmit data stream.

10. The apparatus for raw signal reconstruction of an antenna signal according to any one of claims 7 to 9, wherein the process of training a neural network model of the first neural network signal processing module (202) and the second neural network signal processing module (203) mainly comprises:

the cost function J (theta) is calculated,