CN104683074A - Large-scale MIMO system limiting feedback method based on compressive sensing - Google Patents

Abstract

The invention discloses a large-scale MIMO system limiting feedback method based on compressive sensing. In a large-scale MIMO limiting feedback system, an Nh*Nv-dimensional uniform planar array (UPA) is adopted at a base station, and a linear array is adopted at a user terminal. The large-scale MIMO system limiting feedback method comprises the following steps: firstly, seeking a sparse base matched with a large-scale MIMO horizontal-dimensional channel and a vertical-dimensional channel by using a compressive sensing (CS) theory, thereby furthermore obtaining a two-dimensional united sparse base matched with a large-scale MIMO channel according to a Kronecker product; secondly, projecting an observation matrix irrelevant to the sparse base to a low-dimensional space, thereby obtaining an observation signal, quantifying the signal, and feeding the quantified signal back to the base station; finally, reconstructing channel state information at the base station by using an orthogonal matching pursuit (OMP) algorithm. The utility model provides a simple and efficient implementation method for the large-scale MIMO limiting feedback system.

Description

Based on the extensive mimo system limited feedback method of compressed sensing

Technical field

The present invention relates to 5G mobile communication technology field, show in particular a kind of extensive mimo system limited feedback method based on compressed sensing.

Background technology

Along with the sharp increase of amount of communication data, the raising of the availability of frequency spectrum and efficiency is become to the key of current Development of Wireless Communications.As everyone knows, MIMO (Massive-Multiple Input Multiple Output, Massive MIMO, extensive multiple-input and multiple-output) technology can improve link transmission reliability and obtain higher power system capacity, therefore becomes the key technology of 4G wireless communication system.The capacity of mimo system increases along with the increase of number of antennas, and therefore extensive MIMO is configured with the antenna of huge number in base station end, thus obtains better spatial reuse and space diversity effect, and can the more effective interference eliminated between user.From the angle of energy efficiency, the wave beam that extensive MIMO is formed is narrower, and transmitting power can be made more to concentrate on specific user, thus less transmitting power can be utilized to obtain higher receiving terminal signal to noise ratio.Because extensive MIMO has low energy consumption and high performance advantage, receive the extensive concern of industry and become the crucial investigative technique of 5G.The antenna number configured in base station end due to extensive MIMO increases, thus traditional limited feedback method based on code book be not suitable for extensive mimo system, therefore need to design effective feedback algorithm to reduce the feedback overhead of extensive mimo system.

There is a kind of theory-compressed sensing newly in signal transacting and the communications field in recent years, for the compressed feedback studying channel provides new thinking, this theory is pointed out: as long as signal is compressible or is sparse at certain transform domain, so just with transform-based incoherent observing matrix, conversion gained height dimensional signal can be projected on a lower dimensional space with one, then just from these a small amount of projections, original signal can be reconstructed with high probability by solving an optimization problem, can prove that such projection contains the enough information of reconstruction signal.

The rarefaction representation of signal is basis and the prerequisite of compressive sensing theory application, only has and reasonably have selected sparse base, the sparse coefficient number of signal just can be made the least possible, thus ensure signaling protein14-3-3 precision.Conventional Carlow Nan-Luo Yi (Karhunen-Loeve Transform, KLT) conversion is a conversion under statistical significance with lowest mean square meaning, there is optimum decorrelation and extraordinary compression efficiency, but shortage fast algorithm, and transformation matrix is different with signal, unlike signal need calculate different transformation matrixs, and therefore computation complexity is larger.FFT has efficient algorithm, so FFT is except in theory very important, is realizing also playing central role in various digital signal processing algorithm.DCT has very strong " concentration of energy " characteristic: the energy of most sound and picture signal all concentrates on the low frequency part after discrete cosine transform, and when signal has the statistical property close to markoff process, the decorrelation of DCT is close to Karhunen-Loeve transformation.In extensive mimo system, based on the Limited Feedback technology of compressed sensing, academic is all in the Primary Study stage with industrial circle at present.Because extensive MIMO base station antenna number is too much, and adopt UPA, the Limited Feedback technical research therefore based on compressed sensing not only needs to consider how to find applicable sparse base, also needs effectively to reduce feedback overhead.

Summary of the invention

For above the deficiencies in the prior art, the object of the present invention is to provide a kind of extensive mimo system limited feedback method based on compressed sensing, the method can not only find the two-dimentional joint sparse base being suitable for extensive mimo system, channel information can also be recovered in base station end, reduce feedback overhead simultaneously.

Technical scheme of the present invention is as follows:

Based on an extensive mimo system limited feedback method for compressed sensing, it comprises the following steps:

Step 101, base station end are for being configured with N _h× N _vthe UPA of dimension, wherein N _hfor the horizontal dimension number of transmit antennas of UPA, N _vfor the vertical dimension number of transmit antennas of UPA; User side is configured to single antenna, and channel can be expressed as wherein h _hrepresent horizontal dimension channel information, h _vrepresent vertical dimension channel information.

The sparse base Ψ of horizontal dimension of step 102, design H _hbase Ψ sparse in vertical dimension _v, wherein Ψ _hadopt discrete cosine transform (discrete cosine transform, DCT) sparse base, Ψ _vadopt fast Fourier transform (Fast Fourier Transform, FFT) sparse base, then amass theorem through Kronecker and obtain two-dimentional joint sparse base $\mathrm{\Ψ}={\mathrm{\Ψ}}_h\⊗{\mathrm{\Ψ}}_v.$

Step 103, generation random Gaussian observing matrix Φ ∈ R ^{l × N}, wherein element all obeys 0 average, and variance is independent normal distribution, wherein N=N _h× Nv, L represent measured value, and L < N.

Step 104, by observing matrix Φ, channel H is projected to lower dimensional space, obtain the observation signal y of L × N dimension.

Step 105, base station end are according to observation signal y, observing matrix Φ and two-dimentional joint sparse base Ψ, and what employing OMP algorithm was correct recovers channel information

Further, in step 102, horizontal dimension channel rarefaction representation is as follows: .h _h=Ψ _hh ₁., wherein h ₁represent the signal after orthogonal sparse base conversion, the sparse base Ψ of DCT _hconcrete formula as follows:

$F(u,v)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{i=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}f(i,j)\cos \frac{(2i+1)\mathrm{u\&pi;}}{2N_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{2N_h}\right]$

Inverse transformation IDCT formula is:

$F(i,j)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{u=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}F(u,v)\cos \frac{(2i+1)\mathrm{u\&pi;}}{2N_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{2N_h}\right]$

Wherein N _hfor the horizontal dimension number of transmit antennas of UPA, i, j are space coordinatess, i, j=0,1 ... N _h-1; U, v are DCT space coordinatess, u, v=0,1 ... N _h-1.Variable coefficient c (i)=1 (i=1,2,3...N _h-1).

In like manner, vertical dimension channel rarefaction representation is as follows: h _v=Ψ _vh ₂, wherein h ₂represent the signal after orthogonal sparse base conversion, the sparse base Ψ of FFT _vconcrete formula as follows:

$X\left(m\right)=X_1\left(m\right)+W_{N_v}^m{X}_1\left(m\right),m=\mathrm{0,1},...\frac{N_v}{2}-1$

$X(m+\frac{N_v}{2})=X_1\left(m\right)-W_{N_v}^m{X}_1\left(m\right),m=\mathrm{0,1},...\frac{N_v}{2}-1$

Wherein N _vfor the vertical dimension number of transmit antennas of base station end uniform surface battle array,

$X_1\left(m\right)=\underset{r=0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x\left(2r\right)W_{\frac{N_v}{2}}^{\mathrm{rm}},X_2\left(m\right)=\underset{r=0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x(2r+1)W_{\frac{N_v}{2}}^{\mathrm{rm}}$ And $W_{N_v}^m=e^{-j\frac{2\mathrm{\&pi;}}{N_v}m}.$

Theorem is amassed according to Kronecker: $(A\&CircleTimes;B)(C\&CircleTimes;D)=\mathrm{AC}\&CircleTimes;\mathrm{BD},$ And the rarefaction representation of horizontal dimension channel and vertical dimension channel is substituted in extensive mimo channel model, and then obtain two-dimentional joint sparse base Ψ, be specifically expressed as follows:

$H=h_h\&CircleTimes;h_v={\mathrm{\&Psi;}}_h{h}_1\&CircleTimes;{\mathrm{\&Psi;}}_v{h}_2=({\mathrm{\&Psi;}}_h\&CircleTimes;{\mathrm{\&Psi;}}_v)(h_1\&CircleTimes;h_2)=\mathrm{\&Psi;h}$

Wherein H represents the extensive mimo channel information of N × 1, h _hrepresent N _hthe extensive MIMO horizontal dimension channel information of × 1, h _vrepresent N _vthe extensive MIMO vertical dimension channel information of × 1, Ψ _hrepresent N _h× N _hthe sparse base of DCT, Ψ _hrepresent N _h× N _hthe sparse base of FFT, Ψ represents N × N two dimension joint sparse base, h ₁, h ₂, h represents the signal after orthogonal sparse base conversion.

Further, the random Gaussian matrix Φ in step 103 has an important character: for the random Gaussian matrix Φ of a L × N, and as L >=cK log (N/K), Φ Ψ has RIP character under very large probability:

$(1-\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2\&le;{\left|\right|\mathrm{\&Phi;h}\left|\right|}_2^2\&le;(1+\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2,$

Wherein δ ∈ (0,1), c represents a very little constant, and K represents the degree of rarefication of channel, and N represents the length of channel, and N=N _h× N _v.

Further, in step 104, observation signal can be expressed as following form:

y＝ΦH＝ΦΨh

Wherein Φ represents the random Gaussian matrix of L × N, and H represents the extensive mimo channel information of N × 1, and Ψ represents N × N two dimension joint sparse base, and h represents the signal after two-dimentional joint sparse base conversion.

Further, the channel recovered in step 105 and the normalized mean squared error (Mean Squared Error, MSE) between extensive mimo channel H, try to achieve by following formula:

$\mathrm{\&theta;}=\frac{{\left|\right|H-\tilde{H}\left|\right|}_2^2}{{\left|\right|H\left|\right|}_2^2}$

Contemplated by the invention the feature that extensive mimo antenna number is too much, limited feedback method based on compressed sensing is proposed, first, CS theory is utilized to find the two-dimentional joint sparse base matched with extensive mimo channel, then, utilize observing matrix that condition of sparse channel is carried out linear projection from higher-dimension to low-dimensional, finally, in base station end, recovered the channel condition information of feedback by OMP algorithm.The extensive mimo system limited feedback method based on compressed sensing that the present invention proposes, compared to traditional MIMO Limited Feedback technology, not only more mate extensive mimo channel, and feedback overhead and computation complexity can be effectively reduced, thus, this invention for extensive MIMO Limited Feedback provides a kind of simple and implementation efficiently.

Accompanying drawing explanation

Fig. 1 is the system block diagram of the extensive mimo system limited feedback method based on compressed sensing that the present invention proposes;

Fig. 2 is the uniform surface battle array schematic diagram that base station adopts vertical polarized antenna;

Fig. 3 is compressed sensing basic operation flow chart;

Fig. 4 is the Limited Feedback flow chart that the present invention is based on compressed sensing.

Embodiment

In extensive MIMO Limited Feedback scheme, by observing matrix, condition of sparse channel is projected to lower dimensional space, and the observation signal obtained is fed back to base station, base station utilizes OMP algorithm to rebuild the channel condition information of feedback, like this, computation complexity and feedback overhead can be reduced.

As shown in Figure 1, Fig. 1 is the system block diagram of the extensive mimo system Limited Feedback based on compressed sensing that the present invention proposes.At base station end configuration N _h× N _vthe UPA of dimension, wherein N _hfor the horizontal dimension number of transmit antennas of UPA, N _vfor the vertical dimension number of transmit antennas of UPA, M user of receiving terminal configuration single antenna, user side carries out channel estimating and obtains 3D channel matrix H, and H can be expressed as wherein h _hrepresent horizontal dimension channel information, h _vrepresent vertical dimension channel information.

Fig. 2 is the uniform surface battle array model that the present invention proposes, and considers the restriction of antenna physical size, and this model adopts UPA structure as shown in the figure.

Fig. 3 is compressed sensing basic operation flow chart.As shown in the figure, compressive sensing theory mainly comprises three parts, first, and the rarefaction representation of signal, then, design observing matrix, will ensure that while reducing dimension the information loss of primary signal is minimum, finally, modelled signal recovery algorithms, utilizes measured value to recover primary signal undistortedly.

Fig. 4 is the Limited Feedback flow chart that the present invention proposes based on compressed sensing.

Step 41: base station end configuration N _h× N _vthe UPA of dimension vertical polarized antenna, wherein N _hfor the horizontal dimension number of transmit antennas of UPA, N _vfor the vertical dimension number of transmit antennas of UPA, M user of receiving terminal configuration single antenna.

Step 42: dct transform is carried out to horizontal dimension channel, that is:

h _h＝Ψ _hh ₁

Wherein h ₁represent the signal after orthogonal sparse base conversion, the sparse base Ψ of DCT _hconcrete formula as follows:

$F(u,v)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{i=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}f(i,j)\cos \frac{(2i+1)\mathrm{u\&pi;}}{2N_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{2N_h}\right]$

Inverse transformation IDCT formula is:

$F(i,j)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{u=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}F(u,v)\cos \frac{(2i+1)\mathrm{u\&pi;}}{2N_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{2N_h}\right]$

Wherein N _hfor the horizontal dimension number of transmit antennas of uniform surface battle array UPA, i, j are space coordinatess, i, j=0,1 ... N _h-1; U, v are DCT space coordinatess, u, v=0,1 ... N _h-1.Variable coefficient c (i)=1 (i=1,2,3...N _h-1) _.

Step 43: orthogonal sparse base conversion is carried out to vertical dimension channel, obtains sparse signal h ₂, that is:

h _v＝Ψ _vh ₂

The wherein sparse base Ψ of FFT _vconcrete formula as follows:

$X\left(m\right)=X_1\left(m\right)+W_{N_v}^m{X}_1\left(m\right),m=\mathrm{0,1},...\frac{N_v}{2}-1$

$X(m+\frac{N_v}{2})=X_1\left(m\right)-W_{N_v}^m{X}_1\left(m\right),m=\mathrm{0,1},...\frac{N_v}{2}-1$

Wherein

$X_1\left(m\right)=\underset{r=0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x\left(2r\right)W_{\frac{N_v}{2}}^{\mathrm{rm}},X_2\left(m\right)=\underset{r=0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x(2r+1)W_{\frac{N_v}{2}}^{\mathrm{rm}}$ And $W_{N_v}^m=e^{-j\frac{2\mathrm{\&pi;}}{N_v}m}.$

Step 44: amass theorem according to Kronecker: $(A\&CircleTimes;B)(C\&CircleTimes;D)=\mathrm{AC}\&CircleTimes;\mathrm{BD},$ And the rarefaction representation of horizontal dimension channel and vertical dimension channel is substituted in extensive mimo channel model, and then obtain two-dimentional joint sparse base Ψ, be specifically expressed as follows:

$H=h_h\&CircleTimes;h_v={\mathrm{\&Psi;}}_h{h}_1\&CircleTimes;{\mathrm{\&Psi;}}_v{h}_2=({\mathrm{\&Psi;}}_h\&CircleTimes;{\mathrm{\&Psi;}}_v)(h_1\&CircleTimes;h_2)=\mathrm{\&Psi;h}$

Wherein H represents the extensive mimo channel information of N × 1, h _hrepresent N _hthe extensive MIMO horizontal dimension channel information of × 1, h _vrepresent N _vthe extensive MIMO vertical dimension channel information of × 1, Ψ _hrepresent N _h× N _hthe sparse base of DCT, Ψ _hrepresent N _h× N _hthe sparse base of FFT, Ψ represents N × N two dimension joint sparse base, h ₁, h ₂, h represents the signal after orthogonal sparse base conversion.

Step 45: generate random Gaussian matrix Φ, and an important character must be had: for the random Gaussian matrix Φ of a L × N, for the random Gaussian matrix Φ of a L × N, as L >=cK log (N/K), Φ Ψ has RIP character under very large probability:

$(1-\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2\&le;{\left|\right|\mathrm{\&Phi;h}\left|\right|}_2^2\&le;(1+\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2,$

Wherein δ ∈ (0,1), c represents a very little constant, and K represents the degree of rarefication of channel, and N represents the length of channel, and N=N _h× N _v, h represents the signal after two-dimentional joint sparse base conversion.

Step 46: channel is carried out linear projection by the observing matrix Φ utilizing L × N to tie up from higher-dimension to low-dimensional, obtains the observation signal y of L × N dimension:

y＝ΦH＝ΦΨh

Wherein Φ represents the random Gaussian matrix of L × N, and H represents the extensive mimo channel information of N × 1, and Ψ represents N × N two dimension joint sparse base, and h represents the signal after two-dimentional joint sparse base conversion.

Step 47: base station end, according to observation signal y, observing matrix Φ and two-dimentional joint sparse base Ψ, adopts OMP algorithm to recover channel information and by normalization MSE that following formula is asked:

$\mathrm{\&theta;}=\frac{{\left|\right|H-\tilde{H}\left|\right|}_2^2}{{\left|\right|H\left|\right|}_2^2}.$

Visible, the present invention provides one and simple and implementation method efficiently for extensive MIMO Limited feedback systems.

Claims (8)

1., based on an extensive mimo system limited feedback method for compressed sensing, it is characterized in that: comprise the following steps:

Step 101, base station end are for being configured with N _h× N _vthe UPA of dimension, wherein N _hfor the horizontal dimension number of transmit antennas of UPA, N _vfor the vertical dimension number of transmit antennas of UPA; User side is configured to single antenna, and channel can be expressed as wherein h _hrepresent horizontal dimension channel information, h _vrepresent vertical dimension channel information;

The sparse base Ψ of horizontal dimension of step 102, design H _hbase Ψ sparse in vertical dimension _v, wherein Ψ _hadopt discrete cosine transform (DCT) sparse base, Ψ _vadopt fast Fourier transform (FFT) sparse base, then amass theorem through Kronecker and obtain two-dimentional joint sparse base

Step 103, generation random Gaussian observing matrix Φ ∈ R ^{l × N}, wherein element all obeys 0 average, and variance is independent normal distribution, wherein N=N _h× N _v, L represents measured value, and L < N;

Step 104, by observing matrix Φ, channel H is projected to lower dimensional space, obtain the observation signal y of L × N dimension;

Step 105, base station end, according to observation signal y, observing matrix Φ and two-dimentional joint sparse base Ψ, adopt OMP algorithm to recover channel information

2. method according to claim 1, is characterized in that, in step 102, horizontal dimension channel rarefaction representation is as follows: h _h=Ψ _hh ₁, wherein h ₁represent the signal after orthogonal sparse base conversion, Ψ _hrepresent the sparse base of DCT;

Vertical dimension channel rarefaction representation is as follows: h _v=Ψ _vh ₂, wherein h ₂represent the signal after orthogonal sparse base conversion, Ψ _vrepresent the sparse base of FFT.

3. method according to claim 2, is characterized in that, the sparse base Ψ of DCT _hconcrete formula as follows:

$F(u,v)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{i=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}f(i,j)\cos \frac{(2i+1)\mathrm{u\&pi;}}{{2N}_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{{2N}_h}\right]$

Inverse transformation IDCT formula is:

$F(i,j)=\frac{2}{N_h}C\left(u\right)C\left(v\right)\left[\underset{u=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{N_h-1}{\mathrm{\&Sigma;}}}F(u,v)\cos \frac{(2i+1)\mathrm{u\&pi;}}{{2N}_h}\cos \frac{(2i+1)\mathrm{v\&pi;}}{{2N}_h}\right]$

Wherein N _hfor the horizontal dimension number of transmit antennas of UPA, i, j are space coordinatess, i, j=0,1 ... N _h-1; U, v are DCT space coordinatess, u, v=0,1 ... N _h-1.Variable coefficient c (i)=1 (i=1,2,3...N _h-1).

4. method according to claim 2, is characterized in that, the sparse base Ψ of FFT _vconcrete formula as follows:

$X\left(m\right)=X_1\left(m\right)+W_{N_v}^m{X}_1\left(m\right),m=\mathrm{0,1},...\frac{N_v}{2}-1$

$X(m+\frac{N_v}{2})=X_1\left(m\right)-W_{N_v}^m{X}_1\left(m\right),,=\mathrm{0,1},...\frac{N_v}{2}-1$

Wherein N _vfor the vertical dimension number of transmit antennas of base station end uniform surface battle array, $X_1\left(m\right)=\underset{r=-0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x\left(2r\right)W_{\frac{N_v}{2}}^{\mathrm{rm}},X_2\left(m\right)=\underset{r=0}{\overset{\frac{N_v}{2}-1}{\mathrm{\&Sigma;}}}x(2r+1)W_{\frac{N_v}{2}}^{\mathrm{rm}}$ And $W_{N_v}^m=e^{-j\frac{2\mathrm{\&pi;}}{N_v}m}.$

5. method according to claim 3, is characterized in that, amasss theorem according to Kronecker: $(A\&CircleTimes;B)(C\&CircleTimes;B)=\mathrm{AC}\&CircleTimes;\mathrm{BD},$ Horizontal dimension and vertical dimension channel are brought into thus obtain two-dimentional joint sparse base Ψ, namely $H=h_h\&CircleTimes;h_v={\mathrm{\&Psi;}}_h{h}_1\&CircleTimes;{\mathrm{\&Psi;}}_v{h}_2=({\mathrm{\&Psi;}}_h\&CircleTimes;{\mathrm{\&Psi;}}_v)(h_1\&CircleTimes;h_2)=\mathrm{\&Psi;h},$ Wherein h represents the condition of sparse channel after two-dimentional joint sparse base conversion.

6. method according to claim 1, it is characterized in that, in step 103, condition of sparse channel is carried out linear projection by the observing matrix Φ utilizing L × N to tie up from higher-dimension to low-dimensional, the product of its observing matrix and sparse basis array must meet limited equidistant condition (RIP), thus to guarantee from observation signal reconstruction signal accurately, RIP condition is as follows:

$(1-\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2\&le;{\left|\right|\mathrm{\&phi;h}\left|\right|}_2^2\&le;(1+\mathrm{\&delta;}){\left|\right|h\left|\right|}_2^2$

Wherein δ ∈ (0,1), h represent the condition of sparse channel after two-dimentional joint sparse base conversion.

7. method according to claim 1, is characterized in that, in step 104, observation signal can be expressed as following form:

y＝ΦH＝ΦΨh

Wherein Φ represents the random Gaussian matrix of L × N, and H represents the extensive mimo channel information of N × 1, and Ψ represents N × N two dimension joint sparse base, and h represents the signal after two-dimentional joint sparse base conversion.

8. method according to claim 1, is characterized in that, the channel recovered in step 105 and the normalized mean squared error (MSE) between extensive mimo channel H, try to achieve by following formula:

$\mathrm{\&theta;}=\frac{{\left|\right|H-\tilde{H}\left|\right|}_2^2}{{\left|\right|H\left|\right|}_2^2}.$