CN110896383A - Channel estimation method of orthogonal frequency division multiplexing technology - Google Patents
Channel estimation method of orthogonal frequency division multiplexing technology Download PDFInfo
- Publication number
- CN110896383A CN110896383A CN201911300897.9A CN201911300897A CN110896383A CN 110896383 A CN110896383 A CN 110896383A CN 201911300897 A CN201911300897 A CN 201911300897A CN 110896383 A CN110896383 A CN 110896383A
- Authority
- CN
- China
- Prior art keywords
- signal
- image
- resolution
- channel
- sender
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
- H04L25/0224—Channel estimation using sounding signals
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a channel estimation method of orthogonal frequency division multiplexing technology, which comprises the steps that a sender converts a signal to be sent into a plurality of subcarrier signals, mutually orthogonalizes the plurality of subcarrier signals, adds redundant symbol information and transmits the signal; and the receiving part receives the signal to demodulate and form a low-resolution channel image, the resolution of the low-resolution channel image is improved by adopting an image super-resolution algorithm, and the signal is restored by adopting an image restoration technology to obtain a final channel estimation result. The invention effectively applies the depth image technology to the orthogonal frequency division multiplexing system, effectively improves the estimation capability of the channel through the depth image technology, and recovers the transmission signal; therefore, the method can estimate the channel of the orthogonal frequency division multiplexing technology, and has high reliability, good accuracy and high efficiency.
Description
Technical Field
The invention belongs to the field of information security, and particularly relates to a channel estimation method of an orthogonal frequency division multiplexing technology.
Background
Orthogonal frequency division multiplexing techniques are used to solve the problem of frequency selective fading in wireless channels. In a communication channel, the received signal is typically distorted by the channel characteristics. In order to recover the transmitted symbols, the channel effects must be estimated and compensated for at the receiving end. In practical applications, the channel environment is constantly changing. Therefore, how to reasonably estimate the channel is a problem that needs to be solved for commercialization of the ofdm technology.
The conventional channel estimation algorithm mainly includes a least square method and a least mean square error method. The minimum mean square error method yields better performance by utilizing channel statistics and noise variance than the least squares estimation that does not require channel statistics.
However, the currently used least square method and the least mean square error method both have the problems of relatively low estimation accuracy, low algorithm efficiency and the like, thereby affecting the actual estimation effect.
Disclosure of Invention
The invention aims to provide a channel estimation method of an orthogonal frequency division multiplexing technology, which has high reliability, good accuracy and higher efficiency.
The channel estimation method of the orthogonal frequency division multiplexing technology provided by the invention comprises the following steps:
s1, a sender converts a signal to be sent into a plurality of subcarrier signals;
s2, the sender mutually orthogonalizes the plurality of subcarrier signals obtained in the step S1 and adds redundant symbol information;
s3, the sender transmits the signal obtained in the step S2;
s4, the receiving party receives the signal sent by the sending party and demodulates the signal;
s5, forming a low-resolution channel image by the demodulated signal by the receiver;
s6, the receiver adopts an image super-resolution algorithm to improve the resolution of the low-resolution channel image obtained in the step S5;
and S7, the receiving party recovers the signal obtained in the step S6 by adopting an image recovery technology, so that a final channel estimation result is obtained.
The sender of step S1 converts the signal to be sent into a plurality of subcarrier signals, specifically, the sender forms the signal to be sent, and forms a plurality of subcarrier signals after serial-to-parallel conversion; the signal to be transmitted comprises a data part and a pilot part.
The sender in step S2 mutually orthogonalizes the multiple subcarrier signals obtained in step S1 and adds redundant symbol information, specifically, the sender obtains the multiple mutually orthogonal subcarrier signals from the multiple to-be-sent subcarrier signals through inverse fourier transform, and inserts a cyclic prefix CP into the mutually orthogonal subcarrier signals, thereby adding redundant symbol signals.
The sender in step S3 transmits the signal obtained in step S2, specifically, the sender converts the signal obtained in step S2 into an analog signal by parallel-to-serial conversion, and then converts the analog signal into an analog signal by digital-to-analog conversion, and transmits the analog signal.
The receiving side receives the signal sent by the sending side and demodulates the signal, specifically, the receiving side converts the received signal into a digital signal through analog-to-digital conversion, then performs serial-to-parallel conversion, removes a cyclic prefix, performs fourier transform for demodulation, and finally performs serial-to-parallel conversion on the demodulated signal.
