CN1980088A - Uplink receiving method and device in a distributed antenna mobile communication system - Google Patents

Abstract

The present invention relates to an uplink receiving method and device in a distributed antenna mobile communication system, wherein the device comprises at least two distributed wireless access units, each of which comprises at least two antennas, and the device further comprises: a covariance matrix calculation and eigendecomposition module, which is used to calculate the covariance matrix of baseband data corresponding to each distributed wireless access unit and perform eigendecomposition on the covariance matrix to obtain an eigenvector matrix; a projection value calculation module, which calculates projection values using the covariance matrix and the eigenvector matrix; a eigenvector selection module, which selects eigenvectors corresponding to a specific number of large projection values; and a synthesized signal forming module, which performs weighted operation on baseband data using the selected eigenvectors as weighted vectors to form a synthesized signal.

Description

Uplink receiving method and device in a distributed antenna mobile communication system

technical field

The invention relates to a method for receiving an uplink of a distributed antenna mobile communication, and relates to a method for receiving an uplink of a mobile communication in a generalized distributed antenna system (GDAS).

Background technique

With the development of communication technology and people's requirements for high-speed data communication, a technology based on MIMO (Multiple Input Multiple Output) has been applied. In MIMO, multiple different data streams are transmitted from different antennas, and multiple antennas at the receiving end receive and demodulate the data. This multiplexing-based spatial multiple access method can double the spectrum efficiency. However, there are three key factors that limit the improvement of MIMO performance. First, signals received in uplink in normal mobile communications have spatial correlation. Second, the resource limitation of mobile communication spectrum in the future makes the communication system work in a higher frequency band, and the path loss of the system is relatively large, especially when the carrier frequency of the signal is greater than 3GHz, which is not conducive to the design of mobile communication systems including MIMO systems ; Third, there is shadow fading in the mobile communication system. When the mobile terminal is in the deep shadow fading area, the signal-to-noise ratio of the MIMO system is low, and the bit error rate of the signal will increase greatly.

In order to overcome the effects of path loss and shadow fading and make the base station antenna as close as possible to the mobile terminal, a technology called Distributed Antenna System (DAS) has been applied. By centrally processing the sending and receiving signals of several antennas distributed in space and separated by a certain distance, the effects of spatial multiplexing, macro-diversity, micro-diversity and low path loss can be obtained.

In a generalized distributed antenna system, referring to Figure 1, each distributed radio access unit (RAU) has multiple antenna arrays composed of antenna units, and the received signals of all antenna arrays at multiple distributed radio access units It is transmitted to the transceiver base station through optical fiber or coaxial cable for centralized processing, where MT stands for mobile station, but due to the large number of antennas, the amount of data processed by the base station is very large, which affects the feasibility of the hardware design of the base station equipment, and the number of antennas used is due to the equipment The processing power of the computer is also limited, which leads to the limitation of system performance improvement.

Contents of the invention

The purpose of the present invention is to provide a method and device for selecting and receiving characteristic beams for a generalized distributed antenna system, which can achieve the advantages of integrating MIMO, smart antennas and DAS while reducing the amount of calculation of the system, improving the processing speed, and reducing the system cost. The complexity of the requirements. In order to achieve the purpose of the present invention, adopt following technical scheme:

The present invention provides an uplink receiving method in a mobile communication system, wherein the communication system includes at least two distributed wireless access units, each distributed wireless access unit includes at least two antennas, and the receiving method includes A synthetic signal forming method, the synthetic signal forming method comprising the steps of:

1) calculating the covariance matrix corresponding to the baseband data of each distributed wireless access unit;

2) performing eigendecomposition of the covariance matrix to obtain an eigenvector matrix;

3) Utilize the covariance matrix and the eigenvector matrix to calculate the projection value;

4) Selecting a feature vector corresponding to a specific number of large projection values as a weighted vector formed by the synthesized signal;

5) Forming a composite signal using the baseband data and said weighting vector.

A step of iteratively updating the covariance matrix may also be included between the steps 1) and 2).

Between the step 2) and the step 3), a step of storing the eigenvector matrix and using it in a delayed manner may also be included.

use the formula $R_i\left(k\right)=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\Σ}}}{x}_i(j,k)x_i{(j,k)}^h$ Calculate the covariance matrix, where R _i (k) is the covariance matrix corresponding to the kth data block of the distributed wireless access unit i, the superscript H represents the conjugate transposition of the matrix or vector, and x _i (j , k) represents the column vector output of the jth sample of the distributed wireless access unit i in the kth data block, and J is the number of samples required to calculate the covariance matrix of the current data block.

use the formula $P_{\mathrm{ml}}\left(k\right)=u_{\mathrm{ml}}^h(k-1)R_m\left(k\right)u_{\mathrm{ml}}(k-1)$ Calculate the projection value, wherein P _ml (k) is the lth of the eigenvector matrix of the iterative covariance matrix of the k-1th data block output baseband data of the kth data block of the mth distributed wireless access unit The projection value of the column vector, u _ml (kl) is the feature vector of the lth column of the iterative covariance matrix corresponding to the k-1th data block, and R _m (k) is the corresponding distributed wireless access unit m at the kth The covariance matrix of the data block, the superscript H indicates the conjugate transpose of the matrix or vector.

Use the formula R _i (k)=(1-β _i )R _i (k-1)+β _i R _i (k) to iteratively update the covariance matrix, where β _i is the weighting coefficient and R _i (k) is the corresponding The covariance matrix of the kth data block of the distributed wireless access unit i.

The present invention also provides a wireless receiving system, wherein the wireless receiving system includes at least two distributed wireless access units, each distributed wireless access unit includes at least two antennas, and the wireless receiving system further includes:

The covariance matrix calculation and eigendecomposition module is used to calculate the covariance matrix corresponding to the baseband data of each distributed wireless access unit and perform eigendecomposition of the covariance matrix to obtain the eigenvector matrix;

A projection value calculation module, which uses the covariance matrix and the eigenvector matrix to calculate a projection value;

select the eigenvector module, select the eigenvectors corresponding to a certain number of large projection values;

The synthesized signal forming module uses the feature vector selected in the feature vector selection module as a weight vector to perform a weighting operation on the baseband data to form a synthesized signal.

After the covariance matrix calculation and eigendecomposition module calculates the covariance matrix of the current data block according to the baseband data, iteratively updates the covariance matrix.

The covariance matrix calculation and eigendecomposition module stores the obtained eigenvector matrix after performing eigendecomposition and uses it in a delayed manner.

As can be seen from the above scheme, the present invention includes iterative updating of the covariance matrix of the receiving array at each RAU place, first calculates the covariance matrix averaged over the time of the array output of the current data block at each RAU place, and then calculates the previous data block The calculated covariance matrix and the covariance matrix calculated by the current data block are weighted and summed, so as to complete the iterative update of the covariance matrix at each RAU. Since the angle of arrival and angle spread of the spatial direction from the mobile terminal to each RAU is slowly varying, the iterative update of the covariance matrix can effectively track the time-varying channel.

The present invention also includes eigendecomposition and delayed storage of the covariance matrix of the receiving array at each RAU. Eigen-decompose the covariance matrix calculated at all RAUs, and store all eigenvectors and eigenvalues. These eigenvectors are delayed to the next data block for utilization as weight vectors for beamforming. Because the result of eigendecomposition is used in the next data block, the real-time requirement for calculation is reduced. A significant advantage of the present invention is that both the training sequence and the data can be used to solve the covariance matrix, which facilitates high-efficiency data transmission.

The present invention is proposed for the uplink receiving baseband signal processing of the generalized distributed antenna system. It considers the real-time performance of the system, reduces the degree of freedom of the system and takes into account the overall performance requirements of the system, improves the beamforming efficiency, and greatly reduces the The amount of calculation is reduced, and the complexity of the system is reduced, which can save equipment costs and increase capacity.

Other characteristics, objects and effects of the present invention will become clearer and easier to understand through the following description of preferred embodiments of the present invention in conjunction with the accompanying drawings.

Description of drawings

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings, in which:

FIG. 1 is a structural diagram of a mobile communication system applying a generalized distributed antenna system;

Fig. 2 is the structural diagram of the synthetic signal formation system that is used for generalized distributed antenna system among the present invention:

Fig. 3 is the implementation flowchart of covariance matrix calculation and eigendecomposition module in the present invention;

Fig. 4 is the flow chart of projection value calculation and sorting in the present invention;

In all the above drawings, the same reference numerals indicate the same, similar or corresponding features or functions.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

Figure 3 vividly shows the process of calculating the covariance matrix and the eigendecomposition of the matrix by the iterative method with the implementation block diagram and mathematical formulas. In this embodiment, there are two RAUs (RAU1 and RAU2), and each RAU has an example of three antennas or antenna sub-arrays. The mobile station has two transmit antennas, and each antenna transmits different data. Then, there are 100 samples in time to transmit a data block, and each sample is composed of two different signal space multiplexing. The length of the training sequence has 2 samples, corresponding to sample 1 and sample 2. The remaining 98 samples transmit unknown information data, corresponding to samples 3 to 100. Of course, the present invention is not limited to the above-mentioned mobile stations, and any data sent by any kind of mobile stations is suitable for the method of the present invention.

The first step in implementing the invention is to iteratively calculate the covariance matrix and perform eigendecomposition on it.

First, the baseband data stream is input into the covariance matrix calculation and eigendecomposition module, which uses the calculation formula $R_i\left(k\right)=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\Σ}}}{x}_i(j,k)x_i{(j,k)}^h$ Calculate the covariance matrix of the current data block. In the formula, R _i (k) is the covariance matrix of the kth data block corresponding to the distributed wireless access unit i, and the superscript H of the mathematical symbol represents the conjugate transformation of the matrix or vector In this example, _Xi (j, k) represents the column vector output of the i-th sample of RAUi in the k-th data block. When calculating RAU1, i=1, and J is to calculate the covariance matrix of the current data block The number of samples required, J equals 100 in this example.

The formation process of the covariance matrix is described above, and the iterative update process of the covariance matrix is described below. The covariance matrix is iteratively updated using the formula R _i (k)=(1-β _i )R _i (k-1)+β _i R _i (k), where the value of the weighting coefficient β _i is 0.1 or other Figures determined on a case-by-case basis.

Finally, the eigendecomposition is performed on the iterative covariance matrix of the current data block, and its formula is $R_i\left(k\right)=u_i\left(k\right){\mathrm{\Λ}}_i\left(k\right)u_i^h\left(k\right),$ Where U _i (k) is the eigenvector matrix of the iteration covariance matrix corresponding to the kth data block. In this example, 3 columns of eigenvectors are obtained, and each column of eigenvectors has 3 elements. These 3 columns of feature vectors are stored and delayed until the next data block is used. Of course, the method of the present invention can also be used for real-time calculation of feature vectors without delay.

The parameter calculation method of RAU2 is the same as that of RAU1.

The second step in implementing the present invention is projection value calculation and sorting. This step is carried out in the projection value calculation module, wherein the projection value calculation module receives the iterative covariance matrix and its eigenvector matrix data calculated by the covariance matrix calculation and eigendecomposition module. The detailed calculation process is introduced below:

Fig. 4 shows a block diagram and mathematical expressions for realizing calculation and sorting of projection values with M RAUs and each RAU with L antennas. The projection value calculation module adopts the formula

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; ml</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; ml</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; ml</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Calculate the projection value, wherein P _ml (k) is the lth of the eigenvector matrix of the iterative covariance matrix of the k-1th data block output baseband data of the kth data block of the mth distributed wireless access unit The projection value of the column vector, u _ml (k-1) is the characteristic vector of the lth column of the iteration covariance matrix corresponding to the k-1th data block, R _m (k) is the covariance of the corresponding distributed wireless access unit m Variance matrix, the superscript H denotes the conjugate transpose of a matrix or vector.

Now gather all the projection values calculated by RAU together, and select the eigenvectors corresponding to the first N large projection values. In this embodiment, sorting is adopted to select the eigenvectors corresponding to the first N large projection values. , it is also possible to directly compare the sizes without sorting first, select the largest one each time, and then select the largest one from the remaining projection values, and perform N selections, and N can be determined by the specific system complexity.

This step in this embodiment completes the calculation of the six projection values of the two RAUs, and selects one or more maximum projection values and corresponding feature vectors therefrom.

The projection process is described above, and the physical meaning of the projection is explained below. In GDAS, the direction of the signal received by each RAU has a certain angular expansion in space and is slowly changing, so adjacent data blocks are in space. There is a correlation. Therefore, the baseband data output by the RAU is projected onto the feature vector mentioned above to extract the main signal features of each antenna or antenna sub-array, the system capacity loss is very small, and the array gain is obtained at the same time. In addition, the projected value is calculated by using the delayed feature vector, which reduces the speed requirement for real-time calculation.

The third step of the present invention is synthetic signal formation.

In this implementation example, if the second step selects, for example, two feature vectors in the previous data block as the weighted vectors formed by the synthesized signal, two synthesized signals are thus formed. Let's take RAU1 and RAU2 respectively forming a synthetic signal as an example to illustrate the process of forming the synthetic signal.

First, the principle and process of the synthetic signal formation of RAU1 are given. The first data of the current data block on the time sample, that is, sample 1 forms 3 baseband data outputs at RAU1, the 1st, 2nd and 3rd baseband information data of RAU1 and the feature vector that has been selected in the corresponding step The conjugates of the 1st, 2nd and 3rd elements are respectively multiplied, and then the products are added to obtain the synthesized signal output of RAU1. The synthetic signal formation process is completed in the synthetic signal formation module. RAU2 also forms its composite signal output in the same way. All calculation formulas used above in the present invention are well known to those skilled in the art.

In the following we briefly introduce the channel estimation and symbol demodulation after signal formation based on synthesis.

In the channel estimation unit, the synthesized signal output corresponding to the training sequence can be used in the MMSE method to estimate the channel parameters after the synthesized signal is formed. In the symbol detection unit, the channel parameters are combined with the classic zero-forcing algorithm (ZF), the minimum mean square error (MMSE), the continuous interference cancellation algorithm (SUC (Successive Cancellation)), and the sequenced continuous interference cancellation algorithm OSUC ( OrderedSUC) and maximum likelihood estimation (ML) and other algorithms to complete the final symbol detection and separate the multiplexed transmitted signals, thus completing the information symbol demodulation process, the above channel estimation and symbol demodulation methods are all It is a solution commonly used in the prior art.

The combined signal selection receiving technology based on the distributed antenna system of the present invention reduces the amount of calculation compared with the use of all the existing receiving antennas. Especially when the number of RAUs and the number of antennas of each distributed antenna access unit RAU increase, the calculation amount can be saved many times.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications should also be It is regarded as the protection scope of the present invention.

Claims (14)

1. the method for receiving uplink in the mobile communication system, wherein this mobile communication system comprises at least two distributed wireless access units, each distributed wireless access unit comprises at least two antennas, described method of reseptance comprises synthetic signal formation method, it is characterized in that described synthetic signal formation method comprises the steps:

1) calculating is corresponding to the covariance matrix of the base band data of each distributed wireless access unit;

2) described covariance matrix is carried out feature decomposition, obtain feature matrix;

3) utilize described covariance matrix and feature matrix to calculate projection value;

4) the big projection value characteristic of correspondence vector of selection and given number is as the weight vectors of one tunnel signal formation of synthesizing;

5) utilize base band data and described weight vectors to form one tunnel synthetic signal.

2. the method for receiving uplink in the mobile communication system according to claim 1 is characterized in that, in described step 1) and step 2) between comprise that also iteration upgrades the step of covariance matrix.

3. the method for receiving uplink in the mobile communication system according to claim 1 is characterized in that, in described step 2) and step 3) between also comprise the step of described feature matrix being stored and delaying time and utilizing.

4. the method for receiving uplink in the mobile communication system according to claim 2 is characterized in that, in described step 2) and step 3) between also comprise the step of described feature matrix being stored and delaying time and utilizing.

5. according to the method for receiving uplink in each the mobile communication system in the claim 1 to 4, it is characterized in that, adopt formula $R_i\left(k\right)=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{x}_i(j,k)x_i{(j,k)}^H$ Calculate described covariance matrix, wherein R _i(k) be the covariance matrix of corresponding distributed wireless access unit i, the conjugate transpose of subscript H representing matrix or vector, x k data block _i(j, k) expression distributed wireless access unit i is in the column vector output of j sample of k data block, and J calculates the needed sample number of current data block covariance matrix.

6. according to the method for receiving uplink in each the mobile communication system in the claim 1 to 4, it is characterized in that, adopt formula $P_{\mathrm{ml}}\left(k\right)=u_{\mathrm{ml}}^H(k-1)R_m\left(k\right)u_{\mathrm{ml}}(k-1)$ Calculate described projection value, wherein P _Ml(k) be the projection value of k data block output of m distributed wireless access unit base band data, u at the l column vector of the feature matrix of the iteration covariance matrix of k-1 data block _Ml(k-1) be the l row characteristic vector of the iteration covariance matrix of corresponding k-1 data block, R _m(k) be the covariance matrix of corresponding distributed wireless access unit m, the conjugate transpose of subscript H representing matrix or vector k data block.

7. the method for receiving uplink in the mobile communication system according to claim 2 is characterized in that, adopts formula R _i(k)=(1-β _i) R _i(k-1)+β _iR _i(k) iteration is upgraded covariance matrix, wherein, and β _iBe weight coefficient, R _i(k) be the covariance matrix of corresponding distributed wireless access unit i k data block.

8. spaced antenna wireless receiving system, wherein this wireless receiving system comprises at least two distributed wireless access units, and each distributed wireless access unit comprises at least two antennas, it is characterized in that, and described wireless receiving system also comprises:

Covariance matrix calculates and the feature decomposition module, is used to calculate corresponding to the covariance matrix of the base band data of each distributed wireless access unit and described covariance matrix is carried out feature decomposition obtain feature matrix;

The projection value computing module utilizes described covariance matrix and described feature matrix to calculate projection value;

Select the characteristic vector module, select big projection value characteristic of correspondence vector with given number;

Synthetic signal forms module, will compute weighted to base band data as weight vectors at the characteristic vector that described selection characteristic vector module is selected and form synthetic signal.

9. spaced antenna wireless receiving system according to claim 8, it is characterized in that, described covariance matrix calculating and feature decomposition module are also carried out iteration to covariance matrix and are upgraded after having calculated the covariance matrix of current data block according to base band data.

10. spaced antenna wireless receiving system according to claim 8 is characterized in that, described covariance matrix calculates with the feature decomposition module and carries out the utilization of also feature matrix that obtains being stored and delayed time after the feature decomposition.

11. spaced antenna wireless receiving system according to claim 9 is characterized in that, described covariance matrix calculates with the feature decomposition module and carries out the utilization of also feature matrix that obtains being stored and delayed time after the feature decomposition.

12. each spaced antenna wireless receiving system in 11 is characterized in that according to Claim 8, described covariance matrix calculates with the feature decomposition module and adopts formula $R_i\left(k\right)=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{x}_i(j,k)x_i{(j,k)}^H$ Calculate described covariance matrix, wherein R _i(k) be the covariance matrix of corresponding distributed wireless access unit i, the conjugate transpose of subscript H representing matrix or vector, x k data block _i(j, k) expression distributed wireless access unit i is in the column vector output of j sample of k data block, and J calculates the needed sample number of current data block covariance matrix.

13. each spaced antenna wireless receiving system in 11 is characterized in that according to Claim 8, described projection value computing module adopts formula $P_{\mathrm{ml}}\left(k\right)=u_{\mathrm{ml}}^H(k-1)R_m\left(k\right)u_{\mathrm{ml}}(k-1)$ Calculate described projection value, wherein P _Ml(k) be the projection value of k data block output of m distributed wireless access unit base band data, u at the l column vector of the feature matrix of the iteration covariance matrix of k-1 data block _Ml(k-1) be the l row characteristic vector of the iteration covariance matrix of corresponding k-1 data block, R _m(k) be the covariance matrix of corresponding distributed wireless access unit m, the conjugate transpose of subscript H representing matrix or vector k data block.

14. spaced antenna wireless receiving system according to claim 9 is characterized in that, described covariance matrix calculates with the feature decomposition module and adopts formula R _i(k)=(1-β _i) R _i(k-1)+β _iR _i(k) iteration is upgraded covariance matrix, wherein, and β _iBe weight coefficient, R _i(k) be the covariance matrix of corresponding distributed wireless access unit i k data block.

Images