CN102201890B

CN102201890B - Data transmitting method and device

Info

Publication number: CN102201890B
Application number: CN201010147636.0A
Authority: CN
Inventors: 刘晟; 杜颖钢
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2010-03-25
Filing date: 2010-03-25
Publication date: 2014-08-20
Anticipated expiration: 2030-03-25
Also published as: CN102201890A

Abstract

The invention discloses a data transmitting method, which comprises the following steps of: calculating a weighted sum mean square error of each path of data stream waiting to be transmitted according to channel information to obtain an intermediate quantity of a pre-coding vector of each path of data stream and an intermediate quantity of a balance vector of each path of data stream; performing iterative operation on the intermediate quantity of the pre-coding vector and the intermediate quantity of the balance vector to obtain a pre-coding vector of each path of data stream; pre-coding each path of data stream with the pre-coding vector; and transmitting each path of pre-coded data stream. The invention further discloses a data transmitting device. By adopting the method and the device, the problem of the occurrence of wrong floor can be solved, and the error rate performance is improved.

Description

Data transmitting method and device

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a data transmission method and apparatus.

Background

The MIMO (multiple Input multiple Output) technology is a major breakthrough of the smart antenna technology in the field of wireless mobile communication, and the technology can improve the capacity and the spectrum utilization rate of a communication system by multiples without increasing the bandwidth, and is a key technology adopted by a new generation of mobile communication system. The transmitting end and the receiving end of the MIMO system are both provided with a plurality of antennas. When transmitting signals, the transmitting end may transmit a plurality of independent signals through different transmitting antennas in an SM (Spatial Multiplexing ) manner, and the receiving end may receive signals from different receiving antennas and obtain signals transmitted by the transmitting end from the received signals.

However, when the MIMO technology is not used between adjacent cells, the Interference between adjacent cells is difficult to deal with, and there have been considerable research in both academic circles and industrial circles, for example, using Frequency planning, FFR (Fractional Frequency Reuse, partial Frequency Reuse, etc.) to avoid Interference, or using Interference Cancellation techniques, such as PIC (Parallel Interference Cancellation), SIC (Successive Interference Cancellation, etc.) to reduce the influence of Interference, and so on. When the cells adopt the MIM O technology, the interference situation is more complicated, and the existing MU-MIMO (multi user Multiple-Input Multiple-Output, multi user Input-Output), Beamforming (antenna beam forming) and other technologies can avoid the interference among Multiple users, so as to realize multi-user resource sharing, but these technologies stay in the same cell, and the inter-cell interference is still the main factor that limits the capacity of the cellular system. The multi-cell combined MU-MIMO and Beamforming can reduce interference, but all transmitting nodes need to know all channel information and data information to be transmitted, and the problems of scheduling and transmission delay requirements and the like are solved, so that the requirement on Backhaul transmission between the transmitting nodes is very high, and especially under the trend that the network transmission data volume is larger and larger, the requirement is difficult to meet. In the uplink direction, each mobile terminal is completely independent, and one mode is to perform joint data reception processing between each base station, and eliminate interference between users and separate signals of each user through a joint detection mode. The bandwidth required by the interference cancellation in this way is larger than the bandwidth required by the downlink data to be transmitted for sharing, which also causes high cost in an actual system.

In the existing scheme, only respective channel information is shared among transmitting points, but data to be transmitted is not shared, and then interference from all other transmitting nodes is automatically gathered to a signal space which can be processed by a corresponding receiver by adopting an interference alignment (interference alignment) method, so that interference elimination is realized. However, the linear interference alignment technology is still in the initial stage of research at present, and how to effectively realize the interference alignment of any number of users and any number of antennas is not solved.

The prior art provides an interference alignment method, which uses a time extension method to achieve interference alignment, and achieves KM/2 degree of freedom under the configuration of M transmitting antennas of each user of K users, thereby advancing a great step in the interference processing direction. In order to prove that the degree of freedom of KM/2 can be achieved, the method also provides a precoding mode of each transmitting node.

However, in the above prior art, two arbitrary continuous channels are required to be completely independent, and in order to achieve the degree of freedom of KM/2, the two channels need to be infinitely extended in the time domain, which is difficult to implement in practice. In addition, the linear interference alignment method can effectively cancel the interference, so as to obtain the effect of maximizing SIR (Signal to interference Ratio), but because the scheme mainly aims at the interference, the anti-Noise performance is general, so that the ideal performance can be obtained only under the condition of high SNR (Signal to Noise Ratio), but the performance is not good under the condition of medium and low SNR.

Aiming at the defects of the prior art, the prior art also provides two iterative algorithms, so that better detection performance can be obtained under the condition of medium and low SNR.

However, in the process of implementing the present invention, the inventor finds that the iterative algorithm in the prior art still has the following disadvantages:

the first iterative algorithm of the prior art is not good enough to either minimize Interference power at the receiver and leakage power at the transmitter alternately, or maximize SINR (Signal to Interference plus Noise Ratio) at the receiver and SLNR (Signal to Interference Noise Ratio) at the transmitter. The error floor (error floor) in the algorithm occurs earlier, and the performance requirement cannot be met at a lower signal-to-noise ratio. The second iterative algorithm in the prior art mainly utilizes the characteristic of the symmetry of uplink and downlink channels of TDD (Time Division Duplex), and although the channel information is relatively accurate, it is not applicable to FDD (Frequency Division Duplex) systems.

Disclosure of Invention

The embodiment of the invention provides a data transmitting method, which improves the error rate performance and comprises the following steps:

calculating the weighted sum-mean-square error of each path of data stream to be transmitted according to the channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream;

performing iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the equalization vectors to obtain precoding vectors of each path of data stream;

carrying out precoding processing on each path of data stream by adopting the precoding vector;

and transmitting each path of data stream after precoding processing.

The embodiment of the present invention further provides a data transmitting apparatus, for avoiding the occurrence of error floor, and improving the error rate performance, the apparatus includes:

the calculation module is used for calculating the weighted sum-mean-square error of each path of data stream to be transmitted according to the channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream;

the iteration module is used for carrying out iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the balance vectors to obtain precoding vectors of each path of data stream;

the precoding module is used for precoding each path of data stream by adopting the precoding vector;

and the transmitting module is used for transmitting each path of data stream after precoding processing.

In the embodiment of the invention, the weighted sum-mean-square error of each path of data stream to be transmitted is calculated according to channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream; performing iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the equalization vectors to obtain precoding vectors of each path of data stream; the defect that the domain expansion is too long to cause impracticality when the interference alignment is realized by using a time expansion method in the prior art can be avoided, the interference alignment process is completed through rapid convergence under the condition of limited resources, better performance is obtained compared with an iterative algorithm in the prior art, the early occurrence of error floor is avoided, and the performance requirement can be met under the condition of lower signal to noise ratio; meanwhile, the method is widely applied to TDD systems and FDD systems.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts. In the drawings:

FIG. 1 is a flow chart of a data transmission method according to an embodiment of the present invention;

FIG. 2 is a flow chart of an iterative algorithm process according to an embodiment of the present invention;

FIG. 3 is a diagram illustrating the BER performance of the MIMO interference channel of four users according to an embodiment of the present invention;

FIG. 4 is another diagram illustrating the BER performance of the MIMO interference channel of four users according to the embodiment of the present invention;

FIG. 5 is a diagram illustrating the BER performance of the MIMO interference channel of three users according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of BER performance of the wsum-MSE _ IA algorithm at different iteration times according to the embodiment of the present invention;

FIG. 7 is a flow chart of another iterative algorithm process in accordance with an embodiment of the present invention;

fig. 8 is a schematic structural diagram of a data transmitting apparatus according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.

An embodiment of the present invention provides a data transmitting method, a processing flow of which is shown in fig. 1, and the method may include:

step 101, calculating the weighted sum-mean-square error of each path of data stream to be transmitted according to channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream;

102, performing iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the equalization vectors to obtain precoding vectors of each path of data stream;

103, carrying out precoding processing on each path of data stream by adopting the precoding vector;

and step 104, transmitting each path of data stream after precoding processing.

As can be known from the process shown in fig. 1, in the embodiment of the present invention, the sum-mean-square error after weighting of each path of data stream to be transmitted is calculated according to the channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream; performing iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the equalization vectors to obtain precoding vectors of each path of data stream, so that the defect that the domain expansion is too long to be practical when interference alignment is realized by using a time expansion method in the prior art can be avoided, the interference alignment process is completed by fast convergence under the condition of limited resources, better performance is obtained compared with an iterative algorithm in the prior art, the early occurrence of error floors is avoided, and the performance requirement can be met under the condition of lower signal to noise ratio; meanwhile, the method is widely applied to TDD systems and FDD systems.

In specific implementation, the flow shown in fig. 1 may consider the following K user MIMO interference channels:

suppose there are K pairs of transmitter-receiver, each transmitter has M transmitting antennas, each receiver has N receiving antennas, all transmitters know all channel information, each transmitter does not know the data stream to be transmitted by other transmitters, and user M_kHas L_kPath data stream, m_kE {1,2, …, K }, there being a totalThe power of each data stream is P_i(i ∈ {1,2, …, L }), then:

in the received multiple data streams, the kth data stream is:

<math> <mrow> <msub> <mi>y</mi> <mi>k</mi> </msub> <mo>=</mo> <munderover> <mi>Σ</mi> <mrow> <mi>n</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mi>n</mi> <mo>&NotEqual;</mo> <mi>k</mi> </mrow> <mi>L</mi> </munderover> <msub> <mi>H</mi> <mrow> <msub> <mi>m</mi> <mi>k</mi> </msub> <msub> <mi>m</mi> <mi>n</mi> </msub> </mrow> </msub> <msub> <mi>w</mi> <mi>n</mi> </msub> <msub> <mi>x</mi> <mi>n</mi> </msub> <mo>+</mo> <msub> <mi>H</mi> <mrow> <msub> <mi>m</mi> <mi>k</mi> </msub> <msub> <mi>m</mi> <mi>k</mi> </msub> </mrow> </msub> <msub> <mi>w</mi> <mi>k</mi> </msub> <msub> <mi>x</mi> <mi>k</mi> </msub> <mo>+</mo> <msub> <mi>n</mi> <mi>k</mi> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow> </math>

where k ∈ {1,2, …, L }, w ∈ {1,2, …, L }, w }_kFor a data stream x_kThe size of the precoding vector of (1) is M x 1,is composed of a transmitter m_nTo receiver m_kN x M channel matrix, N_kIs N × 1 White Gaussian Noise (AWGN) with variance σ²；

The k-th path of data flow recovered after the equalization processing is as follows:

wherein, g_kFor a data stream x_kOf size N x 1, i.e. for the received signal y_kUsing an Nx 1 receiving equaliser g_kTo recover the data stream.

In order to obtain the optimal estimation of the kth data flow shown in equation (2), in the embodiment of the present invention, a Mean Square Error (MSE) is calculated according to the channel information, that is:

the above-mentioned methods for calculating the mean square error may be various, and the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream may be obtained by calculation. In addition, since simultaneous optimization of multiple data streams is considered, the calculation and mean square error can be used as the optimization target in the embodiment of the present invention. Considering that some large MSEs may have an excessive influence on the sum-mean-square-error, the embodiment of the present invention further weights the mean-square-error of each data stream to be transmitted, and introduces a weighting systemNumber eta_kThus, the weighted sum-mean-square error of each data stream is:

wherein,η_k>0, gamma is to total powerConstrained lagrangian(Lagrange) multiplier.

Similarly, there may be multiple methods for calculating the weighted sum-mean-square error of each path of data stream to be transmitted, and the calculation may be performed to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream. According to the channel information, minimizing the weighted sum-mean-square error of each path of data stream to be transmitted, namely, respectively taking w for J_kAnd g_kIs zero, w of the minimized cost function J can be obtained_kAnd g_kThe optimal solution of (2), that is, the intermediate quantity of the precoding vectors of each path of data stream, is:

the intermediate quantity of the equalization vector of each path of data flow is as follows:

where k ∈ {1,2, …, L }, m_k∈{1,2,…,K}，I_MIs an M-dimensional identity matrix, I_NIs an N-dimensional identity matrix.

If an equalization vector g is assumed_k(k ∈ {1,2, …, L }) is known, then from equation (3), the optimal precoding solution w according to the MMSE criterion can be found_k(k ∈ {1,2, …, L }), and, similarly, assuming that the set of precoding vectors is known, an optimal equalizer solution can also be obtained, again according to the MMSE criterion. According to w_kAnd g_kIn such a coupling relationship, the embodiment of the present invention uses an iterative method to find an optimized precoding vector and an optimized equalization vector.

First, consider computing the lagrange multiplier γ:

given g_kSubstituting equation (3) into the power constraint equation can obtain:

wherein, B_kIs the Hermitian matrix of M-, satisfying:

its eigenvalue decomposition can be expressed as:

B_{k} = U_{k} D_{k} U_{k}^{H}

wherein, U_kIs a unitary matrix of dimension M, D_kIs a diagonal matrix of dimension M with diagonal elements d_kn(ii) a Order to

b_{k} = U_{k}^{H} H_{m_{k} m_{k}} g_{k},

Equation (5) can be simplified as:

and solving the formula (6) to obtain the Lagrange multiplier gamma.

In one embodiment, when the weighting factors are all equal, i.e., η is for all data streams_kWhen γ is a constant value, 1. If the algorithm converges to a pointThen, the following equations (3) and (4) can be obtained:

thus, it is possible to obtain:

order to

c_{nk} = w_{k}^{* H} H_{m_{n} m_{k}}^{H} g_{n}^{*},

Equation (7) may be rewritten as:

all equations for k from 1 to L in equation (8) are added:

it is noted that

Therefore, there are:

the iterative algorithm in the embodiment of the present invention is given below:

since the optimal precoding vector and the equalization vector have a coupling relationship, the embodiments of the present invention alternate according to the equations (3) and (4)Calculating w_kAnd g_k. Can give w_kK is 1,2, …, L is assigned a set of initial values to calculate g according to equation (4)_kK is 1,2, …, L, and then w is recalculated as in equation (3)_kHere, the weight coefficient η_kSetting the initial value to 1, and then iterating according to the following formula to obtain all eta_k：

Then normalized to satisfyIn the formula (10), i is the number of iterations, α⁽ⁱ⁾The step length is variable, and the following conditions are satisfied:

where α is₀∈(0,1]Is an initial value, c₀Is a fixed value. From equation (10), it can be seen that for users with poor MSE (i.e., large MSE), the weight is increased, so that the MSE decreases faster in the next iteration. To make the algorithm converge as soon as possible, let step size weight α, as shown in equation (11)⁽ⁱ⁾Decreases stepwise as the number of iterations increases, so that the weighting factor eta_kThe variation of (c) decreases as the number of iterations increases. Although γ can be calculated by equation (6), it can be obtained by a simpler method using the characteristic of a fixed value shown by equation (9):

since equation (12) converges only at the algorithm and η_kWhen the value is 1, w is corrected according to the formula (3) each time_kAfter the iterative update, the normalization needs to be performed again to meet the requirement of the transmission power. When all w_kAll are obtained, iteration g can be updated_kAnd repeating the process until a predetermined number of iterations is reached or a convergence condition is satisfied, i.e. w for two adjacent iterations_kAnd g_kThe value is sufficiently small.

In one embodiment, the sum-mean-square-error iterative algorithm can be considered as a special case of the above algorithm, i.e. let the ownership coefficient be constantly equal to 1: eta_k1, all k.

In summary, in the embodiment of the present invention, when performing iterative operation on the intermediate quantity of the precoding vector and the intermediate quantity of the equalization vector in step 102 to obtain the precoding vector of each channel of data stream, in specific implementation, as shown in fig. 2, an iterative algorithm for aligning sum-mean-square-difference interference of each channel of data stream after weighting may be provided, and a processing flow of the iterative algorithm may include:

step 201, initialization, including: from any one ofStarting initialization, wherein c₀And alpha₀∈(0,1]For k, 1,2, …, L,the iterative processing is carried out according to the following steps:

step 202, updating the equalization vector, namely: is calculated according to the following formulaWherein i is the number of iterations:

wherein I_NIs an N-dimensional identity matrix;

step 203, calculating a weight coefficient, namely: is calculated according to the following formulaAnd pressAnd (3) carrying out normalization:

wherein the variable step size is

Step 204, updating the precoding vector, namely: is calculated according to the following formula

Wherein I_MIs an M-dimensional identity matrix;

γ is obtained as follows:

push button

To pairCarrying out normalization;

and step 205, returning to step 202 until the convergence condition is met or the preset iteration number is reached.

The convergence of the iterative algorithm described above, i.e. η for all k, is discussed below_k1, or, considerWeighting, but almost constant, i.e. after a certain number of iterations, α, as shown in equation (11)⁽ⁱ⁾Is already small enough.

As shown in FIG. 2, in the ith iteration, for any givenk is 1,2, …, L, and the cost function J may bek is 1,2, …, the minimum is obtained at L. Thus, in obtainingNext, J may be optimized in the next iterationAnd then further reduced, such repeated iterations mean:

<math> <mfenced open='' close=''> <mtable> <mtr> <mtd> <mi>J</mi> <mrow> <mo>(</mo> <msubsup> <mi>w</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>g</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>&GreaterEqual;</mo> <mi>J</mi> <mrow> <mo>(</mo> <msubsup> <mi>w</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>g</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>&GreaterEqual;</mo> <mi>J</mi> <mrow> <mo>(</mo> <msubsup> <mi>w</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>g</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>&GreaterEqual;</mo> </mtd> </mtr> <mtr> <mtd> <mi>J</mi> <mrow> <mo>(</mo> <msubsup> <mi>w</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>g</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>&GreaterEqual;</mo> <mi>J</mi> <mrow> <mo>(</mo> <msubsup> <mi>w</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>g</mi> <mi>k</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>&GreaterEqual;</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> </mtd> </mtr> </mtable> </mfenced> </math>

thus, it is clear that the above algorithm is a monotonically decreasing function, and thus the convergence thereof can be ensured.

The above simulation results are discussed below, and the iterative algorithm for minimizing the weight and the mean square error described in the embodiment of the present invention is denoted as wsum-MSE _ IA, and is compared with other algorithms, where the iterative algorithm for minimizing the weight and the mean square error in the embodiment of the present invention is denoted as sum-MSE, a first iterative algorithm in the prior art is denoted as KVS, and a second iterative algorithm is denoted as SR.

In the simulation, the assumption conditions include: flat fading, all transmitting nodes and receiving nodes know all channel information, each algorithm experiences the same channel, QPSK modulation, the maximum number of iterations is set to 20, the total power is consistent, and the power of each stream is also set to be equal in KVS and SR algorithms.

Fig. 3 shows a comparison of the MIM O interference channel BER (Bit Error Rate) performance of four users, each node is configured with 3 antennas, and each user only transmits one stream, which is represented as: l ═ 1,1, 1. It can be seen that wsum-MSE _ IA and sum-MSE _ IA are far better than the existing KVS and SR algorithms, the sum-MSE _ IA performance is improved by 3-5 dB compared with the KVS performance, under the condition of high signal-to-noise ratio, the wsum-MSE _ IA can have 6-8 dB performance gain relative to the KVS performance gain, and the SR algorithm is also found to have poor effect, so that the algorithm is not considered in the following comparison.

Fig. 4 shows a BER performance comparison of the MIMO interference channel for four users, each node is configured with 4 antennas, the distribution of the streams is L ═ 2,2,1, fig. 5 shows a BER performance comparison of the MIMO interference channel for three users, each node is configured with 4 antennas, and L ═ 2,2, 2. It can be seen that the algorithm of the embodiment of the present invention still has a good improvement effect, and at a high signal-to-noise ratio, KVS and sum-MSE _ IA both appear to have an error floor, but wsum-MSE _ IA still can work well.

Figure 6 shows the BER performance of the wsum-MSE _ IA algorithm at different iteration numbers, it can be seen that the convergence rate is high, and the algorithm can be considered to be converged after 20 iterations.

In the above embodiment, it is initialized firstWe can also initialize firstThat is, as shown in fig. 7, the iterative process flow may include:

step 701, initialization, including: from any one ofStarting initialization, wherein c₀And alpha₀∈(0,1]For k, 1,2, …, L,the iterative processing is carried out according to the following steps:

step 702, updating the precoding vector, that is: is calculated according to the following formulaWherein i is the number of iterations:

wherein I_MIs an M-dimensional identity matrix;

γ is obtained as follows:

push button

To pairCarrying out normalization;

step 703, updating the equalization vector, that is: is calculated according to the following formula

Wherein I_NIs an N-dimensional identity matrix;

step 704, calculating a weight coefficient, that is: is calculated according to the following formulaAnd pressAnd (3) carrying out normalization:

wherein the variable step size is

Step 705, returning to step 702 until the convergence condition is met or the preset iteration number is reached.

Those skilled in the art will appreciate that all or part of the steps in the method according to the above embodiments may be implemented by hardware instructions related to a program, where the program may be stored in a computer-readable storage medium, and when executed, the program may include all or part of the steps in the method according to the above embodiments, and the storage medium may include: ROM, RAM, magnetic disks, optical disks, and the like.

The embodiment of the invention also provides a data transmitting device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to the data transmission method, the implementation of the device can refer to the implementation of the method, and repeated details are not repeated.

An embodiment of the present invention provides a data transmitting apparatus, whose structure is shown in fig. 8, where the apparatus may include:

a calculating module 801, configured to calculate a sum-mean-square error after weighting of each path of data stream to be transmitted according to channel information to obtain an intermediate quantity of a precoding vector of each path of data stream and an intermediate quantity of a balance vector of each path of data stream;

an iteration module 802, configured to perform iterative operation on the intermediate quantity of the precoding vector and the intermediate quantity of the equalization vector to obtain a precoding vector of each path of data stream;

a precoding module 803, configured to perform precoding processing on each path of data stream by using the precoding vector;

a transmitting module 804, configured to transmit each path of data stream after precoding processing.

In one embodiment, the calculation module 801 may be specifically configured to: and according to the channel information, minimizing the weighted sum-mean-square error of each path of data stream to be transmitted, and obtaining the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream.

In one embodiment, the weighted sum mean square error of each data stream to be transmitted may be:

wherein K is the logarithm of the transmitter and receiver, M is the number of transmitting antennas of each transmitter, N is the number of receiving antennas of each receiver, all transmitters acquire all channel information, and L_kFor user m_kNumber of data flow paths, m_k∈{1,2,…,K}，P_i(i ∈ {1,2, …, L }) is the power of each data stream;

k∈{1,2,…,L}，w_kfor a data stream x_kThe size of the precoding vector of (1) is M x 1,is composed of a transmitter m_nTo receiver m_kN x M channel matrix, N_kIs Nx 1 Gaussian white noise with variance σ²；

g_kFor a data stream x_kThe size of the equalization vector of (1) is N × 1;

η_k>0, gamma is to total powerConstrained lagrange multipliers are performed.

In one embodiment, the intermediate quantity of the precoding vectors of each path of data stream is:

wherein I_MIs an M-dimensional identity matrix;

wherein I_NIs an N-dimensional identity matrix.

In one embodiment, the iteration module 802 may be specifically configured to:

from any one ofStarting initialization, wherein c₀And alpha₀∈(0,1]For k, 1,2, …, L,the iterative processing is carried out according to the following steps:

according to

ComputingWhere I is the number of iterations, I_NIs an N-dimensional identity matrix;

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

According to

ComputingWherein I_MIs an M-dimensional identity matrix;

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

and returning to execute the iteration processing step until the convergence condition meets or reaches the preset iteration times.

In one embodiment, the iteration module 802 may be specifically configured to:

according to

ComputingWhere I is the number of iterations, I_MIs an M-dimensional identity matrix;

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

according to

ComputingWherein I_NIs an N-dimensional identity matrix;

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

In summary, in the embodiment of the present invention, the weighted sum-mean-square error of each path of data stream to be transmitted is calculated according to the channel information to obtain the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream; performing iterative operation on the intermediate quantity of the precoding vectors and the intermediate quantity of the equalization vectors to obtain precoding vectors of each path of data stream; carrying out precoding processing on each path of data stream by adopting the precoding vector; transmitting each path of data flow after precoding processing, so that the defect that the domain expansion is too long to be practical when the interference alignment is realized by using a time expansion method in the prior art can be avoided, the interference alignment process is completed by fast convergence under the condition of limited resources, better performance than an iterative algorithm in the prior art is obtained, the occurrence of error floor is avoided earlier, and the performance requirement can be met under the condition of lower signal-to-noise ratio; meanwhile, the method is widely applied to TDD systems and FDD systems.

In the embodiment of the invention, a weighting factor is introduced into a minimization and mean square error algorithm; under the control of variable step length, the iterative algorithm can ensure monotonic decrease and convergence; the optimal precoding vector and the optimal equalization vector can be found through an iterative algorithm, so that the problem of interference alignment is solved, and the performance of BER is improved.

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims

1. A method for transmitting data, the method comprising:

calculating the weighted sum-mean-square error of each path of data stream to be transmitted according to the channel information;

obtaining the intermediate quantity of the precoding vector of each path of data flow and the intermediate quantity of the balance vector of each path of data flow by calculating the sum-mean-square error;

and transmitting each path of data stream after precoding processing.

2. The method of claim 1, wherein the weighted sum mean square error of each data stream to be transmitted is calculated according to channel information; obtaining the intermediate quantity of the precoding vectors of each path of data stream and the intermediate quantity of the equalization vectors of each path of data stream by calculating the sum-mean-square error, comprising the following steps:

and minimizing the weighted sum-mean-square error of each path of data stream to be transmitted according to the channel information, and obtaining the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream by calculating the sum-mean-square error.

3. The method of claim 1 or 2, wherein the weighted sum mean square error of each data stream to be transmitted is:

wherein K is the logarithm of the transmitter and receiver, M is the number of transmitting antennas of each transmitter, N is the number of receiving antennas of each receiver, all transmitters acquire all channel information, and L_kFor user m_kThe number of lanes of the data stream,path data stream, m_k∈{1,2,…,K}，P_i(i ∈ {1,2, …, L }) is the power of each data stream;

g_kFor a data stream x_kThe size of the equalization vector of (1) is N × 1;

η_kis a weighting coefficient, η_k>0, gamma is to total powerConstrained lagrange multipliers are performed.

4. The method of claim 3, wherein the intermediate quantity of the precoding vectors for each data stream is:

wherein I_MIs an M-dimensional identity matrix;

wherein I_NIs an N-dimensional identity matrix.

5. The method of claim 4, wherein iteratively operating the intermediate quantity of the precoding vector and the intermediate quantity of the equalization vector to obtain the precoding vector for each data stream, comprises:

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

Mean square error

And (3) the k-th path of data flow recovered after equalization processing is as follows:

wherein, y_kIs the k-th data stream in the received multi-path data stream;

according to

ComputingWherein I_MIs an M-dimensional identity matrix;

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

6. The method of claim 4, wherein iteratively operating the intermediate quantity of the precoding vector and the intermediate quantity of the equalization vector to obtain the precoding vector for each data stream, comprises:

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

according to

ComputingWherein I_NIs an N-dimensional identity matrix;

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

Mean square error

wherein, y_kIs the k-th data stream in the received multi-path data stream;

7. A data transmitting apparatus, characterized in that the apparatus comprises:

the calculation module is used for calculating the weighted sum-mean-square error of each path of data stream to be transmitted according to the channel information, and acquiring the intermediate quantity of the precoding vector of each path of data stream and the intermediate quantity of the equalization vector of each path of data stream by calculating the sum-mean-square error;

8. The apparatus of claim 7, wherein the computing module is specifically configured to:

9. The apparatus of claim 7 or 8, wherein the weighted sum mean square error of each data stream to be transmitted is:

where K is the logarithm of the transmitter-receiver, M is the number of transmit antennas per transmitter, N is the number of receive antennas per receiver, all transmittersAll channel information, L, is known_kFor user m_kThe number of lanes of the data stream,path data stream, m_k∈{1,2,…,K}，P_i(i ∈ {1,2, …, L }) is the power of each data stream;

g_kFor a data stream x_kThe size of the equalization vector of (1) is N × 1;

10. The apparatus of claim 9, wherein the intermediate amount of the precoding vector for each data stream is:

wherein I_MIs an M-dimensional identity matrix;

wherein I_NIs an N-dimensional identity matrix.

11. The apparatus of claim 10, wherein the iteration module is specifically configured to:

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

Mean square error

wherein, y_kIs the k-th data stream in the received multi-path data stream;

according to

ComputingWherein I_MIs an M-dimensional identity matrix;

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

12. The apparatus of claim 10, wherein the iteration module is specifically configured to:

according to

Obtaining gamma;

push button

To pairCarrying out normalization;

according to

ComputingWherein I_NIs an N-dimensional identity matrix;

according to

ComputingAnd press

Normalization is performed, wherein the variable step size is

Mean square error

wherein, y_kIs the k-th data stream in the received multi-path data stream;