The receiving party in the step S5 forms the demodulated signal into a low resolution channel image, specifically, models a time-frequency grid of channel response into a two-dimensional image; the two-dimensional image is a two-dimensional image known only at pilot locations.
The receiving side in step S6 adopts an image super-resolution algorithm to improve the resolution of the low-resolution channel image obtained in step S5, specifically, based on the known SRCNN algorithm, first uses an interpolation scheme to find an approximate value of the high-resolution image, and then uses a three-layer convolutional network to improve the resolution: the first convolutional layer uses 64 filters of size 9 × 9 and the second layer uses 32 filters of size 1 × 1, both activated using a linear rectification function (ReLU); the last layer reconstructs the image using only one filter of size 5 x 5.
The receiving side in step S7 restores the signal obtained in step S6 by using an image restoration technique, specifically, a network based on residual learning, which is composed of 20 convolutional layers: the first layer of convolutional layer adopts 64 filters with the size of 3 multiplied by 1, and is followed by a ReLU; the next 18 convolutional layers used 64 filters of size 3 × 3 × 64, followed by batch normalization and ReLU; the last convolutional layer uses a 3 × 3 × 64 filter to reconstruct the output.
The channel estimation method of the orthogonal frequency division multiplexing technology effectively applies the depth image technology to the orthogonal frequency division multiplexing system, effectively improves the estimation capability of the channel through the depth image technology, and recovers the transmission signal; therefore, the method can estimate the channel of the orthogonal frequency division multiplexing technology, and has high reliability, good accuracy and high efficiency.
Drawings
FIG. 1 is a schematic process flow diagram of the process of the present invention.
Detailed Description
FIG. 1 is a schematic flow chart of the method of the present invention: the channel estimation method of the orthogonal frequency division multiplexing technology provided by the invention comprises the following steps:
s1, a sender converts a signal to be sent into a plurality of subcarrier signals; specifically, a signal to be sent is formed for a sender, and after serial-to-parallel conversion, a plurality of subcarrier signals are formed; the signal to be transmitted comprises a data part and a pilot frequency part;
s2, the sender mutually orthogonalizes the plurality of subcarrier signals obtained in the step S1 and adds redundant symbol information; specifically, a sender obtains a plurality of orthogonal subcarrier signals by performing inverse Fourier transform on a plurality of subcarrier signals to be sent, and inserts a Cyclic Prefix (CP) into the orthogonal subcarrier signals, thereby increasing redundant symbol signals;
s3, the sender transmits the signal obtained in the step S2; specifically, the sender converts the signal obtained in step S2 from parallel to serial, converts the converted signal into an analog signal by digital-to-analog conversion, and transmits the obtained analog signal;
s4, the receiving party receives the signal sent by the sending party and demodulates the signal; specifically, a receiving party converts a received signal into a digital signal through analog-to-digital conversion, then removes a cyclic prefix after serial-to-parallel conversion, performs Fourier transform for demodulation, and finally performs serial-to-parallel conversion on the demodulated signal;
s5, forming a low-resolution channel image by the demodulated signal by the receiver; specifically, modeling a time-frequency grid of channel response into a two-dimensional image; the two-dimensional image is a two-dimensional image only known at the pilot frequency position;
s6, the receiver adopts an image super-resolution algorithm to improve the resolution of the low-resolution channel image obtained in the step S5; specifically, based on the known SRCNN algorithm, an interpolation scheme is used to find an approximate value of a high-resolution image, and then three layers of convolutional networks are used to improve the resolution: the first convolutional layer uses 64 filters of size 9 × 9 and the second layer uses 32 filters of size 1 × 1, both activated using a linear rectification function (ReLU); the last layer reconstructs the image using only one filter of size 5 x 5;
s7, the receiver recovers the signal obtained in the step S6 by adopting an image recovery technology, so that a final channel estimation result is obtained; specifically, the network based on residual learning is composed of 20 convolutional layers: the first layer of convolutional layer adopts 64 filters with the size of 3 multiplied by 1, and is followed by a ReLU; the next 18 convolutional layers used 64 filters of size 3 × 3 × 64, followed by batch normalization and ReLU; the last convolutional layer uses a 3 × 3 × 64 filter to reconstruct the output.
