CN109194595B - Neural network-based channel environment self-adaptive OFDM receiving method - Google Patents

Neural network-based channel environment self-adaptive OFDM receiving method Download PDF

Info

Publication number
CN109194595B
CN109194595B CN201811123542.2A CN201811123542A CN109194595B CN 109194595 B CN109194595 B CN 109194595B CN 201811123542 A CN201811123542 A CN 201811123542A CN 109194595 B CN109194595 B CN 109194595B
Authority
CN
China
Prior art keywords
neural network
deep neural
channel
channel estimation
main
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.)
Active
Application number
CN201811123542.2A
Other languages
Chinese (zh)
Other versions
CN109194595A (en
Inventor
姜培文
徐靖
李沙志远
陈翔宇
陈慕涵
金石
温朝凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN201811123542.2A priority Critical patent/CN109194595B/en
Publication of CN109194595A publication Critical patent/CN109194595A/en
Application granted granted Critical
Publication of CN109194595B publication Critical patent/CN109194595B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • H04L25/0204—Channel estimation of multiple channels
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • 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
    • 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
    • H04L25/0256—Channel estimation using minimum mean square error criteria
    • 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
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L25/03159—Arrangements for removing intersymbol interference operating in the frequency domain

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)

Abstract

The invention discloses a channel environment self-adaptive OFDM receiving method based on a neural network, which comprises the following steps: selecting a plurality of different channel environments to obtain each trained deep neural network; inputting a receiving frequency domain pilot frequency and a local frequency domain pilot frequency to estimate to obtain a noisy channel estimation result, inputting the noisy channel estimation result into a trained main wiener filtering deep neural network and outputting a main channel estimation result, and obtaining each auxiliary channel estimation result, multiplying the auxiliary channel estimation result by weight and adding the auxiliary channel estimation results to obtain a final channel estimation result; and inputting the final channel estimation result and the received frequency domain data to obtain a zero-forcing equalization result, inputting the zero-forcing equalization result into the trained main and auxiliary equalization deep neural networks to obtain output results of the networks, multiplying the output results by a weight coefficient, adding the multiplied output results, and outputting the sum after hard decision to obtain an estimated bit stream. And acquiring the known frequency domain pilot frequency or channel parameters in actual transmission in real time, and dynamically adjusting the weight coefficient. The invention has the advantages of few parameters, high working efficiency, flexible and rapid on-line switching and the like.

Description

Neural network-based channel environment self-adaptive OFDM receiving method
Technical Field
The invention relates to a channel environment self-adaptive OFDM receiving method based on a neural network, belonging to the technical field of wireless communication.
Background
Smart communication is considered to be one of the main research directions of next-generation mobile communication. In recent years, as a main branch of machine learning, deep learning technology has been widely applied to wireless communication physical layer research, and has been greatly improved in performance. Research directions include the complete replacement of the entire communication system by an end-to-end deep neural network, or the replacement of only a portion of the modules of the communication system, such as encoders, decoders, detectors, etc., by a deep neural network.
The OFDM receiving method is one of the key technologies of 4G and 5G, and there are a lot of research results in both the channel estimation and channel equalization directions of the receiver. However, the currently adopted method is to treat the whole neural network as a complete black box system, and simply replace different modules of the traditional communication to achieve the performance improvement. The network relies on a large amount of simulation or air interface to acquire data and complete off-line training, and the quality of training data seriously influences the reliability and robustness of the network after actual deployment. And after the network actually operates, the parameters are not changed, and better performance can not be obtained under different environments.
Disclosure of Invention
The technical problem to be solved by the present invention is to overcome the defects of the prior art, and provide a channel environment adaptive OFDM receiving method based on a neural network, which solves the problems that the network in the prior art has fixed parameters during actual operation and cannot adaptively improve the performance according to the operation environment.
The invention specifically adopts the following technical scheme to solve the technical problems:
a channel environment self-adaptive OFDM receiving method based on a neural network comprises the following steps:
step 1, constructing a channel filter network consisting of a main wiener filter deep neural network and a plurality of auxiliary wiener filter deep neural networks connected in parallel in a channel estimation module, and constructing a channel equalization network consisting of a main equalization deep neural network and a plurality of auxiliary equalization deep neural networks connected in parallel in a signal detection module;
selecting a plurality of different channel environments, wherein one channel environment is used as a main training environment and the other channel environments are used as auxiliary training environments; respectively training and determining parameters in a main wiener filtering deep neural network and a main equilibrium deep neural network by using a main training environment so as to obtain a trained main wiener filtering deep neural network and a trained main equilibrium deep neural network; respectively training and determining parameters in each auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network by using different auxiliary training environments to obtain each trained auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network;
step 2, inputting the receiving frequency domain pilot frequency and the local frequency domain pilot frequency for the channel estimation module, and estimating by using a least square method to obtain a noisy channel estimation result
Figure BDA0001811801890000021
Estimating the result of the channel with noise
Figure BDA0001811801890000022
Inputting the trained main wiener filtering deep neural network and outputting a main channel estimation result
Figure BDA0001811801890000023
And estimating the main channel
Figure BDA0001811801890000024
And the sub wiener filtering deep neural network obtains each sub channel estimation result, and the sub channel estimation results are multiplied by the weight coefficient alpha respectively and then added to obtain the final channel estimation result
Figure BDA0001811801890000025
Step 3, inputting the final channel estimation result obtained by the channel estimation module to the signal detection module
Figure BDA0001811801890000026
And receiving the frequency domain data Y, and estimating by using a zero-forcing equalization method to obtain a zero-forcing equalization result
Figure BDA0001811801890000027
Zero-forcing equalization result
Figure BDA0001811801890000028
Adding the result of the main equalization deep neural network after input training and the result of multiplying the sub equalization deep neural network after each training by the weight coefficient beta, and outputting after hard decision to obtain an estimated bit stream
Figure BDA0001811801890000029
And 4, acquiring known frequency domain pilot frequency and frequency domain data in the actual transmission process in real time, and dynamically adjusting the weight coefficients alpha and beta.
Further, as a preferred technical solution of the present invention, the final channel estimation result obtained in step 2
Figure BDA00018118018900000210
Comprises the following steps:
Figure BDA00018118018900000211
wherein,
Figure BDA00018118018900000212
estimating the result for each sub-channel; alpha is alpha1、α2…αnThe weight coefficient of each sub-channel estimation result.
Further, as a preferred technical solution of the present invention, the loss function of the channel filtering network in step 1 is:
Figure BDA00018118018900000213
where N is the number of subcarriers, H (k) is the actual kth subcarrier frequency domain channel,
Figure BDA00018118018900000214
is the k-th subcarrier channel estimation result.
Further, as a preferred technical solution of the present invention, the loss function of the channel equalization network in step 1 is:
Figure BDA00018118018900000215
where B is the number of bits to be estimated, B (i) is the ith bit actually transmitted,
Figure BDA0001811801890000031
is the ith bit estimate that is actually sent.
By adopting the technical scheme, the invention can produce the following technical effects:
the method of the invention combines the channel knowledge of the traditional communication, further improves the performance of the network for the specific channel environment, and exerts the learning optimization function of the deep neural network for the complex channel environment. Compared with the traditional OFDM receiver and a large number of deep learning receivers trained in different environments, the method has better effect.
Therefore, the invention combines the off-line training and the on-line learning, so that the neural network can improve the performance aiming at the specific channel environment and simultaneously can perform on-line learning switching to obtain good robustness. Compared with the traditional receiving method and the receiving method only performing off-line training, the method improves the performance and simultaneously ensures that the system cannot have performance deterioration under different environments due to over optimization of the system to a specific environment. The off-line and on-line combined method is more suitable for the actual deployment requirement of the system.
Drawings
Fig. 1 is a schematic diagram of the channel environment adaptive OFDM reception method based on the neural network according to the present invention.
FIG. 2 is a block diagram of a main wiener filtering deep neural network structure in the present invention.
Fig. 3 is a diagram of pilot and data formats in the present invention.
Detailed Description
The following describes embodiments of the present invention with reference to the drawings.
As shown in fig. 1, the present invention designs a channel environment adaptive OFDM receiving method based on a neural network, including:
step 1, constructing a channel filter network consisting of a main wiener filter deep neural network and a plurality of auxiliary wiener filter deep neural networks connected in parallel in a channel estimation module, and constructing a channel equalization network consisting of a main equalization deep neural network and a plurality of auxiliary equalization deep neural networks connected in parallel in a signal detection module;
selecting a plurality of different channel environments, wherein one channel environment is used as a main training environment and the other channel environments are used as auxiliary training environments; respectively training and determining parameters in the main wiener filtering deep neural network and the main equalization deep neural network by using a main training environment to obtain a trained main wiener filtering deep neural network and a trained main equalization deep neural network; respectively training and determining parameters in each auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network by using different auxiliary training environments to obtain each trained auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network;
specifically, under different channel environments such as a main training environment, an auxiliary training environment 1, and an auxiliary training environment 2, respectively, a corresponding wiener filtering deep neural network in a channel filtering network is trained, for example: under the main training environment, the weight coefficient alpha of each secondary wiener filtering deep neural network is 0, so that the network only trains and determines specific parameters for the parameters in the main wiener filtering deep neural network; in the environment x, the weighting factor alpha is fixedx＝1,αiSetting the weight coefficient of one sub wiener filtering deep neural network corresponding to the environment x to be 1 (i ≠ x), setting the weight coefficients of the other sub wiener filtering deep neural networks to be 0, and training the pair of the environment xApplying a parameter of the sub-wiener filtering deep neural network and determining a specific parameter, and taking a loss function as a final channel estimation result
Figure BDA0001811801890000041
The mean square error of (d); and when other environments are switched, the weight coefficient of one sub wiener filtering deep neural network corresponding to the corresponding environment is 1, and the weight coefficients of the other sub wiener filtering deep neural networks are 0, so that the training switching of each sub wiener filtering deep neural network is realized.
Similarly, training the equalization deep neural network in the channel equalization network under different environments specifically includes: under the main training environment, the weight coefficient of the main equilibrium deep neural network is 1, and the weight coefficient beta of each auxiliary equilibrium deep neural network is 0, so that the network only trains the parameters in the main equilibrium deep neural network and determines specific parameters; in the environment x, the weighting factor beta is fixedx＝1,βiSetting the weight coefficient of a corresponding sub-equalization deep neural network under the environment x to be 1 (i ≠ x), setting the weight coefficients of the other sub-equalization deep neural networks to be 0, training the parameters and determining specific parameters of the corresponding sub-equalization deep neural network under the environment x, and setting the loss function as the output estimation bit stream
Figure BDA0001811801890000042
The mean square error function of. Similarly, in various environments, only one corresponding balanced deep neural network is trained, the network weight coefficient beta in the current environment is 1, the weight coefficients beta of the other networks are 0, and the bit estimation performance is optimized.
In this embodiment, the primary operating channel environment is an 802.11b exponential channel model environment, and the secondary training environment 1 is an SUI5 outdoor channel model environment, without any other channel environment that may operate. Thus, only α needs to be measured1When equal to 0, training the main filter network in the channel filter network, alpha1When 1, training environment 1 filter network in channel filter network.
Step 2, inputting and receiving frequency domain pilot frequency and local frequency domain pilot frequency for a channel estimation module, and utilizing minimumObtaining the estimation result of the noisy channel by two-multiplication estimation
Figure BDA0001811801890000043
Estimating the result of the channel with noise
Figure BDA0001811801890000044
The real part and the imaginary part of the signal are input into the trained main wiener filtering deep neural network in series and output the estimation result of the main channel
Figure BDA0001811801890000045
And estimating the main channel
Figure BDA0001811801890000046
Respectively inputting the trained sub wiener filtering deep neural network to obtain the estimation result of each sub channel
Figure BDA0001811801890000047
And the obtained signals are multiplied by a weight coefficient alpha respectively and then added to obtain the final channel estimation result
Figure BDA0001811801890000051
Is composed of
Figure BDA0001811801890000052
Wherein a is1、a2…anFor the weight coefficients of each subchannel estimation result, a value can be directly selected as each weight coefficient.
In the channel estimation, each neural network adopts a full connection layer without an activation function, after a plurality of estimation networks are connected in parallel, the estimation networks are multiplied by corresponding switching parameters, and the network result is expressed as:
Figure BDA0001811801890000053
in the formula, the first step is that,
Figure BDA0001811801890000054
Wimultiplicative for filtering neural networks in channel estimationAnd (4) parameters. At initialization, WMaster and slaveEstimating W of a real valued matrix for linear minimum mean square errorLMMSE，n1All weight parameters α are 0.
The network training only trains the corresponding multiplicative parameter W under the environment xxAt this time, αx＝1，αi0(i ≠ x), the channel estimation performance is optimized. Taking the case of the main environment and environment 1 as an example, will
Figure BDA0001811801890000055
Inputting the wiener filter neural network under the environment 1, respectively obtaining estimation results, adding the estimation results according to weight to obtain a final channel estimation result
Figure BDA0001811801890000056
Respectively training corresponding wiener filter networks in a main environment and an environment 1: under the main environment, the weight coefficient alpha of the fixed secondary wiener filtering deep neural network1Training the parameters of the main wiener filtering deep neural network, and taking the loss function as the final channel estimation result
Figure BDA0001811801890000057
The mean square error of (d); fixed weight coefficient alpha under environment 11 Training environment 1 corresponding sub-wiener filtering deep neural network parameters, and the loss function is the final channel estimation result
Figure BDA0001811801890000058
The mean square error of (d).
Specifically, the loss function is as follows:
Figure BDA0001811801890000059
where N is the number of subcarriers, H (k) is the actual kth subcarrier frequency domain channel,
Figure BDA00018118018900000510
is the k-th subcarrier channel estimation result.
Step 3, inputting the final channel estimation result obtained by the channel estimation module to the signal detection module
Figure BDA00018118018900000511
And receiving the frequency domain data Y, and estimating by using a zero-forcing equalization method to obtain a zero-forcing equalization result
Figure BDA00018118018900000512
Zero-forcing equalization result
Figure BDA00018118018900000513
Inputting the trained main equilibrium deep neural network and each trained auxiliary equilibrium deep neural network to obtain the output result of each network, multiplying the output result by a weight coefficient beta respectively, and adding the two output results, namely beta1、β2…βnFor the weight coefficient of each sub-equilibrium deep neural network, the numerical value can be directly selected as each weight coefficient, and the added result is output after hard decision to obtain the estimated bit stream
Figure BDA00018118018900000514
Similarly, taking the case of the main environment and the environment 1 as an example, when the main equilibrium deep neural network in the main environment is trained, the weight parameter β of the auxiliary equilibrium deep neural network is fixed10, the loss function estimates the bitstream for the output
Figure BDA0001811801890000061
A mean square error function of; when training the sub-equilibrium deep neural network under the environment 1, fixing the weight parameter beta of the sub-equilibrium deep neural network1The loss function estimates the bitstream for the output as 1
Figure BDA0001811801890000062
The mean square error function of.
Specifically, the loss function is as follows:
Figure BDA0001811801890000063
where B is the number of bits to be estimated, B (i) is the ith bit actually transmitted,
Figure BDA0001811801890000064
is the ith bit estimate that is actually sent.
And 4, acquiring known frequency domain pilot frequency or frequency domain data of the actual channel environment in real time, and dynamically adjusting the weight coefficients alpha and beta. After the OFDM signal is deployed in an actual scene, all network parameters are unchanged, the specific numerical values of the weight coefficients alpha and beta realize online dynamic adjustment by acquiring known pilot frequency or training sequence data in currently transmitted OFDM symbols in real time, and the loss function is output estimation bit stream
Figure BDA0001811801890000065
The mean square error function cost 2.
As shown in FIG. 2, the wiener filtering deep neural networks in each environment are all one-layer full-connection networks, no activation function is adopted, and multiplicative parameters W of the layer networksMaster and slaveInitialization with the real values of the weight matrix for linear minimum mean square error channel estimation, WMaster and slaveInitialisation to all zero, initialisation of the additive parameter n to all zero, noisy channel estimation using least squares, i.e.
Figure BDA0001811801890000066
And in an indoor environment will WMaster and slaveTo be trainable, the weighting factor α1When the network implements a calculation of 0
Figure BDA0001811801890000067
Because of the minimum mean square error channel estimation
Figure BDA0001811801890000068
And only real-valued calculation can be carried out in the deep learning network, so when the input of the network is
Figure BDA0001811801890000069
Real and imaginary strings ofIn connection with, multiplicative parameter W of network is initialized to weight matrix WLMMSEReal value matrix of
Figure BDA00018118018900000610
Wherein Re {. is a real part, and Im {. is an imaginary part; when the additive parameter n is initialized to be all zero, the initial value of the network output is
Figure BDA00018118018900000611
The real and imaginary parts of (a) are connected in series. The cost1 is optimized, the optimizer is an Adam optimizer, small batch gradient descending is adopted during training, 50 batches are adopted in each round, the size of each batch is 1000 samples, 2000 rounds of training are performed, the learning rate is dynamically adjusted, the initial learning rate is 0.001, and the descending rate is 1/5 in each 100 rounds.
The balanced deep neural network structure is 128 multiplied by 120 multiplied by 16, the cost2 is optimized, the optimizer is an Adam optimizer, small batch gradient descent is adopted during training, 50 batches are adopted in each round, and the size of each batch is 1000 samples. A total training of 20000 rounds with a dynamically adjusted learning rate, with the initial learning rate set to 0.001, with a reduction of one fifth of the current per 1000 rounds.
As shown in fig. 3, in an OFDM system with 128 subcarriers, the data frame format is a pilot OFDM symbol and a data OFDM symbol, the pilot and the data both occupy 64 subcarriers, and the constellation modulation mode is QPSK.
Therefore, the invention combines the deep learning and the communication knowledge, and combines the off-line training and the on-line training, thereby providing a new method for designing the receiver of the OFDM system, and greatly improving the robustness of different environments and the BER performance during stable work compared with the traditional communication receiver and the deep learning receiver of off-line training and on-line deployment. And the method has the advantages of few parameters, high working efficiency, flexible and rapid online switching and the like.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.

Claims (3)

1. A channel environment self-adaptive OFDM receiving method based on a neural network is characterized by comprising the following steps:
step 1, constructing a channel filter network consisting of a main wiener filter deep neural network and a plurality of auxiliary wiener filter deep neural networks connected in parallel in a channel estimation module, and constructing a channel equalization network consisting of a main equalization deep neural network and a plurality of auxiliary equalization deep neural networks connected in parallel in a signal detection module;
selecting a plurality of different channel environments, wherein one channel environment is used as a main training environment and the other channel environments are used as auxiliary training environments; respectively training and determining parameters in a main wiener filtering deep neural network and a main equilibrium deep neural network by using a main training environment so as to obtain a trained main wiener filtering deep neural network and a trained main equilibrium deep neural network; respectively training and determining parameters in each auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network by using different auxiliary training environments to obtain each trained auxiliary wiener filtering deep neural network and each auxiliary equilibrium deep neural network;
step 2, inputting the receiving frequency domain pilot frequency and the local frequency domain pilot frequency for the channel estimation module, and estimating by using a least square method to obtain a noisy channel estimation result
Figure FDA0002721529180000011
Estimating the result of the channel with noise
Figure FDA0002721529180000012
Inputting the trained main wiener filtering deep neural network and outputting a main channel estimation result
Figure FDA0002721529180000013
And estimating the main channel
Figure FDA0002721529180000014
Respectively input eachThe trained sub wiener filtering deep neural network obtains each sub channel estimation result, and each sub channel estimation result is multiplied by a weight coefficient alpha and then is compared with the main channel estimation result
Figure FDA0002721529180000015
Adding to obtain the final channel estimation result
Figure FDA0002721529180000016
The method specifically comprises the following steps:
Figure FDA0002721529180000017
wherein,
Figure FDA0002721529180000018
estimating the result for each sub-channel; alpha is alpha1、α2…αnWeight coefficients for each subchannel estimation result;
step 3, inputting the final channel estimation result obtained by the channel estimation module to the signal detection module
Figure FDA0002721529180000019
And receiving the frequency domain data Y, and estimating by using a zero-forcing equalization method to obtain a zero-forcing equalization result
Figure FDA00027215291800000110
Zero-forcing equalization result
Figure FDA00027215291800000111
Adding the result of the main equalization deep neural network after input training and the result of multiplying the sub equalization deep neural network after each training by the weight coefficient beta, and outputting after hard decision to obtain an estimated bit stream
Figure FDA00027215291800000112
And 4, acquiring known frequency domain pilot frequency and frequency domain data in the actual transmission process in real time, and dynamically adjusting the weight coefficients alpha and beta.
2. The adaptive OFDM receiving method according to claim 1, wherein the loss function of the channel filtering network in step 1 is:
Figure FDA00027215291800000113
where N is the number of subcarriers, H (k) is the actual kth subcarrier frequency domain channel,
Figure FDA0002721529180000021
is the k-th subcarrier channel estimation result.
3. The neural network-based channel environment adaptive OFDM receiving method of claim 1, wherein: the loss function of the channel equalization network in the step 1 is as follows:
Figure FDA0002721529180000022
where B is the number of bits to be estimated, B (i) is the ith bit actually transmitted,
Figure FDA0002721529180000023
is the ith bit estimate that is actually sent.
CN201811123542.2A 2018-09-26 2018-09-26 Neural network-based channel environment self-adaptive OFDM receiving method Active CN109194595B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811123542.2A CN109194595B (en) 2018-09-26 2018-09-26 Neural network-based channel environment self-adaptive OFDM receiving method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811123542.2A CN109194595B (en) 2018-09-26 2018-09-26 Neural network-based channel environment self-adaptive OFDM receiving method

Publications (2)

Publication Number Publication Date
CN109194595A CN109194595A (en) 2019-01-11
CN109194595B true CN109194595B (en) 2020-12-01

Family

ID=64907219

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811123542.2A Active CN109194595B (en) 2018-09-26 2018-09-26 Neural network-based channel environment self-adaptive OFDM receiving method

Country Status (1)

Country Link
CN (1) CN109194595B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109889462B (en) * 2019-01-21 2021-05-28 东南大学 Neural network assisted iterative receiving method suitable for high-speed visible light communication
CN109831400A (en) * 2019-03-06 2019-05-31 深兰科技（上海）有限公司 A kind of received method and apparatus of OFDM signal
EP3726529A1 (en) * 2019-04-16 2020-10-21 Fraunhofer Gesellschaft zur Förderung der Angewand Method and apparatus for determining a deep filter
CN111431831B (en) * 2020-03-20 2021-06-25 厦门大学 Multi-dimensional OFDM environment-based adaptive modulation method and system
CN111641570B (en) * 2020-04-17 2021-04-27 浙江大学 Joint equipment detection and channel estimation method based on deep learning
US20230216724A1 (en) * 2020-06-25 2023-07-06 Telefonaktiebolaget Lm Ericsson (Publ) Context Aware Data Receiver for Communication Signals Based on Machine Learning
CN111865863B (en) * 2020-07-20 2021-07-20 山东大学 RNN neural network-based OFDM signal detection method
CN111884976B (en) * 2020-07-21 2021-09-03 西安电子科技大学 Channel interpolation method based on neural network
CN114465858A (en) * 2020-11-09 2022-05-10 索尼公司 Electronic device, signal processing method, and computer-readable storage medium
CN116455706B (en) * 2023-04-12 2024-05-31 华南理工大学 Channel estimation and equalization design method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1731779A (en) * 2005-08-05 2006-02-08 西安电子科技大学 Wavelet packet multi-carrier spread-spectrum system and control method based on neural net equalizer
CN101867533A (en) * 2010-05-27 2010-10-20 东南大学 Method for estimating pilot frequency and communication channel of space division multiple access (SDMA) multi-antenna transmission down link
CN104022979A (en) * 2014-06-23 2014-09-03 东南大学 Joint sparse channel estimation method, device and system
CN104022978A (en) * 2014-06-18 2014-09-03 中国联合网络通信集团有限公司 Half-blindness channel estimating method and system
CN108462517A (en) * 2018-03-06 2018-08-28 东南大学 A kind of MIMO link self-adaption transmission methods based on machine learning

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1731779A (en) * 2005-08-05 2006-02-08 西安电子科技大学 Wavelet packet multi-carrier spread-spectrum system and control method based on neural net equalizer
CN101867533A (en) * 2010-05-27 2010-10-20 东南大学 Method for estimating pilot frequency and communication channel of space division multiple access (SDMA) multi-antenna transmission down link
CN104022978A (en) * 2014-06-18 2014-09-03 中国联合网络通信集团有限公司 Half-blindness channel estimating method and system
CN104022979A (en) * 2014-06-23 2014-09-03 东南大学 Joint sparse channel estimation method, device and system
CN108462517A (en) * 2018-03-06 2018-08-28 东南大学 A kind of MIMO link self-adaption transmission methods based on machine learning

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Implementation of neural network for channel estimation in OFDM network;Ravinder Kaur等;《2016 2nd International Conference on Contemporary Computing and Informatics (IC3I)》;20170504;第437-441页 *
MIMO-OFDM系统信道估计及容量分析技术研究;李喆;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150415;第I136-329页 *
Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems;Hao Ye等;《IEEE Wireless Communications Letters》;20180228;第114-117页 *
Timing requirements and scenarios for 5G NR Hybrid ARQ;Nokia等;《3GPP TSG-RAN WG1#86 R1-167299 Gothenburg,Sweden》;20160826;全文 *
基于信号互相关性的MIMO系统盲均衡算法研究;王珊珊;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150715;第I136-203页 *

Also Published As

Publication number Publication date
CN109194595A (en) 2019-01-11

Similar Documents

Publication Publication Date Title
CN109194595B (en) Neural network-based channel environment self-adaptive OFDM receiving method
Zemen et al. Iterative joint time-variant channel estimation and multi-user detection for MC-CDMA
Tang et al. Space-time interference cancellation in MIMO-OFDM systems
CN106533590B (en) Uplink channel quality measurement method based on receiving end EVM
CN113225131B (en) Blind detection method of underwater visible light communication system
Simko et al. Optimal pilot symbol power allocation in LTE
CN112215335B (en) System detection method based on deep learning
TWI400902B (en) Method and apparatus for demapping symbol in multi-input multi-output communication system
Drakshayini et al. A review of wireless channel estimation techniques: challenges and solutions
CN109391315B (en) Data model dual-drive MIMO receiver
CN111865844B (en) Channel estimation method and device for large-scale MIMO full-duplex relay system
JP2010206457A (en) Wireless access system, and mobile station apparatus
CN107836100B (en) Method and apparatus for low complexity ISI estimation using sparse discontinuous time domain pilot
CN106789776B (en) High-precision channel estimation method of MIMO-OFDM system
CN108683440A (en) The improved modular algebra method for precoding of greedy algorithm is based under a kind of high-speed mobile
CN110868244B (en) Low-complexity communication signal detection method based on channel puncture
Chen et al. Fast converging semi-blind space-time equalisation for dispersive QAM MIMO systems
Kho et al. MIMO channel estimation and tracking based on polynomial prediction with application to equalization
CN108521311B (en) Signal-to-noise ratio estimation method based on Gray sequence
CN100405790C (en) Self-adaptive channel estimation method based on least squares criterion of two-dimensional iteration
CN112929062A (en) MIMO interference channel network interference alignment method based on group sparsity
CN112491752B (en) Multi-user large-scale MIMO relay network joint channel estimation method
Chang et al. Cancellation of ICI by Doppler effect in OFDM systems
CN100588194C (en) Self-adaptive channel estimation method based on two-dimensional minimum mean square criterion
CN109067444A (en) Improved modular algebra precoding algorithms are dispatched based on user under high-speed mobile scene

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant