WO2015196384A1

WO2015196384A1 - Joint sparse channel estimation method, device and system

Info

Publication number: WO2015196384A1
Application number: PCT/CN2014/080717
Authority: WO
Inventors: 戚晨皓; 朱鹏程
Original assignee: 东南大学
Priority date: 2014-06-23
Filing date: 2014-06-25
Publication date: 2015-12-30
Also published as: CN104022979B; CN104022979A

Abstract

Disclosed are a joint sparse channel estimation method, device and system. The method comprises: building a joint sparse reconstruction model, merging all channels to be estimated into a joint sparse vector, solving the positions of all nonzero element blocks of the joint sparse vector by means of the joint sparse reconstruction model, and solving values of nonzero elements of the channels. By means of the present invention, the channel estimation accuracy can be improved, and the pilot overhead can be reduced.

Description

Joint sparse channel estimation method, device and system

Technical field

The present invention relates to a wireless communication system, and more particularly to a joint sparse channel estimation method, apparatus and system.

Background technique

The rapid development of wireless communication technology and the rapid spread of smartphones have brought about an explosive growth in the demand for wireless data transmission. In the collection of the International Telecommunication Union (ITU) for 4th Generation (4G) mobile communication standard candidate schemes, the uplink and downlink peak data rates are clearly required to reach 1G bps; for this reason, the third generation mobile communication The 3rd Generation Partnership Project (3GPP) organization actively conducts Long Term Evolution (LTE) and UTE-Advanced technology research, and supports Release 8 X 8 and Uplink 4 X in Release 11. The multi-antenna system of 4, which can be foreseen, will provide further support for larger multi-antenna systems in the future.

The basic feature of the multi-antenna wireless communication system is that a certain number of antennas are configured in the base station, and the mobile phone users within the coverage of the base station are only configured with a single antenna because of the size of the mobile phone; multi-input and single output from the base station to the mobile phone (Multi- Downlink transmission of Input Single-Output, MIS0), uplink transmission of Single-Input Multi-Output (SIM0) from mobile phone to base station. For downlink beamforming, the base station needs to obtain downlink channel information. Currently, there are mainly two ways. The first method is that the base station transmits a pilot, and the mobile phone uses the received pilot to perform channel estimation, acquires downlink channel information, and feeds it back to the base station, which is usually used for frequency division duplex (FDD). The second method is that the mobile phone transmits a pilot, and the base station uses the received pilot to perform channel estimation to obtain uplink channel information, because in the Time-Duplex Division (TDD) system, the uplink channel and the downlink channel With reciprocity, the base station also acquires downlink channel information, which is usually used in TDD systems. Regardless of whether it is an FDD system or a TDD system, LTE and LTE-Advanced usually use Orthogonal Frequency Division Multiplexing (OFDM) technology for downlink transmission and single carrier frequency division multiple access for uplink transmission ( Single -carrier Frequency-division Multiple Access, SC-FDMA) technology.

Recent studies have shown that the Channel Impulse Response (CIR) sequence of a wireless channel usually exhibits mostly zero, but only a few non-zero sparsity, where the number of non-zero elements The number of multipaths for the wireless channel. Therefore, Compressed Sensing (CS) technology can be fully utilized to replace the existing Least Squares (LS) and Mean Square Errors (MMSE) channel estimation with sparse channel estimation, and reduce pilots. Overhead, alleviating the shortage of pilot resources in multi-antenna systems. In addition, in a multi-antenna system, the time at which signals from different antennas of the base station are simultaneously transmitted to the mobile phone (Time of Arrival, ToA) are approximately the same, and the signals transmitted from the mobile phone to the different antennas of the base station are approximately the same, that is, corresponding to different base station antennas. The positions of the non-zero elements of the CIR sequence of different channels can be considered to be the same, and the values of the non-zero elements are different. Therefore, the information of the same location of the non-zero elements can be fully utilized, and the joint sparse channel estimation of the multiple channels can be performed to acquire the channel information.

In the prior art, the receiver usually performs separate channel estimation for each channel by using the received pilot and the transmitted pilot, and the related art utilizes the sparsity of the channel for separate sparse channel estimation, but there is no technology yet. The multi-channel joint sparse channel estimation is implemented by using the information that multiple channels have the same non-zero element position. Therefore, the pilot overhead of the prior art is still large.

Summary of the invention

The present invention provides an efficient channel estimation method and apparatus for a multi-antenna wireless communication system, which can perform joint sparse channel estimation on multiple channels, improve channel estimation accuracy, and reduce pilot overhead.

The present invention provides a joint sparse channel estimation method, which includes the following steps:

S1: establishing a joint sparse reconstruction model, combining multiple channels into a joint sparse vector;

S2: acquiring, by using the joint sparse reconstruction model, locations of all non-zero element blocks of the joint sparse vector;

S3: Obtain a value of a non-zero element of each of the channels.

Preferably, in the step S2, the following steps are further included:

S21: initializing residuals are joint observation values of the joint sparse reconstruction model, normalizing each column of the joint observation matrix of the joint sparse reconstruction model, initializing the selection into an empty set, and setting a loop number to 0, wherein Normalization refers to an operation that makes the square of the modulus of all elements of the column one;

S22: determining whether the power of the residual is greater than a product of a noise variance and a square of the number of base station antennas, determining whether the number of cycles is less than the channel length, and if both are, performing S23; otherwise, executing S24;

S23: updating the residual and the selection, and adding 1 to the number of cycles;

S24: All elements in the selection set are sequentially output as positions of the all non-zero element blocks of the joint sparse vector. Preferably, in the step si, the joint sparse reconstruction model is expressed as 2 = Biv + n, where z is defined as the joint observation value of the M channels of the model, n is its joint observation noise, and IV is It is a joint sparse vector, β is its joint observation matrix. Preferably, the joint sparse vector Μ is: IV = [M , M^,..., M ] ^T , where WT represents the 〖th element block of the column vector iv, 1 = 1, 2,... , L. defined as:

_Wl = [h^ , h^ , ..., h^ ^M ], 1 = 1, 2, ..., L, L represents the channel length, M represents the number of antennas of the base station, « represents the ith root of the base station The impulse response sequence of the i-th channel corresponding to the antenna, i = 1, 2, ..., Μ, represents the Zth element of «.

The invention also provides a joint sparse channel estimation apparatus, comprising:

Establishing a model unit for combining multiple channels into a joint sparse vector;

a joint sparse vector computing unit for solving positions of all non-zero element blocks of the joint sparse vector of the joint sparse reconstruction model;

An information acquiring unit, configured to solve a value of a non-zero element of each of the channels.

Preferably, the joint sparse vector computing unit further includes:

An initialization module, configured to initialize joint observations whose residuals are joint sparse reconstruction models, normalize each column of the joint observation matrix of the joint sparse reconstruction model, initialize the selection into an empty set, and set a loop number to 0;

a determining module, configured to determine whether the power of the residual is greater than a product of a noise variance and a square of the number of base station antennas, and determine whether the number of cycles is less than a channel length, and if both are, performing an update module; otherwise, executing an output module;

Update module, used to update the residuals and selections, and the number of cycles is increased by 1;

An output module that sequentially outputs all elements of the ensemble as the locations of all non-zero element blocks of the joint sparse vector.

Preferably, the joint sparse reconstruction model is expressed as 2 = Biv + n, where z is defined as the joint observation of the M channels of the model, n is its joint observation noise, IV is its joint sparse vector, and B is Its joint observation matrix. Preferably, the joint sparse vector M is: IV = [M , M^,..., M ] ^T , where, the table The 〖th element block of the column vector iv, I = 1, 2, ..., L . defined as

_Wl = [h ¹ l), h ² ( , ... , h ^M ], 1 = 1, 2, ..., L, L represents the channel length, M represents the number of antennas of the base station, « represents the base station The impulse response sequence of the i-th channel corresponding to the i-antenna, i = 1, 2, ..., Μ, represents the Zth element of «.

The present invention also provides a joint sparse channel estimation system, comprising: setting the joint sparse channel estimation apparatus in an uplink transmission or a downlink transmission of the system.

Preferably, the uplink transmission includes: the data of the mobile terminal is sequentially transmitted through the constellation point mapping, the fast Fourier transform, the inserted pilot, the subcarrier mapping, the inverse fast Fourier transform, the insertion guard interval, and the up-conversion, and then sent to the wireless channel to reach the base station. After that, the transmission data is extracted after down-conversion, de-protection interval, fast Fourier transform, sub-carrier demapping, joint sparse channel estimation, channel equalization, inverse fast Fourier transform, and constellation point de-mapping.

Preferably, the downlink transmission comprises: the data of the base station is sequentially transmitted through the constellation point mapping, the insertion pilot, the subcarrier mapping, the inverse fast Fourier transform, the insertion protection interval, and the up-conversion, and then sent to the wireless channel, and after reaching the mobile phone, sequentially After down-conversion, de-protection interval, fast Fourier transform, sub-carrier demapping, joint sparse channel estimation, channel equalization, and constellation point de-mapping, the transmitted data is extracted.

The invention has the following beneficial effects:

a) using the present invention to perform joint sparse channel estimation on multiple channels, compared to the existing single-sparse channel estimation for each channel, both using the same number of pilots, the former can more accurately estimate the CIR sequence non-zero The position of the element improves the channel estimation accuracy;

b) Using the present invention to perform joint sparse channel estimation on multiple channels, compared with the existing single-sparse channel estimation for each channel, the two must achieve the same channel estimation accuracy, the former uses fewer pilots, and reduces Pilot overhead.

3) The present invention is used for joint sparse channel estimation on multiple channels. The larger the number of base station antennas, the larger the multi-antenna system scale, the higher the channel estimation accuracy, and the more significant the pilot overhead saved.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the embodiments or the prior art description will be briefly described below. It is obvious that the drawings in the following description are merely Some embodiments of the present invention may also be used to obtain other drawings based on these drawings without departing from the prior art. 1 is a flow chart of a joint sparse channel estimation method according to the present invention;

Figure 2 is a flow chart of S2 of Figure 1 of the present invention;

3 is a schematic structural diagram of a joint sparse channel estimation apparatus according to the present invention;

4 is a schematic diagram of transmission of a SIMO multi-antenna system used in Embodiment 1 of the present invention;

Figure 5 is a block diagram of an SC-FDMA system according to Embodiment 1 of the present invention;

6 is a comparison of mean square error performance of a single sparse channel estimation for each channel in the first embodiment of the present invention;

7 is a schematic diagram of transmission of a MISO multi-antenna system used in Embodiment 2 of the present invention;

8 is a block diagram of an OFDM system according to Embodiment 2 of the present invention;

FIG. 9 is a comparison of the mean square error performance of the second embodiment of the present invention with the single-sparse channel estimation for each channel in the prior art.

detailed description

BRIEF DESCRIPTION OF THE DRAWINGS The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

1 is a flowchart of a joint sparse channel estimation method according to the present invention, which includes the following steps: S1: Establish a joint sparse reconstruction model, and combine multiple channels into one joint sparse vector;

S3: Obtain a value of a non-zero element of each of the channels.

2 is a flow chart of S2 of FIG. 1 of the present invention, which includes the following steps:

S24: All elements in the selection set are sequentially output as positions of the all non-zero element blocks of the joint sparse vector. 3 is a schematic structural diagram of a joint sparse channel estimation apparatus according to the present invention. The device consists of the following 3 units:

(1) Establish a model unit for combining a plurality of channels into a joint sparse vector.

(2) A joint sparse vector calculation unit for solving the positions of all non-zero element blocks of the joint sparse vector of the joint sparse reconstruction model.

(3) An information acquisition unit, configured to solve a value of a non-zero element of each of the channels.

The joint sparse vector computing unit further includes the following four modules:

(a) an initialization module for initializing joint observations whose residuals are joint sparse reconstruction models, normalizing each column of the joint observation matrix of the joint sparse reconstruction model, initializing the selection into an empty set, and setting the number of loops to 0.

(b) a judging module, configured to determine whether the power of the residual is greater than a product of the noise variance and the number of base station antennas, determine whether the number of loops is less than the channel length, and if both are, execute the update module; otherwise, execute the output module.

(c) Update module for updating residuals and selections, plus 1 cycle count.

(d) An output module that sequentially outputs all elements in the ensemble as the locations of all non-zero element blocks of the joint sparse vector.

The invention relates to a joint sparse channel estimation system, characterized in that, in the uplink transmission or the downlink transmission of the system, a device as shown in FIG. 3 is provided, and correspondingly, the system will be in the first embodiment and the second embodiment of the present invention. Explain separately.

The uplink transmission refers to that, in the coverage of the base station, the mobile phone configured with a single antenna transmits a signal, and the base station receives the signal. It is assumed that the base station is configured with M antennas (M is a positive integer and M > 1), and each antenna corresponds to one uplink channel. In order to estimate the uplink channel, the mobile phone transmits a pilot, and the base station estimates the M channels by using the received pilot, and the computational complexity is proportional to M. In the TDD system, the uplink channel and the downlink channel have reciprocity, and once the base station acquires the uplink channel information, the downlink channel information is acquired. Since the base station location is fixed and there is sufficient power supply, the base station power limitation problem is not considered; thus, even for a larger multi-antenna system in the future, that is, when M is large, the channel estimation complexity can still be tolerated. The problem at this time is that pilot resources will become increasingly scarce. Embodiment 1 of the present invention performs joint sparse channel estimation on multiple uplink channels to reduce pilot resource overhead.

The downlink transmission refers to the base station communicating with the mobile phone configured with a single antenna within its coverage range. The signal is sent by the base station, and the mobile phone receives the signal to complete the downlink transmission. It is assumed that the base station is configured with M antennas (M is a positive integer, and M> l), and each antenna corresponds to one downlink channel. FDD is another mainstream technology in addition to TDD. In the FDD system, in order to estimate the downlink channel, the base station transmits pilots, and the mobile phone estimates the M channels by using the received pilot. In order to effectively distinguish the M pilots received by the single antenna of the mobile phone, the M pilots transmitted by the base station must be orthogonal in the time domain, the frequency domain, or the code domain. As M becomes larger and larger, M pilots occupy more and more resources in the time domain, frequency domain or code domain, and the pilot overhead is larger and larger. Embodiment 2 of the present invention performs joint sparse channel estimation on multiple downlink channels, and reduces pilot resource overhead. Embodiment 1:

4 is a schematic diagram of transmission of a snro multi-antenna system according to Embodiment 1 of the present invention. As shown in Figure 4, the signal transmitted by the mobile phone, after reflection from multiple buildings, reaches the base station, forming multipath effects and causing inter-symbol interference. For this reason, LTE and LTE-Advanced adopt SC-FDMA, which can effectively fight wireless. Multipath effects in propagation simplify the design of the equalizer.

Figure 5 is a block diagram of an SC-FDMA system in accordance with a first embodiment of the present invention. The data of the mobile terminal is processed by constellation point mapping, Fast Fourier Transform (FFT), insertion pilot, subcarrier mapping, Inverse Fast Fourier Transform (IFFT), insertion guard interval and up-conversion. After being sent to the wireless uplink channel, after reaching the base station, the transmission data is extracted after being subjected to down-conversion, removal of guard interval, FFT, sub-carrier demapping, joint sparse channel estimation, channel equalization, IFFT, and constellation point de-mapping. Compared with the OFDM technology widely used for downlink transmission in wireless systems, SC-FDMA performs FFT before the IFFT and subcarrier mapping on the transmitting end, which can effectively suppress the peak-to-average ratio of the signal and reduce the burden on the mobile power amplifier. It should be noted that the present invention employs joint sparse channel estimation instead of the individual sparse channel estimation for each channel in the prior art.

1 is a flow chart of a method for acquiring channel information of a multi-antenna wireless communication system according to the present invention. Referring to Figure 1, the method includes:

S1: Establishing a joint sparse reconstruction model, combining all channels to be estimated into one joint sparse direction. In an implementation manner of this embodiment, it is assumed that the number of SC-FDMA subcarriers is N, and the number of pilots used is (0 < Κ ≤ Ν ), the subcarrier index corresponding to each pilot subcarrier is Ρ ₂ , ... , Ρ _Κ ( 1 ≤ < Ρ ₂ <... < ≤ N ), and the pilot symbol transmitted by the mobile phone is represented as χ (Ρ ₁ ), χ(Ρ ₂ ), - , χ(Ρκ) ^ The mobile phone sends a pilot symbol, and the base station will receive a different pilot. Symbols, corresponding to M different upstream channels. Since the base station knows the pilot symbols transmitted by the mobile phone, after receiving the M different pilot symbols, the base station performs channel estimation on the M channels, and uses the channel estimation result for subsequent channel equalization. The pilot symbol received by the ith antenna of the base station is represented as a column vector

= [y ω (P , y « (P ₂ ), ... , y « (P _K ] ^T , i = 1,2,... ,M, where the superscript Γ denotes vector transpose. Suppose each The CIR sequence of the upstream channel is

= [ι ^(ί) (1), ι ^(ί) (2) ., ι ^(ί) (Ζ ] ^Τ , i = l,2,...,M. Due to the sparsity of the wireless channel, L Among the elements, most are zero, and only a few are non-zero, and the number of non-zero elements is the number of multipaths of the wireless channel. The existing related literature indicates that for the same transmitted signal, the ToA of the received signals of different antennas of the base station Similarly, it can be considered that the lengths of CIR sequences of different channels are the same, and the positions of non-zero elements in the CIR sequence are the same, and the values of non-zero elements are different. Suppose D is a diagonal matrix of rows and columns, and the diagonal elements are in turn ( ⁵ ₁ ), ( ⁵ ₂ ),..., (/^), Thus, for each base station antenna, the relationship between the transmit pilot and the receive pilot can be established as follows

y ⁽⁰ = DFh ⁱ⁾ + η , i = 1,2,..., M (1) where ? ^ω represents the Gaussian white noise of the i-th upstream channel, ? ^ω is a one-dimensional column vector, each element is independent and subject to a complex Gaussian distribution with a mean of 0 and a variance of σ ² ; for a standard Fourier matrix of N rows and N columns, the first L column and index are extracted, P ₂ , the line constitutes the Fourier leaf matrix. Defining the observation matrix 4 = DF, the formula (1) can be further simplified as

y(i) _{= Ah} (i) _{+ η} (ί) _ι i = H.., M (2) The essence of channel estimation is the process used to solve in the case of noise " ^ω ". Compared to the LS channel It is estimated that sparse channel estimation can achieve the same channel estimation performance as LS using fewer pilots. The present invention combines ^ = 1, 2, ..., Μ into a joint sparse vector iv of one ML dimension as follows

Where Mf represents the Zth element block of the column vector iv, I = 1, 2, ..., L, and the row vector 11^ is defined as follows

_Wl = [i ⁽¹⁾ (Z: ⁽²⁾ (U ^(M) (0), l = 1,2,... ,L Notice that the position of the non-zero element for the different i, is the same, non-zero The values of the elements are different, either the whole block is zero, or the whole block is non-zero, and M is presented as a block sparse structure, so the position of the non-zero element in M can be used to represent the position of the non-zero element in M. , defining a joint view of M channels The measured value Z is as follows

Where the Z-th element block representing the column vector z, 1 = 1, 2, ..., K, and the row vector ^ is defined as follows

Zi = [y ⁽¹ ( , y ⁽² ( y ^(M) ( l 1 = , 2 κ ) defines the joint observation noise η as follows

Where the first element block of the column vector η is represented, l = l, 2, ..., K, and the row vector 7^ is defined as follows

η _χ = [7 ^ω (θΌ( ² )(0,...Ό( ^Μ )(0], I = 1,2,... ,Κ The construction of the joint observation matrix β can be made by element-by-element of matrix 4 Instead of forming, the ith row and the 'column element of matrix 4 are represented as >l(i, ), which will be replaced by i = 1, 2, ..., K, j = 1, 2, ..., L , constitute a MAT row, ML column joint observation matrix β, where / _Μ represents the unitary matrix of the dimension.

The joint sparse reconstruction model can be expressed as

z = Bw + n (3) The present invention first uses the joint observation value Z and the joint observation matrix β to solve the positions of all non-zero element blocks of the joint sparse vector IV, and then separately solves the values of the non-zero elements of each channel.

S2: Solving the position of all non-zero element blocks of the joint sparse vector using the joint sparse reconstruction model. In the first embodiment of the present invention, the base station uses the joint sparse reconstruction model formula (3) to solve the positions of all the non-zero element blocks of the joint sparse vector ,. The flow refers to FIG. 2, and the method includes:

S21: Initializing the residual is a joint observation value of the joint sparse reconstruction model, normalizing each column of the joint observation matrix of the model, initializing the selection into an empty set, and setting the number of loops to 0.

Define the residual r as a column vector of the MAT dimension and initialize it to the joint observation z, ie r = τ. Normalize each column of the joint observation matrix β, where normalization is an operation in which the two norm of each column of β is 1, and the second norm of a vector is defined as the square of the modulus of all elements of the vector. with. Suppose that after normalizing each column of β, a matrix ρ of MAT rows and ML columns is obtained, so that the two norms of each column of ρ are 1. Specifically can be expressed as

B = QG (4) where G is a diagonal matrix of ML rows and ML columns, and each diagonal element of G is a real number greater than zero and a normalization factor corresponding to each column of β. Substituting the formula (4) into the formula (3), z = QGw + n

Definition 17 = GW, does not change the position of the non-zero element of M, get

z = Qv + n (5) Solve the position of all non-zero element blocks of the joint sparse vector M and convert to the position of all non-zero element blocks solved.

Define an anthology Λ to store the locations of non-zero element blocks that are obtained in turn. Since 17 and IV exhibit the same block-like sparse structure, the position of the non-zero element can be represented by the index of the non-zero element block, thus,

The index of a non-zero element block in V directly corresponds to the index of the non-zero element in Central. Initialization Λ is an empty set, ie Λ = 0. Set the number of loops Γ = 0.

S22: Determine whether the power of the residual is greater than a product of the noise variance and the square of the number of base station antennas, and determine whether the number of cycles is less than the channel length. If both are, perform S23; otherwise, execute S24.

Define the residual power as ||r|| , which represents the sum of the squares of the absolute values of all elements in r. If ||r||| > Μ ² σ ² , and Γ < L, then S23 is performed; otherwise, S24 is executed.

S23: Update the residual and the selection, and increase the number of cycles by one.

The column defining the matrix Q is = 1, 2, ..., / ^ Since each block of 17 = 1, 2, ..., either the entire block is zero, or the entire block is non-zero, V is rendered as a block sparse structure Correspondingly, the Q is divided into columns by column. Define the third block of Q as =

= 1,2,... ,L. From the complement of Λ Φ = {1, 2, ..., Z \A, find an element 'E 4>, so that

The element that satisfies the above condition is denoted by /, adding / adding to the selection and updating the selection Λ Λ U {/}, where the superscript - 1 indicates the matrix inversion and the superscript indicates the conjugate transpose. The definition ρ _{Λ is} a matrix composed of blocks of ρ corresponding to the elements in the selection ,, and the new residual is simultaneously, and the number of cycles is increased by 1, that is, Γ^Γ + 1.

S24: sequentially output all elements in the ensemble as all non-zero element blocks of the joint sparse vector The final element contained in the selection , is the position of the non-zero element block in V, which is also the position of the non-zero element block in Μ, which is also = 1, 2, ..., Μ the position of the common non-zero element . All the elements in the selection are output in sequence.

S3: Solve the value of the non-zero element of each channel.

The definition ^ is a matrix composed of 4 columns corresponding to the elements in the selection ,, and the column vector composed of the non-zero elements of the i-th upstream channel is

That is, the value of the non-zero element of the obtained i-th channel.

3 is a schematic structural diagram of a joint sparse channel estimation apparatus according to the present invention. The device consists of the following 3 units:

(c) Update module for updating residuals and selections, plus 1 cycle count.

In the simulation test, the number of base station antennas is M = 8. The number of SC-FDMA subcarriers is N = 256, the number of pilot subcarriers = 16, the pilot subcarrier index ⁵ ^ , ., . , is [8, 40, 48, 52, 72, 82, 99, 142, 145, 154, 158, 161, 183, 209, 212, 230]. QPSK modulation is used. Assume that the channel CIR sequence length is L = 60, of which only 5 = 12 non-zero elements, and the position of the CIR sequence is [2, 13, 21, 24, 29, 33, 41, 42, 43, 53, 54, 60]. Mobile phone send 1 For the pilot symbols, the base station receives 8 pilot symbols at the same time, and the base station needs to estimate the values of the non-zero elements of the 8 channels and the values of the non-zero elements.

Table 1 Comparison of Joint Sparse Channel Estimation and Separate Sparse Channel Estimation for Each Channel in Embodiment 1 of the Invention

Table 1 compares the joint sparse channel estimate of the present invention with individual channel sparse channel estimates for each channel. Set the signal to noise ratio to 27dB. It can be seen that when the joint sparse channel estimation is performed on eight channels by using the present invention, the position of the obtained non-zero element is consistent with the position of the non-zero element of the real channel. However, using the prior art to implement separate sparse channel estimation for 8 channels, it is impossible to accurately estimate the position of non-zero elements. Therefore, according to the theory of compressed sensing, it is necessary to estimate the position and value of 12 non-zero elements, at least 12 x 2 = 24 pilot symbols, but only = 16 pilot symbols are actually used, which is less than the number of unknown variables. Therefore, when each channel is separately subjected to sparse channel estimation, it is impossible to accurately obtain non-zero elements in the CIR sequence. s position. In addition, Table 1 also shows the performance comparison when using the present invention to perform joint sparse channel estimation for 2 out of 8 channels, 4 out of 8 channels, and 6 out of 8 channels, it is not difficult to find that The more the number of channels estimated by the joint sparse channel, the easier it is to accurately estimate the position of the channel non-zero element, indicating that the larger the size of the antenna array system, the more obvious the beneficial effect of the present invention, because it utilizes multiple sparse channels non-zero. The a priori information of the same element position, so that the position of the non-zero element can be obtained more accurately.

FIG. 6 is a comparison of mean square error performance of a single sparse channel estimation for each channel of the present invention and the prior art. According to the position of the non-zero element of the channel CIR sequence obtained according to Table 1, the value of the non-zero element is obtained. Define Mean Square Errors (MSE) as

Where is the channel estimation result. The MSE of each channel in Figure 6 for performing sparse channel estimation alone represents the average of the MSEs of the 8 channels separately performing sparse channel estimation. It is not difficult to see that the performance of joint sparse channel estimation for eight channels is much better than that of single sparse channel estimation. Similar to Table 1, FIG. 6 also shows the performance comparison of the joint sparse channel estimation for two of the eight channels, four of the eight channels, and six of the eight channels using the present invention. It can be seen that the more channels that perform joint sparse channel estimation, the better the MSE performance.

In addition, comparing the eight channel joint sparse channel estimates of the present invention with the individual sparse channel estimates using different pilot numbers, it is found that when the number of pilots used by the latter reaches = 30, for example, the pilot subcarrier index ⁵ is [4] , 8, 12, 16, 24, 27, 34, 39, 49, 74, 76, 81, 88, 101, 104, 109, 125, 129, 133, 146, 171, 189, 202, 205, 214, 222 , 234, 244, 252, 256], can accurately estimate the number of non-zero elements of the channel under the same 27dB SNR. Therefore, the method of the present invention can reduce the pilot overhead of (30 - 16) / 16 = 87.5%, and the antenna array is The larger the scale, the more significant the pilot overhead savings.

Embodiment 2:

FIG. 7 is a schematic diagram of transmission of a MIS0 multi-antenna system according to Embodiment 2 of the present invention. As shown in Figure 7, the signals transmitted by the antennas of the base station are reflected by multiple buildings and reach the mobile phone, forming multipath effects and causing intersymbol interference. Therefore, LTE and LTE-Advanced adopt OFDM, which can effectively combat wireless. Multipath effects in propagation simplify the design of the equalizer.

Figure 8 is a block diagram of an OFDM system according to a second embodiment of the present invention. The data of the base station is processed by the constellation point mapping, the insertion pilot, the subcarrier mapping, the IFFT, the insertion protection interval, and the up-conversion, and then sent to the wireless downlink channel. After reaching the mobile phone, the system performs the down-conversion, the guard interval, the FFT, and the FFT. After processing such as subcarrier demapping, joint sparse channel estimation, channel equalization, and constellation point demapping, the transmitted data is extracted. In order to enable the mobile phone to effectively distinguish after receiving pilots from different antennas, for different base station transmit antennas, the transmitted pilots must be orthogonal in the time domain, frequency domain, or code domain. It should be noted that the present invention uses channel joint sparse channel estimation instead of each channel individual channel estimation in the prior art.

1 is a flow chart of a joint sparse channel estimation method of the present invention. Referring to Figure 1, the method includes:

S1: Establish a joint sparse reconstruction model, and combine all the channels to be estimated into one joint sparse direction. In an implementation manner of this embodiment, it is assumed that the number of OFDM subcarriers is N, and the number of pilots used is AT (KM ≤ N The M different antennas of the base station use M pilot sequences that are mutually orthogonal in the frequency domain, and the pilot sequence of the ith antenna is Ρ ^(ί ), corresponding to the index of different OFDM pilot subcarriers, and

Ρ ^(ί) n = 0, i≠ j, where r represents the intersection of the two sets. Assuming that the OFDM symbol transmitted by the ith antenna of the base station is represented as χ , ΐ = 1, 2, ..., Μ, the pilot symbol sequence transmitted by the antenna is expressed as ((p(0) _; j ₌ 1, 2, ..., M. Since the base station simultaneously transmits M pilot sequences that are mutually orthogonal in the frequency domain, after receiving the signal, the mobile phone can extract the receiving guide corresponding to the transmitting antenna of the ith base station according to the position of different pilot subcarriers. The frequency sequence y(P ⁽ⁱ )), where y represents an OFDM symbol received by the handset. Definition y(p( ) _; i = 1, 2, ..., M. Each antenna of the base station corresponds to one downlink channel The relationship between the ith downlink channel transmission pilot and the reception pilot can be established as follows

y(0 = /)(0 (0^(0 + _Ί) (ι) _> i = _>2> ..., M (6) where 3 ₈ ^:«( ^ denotes a diagonal matrix of rows and columns, the pair The horns are in turn _{Element of 3⁄4χ} ω( _Ρ ω); Gaussian white noise representing the i-th downlink channel, ? 7« is an A: dimension column vector, each element is independent and obeys a complex Gaussian distribution with a mean of 0 and a variance of σ ² ; F is a pre-L column extracted from a standard Fourier matrix of N columns The Fourier matrix formed by the rows indexed; h ⁱ = [ι ^(ί) (1), ι ^(ί) (2) ., ι ^(ί) (Ζ ] ^Τ , i = 1,2, ..., Μ is the CIR sequence of the downlink channel corresponding to each antenna of the base station, where the superscript Γ indicates vector transposition. Due to the sparsity of the radio channel, most of the L elements are zero, and only a few are non-zero, where The number of zero elements is the number of multipaths of the wireless channel. The existing related literature indicates that for the same transmitted signal, the ToA of the received signals of different antennas of the base station are similar, and it can be considered that the lengths of the CIR sequences of different channels are the same, and the CIR sequence is the same. The position of the zero element in the middle is the same, and the value of the non-zero element is different.

Defining the observation matrix = D ⁱ F^ can further simplify the formula (6) as

y = A^h^ + η , i = 1,2,..., M (7) The essence of channel estimation is the process of solving with noise " ^ω ", using ^) and 4«. Compared to LS Channel estimation, sparse channel estimation can achieve the same channel estimation performance as LS using fewer pilots. Since the locations of different non-zero elements are the same, i = 1, 2, ..., Μ; ^ ι ^ω merges into a ML dimension of the joint sparse vector iv as follows

_Wl = [i ⁽¹⁾ (Z: ⁽²⁾ (U ^(M) (0), l = 1,2,... ,L Notice that the position of the non-zero element for the different i, is the same, non-zero The values of the elements are different, either the whole block is zero, or the whole block is non-zero, and M is presented as a block sparse structure, so the position of the non-zero element in M can be used to represent the position of the non-zero element in M. , the joint observations defining the M channels are as follows

Where the Z-th element block representing the column vector z, 1 = 1, 2, ..., K, and the row vector ^ is defined as follows zi = [y ⁽¹ ( , y ⁽² ( y ^(M) ( l ι = κ defines the joint observation noise n as follows

η = ΊΙ2, ..., η^] ^τ Where n is the Zth element block of the column vector n, 1 = 1, 2, ···, !, and the row vector ^ is defined as η _χ = [7 ^ω (θΌ( ² )(0,...Ό( ^Μ )(0), I = 1,2,... ,Κ The construction of the joint observation matrix β can be formed by element-by-element substitution of the matrix E of any row or column L, the first row and the column element of the matrix £■ Expressed as ^ ) with an M row, M column diagonal matrix ά ί3 ₈ {τ! ⁽¹⁾ ( ') ⁽²⁾ ( ') . ^(Μ) ( ')} instead, I = 1, 2,. .., K, j = 1, 2,..., L, constitutes a MAT row, ML column joint observation matrix where the diagonal elements represent the elements of the first row and the first column of the matrix.

The joint sparse reconstruction model can be expressed as

z = Bw + n (8) The present invention first uses the joint observation value Z and the joint observation matrix β to solve the positions of all non-zero element blocks of the joint sparse vector IV, and then separately solves the values of the non-zero elements of each channel.

S2: Solving the position of all non-zero element blocks of the joint sparse vector using the joint sparse reconstruction model. In the second embodiment of the present invention, the base station uses the joint sparse reconstruction model formula (8) to solve the positions of all the non-zero element blocks of the joint sparse vector ,. The flow refers to FIG. 4, and the method includes:

B = QG (9) where G is a diagonal matrix of ML rows and ML columns, and each diagonal element of G is a real number greater than zero, and a normalization factor corresponding to each column of β. Substituting equation (9) into equation (8),

z = QGw + n

z = Qv + n (10) Solve the position of all non-zero element blocks of the joint sparse vector M and convert them to all non-zero element blocks solved s position.

Define an anthology Λ to store the locations of non-zero element blocks that are obtained in turn. Since 1? and IV exhibit the same block sparse structure, the position of the non-zero element can be represented by the index of the non-zero element block, so that

= 1,2, from the complement of Λ Φ = {1,2,...,Z \A, find an element 'E 4>, so that

(QfQj) Qfr is the largest, specifically can be expressed as

S24: sequentially output all the elements in the ensemble as the final elements of the selection of all non-zero element blocks of the joint sparse vector, that is, the position of the non-zero element block in the obtained V, which is also the non-zero element block in the Μ The position, also = 1, 2, ..., Μ common non-zero element position. All the elements in the selection Λ are output in sequence.

S3: Solve the value of the non-zero element of each channel. Defined as a matrix consisting of columns corresponding to the elements in the selection, then the i-th downlink channel

That is, the value of the non-zero element of the obtained i-th channel.

(c) Update module for updating residuals and selections, plus 1 cycle count.

In the simulation test, the number of base station antennas is M = 8. The number of OFDM subcarriers is N = 256, and the number of pilot subcarriers = 16. Adopt QPSK modulation. Assume that the channel CIR sequence length is L = 60, of which only 5 = 12 non-zero elements, and the position of the CIR sequence is [2, 13, 21, 24, 29, 33, 41, 42, 43, 53, 54, 60]. The base station simultaneously transmits M = 8 frequency domain orthogonal pilot sequences. For the design method, please refer to one of the invention patents that we applied for before: A pilot arrangement determination method and base station, application number: 201310687413. 7, Application date: 2013 December 12th. The M = 8 frequency domain orthogonal pilot sequences used in this simulation test are shown in Table 2.

Table 2 In the second embodiment of the present invention, eight frequency-domain orthogonal pilot sequences simultaneously transmitted by the base station Pilot subcarrier index

The first downlink channel 8, 40, 48, 52, 72, 82, 99, 142, 145, 154, 158, 161, 183,

209, 212, 230 2nd downlink channel 9, 41, 49, 53, 73, 83, 100, 143, 146, 155, 159, 162, 184,

210, 213, 231 3rd downlink channel 10, 42, 50, 54, 74, 84, 101, 144, 147, 156, 160, 163,

185, 211, 214, 232 4th downlink channel 17, 25, 47, 56, 59, 63, 75, 111, 115, 130, 141, 149, 153,

174, 200, 250 5th downlink channel 12, 34, 55, 64, 67, 109, 112, 148, 173, 215, 222, 233,

238, 241, 249, 252 6th downlink channel 2, 15, 45, 58, 62, 66, 96, 103, 107, 132, 165, 181, 186,

189, 204, 206 7th downlink channel 18, 22, 33, 68, 76, 80, 88, 91, 95, 116, 133, 167, 198,

205, 229, 246 8th downlink channel 7, 79, 92, 117, 120, 152, 168, 180, 187, 197, 219, 223,

239, 243, 251, 255 Embodiment 2: Joint Sparse Channel Estimation and Comparison of Individual Sparse Channel Estimation for Each Channel

41, 51, 60

Estimation of individual sparse channels for the sixth channel 2, 7, 12, 21, 24, 26, 33, 41, 42, 49, 54,

60

Estimating the sparse channel for the seventh channel 3, 8, 10, 17, 21, 26, 36, 41, 42, 43, 50,

55, 59

2, 6, 15, 18, 20, 21, 24, 29, 32, 41, 49, single sparse channel estimation for the 8th channel

56, 60

The present invention utilizes two channels 8, 9, 10, 12, 13, 15, 21, 25, 36, 43, 44 of 8 channels for joint sparse channel estimation 50, 56, 60

The present invention performs joint sparse channel estimation using four of the eight channels 2, 10, 12, 13, 19, 21, 24, 41, 47, 50, 53, 54, 57, 60

The present invention utilizes six channels of three channels 3, 6, 7, 13, 14, 23, 29, 33, 40, 41, 42, for joint sparse channel estimation, 43, 51, 53, 60

The present invention utilizes 8 channels to combine sparse letters 2, 13, 21, 24, 29, 33, 41, 42, 43, 53, 54.

After receiving the pilot sequence sent by the base station, the mobile phone needs to estimate the values of the non-zero elements of the eight downlink channels and the values of the non-zero elements. Table 3 compares the multiple channel joint sparse channel estimates of the present invention with individual channel sparse channel estimates for each channel. Set the signal to noise ratio to 27dB. It can be seen that when the joint sparse channel estimation is performed on eight channels by the present invention, the position of the obtained non-zero element is consistent with the position of the non-zero element of the real channel. However, using the prior art to implement separate sparse channel estimation for 8 channels, it is impossible to accurately estimate the position of non-zero elements. Therefore, according to the theory of compressed sensing, it is necessary to estimate the position and value of 12 non-zero elements, at least 12 X 2 = 24 pilot symbols, but only = 16 pilot symbols are actually used, which is less than the number of unknown variables. Therefore, when the channel is separately subjected to sparse channel estimation, the non-zero elements in the CIR sequence cannot be accurately obtained. s position. In addition, Table 3 also shows the performance comparison when using the present invention to perform joint sparse channel estimation for 2 out of 8 channels, 4 out of 8 channels, and 6 out of 8 channels, which is not difficult to find. The more the number of channels estimated by the joint sparse channel, the easier it is to accurately estimate the position of the non-zero element of the channel, indicating that the larger the size of the antenna array system, the more obvious the beneficial effect of the present invention, because it utilizes multiple sparse channels non-zero. The a priori information of the same element position can thus obtain the position of the non-zero element more accurately. FIG. 9 is a comparison of the mean square error performance of the single sparse channel estimation for each channel of the second embodiment of the present invention. According to the position of the non-zero element of the channel CIR sequence obtained according to Table 3, the value of the non-zero element is obtained. Define Mean Square Errors (MSE) as

Where is the channel estimation result. The MSE of each channel in Figure 9 for performing sparse channel estimation alone represents the average of the MSEs of the 8 channels separately performing sparse channel estimation. It is not difficult to see that the performance of joint sparse channel estimation for eight channels is much better than that of single sparse channel estimation. Similar to Table 3, the performance comparison of the joint sparse channel estimation for two of the eight channels, four of the eight channels, and six of the eight channels using the present invention is also shown in FIG. It can be seen that the more channels that perform joint sparse channel estimation, the better the MSE performance.

In addition, comparing the eight channel joint sparse channel estimates of the present invention with the individual sparse channel estimates using different pilot numbers, it is found that when the number of pilots used by the latter reaches = 28, the same 27 dB SNR condition can be obtained. The number of non-zero elements of the channel is accurately estimated. Therefore, the method of the present invention can reduce the pilot overhead of (28-16)/16 = 75%, and the larger the size of the antenna array system, the more significant the pilot overhead saved.

A person skilled in the art can understand that all or part of the process in the above embodiments can be implemented, and the related hardware can be instructed by a computer program, and the program can be stored in a computer readable storage medium, and the program is executed. At the time, the flow of the embodiment of each method as described above may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (Random Access Memory).

The above is only a preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, and thus equivalent changes made in the claims of the present invention are still within the scope of the present invention.

Claims

Claim

A joint sparse channel estimation method, comprising the steps of:

S3: Obtain a value of a non-zero element of each of the channels.

2. The method according to claim 1, wherein in the step S2, the method further comprises the following steps:

S24: All elements in the ensemble are sequentially output as positions of all the non-zero element blocks of the joint sparse vector.

3. The method according to claim 1, wherein: in the step S1, the joint sparse reconstruction model is represented as == ^^ + ·?ϊ, where one channel defined as the model The joint observation is the joint observation noise, W is its joint sparse vector, and β is its joint observation matrix.

4. The method according to claim 3, wherein: the joint sparse vector W is: w = [w ^T _1} .. wlf , where 3⁄4 represents a column element, the first element block, ϊ· ― 1 _? L., defined as:

Degree, ^ denotes the number of antennas of the base station, 3⁄4.·:€· indicates the rush of the ίth channel corresponding to the ith antenna of the base station The weight request response sequence, = 1^ . , represents the fth element of Ji.

5. A joint sparse channel estimation apparatus, comprising:

The apparatus according to claim 5, wherein the joint sparse vector calculation unit further comprises: an initialization module, configured to initialize joint observation values whose residuals are joint sparse reconstruction models, and the joint sparse reconstruction model Each column of the joint observation matrix is normalized, the initial selection is an empty set, and the number of loops is set to 0;

Update module, used to update residuals and selections, plus 1 cycle number;

7. Apparatus according to claim 5 wherein: said joint sparse reconstruction model is represented as z = Bw + n, wherein z is defined as the joint observation of the M channels of said model, ? ι is the joint observation noise, w is its joint sparse vector, and S is its joint observation matrix.

8. The apparatus according to claim 7, wherein: the joint sparse vector w is: w = [w^^ _; ] ^Γ , where 3⁄4i represents the first element block of the column vector, I = 1 _; M3⁄4 Defined as w = ^ ^ω (ΐχ ^{ω ω} (1), ..,., 3⁄4 ^{( )} ( t I = 1,2, -. L, Claim

£ denotes the channel length, M denotes the number of antennas of the base station, _ft io denotes the impulse response sequence of the _ίth channel corresponding to the antenna of the base station, f ₌ i _(2j ,, . , ' denotes the first of ft® element.

A joint sparse channel estimation system, comprising: providing an apparatus according to any one of claims 5-8 in an uplink transmission or a downlink transmission of the system.

10. The system according to claim 9, wherein the uplink transmission comprises: the data of the mobile terminal sequentially passes through constellation point mapping, fast Fourier transform, insertion pilot, subcarrier mapping, inverse fast Fourier transform, and insertion guard interval. After being up-converted, after being sent to the wireless channel, after reaching the base station, it is subjected to down-conversion, removal of guard interval, fast Fourier transform, sub-carrier demapping, joint sparse channel estimation, channel equalization, inverse fast Fourier transform, and constellation point de-mapping. , extract the send data.

The system according to claim 9, wherein the downlink transmission comprises: the data of the base station is sequentially subjected to constellation point mapping, insertion pilot, subcarrier mapping, inverse fast Fourier transform, insertion guard interval, and up-conversion. After transmitting to the wireless channel, after reaching the mobile phone, the transmission data is extracted after down-conversion, de-protection interval, fast Fourier transform, sub-carrier demapping, joint sparse channel estimation, channel equalization, and constellation point de-mapping.