Taking into account the estimated signal matrix
Wherein f isSAnd fRIs a function of the image super-resolution algorithm and the image restoration algorithm respectively, and theta is ═ thetaS,θRIn which θSAnd thetaRParameter value sets of the image super-resolution network and the image restoration network respectively,
is an input to the channel network and represents a pilot position vector;
the Mean Square Error (MSE) between the estimated channel response and the actual channel response is expressed as:
wherein T is a set of training data, | | T | | | represents the size of the training set, H represents the ideal channel;
meanwhile, in order to simplify the calculation training process, the loss of the image super-resolution algorithm is trained to be minimum in two steps
Wherein
Representing the output of the image super-resolution algorithm; then, the training result of the image super-resolution algorithm is unchanged, and the loss of the image recovery algorithm is trained to be minimum again
Wherein
Representing the output of the image restoration algorithm.
Claims (8)
1. A channel estimation method of orthogonal frequency division multiplexing technology comprises the following steps:
s1, a sender converts a signal to be sent into a plurality of subcarrier signals;
s2, the sender mutually orthogonalizes the plurality of subcarrier signals obtained in the step S1 and adds redundant symbol information;
s3, the sender transmits the signal obtained in the step S2;
s4, the receiving party receives the signal sent by the sending party and demodulates the signal;
s5, forming a low-resolution channel image by the demodulated signal by the receiver;
s6, the receiver adopts an image super-resolution algorithm to improve the resolution of the low-resolution channel image obtained in the step S5;
and S7, the receiving party recovers the signal obtained in the step S6 by adopting an image recovery technology, so that a final channel estimation result is obtained.
2. The channel estimation method according to claim 1, wherein the sender converts the signal to be sent into a plurality of subcarrier signals, specifically, the sender forms the signal to be sent and forms a plurality of subcarrier signals after serial-to-parallel conversion, in step S1; the signal to be transmitted comprises a data part and a pilot part.
3. The channel estimation method according to claim 1, wherein the sender in step S2 mutually orthogonalizes the plurality of subcarrier signals obtained in step S1 and adds redundant symbol information, specifically, the sender obtains a plurality of mutually orthogonal subcarrier signals from the plurality of subcarrier signals to be sent through inverse fourier transform, and inserts a cyclic prefix CP into the mutually orthogonal subcarrier signals, thereby adding redundant symbol signals.
4. The channel estimation method according to claim 1, wherein the sender in step S3 transmits the signal obtained in step S2, specifically, the sender converts the signal obtained in step S2 from parallel to serial, converts the converted signal into an analog signal by digital-to-analog conversion, and transmits the analog signal.
5. The channel estimation method according to claim 1, wherein the receiving side receives and demodulates the signal sent by the sending side in step S4, specifically, the receiving side converts the received signal into a digital signal through analog-to-digital conversion, then performs serial-to-parallel conversion, then removes a cyclic prefix, performs fourier transform for demodulation, and finally performs serial-to-parallel conversion on the demodulated signal.
6. The channel estimation method according to one of claims 1 to 5, wherein the receiving side forms the demodulated signal into a low resolution channel image, specifically, models a time-frequency grid of a channel response as a two-dimensional image in step S5; the two-dimensional image is a two-dimensional image known only at pilot locations.
7. The channel estimation method according to claim 6, wherein the receiving side in step S6 adopts an image super-resolution algorithm to improve the resolution of the low-resolution channel image obtained in step S5, and specifically based on the known SRCNN algorithm, first uses an interpolation scheme to find an approximate value of the high-resolution image, and then uses a three-layer convolutional network to improve the resolution: the first convolutional layer uses 64 filters of size 9 × 9 and the second layer uses 32 filters of size 1 × 1, both activated using a linear rectification function (ReLU); the last layer reconstructs the image using only one filter of size 5 x 5.
8. The method of claim 7, wherein the receiver recovers the signal obtained in step S6 by using an image recovery technique in step S7, and specifically, the residual learning-based network comprises 20 convolutional layers: the first layer of convolutional layer adopts 64 filters with the size of 3 multiplied by 1, and is followed by a ReLU; the next 18 convolutional layers used 64 filters of size 3 × 3 × 64, followed by batch normalization and ReLU; the last convolutional layer uses a 3 × 3 × 64 filter to reconstruct the output.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911300897.9A CN110896383A (en) | 2019-12-17 | 2019-12-17 | Channel estimation method of orthogonal frequency division multiplexing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911300897.9A CN110896383A (en) | 2019-12-17 | 2019-12-17 | Channel estimation method of orthogonal frequency division multiplexing technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110896383A true CN110896383A (en) | 2020-03-20 |
Family
ID=69788126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911300897.9A Pending CN110896383A (en) | 2019-12-17 | 2019-12-17 | Channel estimation method of orthogonal frequency division multiplexing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110896383A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2658194A1 (en) * | 2012-04-23 | 2013-10-30 | Nxp B.V. | Reduced latency channel-estimation |
| US20180131486A1 (en) * | 2016-11-04 | 2018-05-10 | Futurewei Technologies, Inc. | System and Method for Transmitting a Sub-Space Selection |
| CN108933745A (en) * | 2018-07-16 | 2018-12-04 | 北京理工大学 | A kind of broad-band channel estimation method estimated based on super-resolution angle and time delay |
| CN109981498A (en) * | 2019-03-12 | 2019-07-05 | 上海大学 | Wi-Fi modular system channel estimation methods based on super-resolution image restoration technology |
-
2019
- 2019-12-17 CN CN201911300897.9A patent/CN110896383A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2658194A1 (en) * | 2012-04-23 | 2013-10-30 | Nxp B.V. | Reduced latency channel-estimation |
| US20180131486A1 (en) * | 2016-11-04 | 2018-05-10 | Futurewei Technologies, Inc. | System and Method for Transmitting a Sub-Space Selection |
| CN108933745A (en) * | 2018-07-16 | 2018-12-04 | 北京理工大学 | A kind of broad-band channel estimation method estimated based on super-resolution angle and time delay |
| CN109981498A (en) * | 2019-03-12 | 2019-07-05 | 上海大学 | Wi-Fi modular system channel estimation methods based on super-resolution image restoration technology |
Non-Patent Citations (1)
| Title |
|---|
| MEHRAN SOLTANI 等: "Deep Learning-Based Channel Estimation", 《 IEEE COMMUNICATIONS LETTERS》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101132388B (en) | Receiving method and device for receiving coded signal assisted by signal channel condition information | |
| CN101958873A (en) | Information transmission method for reducing peak to average power ratio of orthogonal frequency division multiplexing signal | |
| CN105915473B (en) | A kind of estimation of ofdm system parametric channel and equalization methods based on compressed sensing technology | |
| CN101848184B (en) | Selective mapping scrambling method for reducing peak-to-average power ratio of orthogonal frequency division multiplexing system | |
| CN108156101B (en) | MIMO-SCFDE system joint iteration channel estimation and iteration equalization method | |
| CN106941470A (en) | A kind of method of signal peak-to-average ratio in reduction FBMC systems | |
| CN1835484A (en) | Low complexity channel estimation method based on orthogonal sequence design | |
| CN111327551A (en) | Data and pilot frequency domain multiplexing super-Nyquist transmission method and transmission device | |
| CN116132239B (en) | OFDM channel estimation method adopting pre-activation residual error unit and super-resolution network | |
| CN111884761B (en) | Data transmission method for transmitting end of single carrier frequency domain equalization system | |
| CN113971430A (en) | Signal detection and model training method, device, equipment and storage medium | |
| CN111371711A (en) | Fast acquisition and synchronization method of wireless burst communication system | |
| CN114785644A (en) | Mobile underwater acoustic OTFS communication sparse channel estimation method | |
| CN110896383A (en) | Channel estimation method of orthogonal frequency division multiplexing technology | |
| CN102611650A (en) | Frequency-domain channel estimation method and device of generalized multi-carrier system | |
| CN106453189A (en) | Non-orthogonal supported multiple access transmission method | |
| CN113098816B (en) | System and method for dual-mode three-dimensional OFDM index modulation algorithm | |
| CN101383799B (en) | OFDM system synchronization, channel estimation and peak average ratio reduction combined highly effective algorithm | |
| CN102281224B (en) | OFDM receiver and its implementation | |
| CN104243384A (en) | Method for restraining inter-subcarrier interference through precoding based on OFDM system | |
| CN101115047A (en) | OFDM receiving and dispatching system for high speed mobile environment | |
| CN109474412A (en) | A kind of general filtering multi-carrier method based on Choose for user | |
| Li et al. | Signal assisted clipping distortion recovery for OFDM systems based on compressed sensing | |
| CN108768913A (en) | A kind of non-orthogonal multi-carrier modulator approach improving image transmitting performance | |
| CN112332948B (en) | Multi-service transmission sending and receiving method, system, medium 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200320 |