CN104052691B - MIMO-OFDM system channel estimation method based on compressed sensing - Google Patents
MIMO-OFDM system channel estimation method based on compressed sensing Download PDFInfo
- Publication number
- CN104052691B CN104052691B CN201410312859.6A CN201410312859A CN104052691B CN 104052691 B CN104052691 B CN 104052691B CN 201410312859 A CN201410312859 A CN 201410312859A CN 104052691 B CN104052691 B CN 104052691B
- Authority
- CN
- China
- Prior art keywords
- path
- channel
- vector
- angle
- estimated
- 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 34
- 239000013598 vector Substances 0.000 claims abstract description 99
- 230000004044 response Effects 0.000 claims abstract description 60
- 239000011159 matrix material Substances 0.000 claims description 59
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 2
- 239000000969 carrier Substances 0.000 abstract description 3
- 238000005070 sampling Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000005562 fading Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 102000002274 Matrix Metalloproteinases Human genes 0.000 description 1
- 108010000684 Matrix Metalloproteinases Proteins 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Landscapes
- Radio Transmission System (AREA)
Abstract
The invention provides an MIMO-OFDM system channel estimation method based on compressed sensing. The MIMO-OFDM system channel estimation method based on compressed sensing is mainly applied to channel estimation when a receiving terminal is provided with a two-dimensional antenna array. According to the MIMO-OFDM system channel estimation method based on compressed sensing, the time delay, incidence angle and gain of each path of a space channel are estimated in sequence, and channel estimation accuracy can be improved effectively. The MIMO-OFDM system channel estimation method based on compressed sensing comprises the following steps that 1, an initially-estimated value of a channel frequency domain response vector of each pilot frequency sub-carrier is obtained according to the least square criterion; 2, by means of the sparsity of the channel frequency domain response vectors in a time delay domain, the time delay of each path of the channel and an estimated value of a channel time domain response vector of each path of the channel are estimated on the basis of the compressed sensing theory; 3, by means of the sparsity of the channel time domain response vectors in a two-dimensional angle domain, the incidence angle of each path of the channel is estimated on the basis of the compressed sensing theory; 4, the gain coefficient of each path of the channel is estimated according to the least square criterion; 5, estimated values of channel frequency domain responses of all the sub-carriers and antennas are obtained.
Description
Technical Field
The invention relates to a MIMO-OFDM wireless communication system with a two-dimensional antenna array at a receiving end, in particular to a problem of channel estimation of the MIMO-OFDM system based on a compressed sensing theory.
Background
Multiple antennas are configured at a transmitting end and a receiving end by a multiple-input multiple-output (MIMO) technology, and by combining with a multiplexing technology or a diversity technology and the like, multipath characteristics of a scattering channel are fully utilized, multiple independent and parallel data streams are transmitted in space, the capacity and the link reliability of a wireless communication system are improved in multiples under the condition of not increasing the system bandwidth, and the transmission rate of the system is improved. Therefore, the MIMO technology receives a great deal of attention, and is considered to be one of the key technologies inevitably adopted by the modern wireless communication and the future wireless communication. To meet the increasing user data demand, the number of antenna configurations is increasing, such as large-scale MIMO systems, and a two-dimensional antenna array will be a reasonable choice for the antenna configuration of the system.
Orthogonal Frequency Division Multiplexing (OFDM) is an orthogonal multi-carrier modulation technique with efficient spectrum utilization. With the development of wireless communication, in order to meet the increasing demand of people for high-rate services, the system bandwidth is continuously increased, the frequency selectivity of the channel is more prominent, and the conventional equalization technology not only has high complexity, but also is difficult to completely eliminate the intersymbol interference caused by the multipath fading channel. The OFDM system converts a broadband frequency selective fading channel into a series of narrow-band flat fading channels, effectively solves the problem of intersymbol interference, and has unique advantages in the aspects of realizing high-speed data transmission and the like. Therefore, MIMO and OFDM based system architectures are in force.
Compressed sensing, also called compressed sampling, is a new sampling theory, and by developing the sparseness of signals, discrete samples of the signals are obtained by random sampling under the condition that the sampling rate is much less than the Nyquist sampling rate, and then the signals are perfectly reconstructed by a nonlinear reconstruction algorithm. In an actual scattering environment, for a broadband signal, a wireless channel is often composed of several main paths, so that the wireless channel can be regarded as a time delay domain sparse channel, and the channel response can be estimated with a small number of pilots by using a compressed sensing theory. Meanwhile, the number of incident angles from the user to the base station end antenna array is limited, so that the sparsity of the channel in an angle domain is developed, and the estimation of the channel can be obtained by using a compressed sensing theory.
Disclosure of Invention
The technical problem is as follows: the invention provides a channel estimation method of a MIMO-OFDM system based on compressed sensing, which effectively reduces the number of pilot frequencies and has higher estimation resolution.
The technical scheme is as follows: the invention relates to a MIMO-OFDM system channel estimation method based on compressed sensing, which comprises the following steps:
a) obtaining an initial estimation value of a channel frequency domain response vector on each pilot frequency subcarrier by using a least square criterion;
b) estimating the delay of each path of the channel and the estimated value of the channel time domain response vector of each path based on a compressed sensing theory by utilizing the sparsity of the initial estimated value of the channel frequency domain response vector obtained in the step a) in a time delay domain;
c) estimating an incident angle of each path of the channel based on a compressed sensing theory by utilizing the sparsity of the estimated value of the channel time domain response vector estimated in the step b) in a two-dimensional angle domain, wherein the incident angle comprises a vertical pitch angle and a horizontal azimuth angle;
d) estimating a gain coefficient of each path of the channel by using the initial estimation value of the channel frequency domain response vector obtained in the step a), the delay of each path estimated in the step b) and the incidence angle of each path estimated in the step c) and adopting a least square criterion;
e) and substituting the delay of each path estimated in the step b), the incident angle of each path estimated in the step c) and the gain coefficient of each path estimated in the step d) into a multipath channel model of channel frequency domain response to obtain the estimated values of the channel frequency domain response on all subcarriers and all antennas.
In the preferred scheme of the method of the invention, the specific flow of the step b) is as follows:
first, the channel path is delayed from 0 to the maximumThe large path time delay is uniformly discretized to obtain the minimum time delay interval delta tau and the number of the discretized paths NτAnd expressing the channel frequency domain response vector on each subcarrier as the product of the base vector and the channel time domain matrix by utilizing the Fourier transform relation of the channel frequency domain response and the channel time domain response.
Secondly, arranging the initial estimated values of the channel frequency domain response vectors on all the pilot frequency subcarriers into a matrix in sequence, and recording the matrix as a channel frequency domain matrix; arranging all the basis vectors into a matrix according to the same sequence, and recording the matrix as a projection matrix; thereby expressing the channel frequency domain matrix estimation value as:
wherein,representing the estimated value of the channel frequency domain matrix, wherein XI is a projection matrix, A is a channel time domain matrix, and W is a noise matrix;
finally, solving a channel time domain matrix by adopting a compressed sensing theory according to the formula (12), wherein the solved channel time domain matrix is N in totalτAnd searching for non-zero row vectors, wherein the number L of the non-zero row vectors is the total number of the estimated channel paths, the L-th non-zero row vector is the estimated value of the channel time domain response vector of the L-th path, the path delay corresponding to the L-th path is (m-1) delta tau, wherein L is the serial number of the non-zero row vector, and m is the number m of the channel time domain matrix rows where the L-th non-zero row vector is located.
In the preferred scheme of the method of the invention, the specific flow of the step c) is as follows:
firstly, uniformly discretizing the vertical pitch angle to obtain a minimum angle interval delta theta, wherein the number of the discretized angles is Nθ(ii) a The horizontal azimuth angle is discretized evenly to obtain the minimum angle interval delta phi, and the number of the discretized angles is NφAnd using a two-dimensional Fourier transform relation between the channel time domain response and the path incidence angle to express the channel time domain response vector estimation value of each path as:
wherein,is the estimated value of the time domain response vector of the channel of the ith path, P is the projection matrix, gammalIs the path complex gain vector of the l-th path, w is the noise vector;
then the incident angle of each path is calculated according to the following method: for the ith path, using equation (18), solving by using a compressive sensing theory to obtain an estimated value of a path complex gain vector, and calculating an incident angle of the ith path according to the position of the maximum value of an element in the path complex gain vector, wherein the incident angle comprises a vertical pitch angle and a horizontal azimuth angle.
In the above preferred embodiment of the method of the present invention, in step c), the arrangement of elements in each row of the projection matrix and the path complex gain vector γ are determined by a method of the present inventionlThe arrangement modes of the middle elements are the same and are obtained by sequencing according to the sequence of the kronecker product of the paths corresponding to the vertical pitch angle and the horizontal azimuth angle;
the method for calculating the incident angle of the ith path comprises the following steps: according to the position s of the maximum value of the element in the path complex gain vector, the vertical pitch angle of the ith path is obtainedHorizontal azimuth angle ofWherein n isφ＝mod(s,Nφ)，nθ＝(s-nφ)/Nφ+1, mod (x, y) denotes x modulo y.
