CN111628833A - MIMO antenna number estimation method based on convolutional neural network - Google Patents
MIMO antenna number estimation method based on convolutional neural network Download PDFInfo
- Publication number
- CN111628833A CN111628833A CN202010523557.9A CN202010523557A CN111628833A CN 111628833 A CN111628833 A CN 111628833A CN 202010523557 A CN202010523557 A CN 202010523557A CN 111628833 A CN111628833 A CN 111628833A
- Authority
- CN
- China
- Prior art keywords
- data
- antennas
- training
- neural network
- convolutional neural
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/10—Monitoring; Testing of transmitters
- H04B17/101—Monitoring; Testing of transmitters for measurement of specific parameters of the transmitter or components thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/20—Monitoring; Testing of receivers
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Radio Transmission System (AREA)
Abstract
The invention discloses a convolutional neural network-based MIMO antenna number estimation method, which comprises an MIMO antenna system of a cooperative communication party and a third-party non-cooperative communication receiving party, wherein both parties of the cooperative communication MIMO antenna system adopt multiple antennas to transmit and receive signals, the non-cooperative party adopts a single antenna to receive signals, and the non-cooperative receiving party comprises the following steps: 1) classifying and packaging the signals; 2) constructing a model; 3) training; 4) testing; 5) evaluating; 6) and (6) adjusting. The method can improve the estimation accuracy of the number of MIMO transmitting antennas.
Description
Technical Field
The invention belongs to the field of wireless communication, and relates to the estimation of the number of Multiple Input Multiple Output (MIMO) transmitting antennas, in particular to a method for estimating the number of the MIMO antennas based on a convolutional neural network.
Background
The detection of MIMO signals identifies a hot research area in non-cooperative communications, whether in the military or civilian communications fields. The estimation of the number of MIMO antennas is a prerequisite and key for subsequent channel estimation, coding scheme identification and signal demodulation.
The existing algorithm mainly includes a method based on Akaike Information Criterion (AIC for short), a Minimum Description Length (MDL for short), and a method for separating noise characteristic values of random matrix signals. Such an algorithm has a disadvantage in that the number of antennas on the receiving side must be greater than the number of antennas on the transmitting side, and a large number of antennas are required on the receiving side without knowing the number of antennas on the transmitting side.
In recent years, neural networks have been widely used in image processing and natural language processing, and studies on modulation scheme recognition have been advanced in the field of communications. The deep learning convolutional neural network has a feature extraction function, and the number of the antennas can be identified by the network when signals are sent into the convolutional neural network.
Disclosure of Invention
The invention aims to provide a method for estimating the number of MIMO antennas based on a convolutional neural network, aiming at the defects of the prior art. The method can improve the estimation accuracy of the number of MIMO transmitting antennas.
The technical scheme for realizing the purpose of the invention is as follows:
a method for estimating the number of MIMO antennas based on a convolutional neural network comprises an MIMO antenna system of a cooperative communication party and a non-cooperative communication receiving system of a third party, wherein both parties of the cooperative communication MIMO antenna system adopt multiple antennas to transmit and receive signals, the non-cooperative party adopts a single antenna to receive signals, and the non-cooperative receiving party comprises the following steps:
1) and (3) classification and packaging of signals: the single antenna inputs the collected I/Q radio frequency signals to the convolutional neural network for label processing, the collected signal data brand different labels according to the different antenna number conditions of the transmitting end, in order to provide enough information for the neural network, a plurality of groups of data storage and I/Q two-path radio frequency signals acquired by a single antenna need to be acquired, I/Q components are orthogonal and are not coherent, 1024I/Q data are used as a training packet, 2000 groups of data are respectively adopted for training 2 antennas, 3 antennas and 4 antennas to form a data set, information source data are coded and modulated, and then loading data to different antennas through MIMO signal processing, wherein the MIMO signal processing comprises spatial multiplexing and spatial diversity, the spatial multiplexing divides the data into a plurality of data streams, different coding matrixes are adopted for transmission on different antennas, and A is carried out.SMTo code the matrix, S1,S2Represents different symbol data:
space diversity corresponds to transmitting relatively redundant data on different channels, which can improve the stability of the system, and can be divided into receiving diversity and transmitting diversity, and space-time block coding in the diversity technology applies space and time diversity, such as Alamouti coding, BALTo code the matrix, S1,S2Represents the data of different symbols, -represents the inverse, represents the conjugate:
in the neural network training model, data of different transmitting antennas with spatial multiplexing and spatial diversity are input simultaneously;
2) constructing a model: inputting the data set obtained in the step 1) into a convolutional neural network model, wherein the model outputs different numbers of transmitting antennas of signals in a classified manner, the convolutional neural network model is provided with an input layer, a first convolutional layer, a second convolutional layer, a first dense connection layer, a second dense connection layer and an output layer which are sequentially connected, wherein an activation function in the first convolutional layer and the second convolutional layer is a relu function, an activation function selected by the first dense connection layer is a relu function, a softmax classified activation function is selected by the second dense connection layer, and the output layer outputs 3 nodes which respectively correspond to 3 conditions of 2 transmitting antennas, 3 transmitting antennas and 4 transmitting antennas;
3) training: according to the data collected in 1), according to the training data: inputting training data into a convolutional neural network for training, wherein the test data is 7: 3;
4) and (3) testing: inputting the test data set into the convolutional neural network trained in the step 3), estimating that the value on the diagonal line in the convolutional network test confusion matrix is greater than 0.8 according to the number of the antennas, and the fitting degree precision of curves of the training precision and the test precision in the training performance is more than 0.8, so that the performance effect of the test data passing through the training model is considered to be good;
5) evaluation: according to the performance in the test in the step 4), if the value on the diagonal line in the confusion matrix is greater than 0.8 or the precision in the training performance is greater than 0.8, the performance of the convolutional neural network is considered to meet the requirement of the antenna number estimation classification, and the performance is better when the value is approximate to 1;
6) adjusting: and (3) evaluating according to the test in the step 4) and the performance according to the step 5), if the loss degree of the test is reduced along with the loss degree of the training, the network is proved to be in learning, which is best, so as to determine whether to change the number and size parameters of the convolution kernels of the model in the step 2), and improve the recognition rate.
In the technical scheme, the condition that the number of receiving antennas is greater than the number of transmitting antennas does not need to be set at a receiving end, the condition that the estimation performance is rapidly deteriorated when the number of the receiving antennas at the receiving end is less than the number of the transmitting antennas at the transmitting end can be avoided, and meanwhile, the convolutional neural network ensures higher estimation performance.
The method can improve the estimation accuracy of the number of MIMO transmitting antennas.
Drawings
Fig. 1 is a schematic diagram of an identification structure of a MIMO antenna according to an embodiment;
FIG. 2 is a diagram illustrating a MIMO-OFDM system according to an embodiment;
FIG. 3 is a schematic flow chart of an embodiment of a method;
FIG. 4 is a diagram showing a structure of a convolutional neural network in an embodiment;
FIG. 5 is a diagram illustrating the classification accuracy of the antenna number estimation in the embodiment;
FIG. 6 is a schematic diagram of an embodiment of a convolutional network test confusion matrix for antenna number estimation;
fig. 7 is a diagram showing the performance of training in the embodiment.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples, but the present invention is not limited thereto.
Example (b):
as shown in fig. 3, the method for estimating the number of MIMO antennas based on convolutional neural network includes a MIMO antenna system of a cooperative communication party and a third-party non-cooperative communication receiving system, where both parties of the cooperative communication MIMO antenna system transmit and receive signals using multiple antennas, and the non-cooperative party receives signals using a single antenna, and as shown in fig. 1, the non-cooperative receiving party includes the following steps:
the method comprises the following steps:
1) and (3) classification and packaging of signals: the method comprises the steps that a single antenna carries out label processing on collected I/Q radio frequency signals before the signals are input into a convolutional neural network, collected signal data are printed with different labels according to the number of different antennas of a transmitting end, multiple groups of data are required to be collected and stored in order to provide enough information for the neural network, two paths of I/Q signals collected by the single antenna are orthogonal and are not coherent, 1024I/Q data are used as a training packet, and 2 antennas, 3 antennas and 4 antennas transmit signalsIn the case of transmission, 2000 groups are collected to form a data set, as shown in fig. 2, source data is coded and modulated, and then data is loaded to different antennas through MIMO signal processing, which includes spatial multiplexing and spatial diversity, the spatial multiplexing divides data into several data streams, different coding matrices are used for transmission on different antennas, aSMRepresenting a coding matrix, S1,S2Represents different symbol data:
space diversity corresponds to transmitting relatively redundant data in different channels, which can improve the stability of the system, and can be divided into receiving diversity and transmitting diversity, the space-time block coding in the diversity technology applies space and time diversity, in this case, Alamouti coding, BALTo code the matrix, S1,S2Represents the data of different symbols, -represents the inverse, represents the conjugate:
in the neural network training model, data of different transmitting antennas with spatial multiplexing and spatial diversity are input simultaneously;
2) constructing a model: inputting the data set obtained in step 1) into a convolutional neural network model, which is provided with an input layer, a first convolutional layer, a second convolutional layer, a first dense connection layer, a second dense connection layer and an output layer that are sequentially connected, and the model outputs different numbers of transmitting antennas of signals, wherein an activation function in the first convolutional layer and the second convolutional layer is a relu function, an activation function selected by the first dense connection layer is a relu function, a softmax classification activation function is selected by the second dense connection layer, the output layer outputs 3 nodes corresponding to the situations of 2 transmitting antennas, 3 transmitting antennas and 4 transmitting antennas, in this example, as shown in fig. 4, the dropoff of the convolutional neural network is set to be 0.5, the number of convolutional cores of the first convolutional layer is 170, the size of the first convolutional layer is 1 × 3, the number of convolutional cores of the second convolutional layer is 75, the size of the second convolutional layer is 2 × 3, the number of the first full-connection layer neurons is 256, the number of the second full-connection layer neurons is 3, and finally, the number of the output 3 nodes corresponds to the number of the antennas of the test antenna;
3) training: according to the data collected in 1), according to the training data: inputting training data into a convolutional neural network for training, wherein the test data is 7: 3;
4) and (3) testing: inputting the test data set into the convolutional neural network trained in the step 3), estimating that the value on the diagonal line in the convolutional network test confusion matrix is greater than 0.8 according to the number of the antennas, and the fitting degree precision of curves of the training precision and the test precision in the training performance is more than 0.8, so that the performance effect of the test data passing through the training model is considered to be good;
5) evaluation: according to the performance in the test in the step 4), if the value on the diagonal line in the confusion matrix is greater than 0.8 or the precision in the training performance is greater than 0.8, the performance of the convolutional neural network is considered to meet the requirement of the antenna number estimation classification, and the performance is better when the value is approximate to 1;
6) adjusting: and (3) evaluating according to the test in the step 4) and the performance according to the step 5), if the loss degree of the test is reduced along with the loss degree of the training, the network is proved to be in learning, which is best, so as to determine whether to change the number and size parameters of the convolution kernels of the model in the step 2), and improve the recognition rate.
As shown in fig. 5, under the condition of different signal-to-noise ratios, the classification progress of the data is schematically illustrated by the convolutional neural network model, and it can be seen that the accuracy of the model for classification can reach ninety-nine percent.
As shown in fig. 6, the antenna number estimation convolutional network tests a confusion matrix diagram, and through the matrix diagram, it is easy to observe that crosstalk misrecognition occurs in detection of different antenna numbers under low signal-to-noise ratio, but the probability of misrecognition is low.
As shown in fig. 7, in the training performance diagram of this embodiment, the horizontal axis is the number of epochs for training, the vertical axis is a loss value or an error, one epoch is data to complete one training, during the training process, the model continuously adjusts parameters by itself, so that the loss value loss continuously decreases, and the accuracy acc continuously increases.
Claims (1)
1. The method for estimating the number of the MIMO antennas based on the convolutional neural network comprises an MIMO antenna system of a cooperative communication party and a third-party non-cooperative communication receiving party, wherein both the cooperative communication MIMO antenna system and the non-cooperative communication receiving party adopt multiple antennas to transmit and receive signals, the non-cooperative communication receiving party adopts a single antenna to receive the signals, and the non-cooperative receiving party comprises the following steps:
1) and (3) classification and packaging of signals: the single antenna carries out label processing on the acquired I/Q radio frequency signals, the acquired I/Q radio frequency signal data are branded with different labels according to the conditions of different antenna numbers of the transmitting end, a plurality of groups of data are acquired and stored, I/Q components of the I/Q radio frequency signals acquired by the single antenna are orthogonal and are not coherent, 1024I/Q data are used as a training packet, 2000 groups are adopted during training of 2 antennas, 3 antennas and 4 antennas, and a data set is formed. Source data is coded and modulated, then data is loaded to different antennas through MIMO signal processing, the MIMO signal processing comprises space multiplexing and space diversity, the space multiplexing divides the data into several data streams, different coding matrixes are adopted for transmission on different antennas, and S1,S2Represents different symbol data:
the space diversity is to send the relative redundant data in different channels, and the space diversity can be divided into receiving diversity and transmitting diversity, the space-time block coding in the diversity technology applies the space and time diversity, and the data of different transmitting antennas of the space multiplexing and the space diversity is input in the convolutional neural network training model;
2) constructing a model: inputting the data set obtained in the step 1) into a convolutional neural network model, wherein the model outputs different numbers of transmitting antennas of signals in a classified manner, the convolutional neural network model is provided with an input layer, a first convolutional layer, a second convolutional layer, a first dense connection layer, a second dense connection layer and an output layer which are sequentially connected, wherein an activation function in the first convolutional layer and the second convolutional layer is a relu function, an activation function selected by the first dense connection layer is a relu function, a softmax classified activation function is selected by the second dense connection layer, and the output layer outputs 3 nodes which respectively correspond to 3 conditions of 2 transmitting antennas, 3 transmitting antennas and 4 transmitting antennas;
3) training: according to the data collected in step 1), according to the training data: inputting training data into a convolutional neural network for training, wherein the test data is 7: 3;
4) and (3) testing: inputting the test data set into the convolutional neural network trained in the step 3), estimating that the value on the diagonal line in the convolutional network test confusion matrix is greater than 0.8 according to the number of the antennas, and the fitting degree precision of curves of the training precision and the test precision in the training performance is more than 0.8, so that the performance effect of the test data passing through the training model is considered to be good;
5) evaluation: according to the performance in the test in the step 4), if the value on the diagonal line in the confusion matrix is greater than 0.8 or the precision in the training performance is greater than 0.8, the performance of the convolutional neural network is considered to meet the requirement of the antenna number estimation classification, and the performance is better when the value is approximate to 1;
6) adjusting: and (3) evaluating according to the test in the step 4) and the performance according to the step 5), if the loss degree of the test is reduced along with the loss degree of the training, the network is proved to be in learning, which is best, so as to determine whether to change the number and size parameters of the convolution kernels of the model in the step 2), and improve the recognition rate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010523557.9A CN111628833B (en) | 2020-06-10 | 2020-06-10 | MIMO antenna number estimation method based on convolutional neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010523557.9A CN111628833B (en) | 2020-06-10 | 2020-06-10 | MIMO antenna number estimation method based on convolutional neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111628833A true CN111628833A (en) | 2020-09-04 |
CN111628833B CN111628833B (en) | 2022-02-08 |
Family
ID=72271996
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010523557.9A Active CN111628833B (en) | 2020-06-10 | 2020-06-10 | MIMO antenna number estimation method based on convolutional neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111628833B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111368384A (en) * | 2018-12-07 | 2020-07-03 | 华为技术有限公司 | Method and equipment for predicting antenna engineering parameters |
CN113300746A (en) * | 2021-05-24 | 2021-08-24 | 内蒙古大学 | Millimeter wave MIMO antenna and hybrid beam forming optimization method and system |
CN113556157A (en) * | 2021-06-08 | 2021-10-26 | 西安电子科技大学 | Method and system for estimating number of transmitting antennas of MIMO system under non-Gaussian interference |
WO2023168767A1 (en) * | 2022-03-10 | 2023-09-14 | 华南理工大学 | Multi-mode transmission line-based wired cmimo signal transmission method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050271157A1 (en) * | 2004-05-27 | 2005-12-08 | Airgo Networks, Inc. | Detecting the number of transmit antennas in wireless communication systems |
EP2517037A1 (en) * | 2009-12-21 | 2012-10-31 | Thales | Method for estimating the number of incident sources in a sensor array by means of estimating noise statistics |
CN106713190A (en) * | 2017-01-05 | 2017-05-24 | 西安电子科技大学 | MIMO (Multiple Input Multiple Output) transmitting antenna number blind estimation algorithm based on random matrix theory and feature threshold estimation |
CN108494710A (en) * | 2018-03-30 | 2018-09-04 | 中南民族大学 | Visible light communication MIMO anti-interference noise-reduction methods based on BP neural network |
CN108616470A (en) * | 2018-03-26 | 2018-10-02 | 天津大学 | Modulation Signals Recognition method based on convolutional neural networks |
US20180324595A1 (en) * | 2017-05-05 | 2018-11-08 | Ball Aerospace & Technologies Corp. | Spectral sensing and allocation using deep machine learning |
CN110266620A (en) * | 2019-07-08 | 2019-09-20 | 电子科技大学 | 3D MIMO-OFDM system channel estimation method based on convolutional neural networks |
US10637544B1 (en) * | 2018-04-24 | 2020-04-28 | Genghiscomm Holdings, LLC | Distributed radio system |
-
2020
- 2020-06-10 CN CN202010523557.9A patent/CN111628833B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050271157A1 (en) * | 2004-05-27 | 2005-12-08 | Airgo Networks, Inc. | Detecting the number of transmit antennas in wireless communication systems |
EP2517037A1 (en) * | 2009-12-21 | 2012-10-31 | Thales | Method for estimating the number of incident sources in a sensor array by means of estimating noise statistics |
CN106713190A (en) * | 2017-01-05 | 2017-05-24 | 西安电子科技大学 | MIMO (Multiple Input Multiple Output) transmitting antenna number blind estimation algorithm based on random matrix theory and feature threshold estimation |
US20180324595A1 (en) * | 2017-05-05 | 2018-11-08 | Ball Aerospace & Technologies Corp. | Spectral sensing and allocation using deep machine learning |
CN108616470A (en) * | 2018-03-26 | 2018-10-02 | 天津大学 | Modulation Signals Recognition method based on convolutional neural networks |
CN108494710A (en) * | 2018-03-30 | 2018-09-04 | 中南民族大学 | Visible light communication MIMO anti-interference noise-reduction methods based on BP neural network |
US10637544B1 (en) * | 2018-04-24 | 2020-04-28 | Genghiscomm Holdings, LLC | Distributed radio system |
CN110266620A (en) * | 2019-07-08 | 2019-09-20 | 电子科技大学 | 3D MIMO-OFDM system channel estimation method based on convolutional neural networks |
Non-Patent Citations (3)
Title |
---|
AKRAM MARSEET 等: "Application of complex-valued convolutional neural network for next generation wireless networks", 《2017 IEEE WESTERN NEW YORK IMAGE AND SIGNAL PROCESSING WORKSHOP (WNYISPW)》 * |
YAN JUN JOHN TEOH 等: "RF and Network Signature-based Machine Learning on Detection of Wireless Controlled Drone", 《2019 PHOTONICS & ELECTROMAGNETICS RESEARCH SYMPOSIUM - SPRING (PIERS-SPRING)》 * |
黄雨迪: "基于机器学习的MIMO***收发机设计理论与方法", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111368384A (en) * | 2018-12-07 | 2020-07-03 | 华为技术有限公司 | Method and equipment for predicting antenna engineering parameters |
CN113300746A (en) * | 2021-05-24 | 2021-08-24 | 内蒙古大学 | Millimeter wave MIMO antenna and hybrid beam forming optimization method and system |
CN113300746B (en) * | 2021-05-24 | 2022-04-15 | 内蒙古大学 | Millimeter wave MIMO antenna and hybrid beam forming optimization method and system |
CN113556157A (en) * | 2021-06-08 | 2021-10-26 | 西安电子科技大学 | Method and system for estimating number of transmitting antennas of MIMO system under non-Gaussian interference |
WO2023168767A1 (en) * | 2022-03-10 | 2023-09-14 | 华南理工大学 | Multi-mode transmission line-based wired cmimo signal transmission method |
Also Published As
Publication number | Publication date |
---|---|
CN111628833B (en) | 2022-02-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111628833B (en) | MIMO antenna number estimation method based on convolutional neural network | |
CN106059972B (en) | A kind of Modulation Identification method under MIMO correlated channels based on machine learning algorithm | |
CN1941659B (en) | Spatial multiplexing detection apparatus and method in MIMO system | |
EP1548971A2 (en) | Constellation-rotating orthogonal space-time block coding technique | |
CN107835068B (en) | Low-complexity orthogonal space modulation spherical decoding detection algorithm with transmit diversity | |
CN106059639B (en) | Transmitting antenna number blindness estimation method based on your circle of matrix lid | |
Wang et al. | A deep learning-based intelligent receiver for improving the reliability of the MIMO wireless communication system | |
CN114745230B (en) | OTFS signal receiving and recovering method based on deep neural network structure | |
CN110659684A (en) | Convolutional neural network-based STBC signal identification method | |
CN110086743A (en) | A kind of short burst MIMO-OFDM communication system and method based on differential encoding | |
CN111327381A (en) | Joint optimization method of wireless communication physical layer transmitting and receiving end based on deep learning | |
CN102577163B (en) | Signal detection apparatus and method in spatial multiplexing system | |
CN114745248B (en) | DM-GSM signal detection method based on convolutional neural network | |
CN113971430A (en) | Signal detection and model training method, device, equipment and storage medium | |
CN101888287A (en) | Signal detection method and device for multi-input multi-output receiver | |
Ye et al. | Bilinear convolutional auto-encoder based pilot-free end-to-end communication systems | |
CN115296759A (en) | Interference identification method based on deep learning | |
CN113114603B (en) | Information recovery method and device for MIMO-OFDM system | |
CN109286587A (en) | A kind of how active generalized space method for modulation detection | |
KR101911168B1 (en) | Adaptive signal detection method using MIMO-OFDM system and apparatus thereof | |
CN111314255B (en) | Low-complexity SISO and MIMO receiver generation method | |
CN106911429A (en) | For the signal detecting method of gsm communication system | |
KR101937559B1 (en) | Linear approximation signal detection apparatus using MIMO-OFDM system and method thereof | |
CN110912585A (en) | Antenna selection method based on channel factors | |
Zhang et al. | Joint estimation and detection for MIMO-STBC system based on deep neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |