CN112866150A - Wireless channel feature extraction and dimension reduction method and system - Google Patents
Wireless channel feature extraction and dimension reduction method and system Download PDFInfo
- Publication number
- CN112866150A CN112866150A CN202110044042.5A CN202110044042A CN112866150A CN 112866150 A CN112866150 A CN 112866150A CN 202110044042 A CN202110044042 A CN 202110044042A CN 112866150 A CN112866150 A CN 112866150A
- Authority
- CN
- China
- Prior art keywords
- channel
- frequency domain
- domain channel
- data
- time domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
- H04L25/024—Channel estimation channel estimation algorithms
- H04L25/0254—Channel estimation channel estimation algorithms using neural network algorithms
Abstract
The invention relates to a method and a system for extracting wireless channel characteristics and reducing dimension in the field of channel classification, wherein the method comprises the following steps: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver; reducing the dimensionality of the data to be detected or trained through data interception; converting the intercepted channel from a time domain channel to a frequency domain channel; estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion; the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics; and finally, the extracted frequency domain channel characteristics are sent to a channel classification algorithm module to finish the identification and classification of the channel environment where the communication transceiver is located. The invention can effectively recover the channel characteristics by extracting the time domain characteristics and converting the time domain characteristics into the frequency domain.
Description
Technical Field
The invention relates to the field of channel classification, in particular to a wireless channel feature extraction and dimension reduction method and a system, and more particularly to a wireless channel feature extraction and dimension reduction method and a channel classification sensing method.
Background
Typical channel models exist as follows: low altitude channel, urban channel, rural channel and mountainous channel models. Due to the fact that channel environments in wireless communication are complex and changeable, the signal transmission process is influenced by surrounding complex physical environments, and therefore signals under different channel scenes have obvious differences in the aspects of energy, power, time delay, channel response and the like.
The signal characteristic extraction method mainly comprises the time domain characteristics such as time delay and frequency deviation matching degree characteristics, high-order statistics, cyclostationarity, energy ratio of each order component of wavelet transform extraction signals and the like. By estimating the impulse response of the radio channel, the environment in which the communication transceiver is located is perceived, with the goal of accuracy of classification. If only one feature, such as a time domain feature, is used, the classification accuracy may be insufficient, and especially in a dynamic channel scenario, the channel may have time-selective fading, so that the accuracy of classification based on only the time domain feature is reduced.
If the feature extraction of a time domain or a frequency domain (the frequency domain can be a generalized frequency domain and can be a wavelet transform besides an FFT transform) can be combined, the accuracy and the reliability of the classification can be guaranteed. The existing communication waveform is common by adopting the OFDM technology, and the characteristic extraction by using a frequency domain is easier.
The data dimension reduction method mainly comprises a principal component analysis algorithm (PCA), a supervised linear dimension reduction algorithm (LDA), Local Linear Embedding (LLE), Laplace feature mapping and the like. PCA is the most commonly used linear dimensionality reduction method, which maps high-dimensional data into a low-dimensional space for representation through some linear projection, and expects the variance of the data to be maximum in the projected dimension, so that fewer data dimensions are used, and more original data characteristics are reserved.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a channel feature extraction and dimension reduction method and system.
The invention provides a wireless channel feature extraction and dimension reduction method, which comprises the following steps:
time domain channel characteristic estimation step: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
and (3) intercepting the time domain channel characteristics: the dimensionality of the data to be detected or trained is reduced by intercepting the time domain channel characteristic estimation data;
time-frequency domain channel conversion step: converting the intercepted channel from a time domain channel to a frequency domain channel;
estimating the frequency domain channel characteristics: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
and a frequency domain channel characteristic extraction step: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
the channel classification algorithm comprises the following steps: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
In some embodiments, in the time domain channel characteristic estimation step, the extraction of the signal characteristic is completed by performing cross-correlation between a pilot reference signal in the received signal and a local reference sequence, and performing normalization and gain combination operations.
In some embodiments, the time domain channel feature intercepting step is to intercept a portion meeting a predetermined requirement by observing a signal after feature extraction, and reduce a dimensionality of data, thereby reducing a complexity of the data.
In some embodiments, in the step of converting the time-frequency domain channel, the converting of the time-domain channel into the frequency-domain channel is performed by fast fourier transform.
In some embodiments, in the step of channel classification algorithm, the processing of the extracted frequency domain channel characteristics is completed by KNN, SVM, random forest, neural network or CNN algorithm.
The invention also provides a wireless channel feature extraction and dimension reduction system, which comprises:
a time domain channel characteristic estimation module: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
a time domain channel characteristic intercepting module: reducing the dimensionality of the data to be detected or trained through data interception;
the time-frequency domain channel conversion module: converting the intercepted channel from a time domain channel to a frequency domain channel;
a frequency domain channel characteristic estimation module: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
a frequency domain channel characteristic extraction module: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
a channel classification algorithm module: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
In some embodiments, in the time domain channel characteristic estimation module, the extraction of the signal characteristic is completed by performing cross-correlation between a pilot reference signal in the received signal and a local reference sequence, and performing normalization and gain combination operations.
In some embodiments, the time domain channel feature intercepting module intercepts a part with obvious feature difference by observing a signal after feature extraction, and reduces the dimensionality of data, thereby reducing the complexity of the data.
In some embodiments, in the time-frequency domain channel conversion module, the conversion of the time domain channel into the frequency domain channel is performed by Fast Fourier Transform (FFT).
In some embodiments, the channel classification algorithm module performs processing on the extracted frequency domain channel features through a KNN, SVM, random forest, neural network, or CNN algorithm.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention can effectively recover the channel characteristics after extracting the time domain characteristics and converting the time domain characteristics into the frequency domain;
2. the complexity of the data after the dimension reduction processing can be effectively reduced;
3. the invention can carry out classification training on modeling data or collected real environment data, and channel information can be effectively obtained after the modeling data or the collected real environment data pass through the channel classification module.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a block diagram of the algorithm flow of the present invention;
FIG. 2 is a time domain impulse response of a channel;
FIG. 3 is a truncated and dimension-reduced time domain impulse response of a channel;
FIG. 4 is a frequency domain impulse response of a channel;
fig. 5 shows the channel frequency domain impulse response after decimation.
Fig. 6 is a curve of accuracy of classification output by the KNN algorithm along with change of signal-to-noise ratio after 200-point time domain data is transformed to a frequency domain and feature extraction is performed.
Fig. 7 is a curve of accuracy of classification output by the KNN algorithm along with the change of the signal-to-noise ratio after 200-point time domain data is transformed to the frequency domain and then feature extraction, dimension reduction extraction, and the KNN algorithm is performed.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
The invention provides a channel characteristic extraction and dimension reduction method, which belongs to the classification of wireless channels, and the method is mainly based on a known pilot frequency reference signal, a time domain channel response is obtained by cross-correlation between an unknown channel and a local reference sequence, the obtained time domain channel response is truncated, the frequency domain channel response is converted into a frequency domain through FFT (fast Fourier transform), extraction operation is carried out on the frequency domain channel response to obtain dimension reduced data, and the channel classification process is carried out by obtaining the relevant characteristics of the channel.
The characteristic extraction to be adopted by the invention is to carry out normalization, gain combination and other operations on the cross-correlation sequence of the pilot frequency reference signal and the local signal in the time domain. The data dimension reduction method adopted by the invention is to extract a certain proportion of data from the frequency domain impulse response in the frequency domain, thereby reducing the dimension of the data. And carrying out classification training on the data subjected to dimension reduction by using a machine learning method, so that the physical environment where the wireless communication transceiver is located can be sensed in a new environment. The methods to be adopted include KNN, CNN, SVM, random forest, neural network or CNN algorithm and the like.
Example 1
As shown in fig. 1-7, the present invention provides a channel feature extraction and dimension reduction method, which includes:
time domain channel characteristic estimation step: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
and (3) intercepting the time domain channel characteristics: reducing the dimensionality of the data to be detected or trained through data interception;
time-frequency domain channel conversion step: converting the intercepted channel from a time domain channel to a frequency domain channel;
estimating the frequency domain channel characteristics: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
and a frequency domain channel characteristic extraction step: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
the channel classification algorithm comprises the following steps: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
In the time domain channel characteristic estimation step, the extraction of the signal characteristics is completed by performing normalization and gain combination operations through the cross correlation of the pilot frequency reference signal in the received signal and the local reference sequence.
The time domain channel feature intercepting step is to intercept the part with obvious feature difference by observing the signal after feature extraction, and reduce the dimensionality of data, thereby reducing the complexity of the data. And the significant features in the portions with significant feature differences mainly refer to the energies of the multipaths.
In the time-frequency domain channel conversion step, the conversion of the time domain channel into the frequency domain channel is completed by Fast Fourier Transform (FFT).
In the step of the channel classification algorithm, the extracted frequency domain channel characteristics are processed through KNN, SVM, random forest, neural network or CNN algorithm.
The KNN algorithm determines the category of the classified sample according to the category of the nearest sample or samples, and the selected neighbors are all the objects which are classified correctly;
the SVM algorithm is a binary classification model, a basic model is defined as a linear classifier with the maximum interval on a feature space, and a learning strategy of the SVM algorithm is interval maximization and can be finally converted into the solution of a convex quadratic programming problem.
The random forest algorithm randomly selects a subset containing k attributes from the attribute set of the node, and then selects an optimal attribute from the subset for division.
The neural network algorithm means that a neuron receives input signals transmitted by eta other neurons, the input signals are transmitted through weighted connections, total input values received by the neuron are compared with threshold values of the neuron, and then the total input values are processed through an activation function to generate output of the neuron.
After the CNN algorithm is processed by networks such as a convolutional layer and a pooling layer, an image becomes smaller and smaller, but the convolutional layer becomes deeper and deeper, a conventional feedforward neural network consisting of a plurality of fully-connected layers is added at the top of a stack, and the final layer outputs prediction.
Example 2
As shown in FIGS. 1-7, the present invention provides a channel feature extraction and dimension reduction system, which comprises
A time domain channel characteristic estimation module: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
a time domain channel characteristic intercepting module: reducing the dimensionality of the data to be detected or trained through data interception;
the time-frequency domain channel conversion module: converting the intercepted channel from a time domain channel to a frequency domain channel;
a frequency domain channel characteristic estimation module: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
a frequency domain channel characteristic extraction module: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
a channel classification algorithm module: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
In the time domain channel characteristic estimation module, the extraction of the signal characteristics is completed by performing cross correlation on the pilot frequency reference signal in the received signal and a local reference sequence, and performing normalization and gain combination operations.
The time domain channel feature intercepting module intercepts parts with obvious feature differences by observing signals after feature extraction, reduces the dimensionality of data and further reduces the complexity of the data. And the significant features in the portions with significant feature differences mainly refer to the energies of the multipaths.
In the time-frequency domain channel conversion module, the conversion of the time domain channel into the frequency domain channel is completed through Fast Fourier Transform (FFT).
And in the channel classification algorithm module, the extracted frequency domain channel characteristics are processed by KNN, SVM, random forest, neural network or CNN algorithm.
The KNN algorithm determines the category of the classified sample according to the category of the nearest sample or samples, and the selected neighbors are all the objects which are classified correctly;
the SVM algorithm is a binary classification model, a basic model is defined as a linear classifier with the maximum interval on a feature space, and a learning strategy of the SVM algorithm is interval maximization and can be finally converted into the solution of a convex quadratic programming problem.
The random forest algorithm randomly selects a subset containing k attributes from the attribute set of the node, and then selects an optimal attribute from the subset for division.
The neural network algorithm means that a neuron receives input signals transmitted by eta other neurons, the input signals are transmitted through weighted connections, total input values received by the neuron are compared with threshold values of the neuron, and then the total input values are processed through an activation function to generate output of the neuron.
After the CNN algorithm is processed by networks such as a convolutional layer and a pooling layer, an image becomes smaller and smaller, but the convolutional layer becomes deeper and deeper, a conventional feedforward neural network consisting of a plurality of fully-connected layers is added at the top of a stack, and the final layer outputs prediction.
Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A wireless channel feature extraction and dimension reduction method is characterized by comprising the following steps:
time domain channel characteristic estimation step: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
and (3) intercepting the time domain channel characteristics: the dimensionality of the data to be detected or trained is reduced by intercepting the time domain channel characteristic estimation data;
time-frequency domain channel conversion step: converting the intercepted channel from a time domain channel to a frequency domain channel;
estimating the frequency domain channel characteristics: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
and a frequency domain channel characteristic extraction step: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
the channel classification algorithm comprises the following steps: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
2. The method as claimed in claim 1, wherein the time domain channel characteristic estimating step performs cross-correlation between a pilot reference signal in the received signal and a local reference sequence, and performs normalization and gain combination to complete the extraction of the signal characteristics.
3. The method for extracting and reducing the dimension of the wireless channel feature of claim 1, wherein the step of intercepting the time-domain channel feature is to intercept a part meeting a predetermined requirement by observing a signal after feature extraction, so as to reduce the dimension of data, thereby reducing the complexity of data.
4. The method for extracting characteristics of wireless channels and reducing the dimensions of the wireless channels according to claim 1, wherein in the step of converting time-frequency domain channels, the conversion of time-frequency domain channels into frequency domain channels is performed by fast fourier transform.
5. The method as claimed in claim 1, wherein the step of channel classification algorithm comprises processing the extracted frequency domain channel features by KNN, SVM, random forest, neural network or CNN algorithm.
6. A wireless channel feature extraction and dimension reduction system, comprising:
a time domain channel characteristic estimation module: estimating channel impact response characteristics under the environment of the wireless receiver according to the time domain baseband signals received by the communication receiver;
a time domain channel characteristic intercepting module: reducing the dimensionality of the data to be detected or trained through data interception;
the time-frequency domain channel conversion module: converting the intercepted channel from a time domain channel to a frequency domain channel;
a frequency domain channel characteristic estimation module: estimating the channel impact response characteristics of the wireless receiver under the environment according to the frequency domain channel obtained by conversion;
a frequency domain channel characteristic extraction module: the dimensionality of the data to be detected or trained is reduced by extracting the frequency domain channel characteristics;
a channel classification algorithm module: and sending the extracted frequency domain channel characteristics to a channel classification algorithm module to obtain a classification result of the channel, and completing the identification and classification of the channel environment where the communication transceiver is located.
7. The wireless channel feature extraction and dimension reduction system according to claim 6, wherein the time domain channel feature estimation module performs cross-correlation between a pilot reference signal in the received signal and a local reference sequence, and performs normalization and gain combination to complete signal feature extraction.
8. The wireless channel feature extraction and dimension reduction system according to claim 6, wherein the time domain channel feature extraction module is configured to observe the signal after feature extraction, extract a portion with significant feature difference, reduce the dimension of data, and thereby reduce the complexity of data.
9. The wireless channel feature extraction and dimension reduction system of claim 6, wherein the time-frequency domain channel conversion module converts the time-domain channel into the frequency-domain channel by Fast Fourier Transform (FFT).
10. The wireless channel feature extraction and dimension reduction system according to claim 6, wherein the channel classification algorithm module performs processing on the extracted frequency domain channel features through KNN, SVM, random forest, neural network or CNN algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110044042.5A CN112866150A (en) | 2021-01-13 | 2021-01-13 | Wireless channel feature extraction and dimension reduction method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110044042.5A CN112866150A (en) | 2021-01-13 | 2021-01-13 | Wireless channel feature extraction and dimension reduction method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112866150A true CN112866150A (en) | 2021-05-28 |
Family
ID=76003459
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110044042.5A Pending CN112866150A (en) | 2021-01-13 | 2021-01-13 | Wireless channel feature extraction and dimension reduction method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112866150A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113765643A (en) * | 2021-10-05 | 2021-12-07 | 北京遥感设备研究所 | Channel estimation method and system |
Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030058787A1 (en) * | 2001-09-27 | 2003-03-27 | Resonext Communications, Inc. | Method and apparatus for channel estimation |
KR20060008574A (en) * | 2004-07-21 | 2006-01-27 | 삼성전자주식회사 | Apparatus and method for channel estimation in an orthogonal frequency division multiplexing communication system |
US20060114981A1 (en) * | 2002-08-13 | 2006-06-01 | Koninklijke Philips Electronics N.V. | Joint channel and noise variance estimation in a wideband ofdm system |
KR20060095256A (en) * | 2005-02-28 | 2006-08-31 | 주식회사 팬택앤큐리텔 | Channel estimation apparatus using conversion of frequency domain and time domain |
CN1921466A (en) * | 2006-09-08 | 2007-02-28 | 清华大学 | Channel estimation method for solving OFDM interception position hopping using rotating technology |
CN101299733A (en) * | 2008-03-05 | 2008-11-05 | 中科院嘉兴中心微系统所分中心 | Minimum mean-square error channel estimation apparatus for wireless sensing network |
CN101378371A (en) * | 2007-08-27 | 2009-03-04 | 株式会社Ntt都科摩 | Method for channel estimation of wideband wireless mobile commutation system and channel estimator |
US20090141819A1 (en) * | 2007-11-29 | 2009-06-04 | Nokia Corporation | Method and apparatus of recursive time-frequency channel estimation |
CN101702696A (en) * | 2009-11-25 | 2010-05-05 | 北京天碁科技有限公司 | Implement method and device of channel estimation |
CN101753498A (en) * | 2008-12-05 | 2010-06-23 | 中兴通讯股份有限公司 | Method for filtering orthogonal frequency division multiplexing channel estimation results and the device thereof |
US20110142118A1 (en) * | 2008-08-28 | 2011-06-16 | Jae-Hyun Seo | Apparatus and method for equalizing channel based on channel estimation |
CN102130871A (en) * | 2010-01-15 | 2011-07-20 | 无锡百阳科技有限公司 | Channel estimation method and device |
CN102238110A (en) * | 2010-04-23 | 2011-11-09 | 中兴通讯股份有限公司 | Multi-user channel estimation method and device |
CN103227760A (en) * | 2013-04-28 | 2013-07-31 | 中国铁路通信信号股份有限公司 | Channel estimation method under high-speed mobile environment |
CN109067688A (en) * | 2018-07-09 | 2018-12-21 | 东南大学 | A kind of OFDM method of reseptance of data model double drive |
CN109450830A (en) * | 2018-12-26 | 2019-03-08 | 重庆大学 | Channel estimation methods based on deep learning under a kind of high-speed mobile environment |
CN109617847A (en) * | 2018-11-26 | 2019-04-12 | 东南大学 | A kind of non-cycle prefix OFDM method of reseptance based on model-driven deep learning |
CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
CN111510402A (en) * | 2020-03-12 | 2020-08-07 | 西安电子科技大学 | OFDM channel estimation method based on deep learning |
CN111817990A (en) * | 2020-06-22 | 2020-10-23 | 重庆邮电大学 | Channel estimation improvement algorithm based on minimum mean square error in OFDM system |
CN111835444A (en) * | 2020-06-17 | 2020-10-27 | 武汉大学 | Wireless channel scene identification method and system |
-
2021
- 2021-01-13 CN CN202110044042.5A patent/CN112866150A/en active Pending
Patent Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030058787A1 (en) * | 2001-09-27 | 2003-03-27 | Resonext Communications, Inc. | Method and apparatus for channel estimation |
US20060114981A1 (en) * | 2002-08-13 | 2006-06-01 | Koninklijke Philips Electronics N.V. | Joint channel and noise variance estimation in a wideband ofdm system |
KR20060008574A (en) * | 2004-07-21 | 2006-01-27 | 삼성전자주식회사 | Apparatus and method for channel estimation in an orthogonal frequency division multiplexing communication system |
KR20060095256A (en) * | 2005-02-28 | 2006-08-31 | 주식회사 팬택앤큐리텔 | Channel estimation apparatus using conversion of frequency domain and time domain |
CN1921466A (en) * | 2006-09-08 | 2007-02-28 | 清华大学 | Channel estimation method for solving OFDM interception position hopping using rotating technology |
CN101378371A (en) * | 2007-08-27 | 2009-03-04 | 株式会社Ntt都科摩 | Method for channel estimation of wideband wireless mobile commutation system and channel estimator |
US20090141819A1 (en) * | 2007-11-29 | 2009-06-04 | Nokia Corporation | Method and apparatus of recursive time-frequency channel estimation |
CN101299733A (en) * | 2008-03-05 | 2008-11-05 | 中科院嘉兴中心微系统所分中心 | Minimum mean-square error channel estimation apparatus for wireless sensing network |
US20110142118A1 (en) * | 2008-08-28 | 2011-06-16 | Jae-Hyun Seo | Apparatus and method for equalizing channel based on channel estimation |
CN101753498A (en) * | 2008-12-05 | 2010-06-23 | 中兴通讯股份有限公司 | Method for filtering orthogonal frequency division multiplexing channel estimation results and the device thereof |
CN101702696A (en) * | 2009-11-25 | 2010-05-05 | 北京天碁科技有限公司 | Implement method and device of channel estimation |
CN102130871A (en) * | 2010-01-15 | 2011-07-20 | 无锡百阳科技有限公司 | Channel estimation method and device |
CN102238110A (en) * | 2010-04-23 | 2011-11-09 | 中兴通讯股份有限公司 | Multi-user channel estimation method and device |
CN103227760A (en) * | 2013-04-28 | 2013-07-31 | 中国铁路通信信号股份有限公司 | Channel estimation method under high-speed mobile environment |
CN109067688A (en) * | 2018-07-09 | 2018-12-21 | 东南大学 | A kind of OFDM method of reseptance of data model double drive |
CN109617847A (en) * | 2018-11-26 | 2019-04-12 | 东南大学 | A kind of non-cycle prefix OFDM method of reseptance based on model-driven deep learning |
CN109450830A (en) * | 2018-12-26 | 2019-03-08 | 重庆大学 | Channel estimation methods based on deep learning under a kind of high-speed mobile environment |
CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
CN111510402A (en) * | 2020-03-12 | 2020-08-07 | 西安电子科技大学 | OFDM channel estimation method based on deep learning |
CN111835444A (en) * | 2020-06-17 | 2020-10-27 | 武汉大学 | Wireless channel scene identification method and system |
CN111817990A (en) * | 2020-06-22 | 2020-10-23 | 重庆邮电大学 | Channel estimation improvement algorithm based on minimum mean square error in OFDM system |
Non-Patent Citations (4)
Title |
---|
NTT DOCOMO: "Frequency Domain Channel-Dependent Scheduling Considering Interference to Neighbouring Cell for E-UTRA Uplink", 3GPP TSG RAN WG1 MEETING #47BIS R1-070099 * |
PANASONIC: "Pilot channel multiplexing method for multi-antenna transmission in EUTRA OFDMA based downlink", TSG-RAN WG1 #42 R1-050828 * |
姜俊迪;林如俭;: "一种OFDM时域导频插入的最小二乘估计方法", 电子测量技术, no. 03 * |
张兴;周兰花;王帅;廖勇;: "深度学习在无线通信系统信道估计中的应用", 信息通信, no. 06 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113765643A (en) * | 2021-10-05 | 2021-12-07 | 北京遥感设备研究所 | Channel estimation method and system |
CN113765643B (en) * | 2021-10-05 | 2023-11-14 | 北京遥感设备研究所 | Channel estimation method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wong et al. | Clustering learned CNN features from raw I/Q data for emitter identification | |
CN110826630A (en) | Radar interference signal feature level fusion identification method based on deep convolutional neural network | |
Ureten et al. | Wireless security through RF fingerprinting | |
CN110166387B (en) | Method and system for identifying signal modulation mode based on convolutional neural network | |
CN112818891B (en) | Intelligent identification method for communication interference signal type | |
CN109766791B (en) | Communication signal modulation identification method based on self-encoder | |
CN114337880B (en) | Spectrum sensing method considering random arrival and departure of main user signals | |
CN112910811B (en) | Blind modulation identification method and device under unknown noise level condition based on joint learning | |
Jeong et al. | Spectrogram-based automatic modulation recognition using convolutional neural network | |
Che et al. | Spatial-temporal hybrid feature extraction network for few-shot automatic modulation classification | |
Mirmohammadsadeghi et al. | Modulation classification using convolutional neural networks and spatial transformer networks | |
CN111680737A (en) | Radar radiation source individual identification method under differential signal-to-noise ratio condition | |
CN115661869A (en) | Radio frequency signal fingerprint identification method and system based on multi-dimensional feature intelligent fusion | |
CN112866150A (en) | Wireless channel feature extraction and dimension reduction method and system | |
CN113449682B (en) | Method for identifying radio frequency fingerprints in civil aviation field based on dynamic fusion model | |
Wang et al. | A cooperative spectrum sensing method based on a feature and clustering algorithm | |
CN110929242A (en) | Method and system for carrying out attitude-independent continuous user authentication based on wireless signals | |
CN105553574A (en) | Support-vector-machine-based MAC protocol identification method in cognitive radio | |
CN110472584A (en) | A kind of communication equipment personal identification method, electronic equipment and computer program product | |
Ali et al. | Modulation format identification using supervised learning and high-dimensional features | |
Wang et al. | A new method of automatic modulation recognition based on dimension reduction | |
CN113205140A (en) | Semi-supervised specific radiation source individual identification method based on generative countermeasure network | |
Bkassiny | A deep learning-based signal classification approach for spectrum sensing using long short-term memory (LSTM) networks | |
CN113037668A (en) | Millimeter wave point-to-point communication channel equalization method | |
CN104408474A (en) | Radiation source recognition method and device |
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 | ||
CB03 | Change of inventor or designer information | ||
CB03 | Change of inventor or designer information |
Inventor after: Xu Jun Inventor after: Li Yuntian Inventor after: Xiong Jian Inventor after: Li Xia Inventor before: Xu Jun Inventor before: Li Yuntian |