In another preferred embodiment of the method of the present invention, in step c), the element arrangement and path complex gain vector γ in the projection matrixlThe arrangement modes of the middle elements are the same and are the sequence of the kronecker product of the paths corresponding to the horizontal azimuth angle and the vertical pitch angle;
the method for calculating the incident angle of the ith path comprises the following steps: according to the position s of the maximum value of the element in the path complex gain vector, the vertical pitch angle of the ith path is obtainedHorizontal azimuth angle ofWherein n isθ＝mod(s,Nθ)，nφ＝(s-nθ)/Nθ+1。
Has the advantages that: compared with the prior art, the invention has the following advantages:
1. the channel estimation method further estimates parameters (including the delay, the incidence angle and the gain coefficient of a path) for characterizing the channel characteristics on the basis of least square estimation, and the channel estimation result is remarkably improved compared with the least square estimation.
2. In a conventional channel estimation scheme of the MIMO-OFDM system, such as least square estimation, the number of pilots in one OFDM symbol must not be less than the length of the time domain of the channel, so as to ensure that the distribution interval of the pilots in the frequency domain is less than the coherent frequency interval, otherwise aliasing of the channel estimation result in the time domain may be caused. The invention utilizes the sparsity of the channel in the time delay domain and combines the theory of compressed sensing, and the number of the pilot frequency can be smaller than the time domain length of the channel, thereby reducing the overhead of the pilot frequency.
3. Traditional subspace algorithm-based incidence angle estimation schemes, such as the MUSIC algorithm, require a large number of sample estimation covariance matrices. When the number of samples is limited, or the samples have correlation, the estimation performance is poor, and the number of incident angles that can be resolved is smaller than the number of antennas. The invention utilizes the sparsity of the channel in a two-dimensional angle domain and combines the theory of compressed sensing to estimate the incidence angle, thereby not only effectively reducing the number of observation samples, but also being not influenced by the sample correlation and the number of antennas, and having wider application.
The invention adopts a method based on compressed sensing to obtain the estimation of the channel parameters, fully utilizes the sparsity of the channel in a time delay domain and a two-dimensional angle domain, has obviously improved estimation result compared with least square estimation, reduces the pilot frequency overhead, and has wide application prospect for broadband and large-scale MIMO systems.
Drawings
Fig. 1 is a schematic diagram of a receiving-end antenna array configuration.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the following description and the accompanying drawings are used for clearly and completely describing the technical solution of the present invention, and it should be understood that these examples are only used for illustrating the specific implementation manner of the technical solution of the present invention, and are not used for limiting the scope of the present invention. Various equivalent modifications and alterations of this invention will occur to those skilled in the art after reading this disclosure, and it is intended to cover such alternatives and modifications as fall within the scope of the invention as defined by the appended claims.
Example 1:
1. system model
The scheme considers a MIMO-OFDM system. FIG. 1 is a schematic diagram of a receiving-end antenna configuration, not generally, using a two-dimensional uniform area array of antennas, where N is horizontalhRoot antenna, with N in vertical directionvAnd the distance between adjacent antennas is d. Time domain multipleThe path channel response matrix is
Wherein (tau) represents Kronecker pulse function, L is the path number of multipath channel, taul、βlRespectively the delay and gain factor of the ith path.Is the time domain channel matrix of the user to antenna array. If the user is far enough away from the antenna array, i.e. the incident wave is a plane wave, the vertical and horizontal steering vectors of the first path are
Wherein, thetal∈[0,π/2],φl∈[0,π]The vertical pitch angle and the horizontal azimuth angle of the first path are respectively, and lambda is the wavelength of incident waves. The multipath channel model for obtaining the channel frequency domain response by utilizing Fourier transform is as follows:
wherein, tsIs a sampling period, NcIs the number of OFDM symbol sub-carriers, HkThe channel frequency domain matrix of the k subcarrier has the elements of the v row and h column representing the channel frequency domain response of the user to the v antenna in the vertical direction and the h antenna in the horizontal direction on the k subcarrier, and is expressed as
Wherein, the channel time domain response of the user to the v-th antenna in the vertical direction and the h-th antenna in the horizontal direction on the l path is
Assuming that the number of users is 1 and single antenna is provided, the system model is
Yk＝HkXk+Zk(5)
Wherein,Xkrespectively representing received data and transmitted data on the k-th sub-carrier, ZkIs an additive white gaussian noise matrix. Vectorizing the left and right matrixes with equal signs of the formula (5) to obtain
yk＝hkXk+zk(6)
Wherein, yk＝vec(Yk)，hk＝vec(Hk)，zk＝vec(Zk) Vec (·) represents a matrix vectorization operator.
2. Initial estimation of channel
Suppose that the user sends a pilot symbol of XkThe receiving end receives the pilot data as ykAccording to the system model formula (6), the initial estimation value of the channel frequency domain response vector on each pilot frequency subcarrier is obtained by using the least square criterion
It is assumed that the pilots are uniformly distributed in the frequency domain, where the pilot interval is kpThe number of pilot frequencies is NpThen the pilot set is {0, k }p,…，(Np-1)kp}. This channel estimation method is simple but has limited accuracy. Then, the scheme estimates the delay, the incident angle and the gain coefficient of each path of the multipath channel in turn based on the initial estimation result, thereby obtaining a more accurate channel estimation result.
3. Delay estimation
Let the maximum delay of the channel path be taumaxBecause the delay of each channel path is 0-taumaxAre randomly distributed, and can delay the channel path from 0 to taumaxUniformly discretizing the two paths, wherein the discrete interval is delta tau, and the number of the discretized paths isWherein,expressed as the result of rounding up a, the frequency domain response of the channel obtained from equation (3) is expressed as
Wherein the delay set is selected as
＝{0,Δτ,2Δτ,…,(Nτ-1)Δτ} (9)
The channel frequency domain responses of all antennas on the kth subcarrier are collected together and recorded as a channel frequency domain response vector, which is expressed as
hk＝Auk(10)
Wherein the channel frequency domain vector of the k sub-carrierBase vectorChannel time domain matrix
According to equations (7) and (10), the initial estimate of the channel frequency domain response vector is expressed as
Arranging initial estimated values of channel frequency domain response vectors on all pilot frequency subcarriers into a channel frequency domain matrix in sequenceArranging all basis vectors in the same order as a projection matrix xi results in an estimated value of the channel frequency domain matrix expressed as:
wherein, since the majority element of the channel time domain matrix A is 0 and is a sparse matrix, the formula (12) is the sparse representation of the initial estimation value of the channel frequency domain response vector in the time delay domain, and the channel time domain matrix A is solved based on the compressed sensing theory
Wherein μ is regularizationParameters, the selection of which affects the accuracy of signal reconstruction, | · | | luminance1、||·||2Representing the 1-norm and 2-norm of the matrix, respectively. In the solved channel time domain matrixIn the method, if a certain row is not a zero vector, a corresponding delay path exists, such as a channel time domain matrixIs the l non-zero row vector, the estimated value of the time domain response vector of the l path is
The corresponding path delay is estimated asHypothetical channel time domain matrixMedium non-zero row vectorThen the total number of estimated channel paths isThe estimated corresponding set of delays is
4. Incident angle estimation
The vertical pitch angles are randomly distributed between 0 pi and 0.5 pi, so that the vertical pitch angles are uniformly discretized, the minimum angle interval is delta theta, and the number of the discretized angles isWherein,expressed as a result of rounding down a; because the horizontal azimuth angles are randomly distributed between 0 and pi, the horizontal azimuth angles are uniformly discretized, the minimum angle interval is delta phi, and the number of the discretized angles isThe channel time domain response can be represented as (4) by
Wherein the angle set theta and phi are selected as
From the estimation result of equation (14) and equation (16), the estimated value of the estimated time domain response vector of the ith path is expressed as
Wherein the path complex gain vectorw is the noise vector and P is the projection matrix, expressed as
Wherein,φs＝sΔφ，θqq Δ θ. Due to the path complex gain vector ylOnly one element in the channel is not 0 and is a sparse vector, so the formula (18) is sparse representation of the estimated value of the channel time domain response vector in a two-dimensional angle domain, and a path complex gain vector gamma is solved based on a compressed sensing theorylIs composed of
Estimated path complex gain vectorThe position of the maximum of the middle element reflects the size of the incident angle. According to the vector gammalThe arrangement mode of the medium elements can be calculated to obtain a vertical pitch angle and a horizontal azimuth angle. If the estimated path complex gain vectorThe position of the maximum of the middle element is s, i.e.Then
(1) If path complex gain vector gammalThe arrangement mode of the medium elements is the sequence of the kronecker product of the paths corresponding to the vertical pitch angle and the horizontal azimuth angle, namely, firstly, for each vertical pitch angle, the path complex gains are arranged according to the sequence from the small to the big of the horizontal azimuth angle to obtain a path complex gain with the length of NφThe vector of (a); then, the lengths N are arranged in the order of the vertical pitch angle from small to largeφObtaining a vector of length NθNφIs expressed as
The vertical pitch angle and the horizontal azimuth angle corresponding to the ith path are estimated as
Wherein n isφ＝mod(s,Nφ)，nθ＝(s-nφ)/Nφ+1, mod (x, y) denotes x modulo y.
(2) If path complex gain vector gammalThe arrangement mode of the medium elements is the sequence of the kronecker product of the paths corresponding to the horizontal azimuth angle and the vertical pitch angle, namely, firstly, for each horizontal azimuth angle, the path complex gains are arranged according to the sequence of the vertical pitch angle from small to large to obtain a path complex gain with the length of NθThe vector of (a); then, the lengths N are arranged in the order of the horizontal azimuth angles from small to largeθObtaining a vector of length NθNφIs expressed as
The vertical pitch angle and the horizontal azimuth angle corresponding to the ith path are estimated as
Wherein n isθ＝mod(s,Nθ)，nφ＝(s-nθ)/Nθ+1。
Estimate out in turnThe set of incidence angles (vertical pitch and horizontal azimuth) of the swath path is
5. Gain coefficient estimation
Based on the initial estimation value of the channel frequency domain response, the estimation value of the delay of each path, and the estimation value of the incident angle of each path, equations (3) and (4) can be used to obtain
Wherein,Wv,h,krespectively the initial estimation value of the channel frequency domain response of the user to the h antenna in the horizontal direction and the v antenna in the vertical direction on the k subcarrier and the noise,respectively the estimated delay, azimuth and elevation of the ith path, βlIs the gain factor of the l-th path. Arranging the initial estimated values of the channel frequency domain response on all the pilot frequency sub-carriers into a column vector, and obtaining the channel frequency domain response according to the formula (22)
Wherein,from the path delays and incident angles obtained by equations (15) and (21), the matrix E is constructed as follows
Wherein,using the least square criterion to obtain a channel gain coefficient of
6. Channel estimation
And substituting the estimated path delay, the estimated incidence angle and the estimated gain coefficient into a multipath channel model (2) of channel frequency domain response, and obtaining the estimated values of the channel frequency domain response on all subcarriers and antennas, wherein the estimated values are as follows:
example 2:
according to the compressed sensing theory, the equations (13) and (20) can be obtained, and the optimization algorithm can be used for solving. This embodiment provides an OMP algorithm solving equation (13) and equation (20), which specifically include:
a. initialization residualWherein k ispIs a pilot interval, NpFor the number of pilots,and the initial estimation value of the channel frequency domain response of the kth antenna and the vth antenna in the horizontal direction in the kth subcarrier is from the user to the vth antenna in the vertical direction. Initializing a set of delay indicesThe iteration number t is 1. Let Ψ(0)Is an empty matrix. Assume that the number of multipaths L has already beenAnd (5) knowing.
b. In the t iteration, the position index of the t delay path is found by solving the following formula:
wherein,is the jth column of the projection matrix xi, denoted as
Wherein,tsis the sampling interval.
c. Saving the estimation result of the t iteration:
d. solving channel time domain response vector of path with finished delay estimation
e. Updating the residual
f. Let t be t + 1. If t is less than or equal to L, returning to the step b for next iteration; otherwise, the process continues.
g. The steps b to f search out L paths, wherein the estimated path delay position index set is Λ(L)For any l ∈Λ(L)Then there is a path with a delay ofCorresponding path gain ofThe estimated value of the channel time domain response vector of the first path from the user to all the antennas is
The set of delay estimates is ordered as
h. For the estimated L paths, searching corresponding incident angles in sequence, wherein the incident angle index value of the ith path is
Wherein p isjIs the j-th column of the projection matrix P. The projection matrix P is given by
Wherein,θqd is the minimum antenna spacing of the uniform antenna area array, and λ is the carrier wavelength. According to the sorting mode of row vector elements in the projection matrix P, the corresponding incident angle can be obtained, wherein the arrangement mode of each row element in the projection matrix P and the path complex gain vector gamma of the previous textlThe arrangement of the medium elements is the same, so
(1) If path complex gain vector gammalThe arrangement mode of the medium elements is the sequence of the kronecker product of the paths corresponding to the vertical pitch angle and the horizontal azimuth angle, namely, firstly, for each vertical pitch angle, the path complex gains are arranged according to the sequence from the small to the big of the horizontal azimuth angle to obtain a path complex gain with the length of NφThe vector of (a); then, the lengths N are arranged in the order of the vertical pitch angle from small to largeφObtaining a vector of length NθNφIs expressed as
The vertical pitch angle and the horizontal azimuth angle corresponding to the ith path are estimated as
Wherein n isφ＝mod(s,Nφ)，nθ＝(s-nφ)/Nφ+1, mod (x, y) denotes x modulo y.
(2) If the arrangement of the elements in the path complex gain vector γ l is the order of the kronecker product of the paths corresponding to the horizontal azimuth angle and the vertical pitch angle, that is, first, for each horizontal azimuth angle, the path complex gains are arranged in the order of the vertical pitch angle from small to large, and a length N is obtainedθThe vector of (a); then, the lengths N are arranged in the order of the horizontal azimuth angles from small to largeθObtaining a vector of length NθNφVector of (2)Is shown as
The vertical pitch angle and the horizontal azimuth angle corresponding to the ith path are estimated as
Wherein n isθ＝mod(s,Nθ)，nφ＝(s-nθ)/Nθ+1。
The set of incidence angles (vertical pitch and horizontal azimuth) of the L paths is estimated sequentially as
Claims (5)
1. A MIMO-OFDM system channel estimation method based on compressed sensing is characterized in that the method comprises the following steps:
a) obtaining an initial estimation value of a channel frequency domain response vector on each pilot frequency subcarrier by using a least square criterion;
b) estimating the delay of each path of the channel and the estimated value of the channel time domain response vector of each path based on a compressed sensing theory by utilizing the sparsity of the initial estimated value of the channel frequency domain response vector obtained in the step a) in a time delay domain;
c) estimating an incident angle of each path of the channel based on a compressed sensing theory by utilizing the sparsity of the estimated value of the channel time domain response vector estimated in the step b) in a two-dimensional angle domain, wherein the incident angle comprises a vertical pitch angle and a horizontal azimuth angle;
d) estimating a gain coefficient of each path of the channel by using the initial estimation value of the channel frequency domain response vector obtained in the step a), the delay of each path estimated in the step b) and the incidence angle of each path estimated in the step c) and adopting a least square criterion;
e) and substituting the delay of each path estimated in the step b), the incident angle of each path estimated in the step c) and the gain coefficient of each path estimated in the step d) into a multipath channel model of the channel frequency domain response to obtain the estimated values of the channel frequency domain responses on all subcarriers and antennas.
2. The method for estimating the channel of the MIMO-OFDM system according to claim 1, wherein the specific process of step b) is as follows:
firstly, the channel path delay is uniformly discretized from 0 to the maximum path delay to obtain the minimum delay interval of △ tau and the number of discretized paths of NτExpressing the channel frequency domain response vector on each subcarrier as the product of a base vector and a channel time domain matrix by utilizing the Fourier transform relation of the channel frequency domain response and the channel time domain response;
secondly, arranging the initial estimated values of the channel frequency domain response vectors on all the pilot frequency subcarriers into a matrix in sequence, and recording the matrix as a channel frequency domain matrix; arranging all the basis vectors into a matrix according to the same sequence, and recording the matrix as a projection matrix; thereby expressing the channel frequency domain matrix estimation value as:
wherein,representing the estimated value of the channel frequency domain matrix, wherein XI is a projection matrix, A is a channel time domain matrix, and W is a noise matrix;
finally, solving a channel time domain matrix by adopting a compressed sensing theory according to the formula (12), wherein the solved channel time domain matrix is N in totalτLine, searching non-zero line vector, the number L of said non-zero line vector is the estimated channel path total number, the secondlA non-zero row vector is the firstlAn estimate of the channel time domain response vector for each path, e.glThe path delay corresponding to each path is (m-1) △ tau, where,lis the sequence number of a non-zero row vector, m is the numberlAnd the row number m of the channel time domain matrix where the non-zero row vector is located.
3. The method for estimating the channel of the MIMO-OFDM system according to claim 1 or 2, wherein the specific procedure of the step c) is as follows:
firstly, the vertical pitch angle is uniformly discretized to obtain △ theta minimum angle intervals, and the number of the discretized angles is NθUniformly discretizing the horizontal azimuth to obtain a minimum angle interval of △ phi, wherein the number of the discretized angles is NφAnd using a two-dimensional Fourier transform relation between the channel time domain response and the path incidence angle to express the channel time domain response vector estimation value of each path as:
wherein,is as followslChannel time domain response vector estimation of each path, P is projection matrix, gamma l Is as followslA path complex gain vector of each path, w being a noise vector;
then the incident angle of each path is calculated according to the following method: for the firstlThe path is obtained by solving the path by using a compressed sensing theory according to an equation (18)The estimated value of the path complex gain vector is calculated according to the position of the maximum value of the element in the path complex gain vectorlThe angle of incidence for each path, including vertical pitch and horizontal azimuth.
4. The method as claimed in claim 3, wherein in step c), the element arrangement and path complex gain vector γ of each row in the projection matrix P is determined according to the channel estimation method of the MIMO-OFDM system based on compressed sensing l The medium elements are arranged in the same way and are obtained by sequencing according to the kronecker product sequence of the paths corresponding to the vertical pitch angle and the horizontal azimuth angle, namely, firstly, for each vertical pitch angle, the path complex gains are arranged according to the sequence of the horizontal azimuth angle from small to large to obtain a path complex gain with the length of NφThe vector of (a); then, the lengths N are arranged in the order of the vertical pitch angle from small to largeφObtaining a vector of length NθNφIs represented as:
calculate the firstlThe method of the incident angle of each path is as follows: according to the position s of the maximum value of the element in the path complex gain vector, the first one is obtainedlVertical pitch angle of each pathHorizontal azimuth angle ofWherein n isφ＝mod(s,Nφ)，nθ＝(s-nφ)/Nφ+1, mod (x, y) denotes x modulo y.
5. The method as claimed in claim 3, wherein in step c), each row of elements in the projection matrix P is arranged as a row elementColumn wise and path complex gain vector gamma l The medium elements are arranged in the same mode and are obtained by sequencing according to the sequence of the kronecker products of the paths corresponding to the horizontal azimuth angle and the vertical pitch angle, namely, firstly, for each horizontal azimuth angle, the path complex gains are arranged according to the sequence of the vertical pitch angle from small to large to obtain a path complex gain with the length of NθThe vector of (a); then, the lengths N are arranged in the order of the horizontal azimuth angles from small to largeθObtaining a vector of length NθNφIs expressed as
Calculate the firstlThe method of the incident angle of each path is as follows: according to the position s of the maximum value of the element in the path complex gain vector, the first one is obtainedlVertical pitch angle of each pathHorizontal azimuth angle ofWherein n isθ＝mod(s,Nθ)，nφ＝(s-nθ)/Nθ+1。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410312859.6A CN104052691B (en) | 2014-07-02 | 2014-07-02 | MIMO-OFDM system channel estimation method based on compressed sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410312859.6A CN104052691B (en) | 2014-07-02 | 2014-07-02 | MIMO-OFDM system channel estimation method based on compressed sensing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104052691A CN104052691A (en) | 2014-09-17 |
| CN104052691B true CN104052691B (en) | 2017-02-15 |
Family
ID=51505071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410312859.6A Active CN104052691B (en) | 2014-07-02 | 2014-07-02 | MIMO-OFDM system channel estimation method based on compressed sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104052691B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104539335B (en) * | 2014-12-24 | 2017-11-03 | 无锡北邮感知技术产业研究院有限公司 | A kind of limited feedback method and device of extensive antenna system |
| CN104954310B (en) * | 2015-04-24 | 2018-12-11 | 清华大学 | Impulse noise removing method and device |
| CN104869086B (en) * | 2015-05-27 | 2017-11-14 | 东南大学 | MIMO ofdm communication system down channels method of estimation, device based on two dimensional compaction perception |
| CN105610477B (en) * | 2016-01-28 | 2018-06-19 | 西南交通大学 | Multiple-input multiple-output system enhancing method of signal multiplexing based on compressed sensing |
| CN107276924A (en) * | 2016-04-08 | 2017-10-20 | 北京信威通信技术股份有限公司 | The channel estimation methods and device of extensive multiple-input and multiple-output |
| CN107872416A (en) * | 2016-09-27 | 2018-04-03 | 中国移动通信有限公司研究院 | The processing method and baseband processing unit of baseband signal |
| CN106453190A (en) * | 2016-10-10 | 2017-02-22 | 东南大学 | Multi-user sub-carrier index modulation orthogonal frequency-division multiplexing (SIM-OFDM) transmission method |
| CN106487432B (en) * | 2016-10-28 | 2020-01-03 | 天津大学 | Massive MIMO (multiple input multiple output) safe transmission method based on compressed sensing |
| CN108242943B (en) * | 2016-12-23 | 2021-04-13 | 上海诺基亚贝尔股份有限公司 | Method and device for precoding in communication |
| US10014026B1 (en) * | 2017-06-20 | 2018-07-03 | Seagate Technology Llc | Head delay calibration and tracking in MSMR systems |
| KR102266761B1 (en) | 2017-07-14 | 2021-06-21 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Beamforming method and device |
| CN107465636B (en) * | 2017-08-21 | 2020-05-01 | 清华大学 | Channel estimation method of millimeter wave large-scale array space-frequency double-broadband system |
| CN108881076B (en) * | 2018-06-30 | 2021-05-04 | 宜春学院 | MIMO-FBMC/OQAM system channel estimation method based on compressed sensing |
| CN109005133B (en) * | 2018-07-12 | 2021-04-16 | 南京邮电大学 | Double-sparse multi-path channel model and channel estimation method based on model |
| CN111385006A (en) * | 2018-12-27 | 2020-07-07 | 财团法人工业技术研究院 | Millimeter wave channel estimation method |
| EP4147508A4 (en) * | 2020-05-08 | 2024-01-31 | QUALCOMM Incorporated | Signaling design for type ii csi-rs spatial domain and frequency domain basis configuration |
| CN111988046B (en) * | 2020-08-19 | 2021-11-19 | 中国电子科技集团公司第三十六研究所 | Nonlinear system correction method and device and electronic equipment |
| CN114039639B (en) * | 2021-11-04 | 2023-04-25 | 电子科技大学 | Combined channel estimation and user detection method under large connection scene |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102891823A (en) * | 2012-10-17 | 2013-01-23 | 东南大学 | Channel estimation method of orthogonal frequency division multi-address mobile communication system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8774294B2 (en) * | 2010-04-27 | 2014-07-08 | Qualcomm Incorporated | Compressed sensing channel estimation in OFDM communication systems |
-
2014
- 2014-07-02 CN CN201410312859.6A patent/CN104052691B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102891823A (en) * | 2012-10-17 | 2013-01-23 | 东南大学 | Channel estimation method of orthogonal frequency division multi-address mobile communication system |
Non-Patent Citations (1)
| Title |
|---|
| "Compressive Estimation of Doubly Selective Channels in Multicarrier Systems: Leakage Effects and Sparsity-Enhancing Processing";Georg Tauböck等;《IEEE Journal of Selected Topics in Signal Processing》;20100430;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104052691A (en) | 2014-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104052691B (en) | MIMO-OFDM system channel estimation method based on compressed sensing | |
| US8594221B2 (en) | Model-based channel estimator for correlated fading channels and channel estimation method thereof | |
| CN106453162B (en) | For the channel estimation methods of multi-input multi-output orthogonal frequency division multiplexing system | |
| CN104869086B (en) | MIMO ofdm communication system down channels method of estimation, device based on two dimensional compaction perception | |
| CN101841375A (en) | Testing method and device for multi-input multi-output single carrier block transmission system | |
| CN108881076B (en) | MIMO-FBMC/OQAM system channel estimation method based on compressed sensing | |
| CN107465636B (en) | Channel estimation method of millimeter wave large-scale array space-frequency double-broadband system | |
| CN113922848B (en) | Signal sending method, channel estimation method, sending end equipment and receiving end equipment | |
| CN106506415B (en) | A kind of method of multi-user MIMO-OFDM system channel estimation | |
| CN103763223B (en) | Sparse MIMO-OFDM channel estimation method based on space-time correlation of channel | |
| CN109005133B (en) | Double-sparse multi-path channel model and channel estimation method based on model | |
| US11063724B1 (en) | Reduced channel-sounding in MU-MIMO WLANS | |
| CN106233685B (en) | The method of hybrid analog-digital simulation digital precode for extensive mimo system | |
| Uwaechia et al. | Compressed channel estimation for massive MIMO-OFDM systems over doubly selective channels | |
| CN104618293B (en) | A kind of optimization method of the unitary transformation matrix of smooth singular value decomposition | |
| CN115378529A (en) | Sky wave large-scale MIMO signal detection method based on Slepian transformation | |
| Bhoyar et al. | Leaky least mean square (LLMS) algorithm for channel estimation in BPSK-QPSK-PSK MIMO-OFDM system | |
| CN114553640B (en) | Cross-frequency-band statistical channel state information estimation method in multi-frequency-band large-scale MIMO system | |
| Tu et al. | Subspace blind MIMO-OFDM channel estimation with short averaging periods: Performance analysis | |
| CN113411108B (en) | Method, apparatus and storage medium for signal modulation and demodulation | |
| Shen et al. | Differential CSIT acquisition based on compressive sensing for FDD massive MIMO systems | |
| Xu et al. | Research on sparse channel estimation algorithm based on MIMO-OFDM system | |
| CN113395222B (en) | Channel prediction throughput optimization method based on non-uniform pilot frequency | |
| Sun et al. | Pilot Reduction Using Spatial and Temporal Correlation in Massive MIMO-OFDM Systems | |
| Diallo et al. | A limited and efficient quantized feedback for IEEE 802.11 n evolution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |