CN112769462B - Millimeter wave MIMO broadband channel estimation method based on joint parameter learning - Google Patents

Millimeter wave MIMO broadband channel estimation method based on joint parameter learning Download PDF

Info

Publication number
CN112769462B
CN112769462B CN202110016400.1A CN202110016400A CN112769462B CN 112769462 B CN112769462 B CN 112769462B CN 202110016400 A CN202110016400 A CN 202110016400A CN 112769462 B CN112769462 B CN 112769462B
Authority
CN
China
Prior art keywords
channel
representing
millimeter wave
antenna
parameter
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
CN202110016400.1A
Other languages
Chinese (zh)
Other versions
CN112769462A (en
Inventor
程龙
伍怡
岳光荣
李少谦
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.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
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 University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202110016400.1A priority Critical patent/CN112769462B/en
Publication of CN112769462A publication Critical patent/CN112769462A/en
Application granted granted Critical
Publication of CN112769462B publication Critical patent/CN112769462B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • 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
    • H04L25/0242—Channel estimation channel estimation algorithms using matrix methods
    • 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

Abstract

The invention belongs to the field of array signal processing in the communication technology, and particularly relates to a millimeter wave MIMO broadband channel estimation method based on joint parameter learning. The invention is based on a novel millimeter wave large-scale multiple-input multiple-output channel model, extracts channel information from a space domain and a frequency domain through part of received signals, and converts a channel estimation problem into a line spectrum estimation problem. Then, a joint parameter learning algorithm is proposed to estimate all channel parameters. Therefore, the channel can be finally recovered, and the method is a low-complexity and high-performance channel estimation method.

Description

Millimeter wave MIMO broadband channel estimation method based on joint parameter learning
Technical Field
The invention belongs to the field of array signal processing in the communication technology, and particularly relates to a millimeter wave MIMO broadband channel estimation method based on joint parameter learning.
Background
Large-scale multiple-input multiple-output (MIMO) and millimeter-wave (mmWave) are recognized as promising technologies for future fifth generation (5G) wireless communication. In one aspect, the millimeter wave spectrum is capable of providing transmission bandwidths of approximately 2 GHz. On the other hand, massive MIMO may provide high beamforming gain to compensate for path loss of the millimeter wave channel and improve spectrum and energy efficiency. Therefore, mmWave massive MIMO systems can achieve 1000 times the data traffic growth for future 5G cellular networks.
However, the practical implementation of mmWave massive MIMO may not be an easy task. One challenging problem is massive MIMO channel estimation due to the large number of antennas. The conventional solution is to set the beam codeword and then traverse all directions. Although these schemes are computationally inexpensive, they either require frequent feedback of training information or have a high training overhead. In the past few years, many effective channel estimation algorithms have been proposed, taking advantage of the sparse nature of mmWave channels. In particular, the channel estimation problem is translated into a sparse recovery problem using compressed sensing techniques. In addition, by modeling the received signal as a third order tensor, all channel parameters can be obtained based on a scheme of tensor decomposition. In fact, most of the existing research is based on conventional MIMO channel models, simply assuming that the channel size is large. However, if the number of antennas is large and the transmission bandwidth is very wide, different antenna elements will receive different symbols at the same sampling time. This phenomenon, known as beam squint in the field of array signal processing, produces a spatial broadband effect for mmWave massive MIMO systems. There are studies that propose a novel channel model in which the spatial broadband effect is taken into account. However, existing channel estimation schemes have high computational complexity and require sampling data from all antennas.
Disclosure of Invention
The invention aims to provide a millimeter wave MIMO low-complexity broadband channel estimation method based on joint parameter learning. The invention considers that a method with low complexity is designed to be suitable for engineering design, and meanwhile, the robustness of channel estimation needs to be improved on the basis of coping with the broadband effect.
The invention provides a millimeter wave MIMO low-complexity broadband channel estimation scheme based on joint parameter learning based on a novel millimeter wave large-scale MIMO channel model. In particular, by extracting channel information from the spatial and frequency domains. The channel estimation problem is converted into a line spectrum estimation problem. Then, a joint parameter learning algorithm is proposed to estimate all channel parameters. Therefore, the channel can be finally recovered, and the method is a low-complexity and high-performance channel estimation method.
The core idea of the invention is that the channel estimation problem is converted into a two-dimensional linear spectrum estimation problem, and channel parameters are extracted from the received signals by using a joint parameter learning algorithm.
For ease of understanding, the model used in the present invention is first introduced:
the method is used for the millimeter wave channel model considering the space broadband effect. In a millimeter-wave massive MIMO system with Uniform Linear Arrays (ULAs), a Base Station (BS) has N antennas, and a Mobile Station (MS) is equipped with 1 antenna. For each example, consider a millimeter wave orthogonal frequency division multiple access (OFDM) transmission scheme. Tau isk,nIndicating the k-th path from the MS to the n-th antenna of the BS. The baseband time domain signal received by the nth antenna can be represented as:
Figure BDA0002886845420000021
where K is the number of independent physical paths, αkRepresenting the complex gain of the k-th path, x (t) being the transmitted baseband signal, fcRepresenting the carrier frequency. Based on array signal theory, we have:
Figure BDA0002886845420000022
wherein the content of the first and second substances,
Figure BDA0002886845420000023
representing the time delay, tau, of adjacent antennask,1Can be regarded as a multipath time delay, abbreviated as tauk，θkIndicates the angle of arrival (AOA) of the kth path, d is the antenna spacing, and c indicates the speed of light.
Thus, the received signal can be described as:
Figure BDA0002886845420000024
wherein
Figure BDA0002886845420000025
Representing the equivalent channel gain.
The millimeter wave MIMO time domain channel for the nth antenna is represented as:
Figure BDA0002886845420000031
obtaining a frequency domain channel model of the nth antenna by utilizing Fourier transform
Figure BDA0002886845420000032
Suppose a MIMO-OFDM system has M subcarriers. The space-frequency channel model for all antennas can be expressed as:
Figure BDA0002886845420000033
wherein the content of the first and second substances,
Figure BDA0002886845420000034
is a spatial guide vector that is a vector of the spatial orientation,
Figure BDA0002886845420000035
can be regarded as a frequency-oriented vector, the sign o denotes the hadamard product, fsIs the transmission bandwidth of the OFDM system,
Figure BDA0002886845420000036
can be thought of as a spatial broadband factor matrix, each of which
The items are:
Figure BDA0002886845420000037
the technical scheme of the invention is as follows:
a low-complexity broadband channel estimation method based on a combined parameter learning millimeter wave large-scale MIMO system is characterized in that a channel estimation problem is converted into a two-dimensional linear spectrum estimation problem, and a spatial broadband channel is finally recovered based on comprehensive perception of channel estimation, and comprises the following steps:
s1, acquiring the sample data from only the first antenna of the BS, and the MS transmits 1 to the BS, simplifying the received data to:
Figure BDA0002886845420000038
wherein, Y1,:A first row representing a received frequency domain signal, n being an additive white gaussian noise vector; the sampling data of the first subcarrier is as follows:
Figure BDA0002886845420000039
s2, considering sparsity of the millimeter wave channel, expressing the linear spectrum estimation problem as:
Figure BDA0002886845420000041
wherein the content of the first and second substances,
Figure BDA0002886845420000042
respectively, estimated channel gain and estimated spatial steering matrix, epsilon represents error tolerance;
s3, by substituting a logarithmic sum function and a data fitting term for l0Norm, further representing the problem of step S2 as an unconstrained optimization problem:
Figure BDA0002886845420000043
wherein the content of the first and second substances,
Figure BDA0002886845420000044
representing the estimated number of paths;
minimization function
Figure BDA0002886845420000045
Equal to:
Figure BDA0002886845420000046
wherein the content of the first and second substances,
Figure BDA0002886845420000047
is the iterative function of the ith iteration, diagonal matrix D(i)：
Figure BDA0002886845420000048
S4 obtaining the optimal value by partial derivation
Figure BDA0002886845420000049
Figure BDA00028868454200000410
Obtaining:
Figure BDA00028868454200000411
where η is the regularization parameter;
s5, updating the regularization parameter by the following method:
Figure BDA00028868454200000412
where r is a constant proportionality parameter, ηmaxIndicating an initial parameter, ξ, that makes the problem well-conditioned(i)Is the noise variance:
Figure BDA0002886845420000051
each iteration is to estimate a new AOA so that the cost function is smaller; by using a gradient descent method, newly estimated
Figure BDA0002886845420000052
Comprises the following steps:
Figure BDA0002886845420000053
wherein γ represents a step size;
s6, obtained by iteration
Figure BDA0002886845420000054
The specific steps are as follows:
s61, S4 and S5
Figure BDA0002886845420000055
As an initial value, according to the formula
Figure BDA0002886845420000056
Calculating the value of eta;
s62, retrieve the function:
Figure BDA0002886845420000057
s63, according to the formula
Figure BDA0002886845420000058
Recalculation
Figure BDA0002886845420000059
S64, according to the formula
Figure BDA00028868454200000510
Recalculating new path gains
Figure BDA00028868454200000511
S65, repeating the process from S61 to S64 until
Figure BDA00028868454200000512
S7, mixing Y:,1、
Figure BDA00028868454200000513
Respectively change into
Figure BDA00028868454200000514
τ repeat the process from S1 to S5, from Y1,:To extract the obtained
Figure BDA00028868454200000515
S8, due to
Figure BDA00028868454200000516
And
Figure BDA00028868454200000517
all represent channel gains, only in different order, according to
Figure BDA00028868454200000518
And
Figure BDA00028868454200000519
in the same parameter, rearranging the vectors
Figure BDA00028868454200000520
The gain, angle, and delay of each path are aligned. (ii) a
S9、
Figure BDA00028868454200000521
Wherein
Figure BDA00028868454200000522
Representing the channel gain obtained in the last iteration,
Figure BDA00028868454200000523
representing the final angular vector obtained by iteration and sequential rearrangement,
Figure BDA00028868454200000524
the delay vector is also obtained through iteration and sequential rearrangement. (ii) a
S10, according to the formula
Figure BDA00028868454200000525
And three parameter vectors have been obtained
Figure BDA00028868454200000526
A channel is constructed.
The invention has the beneficial effects that:
aiming at a novel channel model (considering space broadband influence), the method uses a joint parameter learning algorithm, only uses partial antenna receiving information to complete channel estimation, reduces algorithm complexity, and simultaneously improves system robustness in engineering application.
Drawings
FIG. 1 is a graph of Normalized Mean Square Error (NMSE) and computational complexity at different bandwidths;
figure 2 is a graph of normalized mean square error and computational complexity for different antenna numbers.
Detailed Description
The technical solution of the present invention is described in detail below with reference to the embodiments and the accompanying drawings.
Examples
Taking a millimeter wave broadband massive MIMO system as an example,
Figure BDA0002886845420000061
M-N-128, K-3, carrier frequency fc60GHz, comprising the following steps:
s1, only the sampled data from the first antenna is considered, and the MS transmits 1 to the BS, then the received data can be simplified as:
Figure BDA0002886845420000062
wherein, Y1,:Representing the first row of the received frequency domain signal, n is an additive white gaussian noise vector. The sampling data of the first subcarrier is as follows:
Figure BDA0002886845420000063
s2, the linear spectrum estimation problem is expressed as:
Figure BDA0002886845420000064
wherein the content of the first and second substances,
Figure BDA0002886845420000071
respectively representing an estimated channel gain and an estimated space steering matrix, wherein epsilon represents error tolerance and is 0.05;
s3, by substituting a logarithmic sum function and a data fitting term for l0Norm, this problem can be further expressed as an unconstrained optimization problem:
Figure BDA0002886845420000072
wherein the content of the first and second substances,
Figure BDA0002886845420000073
representing the estimated number of paths. Further, minimizing a function
Figure BDA0002886845420000074
Is equal to
Figure BDA0002886845420000075
Wherein the content of the first and second substances,
Figure BDA0002886845420000076
is the iterative function of the ith iteration, diagonal matrix D(i)：
Figure BDA0002886845420000077
S4, obtaining the optimum through partial derivation
Figure BDA0002886845420000078
Figure BDA0002886845420000079
This makes it possible to obtain:
Figure BDA00028868454200000710
where η is the regularization parameter.
S5, since the regularization parameter η is crucial to controlling the trade-off between sparsity and data fitting error, it can be updated in the following way
Figure BDA00028868454200000711
Wherein r is a constant proportionality parameter of 0.2, etamaxInitial parameters indicating that the problem was well-conditioned were 0.00008, ξ(i)Is the variance of the noise, i.e.
Figure BDA0002886845420000081
Each iteration is to estimate a new AOA so that the cost function is smaller. By using a gradient descent method, newly estimated
Figure BDA0002886845420000082
This gives:
Figure BDA0002886845420000083
wherein gamma represents a step size, and the step size is 0.01;
s6, obtained by iteration
Figure BDA0002886845420000084
The specific steps are as follows;
s61, S4 and S5
Figure BDA0002886845420000085
As an initial value, according to the formula
Figure BDA0002886845420000086
Calculating the value of eta;
s62, retrieve the function:
Figure BDA0002886845420000087
s63, according to the formula
Figure BDA0002886845420000088
Recalculation
Figure BDA0002886845420000089
S64, according to the formula
Figure BDA00028868454200000810
Recalculating new path gains
Figure BDA00028868454200000811
S65, repeating the process from S61 to S64 until
Figure BDA00028868454200000812
S7, mixing Y:,1、
Figure BDA00028868454200000813
Respectively change into
Figure BDA00028868454200000814
τ repeat the process from S1 to S5, from Y1,:Is extracted and found
Figure BDA00028868454200000815
S8, according to
Figure BDA00028868454200000816
And
Figure BDA00028868454200000817
in the same parameter, rearranging the vectors
Figure BDA00028868454200000818
The order of (a);
S9、
Figure BDA00028868454200000819
s10, according to the formula
Figure BDA00028868454200000820
A channel is constructed.
The performance of the method of the invention will be analyzed to further verify the performance of the invention.
Two aspects are used to measure the effectiveness of the algorithm, one is to use a normalized mean square error plot to measure the accuracy of the channel estimate. Another is the complexity of the algorithm using a computational complexity curve metric.
Two scenarios are adopted to embody the superiority of the algorithm. One is the impact on the algorithm at different bandwidths and the other is the impact of different numbers of transmit antennas on the algorithm performance.
Fig. 1 is a graph of Normalized Mean Square Error (NMSE) and computational complexity at different bandwidths, where SNR is 10dB,
Figure BDA0002886845420000091
in the case of P10, the algorithm proposed by the present invention compares the normalized mean square error and the computational complexity with different algorithms. Although the algorithm NMSE provided by the invention is slightly higher than the algorithm based on the beam space, the complexity can be reduced by 88%. When the transmission band is large, conventional compressed sensing-based algorithms such as Orthogonal Matching Pursuit (OMP) do not perform well. Compared with the beam space algorithm, the algorithm can realize 88% complexity reduction and has considerable performance improvement. The result shows that the invention skillfully utilizes the information of space and frequency dimensions and reduces the calculation complexity. Furthermore, when data is sampled from partial antennas (e.g., 100 antennas and 80 antennas), the present algorithm can still effectively estimate the channel, while other algorithms cannot. In addition, in an actual millimeter wave system, the special advantage of the algorithm can be used for reducing the power consumption of the system;
fig. 2 is a graph of normalized mean square error and computational complexity for different antenna counts, where SNR is 10dB,
Figure BDA0002886845420000092
P＝10,fsin the case of 1GHz, as the number of antennas increases, the normalized mean square error of the algorithm based on compressed sensing increases rapidly, while the complexity of the algorithm based on beam space increases rapidly, whereas the normalized mean square error of the algorithm of the present invention may even decrease, and the computational complexity increases only slightly. As the number of antennas increases, the performance of the conventional compressed sensing-based algorithm decreases, and the complexity of the beam space algorithm increases rapidly. However, the algorithm of the present invention still remains less complex compared to the beam space approach. This may be due to the invention making use ofUnaliased information in the beam space domain is obtained.
In summary, the present invention provides a new low-complexity wideband channel estimation scheme for the millimeter wave spatial wideband channel model. Simulation results show that compared with the beam space algorithm of a new channel model, the method can reduce the complexity by 88% and has less performance loss. Furthermore, our method can effectively estimate new channels when sampling data from partial antennas, which has a significant advantage in practical millimeter-wave massive MIMO systems.

Claims (1)

1. A millimeter wave MIMO broadband channel estimation method based on joint parameter learning is used for a millimeter wave large-scale MIMO system, wherein a base station BS in the system is provided with N antennas, a mobile station MS is provided with 1 antenna, and the received signal of the nth antenna of the BS is as follows:
Figure FDA0003289037360000011
wherein the content of the first and second substances,
Figure FDA0003289037360000012
representing equivalent channel gain, x (t) being the transmitted baseband signal, alphakComplex gain, f, of k path representing the n antenna from MS to BScRepresenting the carrier frequency, τk,nDenotes the K-th path from the MS to the n-th antenna of the BS, K being the number of independent physical paths, τkIs the time delay of the multi-path,
Figure FDA0003289037360000013
representing the time delay of adjacent antennas, d is the antenna spacing, c represents the speed of light, thetakRepresenting the arrival angle of the kth path, and lambda is the signal wavelength;
the millimeter wave MIMO time domain channel for the nth antenna of the BS is represented as:
Figure FDA0003289037360000014
where δ (t) is the impulse function, assuming that the system has M subcarriers, the space-frequency channel model for all antennas is expressed as:
Figure FDA0003289037360000015
wherein the content of the first and second substances,
Figure FDA0003289037360000016
is a spatial guide vector that is a vector of the spatial orientation,
Figure FDA0003289037360000017
is a frequency-oriented vector, symbol
Figure FDA0003289037360000018
Representing the Hadamard product, fsIs the transmission bandwidth of the system and,
Figure FDA0003289037360000019
however, the spatial wideband factor matrix, each of its terms is:
Figure FDA00032890373600000110
the channel estimation method is characterized by comprising the following steps:
s1, acquiring the sample data from only the first antenna of the BS, and the MS transmits 1 to the BS, simplifying the received data to:
Figure FDA00032890373600000111
wherein, Y1,:A first row representing a received frequency domain signal, n being an additive white gaussian noise vector; the sampling data of the first subcarrier is as follows:
Figure FDA0003289037360000021
s2, considering sparsity of the millimeter wave channel, expressing the linear spectrum estimation problem as:
Figure FDA0003289037360000022
wherein the content of the first and second substances,
Figure FDA0003289037360000023
respectively, estimated channel gain and estimated spatial steering matrix, epsilon represents error tolerance;
s3, by substituting a logarithmic sum function and a data fitting term for l0Norm, further representing the problem of step S2 as an unconstrained optimization problem:
Figure FDA0003289037360000024
wherein the content of the first and second substances,
Figure FDA0003289037360000025
representing the estimated number of paths;
minimization function
Figure FDA0003289037360000026
Equal to:
Figure FDA0003289037360000027
wherein the content of the first and second substances,
Figure FDA0003289037360000028
is the iterative function of the ith iteration, diagonal matrix D(i)：
Figure FDA0003289037360000029
S4 obtaining the optimal value by partial derivation
Figure FDA00032890373600000210
Figure FDA00032890373600000211
Obtaining:
Figure FDA00032890373600000212
where η is the regularization parameter;
s5, updating the regularization parameter by the following method:
Figure FDA0003289037360000031
where r is a constant proportionality parameter, ηmaxIndicating an initial parameter, ξ, that makes the problem well-conditioned(i)Is the noise variance:
Figure FDA0003289037360000032
each iteration is to estimate a new AOA so that the cost function is smaller; by using a gradient descent method, newly estimated
Figure FDA0003289037360000033
Comprises the following steps:
Figure FDA0003289037360000034
wherein γ represents a step size;
s6, obtained by iteration
Figure FDA0003289037360000035
The specific steps are as follows:
s61, S4 and S5
Figure FDA0003289037360000036
As an initial value, according to the formula
Figure FDA0003289037360000037
Calculating the value of eta;
s62, retrieve the function:
Figure FDA0003289037360000038
s63, according to the formula
Figure FDA0003289037360000039
Recalculation
Figure FDA00032890373600000310
S64, according to the formula
Figure FDA00032890373600000311
Recalculating new path gains
Figure FDA00032890373600000312
S65, repeating the process from S61 to S64 until
Figure FDA00032890373600000313
S7, mixing Y:,1、
Figure FDA00032890373600000314
Respectively change into
Figure FDA00032890373600000315
τ repeat the process from S1 to S5, from Y1,:To extract the obtained
Figure FDA00032890373600000316
S8, due to
Figure FDA00032890373600000317
And
Figure FDA00032890373600000318
all represent channel gains, only in different order, according to
Figure FDA00032890373600000319
And
Figure FDA00032890373600000320
in the same parameter, rearranging the vectors
Figure FDA00032890373600000321
Aligning the gain, angle, and delay of each path;
S9、
Figure FDA0003289037360000041
Figure FDA0003289037360000042
wherein
Figure FDA0003289037360000043
Representing the channel gain obtained in the last iteration,
Figure FDA0003289037360000044
indicating a final pass through the stackThe angle vectors obtained by the generation and sequential rearrangement,
Figure FDA0003289037360000045
is a delay vector obtained through iteration and sequential rearrangement;
s10, according to the formula
Figure FDA0003289037360000046
And three parameter vectors that have been obtained
Figure FDA0003289037360000047
A channel is constructed.
CN202110016400.1A 2021-01-07 2021-01-07 Millimeter wave MIMO broadband channel estimation method based on joint parameter learning Active CN112769462B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110016400.1A CN112769462B (en) 2021-01-07 2021-01-07 Millimeter wave MIMO broadband channel estimation method based on joint parameter learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110016400.1A CN112769462B (en) 2021-01-07 2021-01-07 Millimeter wave MIMO broadband channel estimation method based on joint parameter learning

Publications (2)

Publication Number Publication Date
CN112769462A CN112769462A (en) 2021-05-07
CN112769462B true CN112769462B (en) 2022-02-08

Family

ID=75700357

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110016400.1A Active CN112769462B (en) 2021-01-07 2021-01-07 Millimeter wave MIMO broadband channel estimation method based on joint parameter learning

Country Status (1)

Country Link
CN (1) CN112769462B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113660046B (en) * 2021-08-17 2022-11-11 东南大学 Method for accelerating generation of large-scale wireless channel coefficients
CN115022130B (en) * 2022-05-16 2024-02-06 海南大学 Millimeter wave large-scale MIMO channel estimation method based on quantum mechanical denoising

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106559367A (en) * 2016-12-08 2017-04-05 电子科技大学 MIMO ofdm system millimeter wave channel estimation methods based on low-rank tensor resolution
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
CN108964726A (en) * 2018-09-03 2018-12-07 东南大学 A kind of extensive MIMO uplink transmission channels estimation method of low complex degree
WO2019195426A1 (en) * 2018-04-03 2019-10-10 University Of Southern California Analog channel estimation techniques for beamformer design in massive mimo systems
CN110336761A (en) * 2019-07-12 2019-10-15 电子科技大学 The beam space channel estimation methods of the extensive mimo system of millimeter wave

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015073260A (en) * 2013-09-04 2015-04-16 富士通株式会社 Radio communication system and radio communication method
CN109768943A (en) * 2019-03-05 2019-05-17 北京邮电大学 Based on the channel estimation methods of Variational Bayesian Learning in broadband multiuser millimeter-wave systems

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106559367A (en) * 2016-12-08 2017-04-05 电子科技大学 MIMO ofdm system millimeter wave channel estimation methods based on low-rank tensor resolution
WO2019195426A1 (en) * 2018-04-03 2019-10-10 University Of Southern California Analog channel estimation techniques for beamformer design in massive mimo systems
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
CN108964726A (en) * 2018-09-03 2018-12-07 东南大学 A kind of extensive MIMO uplink transmission channels estimation method of low complex degree
CN110336761A (en) * 2019-07-12 2019-10-15 电子科技大学 The beam space channel estimation methods of the extensive mimo system of millimeter wave

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Millimeter Wave Time-Varying Channel Estimation via Exploiting Block-Sparse and Low-Rank Structures;Long Cheng 等;《IEEE Access》;20190826;第7卷;第123355-123366页 *
Quasi-Static and Time-Selective Channel Estimation for Block-Sparse Millimeter Wave Hybrid MIMO Systems: Sparse Bayesian Learning (SBL) Based Approaches;Suraj Srivastava 等;《IEEE Transactions on Signal Processing》;20181227;第67卷(第5期);第1251-1266页 *
Spatial- and Frequency-Wideband Effects in Millimeter-Wave Massive MIMO Systems;Bolei Wang;《IEEE Transactions on Signal Processing》;20180504;第66卷(第13期);第3393-3406页 *
Tensor Decomposition-Aided Time-Varying Channel Estimation for Millimeter Wave MIMO Systems;Long Cheng 等;《IEEE Wireless Communications Letters》;20190419;第8卷(第4期);第1216-1219页 *

Also Published As

Publication number Publication date
CN112769462A (en) 2021-05-07

Similar Documents

Publication Publication Date Title
CN109743268B (en) Millimeter wave channel estimation and compression method based on deep neural network
Peken et al. Blind channel estimation for massive MIMO
WO2017219389A1 (en) Methods for sending and receiving synchronization signals and signals subjected to perfect omnidirectional pre-coding in large-scale mimo system
CN110213185B (en) Three-dimensional channel parameter estimation method based on atomic norm minimization
CN112769462B (en) Millimeter wave MIMO broadband channel estimation method based on joint parameter learning
CN109005133B (en) Double-sparse multi-path channel model and channel estimation method based on model
CN112565118B (en) Channel estimation method for millimeter wave lens antenna array communication system
Zhang et al. Channel estimation and training design for hybrid multi-carrier mmwave massive MIMO systems: The beamspace ESPRIT approach
CN110650103B (en) Lens antenna array channel estimation method for enhancing sparsity by using redundant dictionary
Gong et al. Block distributed compressive sensing-based doubly selective channel estimation and pilot design for large-scale MIMO systems
CN113411107A (en) Millimeter wave large-scale MIMO system broadband channel estimation method based on beam
CN109861933A (en) A kind of millimeter wave mimo channel estimation method based on MUSIC algorithm and precoding
CN110212951B (en) Large-scale MIMO channel estimation method based on Butler matrix
CN112636794A (en) Wave beam shaping method based on greedy algorithm
Cong et al. Hybrid precoding for multi-user mmWave systems based on MMSE criterion
CN115733530A (en) Combined precoding method for reconfigurable intelligent surface assisted millimeter wave communication
Srivastava et al. MSBL-based simultaneous sparse channel estimation in SC wideband mmWave hybrid MIMO systems
Su et al. Diversity and multiplexing technologies by 3D beams in polarized massive MIMO systems
Bai et al. Dictionary learning based channel estimation and activity detection for mMTC with massive MIMO
CN108900290B (en) Pilot frequency distribution method based on positioning information
Peng et al. Channel estimation for RIS-aided mmWave MIMO system from 1-sparse recovery perspective
Mo et al. Wideband sparse cascaded channel estimation for RIS-assisted wireless systems
Ravindran et al. Direction of arrival and channel estimation using machine learning for multiple input multiple output system
Wang et al. Smoothed Sparse Blind DOA Estimation for MC-CDMA Systems Based on Tensor Decomposition
Cheng et al. Low-complexity wideband channel estimation for millimeter-wave massive MIMO systems via joint parameter learning

Legal Events

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