CN116032699B - Sparse channel estimation method for ultra-large-scale MIMO system - Google Patents
Sparse channel estimation method for ultra-large-scale MIMO system Download PDFInfo
- Publication number
- CN116032699B CN116032699B CN202211694969.4A CN202211694969A CN116032699B CN 116032699 B CN116032699 B CN 116032699B CN 202211694969 A CN202211694969 A CN 202211694969A CN 116032699 B CN116032699 B CN 116032699B
- Authority
- CN
- China
- Prior art keywords
- sparse
- channel
- channel estimation
- scale mimo
- prior model
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 239000013598 vector Substances 0.000 claims abstract description 42
- 230000010287 polarization Effects 0.000 claims abstract description 24
- 238000004891 communication Methods 0.000 claims abstract description 17
- 230000005669 field effect Effects 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 23
- 238000005070 sampling Methods 0.000 claims description 20
- 238000010276 construction Methods 0.000 claims description 4
- 239000000654 additive Substances 0.000 claims description 3
- 230000000996 additive effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 description 12
- 238000004590 computer program Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000014509 gene expression Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- NTSBMKIZRSBFTA-AIDOXSFESA-N Digoxigenin bisdigitoxoside Chemical compound C1[C@H](O)[C@H](O)[C@@H](C)O[C@H]1O[C@@H]1[C@@H](C)O[C@@H](O[C@@H]2C[C@@H]3[C@]([C@@H]4[C@H]([C@]5(CC[C@@H]([C@@]5(C)[C@H](O)C4)C=4COC(=O)C=4)O)CC3)(C)CC2)C[C@@H]1O NTSBMKIZRSBFTA-AIDOXSFESA-N 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Radio Transmission System (AREA)
Abstract
The invention discloses a sparse channel estimation method for a super-large-scale MIMO system, which comprises the following steps: based on the double-layer block sparse structural characteristics of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, constructing a channel sparse prior model, wherein the channel sparse prior model considers a near field effect when being constructed and is modeled based on spherical wave hypothesis; based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal; and based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value. The invention can realize high-precision and low-complexity channel estimation.
Description
Technical Field
The invention relates to the technical field of wireless communication, in particular to a sparse channel estimation method for a super-large-scale MIMO system.
Background
Large-scale Multiple-input Multiple-Output (MIMO) has been widely deployed in current wireless communication networks as one of the key technologies for the new type of air interface of 5G. System gain is obtained by deploying a large antenna array at the Base Station (BS) side, including high spectral efficiency and capacity, high energy efficiency, and low complexity signal processing, etc. In order to meet the capacity and rate requirements of future 6G communication, a super-large-scale MIMO system with a higher dimension antenna array needs to be further deployed and a corresponding new signal processing method needs to be researched, particularly a channel estimation method is used for acquiring accurate channel state information (ChannelStateInformation, CSI), and is a basis for realizing technologies such as channel equalization, precoding, beam forming, resource allocation and the like, and is very important for guaranteeing effective transmission of super-large-scale MIMO signals and acquiring the gain of the super-large-scale MIMO system.
On the one hand, as the future 6G ultra-large-scale MIMO system is provided with a larger-scale antenna array at the base station side, the Rayleigh distance (generally defined as 2D 2/λc, wherein D represents the dimension of the antenna array, lambda c represents the carrier wavelength) is increased, so that the mobile user and the scatterer are within the Rayleigh distance, and the channel sparse channel estimation method based on the planar wavefront assumption, which is widely adopted at present, is not applicable any more, and a channel sparse estimation model needs to be researched and constructed under the spherical wavefront assumption; on the other hand, sparse channel estimation based on compressed sensing (compressed sensing, CS) theory has been widely studied in order to reduce pilot overhead and channel estimation algorithm complexity. Therefore, the invention aims to provide a new channel estimation scheme for a future 6G ultra-large-scale MIMO system.
Accordingly, there is a need for improvement and advancement in the art.
Disclosure of Invention
The invention aims to solve the technical problem of providing a sparse channel estimation method for a super-large-scale MIMO system aiming at the defects of the prior art, and aims to provide a new channel estimation scheme for a future 6G super-large-scale MIMO system.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
In a first aspect, the present invention provides a sparse channel estimation method for a super-large-scale MIMO system, where the method includes:
Based on the double-layer block sparse structural characteristics of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, constructing a channel sparse prior model, wherein the channel sparse prior model considers a near field effect when being constructed and is modeled based on spherical wave hypothesis;
Based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal;
And based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value.
In one implementation, the sparsity of the channel in the time delay and polarization domain is expressed as:
Wherein, W represents a sparse channel vector to be estimated, q=m×i represents the number of polarization domain samples, I represents the number of spatial distance samples, M represents the dimension, L represents the number of multipaths, B represents a sparse dictionary,/>Representing the kronecker product.
In one implementation, w has a block-sparse structure similar to h, referred to as outer layer block sparsity; w l has a block sparse structure, referred to as inner layer block sparsity;
Wherein, The sparse representation vector of the channel under the sparse dictionary B is represented.
In one implementation, the polarization domain is an angle-space distance domain, and the constructing a channel sparse prior model based on a double-layer block sparse structure feature of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system includes:
Sampling in both the angular and spatial dimensions to construct a sparse representation matrix of the channel in near field conditions, where for angular domain sampling, the sampling angle is represented as
For spatial distance sampling, the sampling distance relative to angle θ l,m is expressed as
Wherein,Where β Δ represents a low coherence threshold that guarantees a sparse channel representation matrix.
In one implementation, the method further comprises:
Constructing a system model of ultra-large-scale MIMO, wherein the system model is as follows under near-field channel sparse representation: y=Φw+n,
Wherein,Diagonal matrix representing signal vector x,/>A submatrix representing a discrete Fourier matrix, K being a pilot symbol, wherein the (K, l) th term is/>Representing additive gaussian white noise, satisfying a complex gaussian distribution CN (0, σ -1IMK), where σ represents noise accuracy.
In one implementation, the sparse channel vector w to be estimated satisfies the following a priori distribution:
Wherein, As a non-negative parameter, the outer layer block sparsity of w is controlled, and when γ l =0, the variance of w l is 0, which means that the values in the block are all 0,/>Λ l is a diagonal matrix that controls the inner layer block sparsity of w l, denoted as
Wherein, a l,q＝(αl,q+βαl,q-1+βαl,q+1)-1,αl,q is a non-negative parameter, β characterizes the correlation between the parameter α l,q and its neighboring parameter { α l,q-1,αl,q+1 }.
In one implementation manner, the estimating method based on the non-inversion block sparse variance dbis inference algorithm iteratively updates parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value, and includes:
Presetting the maximum iteration times;
iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inversion-free block sparse variable decibel leaf inference algorithm;
And stopping iteration if the current iteration number meets the maximum iteration number, and obtaining the channel estimation value.
In one implementation manner, the estimating method based on the non-inversion block sparse variance dbis inference algorithm iteratively updates parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value, and includes:
presetting an error value of posterior distribution after updating the parameters to be estimated;
Iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inversion-free block sparse variable decibel leaf-based inference algorithm, and determining actual error values of the posterior distribution after updating the parameter to be estimated;
and stopping iteration if the actual error value meets the preset error value, and obtaining the channel estimation value.
In a second aspect, an embodiment of the present invention further provides a sparse channel estimation system for a super-large-scale MIMO system, where the system includes:
The model construction module is used for constructing a channel sparse prior model based on the double-layer block sparse structural characteristics of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, wherein the channel sparse prior model considers a near field effect when being constructed and is modeled based on spherical wave hypothesis;
the channel receiving module is used for controlling a user terminal to send pilot symbols to a base station terminal based on the channel sparse prior model, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal;
And the channel estimation module is used for carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector based on an inverse block-free sparse-variance decibel leaf-based inference algorithm to obtain a channel estimation value.
In a third aspect, an embodiment of the present invention further provides a terminal device, where the terminal device is applied to a receiving end or a transmitting end, and includes a memory, a processor, and a sparse channel estimation program for a super-large scale MIMO system stored in the memory and capable of running on the processor, where the processor implements the steps of the sparse channel estimation method for a super-large scale MIMO system of any one of the above schemes when executing the sparse channel estimation program for a super-large scale MIMO system.
In a fourth aspect, an embodiment of the present invention further provides a computer readable storage medium, where a sparse channel estimation program for a super-large scale MIMO system is stored on the computer readable storage medium, and when the sparse channel estimation program for the super-large scale MIMO system is executed by a processor, the steps of the sparse channel estimation method for the super-large scale MIMO system according to any one of the above schemes are implemented.
The beneficial effects are that: compared with the prior art, the invention provides a sparse channel estimation method for a super-large-scale MIMO system, which is characterized in that firstly, a channel sparse prior model is constructed based on the double-layer block sparse structure characteristics of a channel in a time delay and polarization domain in an uplink super-large-scale MIMO communication system, wherein the channel sparse prior model considers near field effect when being established and is modeled based on spherical wave hypothesis. And then, based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal. And finally, based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value. The invention can realize high-precision and low-complexity channel estimation.
Drawings
Fig. 1 is a flowchart of a specific implementation of a sparse channel estimation method for a super-large-scale MIMO system according to an embodiment of the present invention.
Fig. 2 is an overall architecture diagram of a sparse channel estimation system for a super-large-scale MIMO system according to an embodiment of the present invention.
Fig. 3 is a schematic block diagram of an internal structure of a terminal device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and more specific, the present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment provides a sparse channel estimation method for a super-large-scale MIMO system, and when the method is specifically applied, the method is firstly based on the double-layer block sparse structure characteristics of a channel in a time delay and polarization domain in an uplink super-large-scale MIMO communication system to construct a channel sparse prior model, wherein the channel sparse prior model considers a near field effect when being established and is modeled based on spherical wave hypothesis. And then, based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal. And finally, based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value. The invention can realize high-precision and low-complexity channel estimation.
Exemplary method
The sparse channel estimation method for the ultra-large-scale MIMO system can be applied to terminal equipment, wherein the terminal equipment is intelligent product equipment such as computers, mobile phones and intelligent televisions. As shown in fig. 1, the sparse channel estimation method for a super-large-scale MIMO system of the present embodiment includes the steps of:
Step S100, a channel sparse prior model is built based on double-layer block sparse structural features of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, wherein the channel sparse prior model considers near field effect and is modeled based on spherical wave hypothesis when being built.
Step 200, based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal;
And step S300, based on a non-inversion block sparse variance dB-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value.
In specific application, the embodiment considers the near field effect caused by the deployment of the high-dimensional antenna array in the 6G ultra-large-scale MIMO system, digs the potential structural characteristics of the channel in the time delay-polarization domain, and constructs the structural sparse channel estimation problem; secondly, designing a sparse prior model for describing a double-layer block channel of a channel in a time delay-polarization domain, wherein the sparse prior model describes a double-layer block channel sparse structure of the channel by using a group of super parameters; finally, aiming at the high computational complexity caused by matrix inversion under the current sparse Bayesian framework, a high-efficiency sparse Bayesian learning algorithm without matrix inversion (lnverse-free, IF) is provided, and high-precision and low-complexity channel estimation is realized.
In particular, the present embodiment contemplates an uplink super-large-scale MIMO communication system. A Uniform linear antenna array (Uniform LINEAR ANTENNA-array, ULA) with dimension of M (M > > 1) is equipped at the base station end to serve single antenna users, and orthogonal frequency division multiplexing (Orthogonal Frequency Division Multiplexing, OFDM) is adopted as a signal transmission mechanism. In order to estimate the uplink channel, the present embodiment adopts a pilot-aided channel estimation method, that is, K (K < < N) subcarriers among N subcarriers included in each OFDM symbol are used for transmitting pilot symbols, and pilot position identification is defined asAccordingly, the signal vector of the transmitted pilot symbol is x= [ x (p 1),x(p2),...,x(pK)]T. Therefore, the received signal containing K pilot symbols received by the antenna array at the base station end is the following formula (1), that is, the system model of the ultra-large-scale MIMO constructed in this embodiment.
In the method, in the process of the invention,Diagonal matrix representing signal vector x,/> A submatrix representing a discrete Fourier matrix, wherein the (k, l) th term is/>Representing the kronecker product, I M represents an identity matrix with dimension M,/> Representing a time domain channel vector, L representing the number of multipaths, Representing additive gaussian white noise, satisfying a complex gaussian distribution CN (0, σ -1IMK), where σ represents noise accuracy.
Further, the present embodiment considers the near field effect,Further modeled under spherical wave assumption as
Where lambda c denotes the carrier wavelength, S l denotes the number of sub-paths included in the first path, g l,s denotes the power of the S-th sub-path in the first path, theta l,s denotes the angle of arrival of the S-th sub-path in the first path to the center of the base station antenna array, r l,s denotes the distance of the S-th sub-path in the first path to the center of the base station antenna array,Representing the distance from the s-th sub-path to the m-th antenna array at the base station side in the first path, can be further calculated as
Wherein delta m represents the identity of the mth antenna element in the antenna array, defined as D represents the distance between adjacent antenna elements.
As can be seen from equations (2) and (3), the antenna response vector b (θ l,s,rl,s) in the near field condition is related to the angle of arrival of the radiation passing through the scatterer and the spatial position between the base station. Therefore, this embodiment requires sampling in both the angular and spatial dimensions to construct a channel sparse representation matrix under near field conditions. Notably, the angle-space distance domain is also referred to as a polarization domain, since the angle and space distance can represent coordinates in a polar coordinate system. For angle domain sampling, the sampling angle is expressed as:
for spatial distance sampling, the sampling distance relative to angle θ l.m is expressed as
In the method, in the process of the invention,Where β Δ denotes the low coherence threshold that guarantees a sparse channel representation matrix, I denotes the number of spatial distance samples.
Channel(s)Can be further sparsely represented as
B＝[B1,...,Bi,...,BI]， (8)
Bi＝[b(θl,0,rl,0(i)),...,b(θl,M-1,rl,M-1(i))]， (9)
In the method, in the process of the invention,Representing the sparse representation vector of the channel under sparse dictionary B, q=m×i represents the number of polarization domain samples. Based on equation (4-7), channel h is further denoted as
In the method, in the process of the invention,
For the sparse channel vector w to be estimated, it has a two-layer block sparse structure for the following reasons: on the one hand, a large number of channel measurements indicate that the number of multipaths with greater power in the time domain channel is small, and therefore,Which may appear as all-zero blocks or all-non-zero blocks, thereby allowing channel h to have a block sparse structure, i.e., |h| 0,K < < L, where, ||·| 0,K represents number of non-zero blocks. Thus, w has a block-sparse structure similar to h, referred to as outer layer block sparsity. On the other hand, the theory of geometric-based random channel modeling shows that the channel multipath is modeled as a scattering cluster, and the scattering cluster comprises a plurality of sub-clusters, wherein the sub-clusters are distributed near the central arrival angle. The position of the non-zero element in w l is related to the angle of arrival of the sub-cluster, and therefore w l has a block-sparse structure, referred to as inner layer block-sparsity. The two aspects are considered, and the sparse vector w to be estimated has a double-layer block sparse structure.
In addition, this embodiment also constructs a system model of ultra-large-scale MIMO, and rewrites the system model shown in formula (1) into one under near-field channel sparse representation
y＝Φw+n， (1 1)
In the method, in the process of the invention,
Further, the channel w has a double-layer block sparse structure in the time delay-polarization domain, and can be used for designing a channel estimation algorithm to improve the performance of large-scale MIMO channel estimation. In order to characterize such a channel sparse structure, a sparse channel vector w to be estimated is defined to satisfy the following a priori distribution:
In the method, in the process of the invention, As a non-negative parameter, the outer layer block sparsity of w is controlled. When γ l =0, the variance of w l is 0, indicating that the values in the block are all 0. Therefore, the number of non-zeros in γ determines the outer block sparsity. At the same time/>A l is a diagonal matrix that controls the inner layer block sparsity of w l, denoted as
Λl＝diag([∧l,1,...,∧l,q,...,∧l,Q]T)， (13)
Where, l,q＝(αl,q+βαl,q-1+βαl,q+1)-1,αl,q is a non-negative parameter, β characterizes the correlation between the parameter α l,q and its neighboring parameter { α l,q-1,αl,q+1 }. Meanwhile, in order to further promote the sparsity of the model, the prior distribution of Gamma l and alpha l,q is defined to meet the Gamma distribution, namely
P(γl|λ)＝λexp(-λγl)， (14)
P(αl,q|ξl)＝ξlexp(-ξlαl,q)， (15)
Where λ is a parameter related to γ l and ζ l is a parameter related to α l,q.
In the traditional sparse Bayesian learning, the defect of high computational complexity of inversion operation in the variance solving process of posterior distribution exists. This embodiment, by taking reference to the concept of non-inversion sparse Bayesian learning (Inverse-FREE SPARSE Bayesian learning, IF-SBL), proposes a lower relaxation likelihood limit (relaxed evidence lower bound, R-ELBO) for w as:
In the method, in the process of the invention, Where eig (·) represents the eigenvalue, (·) H represents the transposed inversion of the matrix. Therefore, the lower limit of the P (y|w, σ) distribution is
Based on the channel prior model to be estimated and the super parameters thereof, the channel to be estimated and the parameter set thereof are Θ{ W, γ, α, ζ, λ, σ }, where/(m)Because the accurate posterior distribution of the parameters to be estimated in Θ is difficult to obtain, the channel estimation algorithm based on the variable decibel leaf reasoning is provided in the embodiment. The Θ posterior approximation distribution factorizes into q (Θ) =q (w) q (γ) q (α) q (ζ) q (λ) q (σ), where q (x) represents the posterior distribution of factor x. Maximizing R-ELB0 is obtained by alternately updating the approximate posterior distribution of the parameters to be estimated, and the optimized expression of q (Θ) is
In the method, in the process of the invention,Θ n represents the nth parameter to be estimated in Θ,/>The expectation of q (Θ)/q (Θ n) is expressed, and q (Θ)/q (Θ n) represents a q (Θ) distribution not including q (Θ n). Based on equation (18), the posterior distribution of each parameter to be estimated is updated step by step while leaving the other parameters unchanged during the iteration. After the iteration converges, the estimated value of each parameter is equal to the expectation of the posterior distribution. Specifically, the present embodiment sets the maximum number of iterations in advance. And then, iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inverse block-free sparse variable decibel leaf inference algorithm. And stopping iteration if the current iteration number meets the maximum iteration number, and obtaining the channel estimation value. Or in another implementation manner, the embodiment presets the error value of the posterior distribution after the parameter to be estimated is updated. And then, iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inverse block-free sparse variable decibel leaf-based inference algorithm, and determining actual error values of the posterior distribution after updating the parameters to be estimated. And stopping iteration if the actual error value meets the preset error value, and obtaining the channel estimation value.
The update rules for each parameter to be estimated are given in detail below.
For update q (w):
lnq(w)∝<lnG(y,w,z,σ)+lnP(w|γ,Λ)>q(γ)q(α)q(σ)∝-<σ)q(σ)wH(Γz+ΦH(y-Φz))-wH(Γ<σ>q(σ)+Ω-1)w, (19)
In the method, in the process of the invention,
As shown in formula (19), q (w) satisfies complex Gaussian distribution, i.e., q (w) to CN (w|mu, Σ), the mean and variance of which are expressed as
μ＝(σ>q(σ)∑(Γz+ΦH(y-Φz))， (20)
∑＝(Γ<σ>q(σ)+Ω-1)-1， (21)
Update q (γ):
Wherein < w l,q|2＞＝|μl,q|2+∑l,q. As shown in formula (22), γ l satisfies the generalized inverse gaussian (Generalized Inverse Gaussian, GIG) distribution. Thus, the mean value of gamma l is calculated as
In the method, in the process of the invention,K v (·) represents a second order modified Bessel function.
Update q (α):
from the formula (24), the term related to α l,q is
Then it is known that the updated expression for α l,q is
Wherein χ is E (0, 1). Notably, here use is made ofApproximately as the mean of the q (α l,q) of the distribution.
Updates q (ζ), q (λ) and q (σ):
the above expression shows that q (ζ l), q (λ) and q (σ) satisfy Gamma distribution, and thus their average value is expressed as
After obtaining the updated expressions of all the unknown parameters in Θ, the present embodiment also proposes to summarize the entire channel estimation procedure using algorithm 1, which is denoted as a super-large-scale MIMO sparse channel estimation method based on Inverse Block-free Block-sparse Variational Bayesian Inference, IB-VBI. The convergence of the IB-VBI algorithm proposed by the present invention is described below. The IB-VBI algorithm proposed in this embodiment is derived under a variational Bayesian inference framework, so that the convergence of the IB-VBI algorithm can be ensured by maximizing the optimized expression (18) of R-ELBO based on the local convergence of the variational Bayesian inference. Algorithm 1 is as follows:
Input: The maximum iteration number Tmax, tolerates the error epsilon.
Initializing:
And (3) circulation:
Updating mu according to formulas (4-20) and (4-21) (t+1)and∑(t+1)
Update/> according to equation (4-23)
Update/> according to equation (4-26)
Update/> according to equation (4-30)
Updating according to (4-31)
Updating according to (4-32)
Until: t > T max or
And (3) outputting: w est = μ
It can be seen that, in this embodiment, a channel sparse prior model is constructed based on a two-layer block sparse structure feature of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, where the channel sparse prior model considers a near-field effect when being established and is modeled based on a spherical wave hypothesis. And then, based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal. And finally, based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value. In the embodiment, a near field effect in the ultra-large-scale MIMO channel is considered, and the invention provides a novel double-layer block sparse prior model for representing a double-layer block sparse structure of an uplink ultra-large-scale MIMO sparse channel in a time delay-polarization domain; aiming at the high computational complexity caused by matrix inversion under the current sparse Bayesian framework, the embodiment provides a high-efficiency sparse Bayesian learning algorithm without matrix inversion (Inverse), namely IB-VBI, so as to realize high-precision and low-complexity channel estimation.
Exemplary System
Based on the above embodiment, the present invention further provides a sparse channel estimation system for a super-large-scale MIMO system, as shown in fig. 2, the system includes: model building block 10, channel receiving block 20 and channel estimating block 30. Specifically, the model building module 10 is configured to build a channel sparse prior model based on a two-layer block sparse structure feature of a channel in a delay and polarization domain in an uplink ultra-large-scale MIMO communication system, where the channel sparse prior model considers a near field effect when building and performs modeling based on a spherical wave hypothesis. The channel receiving module 20 is configured to control a user terminal to send pilot symbols to a base station terminal based on the channel sparse prior model, and receive sparse channel vectors according to the pilot symbols received by the base station terminal. The channel estimation module 30 is configured to iteratively update parameters to be estimated in the channel sparse prior model according to the sparse channel vector based on an inverse block-free sparse variational bayesian inference algorithm, so as to obtain a channel estimation value.
The working principle of each module in the sparse channel estimation system for the ultra-large-scale MIMO system in this embodiment is the same as the execution process of each step in the above method embodiment, and is not described here in detail.
Based on the above embodiment, the present invention further provides a terminal device, and a schematic block diagram of the terminal device may be shown in fig. 3. The terminal device of the present embodiment may include one or more processors 100 (only one is shown in fig. 3), a memory 101, and a computer program 102 stored in the memory 101 and executable on the one or more processors 100, for example, a program for sparse channel estimation for a very large scale MIMO system. The execution of the computer program 102 by one or more processors 100 may implement various steps in an embodiment of a method for sparse channel estimation for a very large scale MIMO system. Or the one or more processors 100, when executing the computer program 102, may implement the functions of the various modules/units in an apparatus embodiment for sparse channel estimation for a very large scale MIMO system, without limitation.
In one embodiment, the Processor 100 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL Processor, DSP), application specific integrated circuits (ApplicationSpecific Integrated Circuit, ASIC), off-the-shelf Programmable gate arrays (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
In one embodiment, the memory 101 may be an internal storage unit of the electronic device, such as a hard disk or a memory of the electronic device. The memory 101 may also be an external storage device of the electronic device, such as a plug-in hard disk provided on the electronic device, a smart memory card (SMARTMEDIACARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), or the like. Further, the memory 101 may also include both an internal storage unit and an external storage device of the electronic device. The memory 101 is used to store computer programs and other programs and data required by the terminal device. The memory 101 may also be used to temporarily store data that has been output or is to be output.
It will be appreciated by persons skilled in the art that the functional block diagram shown in fig. 3 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the terminal device to which the present inventive arrangements are applied, and that a particular terminal device may include more or fewer components than shown, or may combine some of the components, or may have a different arrangement of components.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, operational database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual operation data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A sparse channel estimation method for a super-large-scale MIMO system, the method comprising:
Based on the double-layer block sparse structural characteristics of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, constructing a channel sparse prior model, wherein the channel sparse prior model considers a near field effect when being constructed and is modeled based on spherical wave hypothesis;
Based on the channel sparse prior model, controlling a user terminal to send pilot symbols to a base station terminal, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal;
Based on a non-inversion block sparse variable decibel leaf-based inference algorithm, carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector to obtain a channel estimation value;
the sparse representation of the channel under the time delay and polarization domain is as follows:
Wherein, W represents a sparse channel vector to be estimated, q=m×i represents the number of polarization domain samples, I represents the number of spatial distance samples, M represents the dimension, L represents the number of multipaths, B represents a sparse dictionary,/>Represents the kronecker product;
w has a block sparse structure similar to h, called outer layer block sparsity; w l has a block sparse structure, referred to as inner layer block sparsity;
Wherein, A sparse representation vector representing the channel under the sparse dictionary B;
The polarization domain is an angle-space distance domain, and the construction of the channel sparse prior model based on the double-layer block sparse structural characteristics of the time delay and the polarization domain of the channel in the uplink ultra-large-scale MIMO communication system comprises the following steps:
Sampling in both the angular and spatial dimensions to construct a sparse representation matrix of the channel in near field conditions, where for angular domain sampling, the sampling angle is represented as
For spatial distance sampling, the sampling distance relative to angle θ l,m is expressed as
Wherein,Where β Δ represents a low coherence threshold that guarantees a sparse channel representation matrix.
2. The sparse channel estimation method for a very large scale MIMO system of claim 1, further comprising:
constructing a system model of ultra-large-scale MIMO, wherein the system model is as follows under near-field channel sparse representation: y=Φw+n,
Wherein,Diagonal matrix representing signal vector x,/>A submatrix representing a discrete Fourier matrix, K being a pilot symbol, wherein the (K, l) th term is/>Representing additive gaussian white noise, satisfying a complex gaussian distribution CN (0, σ -1IMK), where σ represents noise accuracy.
3. The sparse channel estimation method for a very large scale MIMO system according to claim 1, wherein the sparse channel vector w to be estimated satisfies the following a priori distribution:
Wherein, As a non-negative parameter, the outer layer block sparsity of w is controlled, and when γ l =0, the variance of w l is 0, which means that the values in the block are all 0,/>Λ l is a diagonal matrix that controls the inner layer block sparsity of w l, denoted as
Λl＝diag([Λ1,1,...,Λ1,q,...,Λ1,Q]T),
Where Λ 1,q＝(α1,q+βαl,q-1+βαl,q+1)-1,αl,q is a non-negative parameter, β characterizes the correlation between parameter α l,q and its neighboring parameter { α l,q-1,αl,q+1 }.
4. The sparse channel estimation method for the ultra-large-scale MIMO system according to claim 1, wherein the iterative updating of parameters to be estimated in the channel sparse prior model according to the sparse channel vector based on the non-inversion block sparse variational bayesian inference algorithm to obtain a channel estimation value comprises:
Presetting the maximum iteration times;
iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inversion-free block sparse variable decibel leaf inference algorithm;
And stopping iteration if the current iteration number meets the maximum iteration number, and obtaining the channel estimation value.
5. The sparse channel estimation method for the ultra-large-scale MIMO system according to claim 1, wherein the iterative updating of parameters to be estimated in the channel sparse prior model according to the sparse channel vector based on the non-inversion block sparse variational bayesian inference algorithm to obtain a channel estimation value comprises:
presetting an error value of posterior distribution after updating the parameters to be estimated;
Iteratively updating posterior distribution of each parameter to be estimated in the channel sparse prior model based on an inversion-free block sparse variable decibel leaf-based inference algorithm, and determining actual error values of the posterior distribution after updating the parameter to be estimated;
and stopping iteration if the actual error value meets the preset error value, and obtaining the channel estimation value.
6. A sparse channel estimation system for a super-large-scale MIMO system, the system comprising:
The model construction module is used for constructing a channel sparse prior model based on the double-layer block sparse structural characteristics of a channel in a time delay and polarization domain in an uplink ultra-large-scale MIMO communication system, wherein the channel sparse prior model considers a near field effect when being constructed and is modeled based on spherical wave hypothesis;
the channel receiving module is used for controlling a user terminal to send pilot symbols to a base station terminal based on the channel sparse prior model, and receiving sparse channel vectors according to the pilot symbols received by the base station terminal;
The channel estimation module is used for carrying out iterative updating on parameters to be estimated in the channel sparse prior model according to the sparse channel vector based on an inverse block-free sparse-variance decibel leaf-based inference algorithm to obtain a channel estimation value;
the sparse representation of the channel under the time delay and polarization domain is as follows:
Wherein, W represents a sparse channel vector to be estimated, q=m×i represents the number of polarization domain samples, I represents the number of spatial distance samples, M represents the dimension, L represents the number of multipaths, B represents a sparse dictionary,/>Represents the kronecker product;
w has a block sparse structure similar to h, called outer layer block sparsity; w l has a block sparse structure, referred to as inner layer block sparsity;
Wherein, A sparse representation vector representing the channel under the sparse dictionary B;
The polarization domain is an angle-space distance domain, and the construction of the channel sparse prior model based on the double-layer block sparse structural characteristics of the time delay and the polarization domain of the channel in the uplink ultra-large-scale MIMO communication system comprises the following steps:
Sampling in both the angular and spatial dimensions to construct a sparse representation matrix of the channel in near field conditions, where for angular domain sampling, the sampling angle is represented as
For spatial distance sampling, the sampling distance relative to angle θ l,m is expressed as
Wherein,Where β Δ represents a low coherence threshold that guarantees a sparse channel representation matrix.
7. A terminal device, characterized in that the terminal device is applied to a receiving end or a transmitting end, and comprises a memory, a processor and a sparse channel estimation program for a super-large scale MIMO system stored in the memory and capable of running on the processor, wherein the processor implements the steps of the sparse channel estimation method for a super-large scale MIMO system according to any one of claims 1-5 when executing the sparse channel estimation program for a super-large scale MIMO system.
8. A computer readable storage medium, wherein a sparse channel estimation program for a super-MIMO system is stored on the computer readable storage medium, and when the sparse channel estimation program for a super-MIMO system is executed by a processor, the steps of the sparse channel estimation method for a super-MIMO system according to any one of claims 1-5 are implemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211694969.4A CN116032699B (en) | 2022-12-28 | 2022-12-28 | Sparse channel estimation method for ultra-large-scale MIMO system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211694969.4A CN116032699B (en) | 2022-12-28 | 2022-12-28 | Sparse channel estimation method for ultra-large-scale MIMO system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116032699A CN116032699A (en) | 2023-04-28 |
| CN116032699B true CN116032699B (en) | 2024-05-14 |
Family
ID=86070068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211694969.4A Active CN116032699B (en) | 2022-12-28 | 2022-12-28 | Sparse channel estimation method for ultra-large-scale MIMO system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116032699B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108111441A (en) * | 2018-01-12 | 2018-06-01 | 电子科技大学 | Channel estimation methods based on variational Bayesian |
| US11190377B1 (en) * | 2020-05-26 | 2021-11-30 | Wuhan University | Time-frequency block-sparse channel estimation method based on compressed sensing |
| CN114465642A (en) * | 2022-01-18 | 2022-05-10 | 清华大学 | Channel estimation method, device, electronic equipment and storage medium |
| CN115118556A (en) * | 2022-06-21 | 2022-09-27 | 厦门大学 | Sparse channel estimation method, device and medium for OFDM underwater acoustic communication system |
| CN115361257A (en) * | 2022-07-14 | 2022-11-18 | 鹏城实验室 | Uplink large-scale MIMO sparse channel estimation method and related equipment |
| CN115361258A (en) * | 2022-07-14 | 2022-11-18 | 鹏城实验室 | Large-scale MIMO sparse channel estimation method and related equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20210109598A (en) * | 2019-01-02 | 2021-09-06 | 코히어 테크놀로지스, 아이엔씨. | Distributed cooperative operation of radio cells based on sparse channel representation |
-
2022
- 2022-12-28 CN CN202211694969.4A patent/CN116032699B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108111441A (en) * | 2018-01-12 | 2018-06-01 | 电子科技大学 | Channel estimation methods based on variational Bayesian |
| US11190377B1 (en) * | 2020-05-26 | 2021-11-30 | Wuhan University | Time-frequency block-sparse channel estimation method based on compressed sensing |
| CN114465642A (en) * | 2022-01-18 | 2022-05-10 | 清华大学 | Channel estimation method, device, electronic equipment and storage medium |
| CN115118556A (en) * | 2022-06-21 | 2022-09-27 | 厦门大学 | Sparse channel estimation method, device and medium for OFDM underwater acoustic communication system |
| CN115361257A (en) * | 2022-07-14 | 2022-11-18 | 鹏城实验室 | Uplink large-scale MIMO sparse channel estimation method and related equipment |
| CN115361258A (en) * | 2022-07-14 | 2022-11-18 | 鹏城实验室 | Large-scale MIMO sparse channel estimation method and related equipment |
Non-Patent Citations (3)
| Title |
|---|
| 基于稀疏MIMO阵列的快速近场二维成像算法;王慧;吴世有;方广有;;电子测量技术;20200508(第09期);全文 * |
| 面向5G的大规模MIMO技术综述;张雷;代红;;电讯技术;20170528(第05期);全文 * |
| 频分双工大规模MIMO系统时变信道估计;杜福德;谢威;夏晓晨;;信号处理;20200325(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116032699A (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wei et al. | Learned conjugate gradient descent network for massive MIMO detection | |
| Jin et al. | Channel estimation for mmWave massive MIMO with convolutional blind denoising network | |
| Wang et al. | Super-resolution channel estimation for arbitrary arrays in hybrid millimeter-wave massive MIMO systems | |
| Zhang et al. | Computationally efficient DOA estimation for monostatic MIMO radar based on covariance matrix reconstruction | |
| US20220393730A1 (en) | Method and system for acquiring massive mimo beam domain statistical channel information | |
| Ye et al. | Sparse channel estimation of MIMO-OFDM systems with unconstrained smoothed l 0-norm-regularized least squares compressed sensing | |
| CN109039403B (en) | Downlink channel estimation method based on redundant dictionary in large-scale MIMO system | |
| Yassine et al. | mpNet: variable depth unfolded neural network for massive MIMO channel estimation | |
| Mai et al. | A novel blind mmWave channel estimation algorithm based on ML-ELM | |
| Hua et al. | On-device federated learning via second-order optimization with over-the-air computation | |
| Kim et al. | Massive data generation for deep learning-aided wireless systems using meta learning and generative adversarial network | |
| CN116032699B (en) | Sparse channel estimation method for ultra-large-scale MIMO system | |
| Guo et al. | Robust deep learning for uplink channel estimation in cellular network under inter-cell interference | |
| Couillet et al. | Robust M-estimation for Array Processing: a random matrix approach | |
| Zou et al. | Analog beamforming for millimeter-wave mimo systems via stochastic optimization | |
| Su et al. | Channel estimation in massive MIMO systems using a modified Bayes-GMM method | |
| CN115361257A (en) | Uplink large-scale MIMO sparse channel estimation method and related equipment | |
| Pellaco et al. | A matrix-inverse-free implementation of the MU-MIMO WMMSE beamforming algorithm | |
| WO2020087339A1 (en) | System and method for adaptive filtering | |
| CN114629533B (en) | Information geometry method and system for large-scale MIMO channel estimation | |
| Zhu et al. | Active user detection and channel estimation via fast admm | |
| Zhao et al. | Two-timescale joint UL/DL dictionary learning and channel estimation in massive MIMO systems | |
| Zhao et al. | Joint UL/DL Dictionary Learning and Channel Estimation via Two-Timescale Optimization in Massive MIMO Systems | |
| Hu et al. | Multiple-compression-rate CSI feedback for mm-wave massive MIMO based on federated learning | |
| CN113810096B (en) | Uplink transmission method and system for large-scale MIMO satellite mobile communication |
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 |