CN110554352A - Method for estimating direction of arrival of interference source of aerospace measurement and control system based on VGG16 neural network - Google Patents
Method for estimating direction of arrival of interference source of aerospace measurement and control system based on VGG16 neural network Download PDFInfo
- Publication number
- CN110554352A CN110554352A CN201910860830.4A CN201910860830A CN110554352A CN 110554352 A CN110554352 A CN 110554352A CN 201910860830 A CN201910860830 A CN 201910860830A CN 110554352 A CN110554352 A CN 110554352A
- Authority
- CN
- China
- Prior art keywords
- neural network
- vgg16
- control system
- arrival
- interference source
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/02—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using radio waves
- G01S3/14—Systems for determining direction or deviation from predetermined direction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/02—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using radio waves
- G01S3/74—Multi-channel systems specially adapted for direction-finding, i.e. having a single antenna system capable of giving simultaneous indications of the directions of different signals
-
- 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Neurology (AREA)
- Variable-Direction Aerials And Aerial Arrays (AREA)
Abstract
A method for estimating the direction of arrival of an interference source of a space flight measurement and control system based on a VGG16 neural network relates to the field of estimation of the direction of arrival of the interference source of the measurement and control system based on the VGG16 neural network. The method aims to overcome the defects of high complexity, low efficiency or large error of the traditional DOA estimation algorithm and the problem of serious limitation caused by human factors in the estimation process of a measurement and control system. The invention effectively utilizes the VGG16 neural network technology, trains the existing data set through the VGG16 neural network method, thereby obtaining the VGG16 neural network model, and further utilizes the VGG16 neural network to estimate the DOA value of the interference source. The calculation method has high efficiency and small error, and is a better solution.
Description
the technical field is as follows:
The invention relates to the field of estimation of the direction of arrival of an interference source of a measurement and control system based on a VGG16 neural network.
Background art:
The comprehensive system for measuring and controlling the space flight target is called a space measurement and control system, and a Tracking and Data Relay Satellite System (TDRSS) is a major breakthrough of the space measurement and control technology in the 20 th century and is the mainstream technology of the current space-based multi-view measurement and control system. With the rapid development of wireless communication and aerospace technologies, the space electromagnetic environment becomes extremely complex, and various cooperative and non-cooperative radio signals exist in the frequency band applied by the space measurement and control system at the same time, so that great interference is caused to the space measurement and control system. The measurement and control communication system is concerned with spacecraft attitude control, orbit control and life and death of the spacecraft, has higher communication reliability and effectiveness requirements compared with a general communication and data transmission system, and how to solve the problem of interference resistance of the space measurement and control system becomes one of main attack directions of space science and technology of main space countries in the world. The airspace anti-interference is the current hot direction. Firstly, estimating the DOA of an interference source in the spatial domain anti-interference process;
At present, the existing method for estimating the DOA direction of the wave arrival of the interference source has the defects of high complexity, low efficiency and large error, and is seriously limited by human factors in the estimation process of a measurement and control system.
Disclosure of Invention
The invention aims to overcome the defects of high complexity, low efficiency or large error of the traditional DOA estimation algorithm and the problem of serious limitation caused by human factors in the estimation process of a measurement and control system, thereby providing a method for estimating the arrival direction of an interference source of a space measurement and control system based on a VGG16 neural network.
The method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network comprises the following steps:
the method comprises the following steps: sample data acquisition: and acquiring interference signals through each array element of the array antenna.
step two: sample feature extraction: in the data set manufacturing process, the DOA value is known, and the DOA value of the interference signal can be obtained by adopting a classic MUSIC algorithm, an improved MUSIC algorithm and the like.
Step three: construction of a neural network: the VGG16 neural network constructed by the invention has 13 convolutional layers and 3 full-connection layers. The number of the neurons in the input layer is determined according to the number of the array elements, and actually, the signals received by the array antenna are transmitted into a VGG16 neural network; the convolution layers have 13 layers, all adopt 3 × 3 convolution kernel sizes, and the maximum pooling sizes adopt 2 × 2; the output layer has a total of 800 neurons: the interference sources are respectively represented by-40 degrees to 40 degrees, the interval is 0.1 degree, and the total number is 800 degrees.
step four: training a neural network: during training, the input value is: x (t), X (t) is an array element input signal matrix, and in the neural network, a transfer function is selected as an s-shaped nonlinear function; the training result of the neural network sets the normalization of an output layer to 0.1; let θkDOA expected output representing a source of interference; and thetakactual network output representing the interference source DOA value;
The errors are therefore:
When the above formula is satisfied, the neural network stops training; otherwise, the neural network reversely transmits and modifies the weight value until the above formula is satisfied, and the training is finished at this moment.
Step five: processing array element data to be detected by using a neural network: and processing the array element data collected by the array antenna by using a trained neural network, extracting the characteristics of a 3 multiplied by 3 area around each data, and calculating an output value by using the neural network. If the output value meets the DOA estimation, namely 0.1 +/-0.1, assigning a value of 100; this makes it easy to implement a histogram display.
Step six: the number of interferers and the DOA value are calculated and displayed in a histogram. The method effectively utilizes the VGG16 neural network technology, trains the existing data set through the VGG16 neural network method, thereby obtaining the VGG16 neural network model, and further utilizes the VGG16 neural network to estimate the interference source DOA (direction of arrival) value. The calculation method has high efficiency and small error, is less limited by human factors in the estimation process of the measurement and control system, and is a better solution.
description of the drawings:
The invention is further illustrated with reference to the figures and examples.
FIG. 1 is a flow chart of a measurement and control system interference source DOA estimation method based on VGG 16.
fig. 2 is a schematic view of sample collection.
FIG. 3 is a diagram of a single neuron architecture.
Fig. 4 is a diagram of a neural network architecture.
the specific implementation mode is as follows:
The invention is further described with reference to the accompanying drawings and the detailed description below:
FIG. 1 is a flow chart of a measurement and control system interference source DOA estimation method based on VGG 16. It comprises the following steps:
the method comprises the following steps: sample data acquisition: interference signals are collected through each array element of the array antenna, and the row number, namely the sampling frequency, is determined.
Step two: sample feature extraction: in the data set manufacturing process, the DOA value is known, and the DOA value of the interference signal can be obtained by adopting a classic MUSIC algorithm, an improved MUSIC algorithm and the like.
step three: construction of a neural network: the VGG16 neural network constructed by the invention has 13 convolutional layers and 3 full-connection layers. The number of the neurons in the input layer is determined according to the number of the array elements, and actually, the signals received by the array antenna are transmitted into a VGG16 neural network; the convolution layers have 13 layers, all adopt 3 × 3 convolution kernel sizes, and the maximum pooling sizes adopt 2 × 2; the output layer has a total of 800 neurons: the interference sources are respectively represented by-40 degrees to 40 degrees, the interval is 0.1 degree, and the total number is 800 degrees.
Step four: training a neural network: during training, the input value is: x (t), X (t) is an array element input signal matrix, and in the neural network, a transfer function is selected as an s-shaped nonlinear function; as a result of the training of the neural network, we set the output layer to normalize to 0.1; let θk *A DOA expected output representing a source of interference; and thetakActual network output representing the interference source DOA value;
therefore, there are:
When the above formula is satisfied, the neural network stops training; otherwise, the neural network reversely transmits and modifies the weight value until the above formula is satisfied, and the training is finished at this moment.
Step five: processing array element data to be detected by using a neural network: and processing the array element data collected by the array antenna by using a trained neural network, extracting the characteristics of a 3 multiplied by 3 area around each data, and calculating an output value by using the neural network. If the output value meets the DOA estimation, namely 0.1 +/-0.1, assigning a value of 100; this makes it easy to implement a histogram display.
Step six: the number of interferers and the DOA value are calculated and displayed in a histogram.
Fig. 2 is a schematic diagram of the relationship between the array antenna and the interference source. The figure is an interference source and an array antenna (only schematic diagram), a plurality of array elements of the array antenna simultaneously receive signals of two interference sources of array elements A and B, and DOA values are A and B respectively.
Fig. 3 is a diagram of a single neuron structure, where the inputs are n eigenvalues, the n inputs are weighted and summed, and then the output value of a single neuron is obtained through a transfer function. In the invention, the transfer function selects an s-type nonlinear function,
Fig. 4 is a structural diagram of the VGG16 neural network. The VGG16 neural network constructed by the invention is divided into three layers: input layer, hidden layer (13 convolutional layers and 3 fully-connected layers) and output layer. The number of neurons in the input layer is the same as that of array elements, and the input of array element data of one array antenna is realized, xkThe kth learning objective for the input layer; volume of convoluting and average miningusing 3 × 3 convolution kernels, wherein pooling layers are subjected to 2 × 2 maximal pooling; the output layer has a total of 800 neurons: representing 800 angles of-40 degrees to 40 degrees, respectively, the fitting problem is converted into a classification problem.
Compared with the prior art, the invention has the following beneficial effects: the traditional DOA estimation method has large calculation amount and low efficiency due to spectrum peak search, and the error of the DOA estimation algorithm with relatively small calculation amount becomes large. The method and the device have the advantages that the problems of computation amount, efficiency, errors and the like are solved, the VGG16 convolutional neural network is used for estimating the interference source DOA, so that the precision can be greatly improved, the existing hardware such as a GPU is conveniently used for acceleration, the efficiency is greatly improved, and the precision of the interference source DOA estimation of the anti-interference scheme of the measurement and control system is ensured.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (5)
1. the method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network is characterized in that: it comprises the following steps:
Step one, sample data acquisition: collecting interference signals through each array element of an array antenna of the aerospace measurement and control system;
step two, sample feature extraction: extracting sample characteristics according to the interference signals acquired in the first step;
step three, building a VGG16 neural network: constructing a VGG16 neural network according to the sample characteristics extracted in the step two;
step four, training a neural network: training the neural network to obtain a trained VGG16 neural network;
Fifthly, estimating the direction of arrival of the interference source of the aerospace measurement and control system: and performing direction-of-arrival estimation on the array element data collected by the array antenna by using the trained neural network, and realizing the direction-of-arrival estimation of the interference source of the aerospace measurement and control system based on the VGG16 neural network.
2. The method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network as claimed in claim 1, wherein in the step of extracting the sample features in the step two: in the data set manufacturing process, the DOA value is used as the known value or the DOA value of the interference signal is obtained by adopting a classical MUSIC algorithm and is used as the sample characteristic of the interference signal.
3. the method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network as claimed in claim 1, wherein the VGG16 neural network constructed in the third step is specifically:
the multilayer ceramic material comprises 13 convolution layers and 3 full-connection layers, wherein: the number of input layer neurons is determined according to the number of array elements, and actually, signals received by the array antenna are transmitted into a VGG16 neural network; the convolution layers have 13 layers, all adopt 3 × 3 convolution kernel sizes, and the maximum pooling sizes adopt 2 × 2; the output layer has a total of 800 neurons: the interference sources are respectively represented by-40 degrees to 40 degrees, the interval is 0.1 degree, and the total number is 800 degrees.
4. The method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network of claim 1, wherein in the fourth step, the neural network is trained to obtain a trained VGG16 neural network; the specific method comprises the following steps:
training a neural network: during training, the input value is: x (t), X (t) is an array element input signal matrix, and in the neural network, a transfer function selects an s-type nonlinear function; setting the normalization of an output layer to be 0.1 according to the training result of the neural network; let θk *A DOA expected output representing a source of interference; and thetakActual network output representing the interference source DOA value;
the errors are therefore:
When the above formula is satisfied, the neural network stops training; otherwise, the neural network reversely transmits and modifies the weight value until the above formula is satisfied, and the training is finished at this moment.
5. the method for estimating the direction of arrival of the interference source of the aerospace measurement and control system based on the VGG16 neural network as claimed in claim 1, wherein in step five, the direction of arrival estimation is performed on the array element data collected by the array antenna by using the trained neural network, specifically:
processing the array element data to be tested by using a VGG16 neural network: processing the array element data collected by the array antenna by using a trained neural network, extracting the characteristics of a 3 multiplied by 3 area around each data, calculating an output value by using the neural network, and assigning a value of 100 if the output value accords with DOA estimation, namely 0.1 +/-0.1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910860830.4A CN110554352A (en) | 2019-09-11 | 2019-09-11 | Method for estimating direction of arrival of interference source of aerospace measurement and control system based on VGG16 neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910860830.4A CN110554352A (en) | 2019-09-11 | 2019-09-11 | Method for estimating direction of arrival of interference source of aerospace measurement and control system based on VGG16 neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110554352A true CN110554352A (en) | 2019-12-10 |
Family
ID=68740096
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910860830.4A Pending CN110554352A (en) | 2019-09-11 | 2019-09-11 | Method for estimating direction of arrival of interference source of aerospace measurement and control system based on VGG16 neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110554352A (en) |
Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050259006A1 (en) * | 2004-05-24 | 2005-11-24 | Samsung Electronics Co., Ltd. | Beam forming apparatus and method using interference power estimation in an array antenna system |
US20130329524A1 (en) * | 2012-06-08 | 2013-12-12 | Samsung Electronics Co., Ltd. | Neuromorphic signal processing device and method for locating sound source using a plurality of neuron circuits |
CN105301557A (en) * | 2015-11-06 | 2016-02-03 | 中国石油大学(华东) | Direction-of-arrival estimate configuration method |
CN105425205A (en) * | 2015-11-03 | 2016-03-23 | 天津津航计算技术研究所 | High-resolution circular and non-circular signal mixed incidence DOA estimation method |
US20160097853A1 (en) * | 2014-10-06 | 2016-04-07 | Nidec Elesys Corporation | Neural network-based radar system |
CN105530703A (en) * | 2016-01-28 | 2016-04-27 | 中国铁建电气化局集团北方工程有限公司 | GSM-R interference source positioning method based on wave arrival direction estimate value |
CN105891741A (en) * | 2016-06-20 | 2016-08-24 | 中国科学院电子学研究所 | Noise suppression method for magnetic field sensor network |
CN106355245A (en) * | 2016-09-12 | 2017-01-25 | 哈尔滨工业大学 | Method for integrating array antenna directional images on basis of neural network algorithms |
CN107657633A (en) * | 2017-09-28 | 2018-02-02 | 哈尔滨工业大学 | A kind of soil improving straw mulching rate measuring method based on BP neural network and sensor data acquisition |
CN108169708A (en) * | 2017-12-27 | 2018-06-15 | 中国人民解放军战略支援部队信息工程大学 | The direct localization method of modular neural network |
CN108449123A (en) * | 2018-03-05 | 2018-08-24 | 南京理工大学 | Spread spectrum communication system multi-target detection, identification and two dimension angular method of estimation over the ground |
US20180260702A1 (en) * | 2017-03-09 | 2018-09-13 | Oki Electric Industry Co., Ltd. | Information processing device and information processing method |
CN109085531A (en) * | 2018-08-27 | 2018-12-25 | 西安电子科技大学 | Near field sources angle-of- arrival estimation method neural network based |
CN109255308A (en) * | 2018-11-02 | 2019-01-22 | 陕西理工大学 | There are the neural network angle-of- arrival estimation methods of array error |
CN109862625A (en) * | 2019-01-21 | 2019-06-07 | 中国人民解放军战略支援部队信息工程大学 | A kind of shortwave radio monitor flexible networking method based on deep learning |
CN110007265A (en) * | 2019-04-30 | 2019-07-12 | 哈尔滨工业大学 | A kind of Wave arrival direction estimating method based on deep neural network |
CN110045322A (en) * | 2019-03-21 | 2019-07-23 | 中国人民解放军战略支援部队信息工程大学 | A kind of shortwave automatic direction finding method based on high-resolution direction finding sonagram intelligent recognition |
CN110045321A (en) * | 2019-04-12 | 2019-07-23 | 大连大学 | The steady DOA estimation method restored based on sparse and low-rank |
CN110045323A (en) * | 2019-03-14 | 2019-07-23 | 电子科技大学 | A kind of relatively prime battle array robust adaptive beamforming algorithm based on matrix fill-in |
CN110161453A (en) * | 2019-05-23 | 2019-08-23 | 成都理工大学 | Relatively prime array wide and narrow strip target DOA estimation algorithm based on multi input residual error network |
CN110221241A (en) * | 2019-04-29 | 2019-09-10 | 西安电子科技大学 | A kind of low elevation angle DOA estimation method based on RBF neural |
-
2019
- 2019-09-11 CN CN201910860830.4A patent/CN110554352A/en active Pending
Patent Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050259006A1 (en) * | 2004-05-24 | 2005-11-24 | Samsung Electronics Co., Ltd. | Beam forming apparatus and method using interference power estimation in an array antenna system |
US20130329524A1 (en) * | 2012-06-08 | 2013-12-12 | Samsung Electronics Co., Ltd. | Neuromorphic signal processing device and method for locating sound source using a plurality of neuron circuits |
US20160097853A1 (en) * | 2014-10-06 | 2016-04-07 | Nidec Elesys Corporation | Neural network-based radar system |
CN105425205A (en) * | 2015-11-03 | 2016-03-23 | 天津津航计算技术研究所 | High-resolution circular and non-circular signal mixed incidence DOA estimation method |
CN105301557A (en) * | 2015-11-06 | 2016-02-03 | 中国石油大学(华东) | Direction-of-arrival estimate configuration method |
CN105530703A (en) * | 2016-01-28 | 2016-04-27 | 中国铁建电气化局集团北方工程有限公司 | GSM-R interference source positioning method based on wave arrival direction estimate value |
CN105891741A (en) * | 2016-06-20 | 2016-08-24 | 中国科学院电子学研究所 | Noise suppression method for magnetic field sensor network |
CN106355245A (en) * | 2016-09-12 | 2017-01-25 | 哈尔滨工业大学 | Method for integrating array antenna directional images on basis of neural network algorithms |
US20180260702A1 (en) * | 2017-03-09 | 2018-09-13 | Oki Electric Industry Co., Ltd. | Information processing device and information processing method |
CN107657633A (en) * | 2017-09-28 | 2018-02-02 | 哈尔滨工业大学 | A kind of soil improving straw mulching rate measuring method based on BP neural network and sensor data acquisition |
CN108169708A (en) * | 2017-12-27 | 2018-06-15 | 中国人民解放军战略支援部队信息工程大学 | The direct localization method of modular neural network |
CN108449123A (en) * | 2018-03-05 | 2018-08-24 | 南京理工大学 | Spread spectrum communication system multi-target detection, identification and two dimension angular method of estimation over the ground |
CN109085531A (en) * | 2018-08-27 | 2018-12-25 | 西安电子科技大学 | Near field sources angle-of- arrival estimation method neural network based |
CN109255308A (en) * | 2018-11-02 | 2019-01-22 | 陕西理工大学 | There are the neural network angle-of- arrival estimation methods of array error |
CN109862625A (en) * | 2019-01-21 | 2019-06-07 | 中国人民解放军战略支援部队信息工程大学 | A kind of shortwave radio monitor flexible networking method based on deep learning |
CN110045323A (en) * | 2019-03-14 | 2019-07-23 | 电子科技大学 | A kind of relatively prime battle array robust adaptive beamforming algorithm based on matrix fill-in |
CN110045322A (en) * | 2019-03-21 | 2019-07-23 | 中国人民解放军战略支援部队信息工程大学 | A kind of shortwave automatic direction finding method based on high-resolution direction finding sonagram intelligent recognition |
CN110045321A (en) * | 2019-04-12 | 2019-07-23 | 大连大学 | The steady DOA estimation method restored based on sparse and low-rank |
CN110221241A (en) * | 2019-04-29 | 2019-09-10 | 西安电子科技大学 | A kind of low elevation angle DOA estimation method based on RBF neural |
CN110007265A (en) * | 2019-04-30 | 2019-07-12 | 哈尔滨工业大学 | A kind of Wave arrival direction estimating method based on deep neural network |
CN110161453A (en) * | 2019-05-23 | 2019-08-23 | 成都理工大学 | Relatively prime array wide and narrow strip target DOA estimation algorithm based on multi input residual error network |
Non-Patent Citations (11)
Title |
---|
MARIJA AGATONOVIĆ: "Empirical ANN models for 2D direction of arrival estimation", 《11TH SYMPOSIUM ON NEURAL NETWORK APPLICATIONS IN ELECTRICAL ENGINEERING》 * |
ZHANG-MENG LIU: "Direction-of-Arrival Estimation Based on Deep Neural Networks With Robustness to Array Imperfections", 《IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION ( VOLUME: 66, ISSUE: 12, DECEMBER 2018)》 * |
姚建丽: "应用SAPSO-BP神经网络的DOA估计方法", 《重庆理工大学学报(自然科学版)》 * |
孙涛: "基于改进BP神经网络的调节阀阀芯型面优化设计与研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
巫军卫: "基于RBF神经网络的弱信号DOA估计方法", 《计算机应用研究》 * |
张兴良: "基于二次特征提取的神经网络信号源个数估计方法", 《火力与指挥控制》 * |
张海涛: "基于VGG16的急性淋巴细胞白血病血液细胞显微图像人工智能辅助诊断分类研究", 《中国医疗设备》 * |
朱银兵: "基于空时滤波的弱干扰信号波达方向估计方法", 《海军工程大学学报》 * |
罗争: "基于选择性集成径向基神经网络的来波方向估计", 《新能源进展》 * |
范辉: "基于MUSIC算法的空间谱测向方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
葛晓凯: "利用深度学习方法的相干源DOA估计", 《信号处理》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020253690A1 (en) | Deep learning beam domain channel estimation method based on approximate message passing algorithm | |
CN103886207B (en) | Nested MIMO radar DOA estimation method based on compressed sensing | |
CN110109059B (en) | Radar radiation source signal identification method based on deep learning network | |
CN109655799B (en) | IAA-based covariance matrix vectorization non-uniform sparse array direction finding method | |
CN107436421B (en) | Mixed signal DOA estimation method under sparse Bayesian learning framework | |
Du et al. | Secondary radar signal processing based on deep residual separable neural network | |
CN109471074B (en) | Radar radiation source identification method based on singular value decomposition and one-dimensional CNN network | |
CN110967665A (en) | DOA estimation method of moving target echoes under multiple external radiation sources | |
CN110045323B (en) | Matrix filling-based co-prime matrix robust adaptive beamforming algorithm | |
CN112995892B (en) | Large-scale MIMO fingerprint positioning method based on complex neural network | |
CN107015205B (en) | False target elimination method for distributed MIMO radar detection | |
CN111580151B (en) | SSNet model-based earthquake event time-of-arrival identification method | |
CN105158735B (en) | Null tone Two-Dimensional Spectral Estimation method based on compression sampling array | |
CN108182474A (en) | Based on the direct localization method of multiple target for not correcting array and neural network | |
CN108169708A (en) | The direct localization method of modular neural network | |
CN103885045B (en) | Based on the circulation associating Adaptive beamformer method of Subarray partition | |
CN103428850A (en) | Compressed sensing based distributed multi-zone positioning method | |
CN106019215A (en) | Nested array direction-of-arrival angle estimation method based on fourth-order cumulants | |
CN103731189B (en) | The dynamic Antenna Subarray Division of conformal array antenna and Wave arrival direction estimating method | |
CN106680815A (en) | Tensor-sparse-representation-based MIMO radar's imaging method | |
CN112861813B (en) | Method for identifying human behavior behind wall based on complex value convolution neural network | |
CN112051552B (en) | Main lobe anti-interference method and device based on multiple stations | |
CN111368930A (en) | Radar human body posture identification method and system based on multi-class spectrogram fusion and hierarchical learning | |
CN108398659B (en) | Direction-of-arrival estimation method combining matrix beam and root finding MUSIC | |
CN113673312A (en) | Radar signal intra-pulse modulation identification method based on deep learning |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191210 |
|
WD01 | Invention patent application deemed withdrawn after publication |