CN102315913A - Method for processing multiple-user multiple input multiple output (MU-MIMO) system and base station - Google Patents

Abstract

The invention discloses a method for processing a multiple-user multiple input multiple output (MU-MIMO) system and a base station. The method comprises the following steps that: 202, the base station calculates the correlation of all code words with the same rank in a codebook which is locally stored; 204, determining a main scheduling user from a set of users to be grouped, and adding the main scheduling user into a grouped user set; 206, according a code word index which is fed back by the main scheduling user and the pre-calculated correlation, determining a first set from the set of users to be grouped, and updating the set of users to be grouped by using the first set, wherein the correlation of the first set and the main scheduling users meets a size relation of a predetermined threshold value; and repeatedly performing the steps 204 and 206 until the quantity of users in the grouped user set is equal to a grouped user quantity threshold value, or the set of users to be grouped is a null set. By the invention, the users can be quickly and efficiently grouped.

Description

The processing method of MU-MIMO system and base station

Technical field

The present invention relates to the communications field, in particular to the processing method and the base station of a kind of MU-MIMO (Multiple-User Multiple Input Multiple Output, multi-user's multiple-input and multiple-output) system.

Background technology

MIMO (Multiple Input Multiple Output, multiple-input and multiple-output) technology is the important breakthrough of intelligent antenna technology in the wireless mobile communications field.This technology can improve the capability of communication system and the availability of frequency spectrum exponentially under the situation that does not increase bandwidth; Can utilize multipath to alleviate multipath fading; And can eliminate common road effectively and disturb; Improving the reliability of channel, reduce the error rate, is a kind of key technology that the third generation mobile communication system adopts.

At present ripe or near ripe 3G/B3G wireless communication system in; Base station side and end side are through the many groups of configuration transmit/receive antenna; And combination MIMO signal processing scheme; This method is called as Single User MIMO (SU-MIMO), and its essence still is point-to-point transmission data between base station (BS) and terminal (MS).And the wireless communication protocol standards of a new generation is all supported multiuser MIMO (MU-MIMO).Fig. 1 shows the structural representation according to the MU-MIMO system of correlation technique.As shown in Figure 1; Compare with descending SU-MIMO, descending MU-MIMO is a BS (Base Station, base station) and a MS (Mobile Station; Travelling carriage; Also claim user terminal) between point-to-point channel expansion be the channel of the point-to-multipoint between a BS and a plurality of MS, through the difference of space characteristic in the base station side aerial array that is caused by the user who disperses on the space, the base station can be in identical frequency channels communicate with a plurality of user terminals simultaneously.

For MU-MIMO, owing to need carry out preliminary treatment so that eliminate the interference between the user at transmitting terminal, so the base station needs its corresponding channel coefficients of each terminal feedback.The next generation with OFDM (Orthogonal Frequency Division Multiplexing; OFDM)/OFDMA (Orthogonal Frequency Division MultipleAccess; OFDM) technology be the basis wireless communication system (such as IEEE802.16m, LTE (Long Term Evolution; Long evolving system), LTE-A (Long Term Evolution Advance; Advanced long-term evolution system) etc.), the channel coefficient matrix H of user terminal adopts the form of code book C to quantize and feeds back, and promptly an identical cover code book C is all preserved with each user terminal in the base station; Each user terminal at first passes through the channel coefficient matrix of down-run survey (centre) pilot measurement oneself; Then according to specific criterion in code book set C, select can the optimum quantization channel coefficients code word V, the index I (Precoding Matrix Indicator, pre-coding matrix index) of this code word V in code book C feeds back to the base station the most at last.The corresponding code word V of index I is found in the base station in the code book set C that oneself preserves behind the codewords indexes I that receives the user terminal feedback.Like this, through above process, the base station can obtain the channel coefficient matrix information of each user terminal.

In MU-MIMO, how utilizing the channel information of user terminal to carry out user fast and reasonably is one of MU-MIMO performance key technology that can be not fully exerted in groups.At present, usually adopt the method for traversal to confirm user in groups, one by one with each in groups the user add in user's set in groups.But; When system allows 3 above users in groups; If adopt the method for traversal this moment owing to confirm one at every turn in groups with traveling through every other user per family; Judge its whether can with each user pairing in user's set in groups, a large amount of computational process of needs so just, thus increased the complexity of system.

In addition, another key technology among the MU-MIMO is the calculating of user's pre-coding matrix in groups.At present; For linear predictive coding; The general employing compeled zero (ZF) or piece diagonal angle algorithm (BD), and direct inversion operation or SVD (Singular Value Decomposition, singular value) that such algorithm all relates to large matrix decompose; And these two kinds of operation methods are all very complicated, so the practical degree in system is relatively poor.

Summary of the invention

Main purpose of the present invention is to provide the processing method and the base station of a kind of MU-MIMO system, and is above-mentioned when allowing 3 above users in groups in system to solve at least, adopts the method for traversal to need a large amount of computational process, increased the problem of the complexity of system.

According to an aspect of the present invention, the processing method of a kind of multi-user's multiple-input and multiple-output MU-MIMO system is provided, has comprised: step S202, the correlation between all code words that order is identical in the local code book of preserving is calculated in the base station; Step S204 confirms the master scheduling user from treat in groups user's set, and adds the master scheduling user to user's set in groups; Step S206; Codewords indexes and precalculated correlation according to the master scheduling user feedback; From treat the set of user in groups, confirm with the master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with first set and to treat that the user gathers in groups; Repeating step S204 and step S206, the number of users in the set of user in groups reach in groups the number of users threshold value or treat that user's set is empty set in groups.

According to a further aspect in the invention, a kind of base station is provided, has comprised: the correlation calculations module is used for calculating the correlation between all identical code words of local code book order of preserving; Confirm to add module, be used for confirming the master scheduling user, and add the master scheduling user to user's set in groups from treating that the user gathers in groups; Update module; Be used for codewords indexes and precalculated correlation according to the master scheduling user feedback; From treat the set of user in groups, confirm with the master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with first set and to treat that the user gathers in groups; Control module is used to control and confirm adds module and repeat operation separately with update module, and the number of users in user's set in groups reaches in groups the number of users threshold value or treats that user's set is an empty set in groups.

Through the present invention; Owing to precompute the correlation between all code words that order is identical in the code book; When carrying out in groups, can directly utilize the parameter of these correlations to find the user's set that satisfies condition fast; And from this user set, determine the user of pairing, thereby when allowing three above users in groups, can realize fast and efficiently that the user in groups in system.

Description of drawings

Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute improper qualification of the present invention.In the accompanying drawings:

Fig. 1 is the structural representation according to the MU-MIMO system of correlation technique;

Fig. 2 is the flow chart according to the processing method of the MU-MIMO system of the embodiment of the invention;

Fig. 3 is MU-MIMO user's operational flowchart in groups according to the preferred embodiment of the invention;

Fig. 4 is the calculation flow chart of the pre-coding matrix of MU-MIMO user group according to the preferred embodiment of the invention;

Fig. 5 is the sketch map according to the base station of the embodiment of the invention.

Embodiment

Hereinafter will and combine embodiment to specify the present invention with reference to accompanying drawing.Need to prove that under the situation of not conflicting, embodiment and the characteristic among the embodiment among the application can make up each other.

In the MU-MIMO system, comprise base station and user terminal.The base station need be matched to a plurality of users (also claim in groups), after in groups, can use the corresponding pre-coding matrix of user in groups that the data that user institute in groups will send are carried out precoding, with the MU-MIMO transfer of data of realization point-to-multipoint.Following examples of the present invention realize in this system.

Fig. 2 is according to the flow chart of the processing method of the MU-MIMO system of the embodiment of the invention, may further comprise the steps:

Step S202, the correlation between all code words that order is identical in the local code book of preserving is calculated in the base station;

For example, correlation can be coefficient correlation, also can be distance.

Step S204 confirms the master scheduling user from treat in groups user's set, and adds the master scheduling user to user's set in groups;

Step S206; According to the correlation between the codewords indexes of master scheduling user feedback all code words identical with the order in step S202, calculated in advance; From treat the set of user in groups, confirm with the master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with first set and to treat that the user gathers in groups;

For example; When the correlation between all identical code words of order in the code book that calculates among the step S202 is coefficient correlation, from treat the set of user in groups, confirm among the step S206 with the master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold and be specially from treat user's set in groups the coefficient correlation between the definite and master scheduling user and gather less than first of correlation coefficient threshold.

When the correlation between all identical code words of order in the code book that calculates among the step S202 for apart from the time, from treat the set of user in groups, confirm among the step S206 with the master scheduling user between first set satisfied with the magnitude relationship of predetermined threshold of correlation be specially: from treat user's set in groups, confirm and the master scheduling user between code word apart from gathering greater than first of distance threshold.

Step S208 repeats above-mentioned steps S204 and step S206, and the number of users in user's set in groups reaches in groups the number of users threshold value or treats that user's set is an empty set in groups.

Above-mentioned steps S202 can be called MU-MIMO user's method in groups to step S208.

Owing to adopt the method for traversal, when system allows three above users in groups, just need a large amount of computational process in the correlation technique, thereby increased the complexity of system.Present embodiment is owing to precompute the correlation between all code words that order is identical in the code book; When carrying out in groups, can directly utilize the parameter of these correlations to find the user's set that satisfies condition fast; And from this user set, determine the user of pairing; Thereby when allowing three above users in groups, can realize fast and efficiently that the user in groups in system.

In the down link of wireless communication system, dispose N _tThe base station of individual descending transmitting antenna adopts following step to accomplish MU-MIMO user in groups:

Step 1, for n=1,2 ... R _H, the base station is calculated rank among the code book C (order) respectively and is coefficient correlation or distance between all code words of n, and the result is kept at form T _nIn, form T wherein _nLine index be codewords indexes i=1,2 ... N _n, column index is codewords indexes j=1,2 ... N _n, form T _nIn the element representation of the capable and j row intersection of i: at rank is in all code words of n, and index is the coefficient correlation between the code word of i and the code word that index is j Perhaps distance

Wherein, r _HPreestablish by system; Code book C is the vector sum set of matrices that system preserves in advance, is used for the quantized channel coefficient; Rank is the number that n representes the column vector that the code word among the code book C comprises.

Step 2, the base station is with the r that calculates in the step 1 _HIndividual form T _n, n=1,2 ... R _H, be kept in the local memory;

The step S202 that above-mentioned steps 1 is corresponding above-mentioned with step 2.Preserve the coefficient correlation between all identical code words of the order that calculates with the form of form, like this, search easily and search fast.

Step 3 is confirmed user's set in groups, and is as shown in Figure 3, comprises the steps:

Step S302, user's set omega={ U is in groups treated in the base station initialization ₁, U ₂..., U _MCarry out MU-MIMO user in groups, and initialization MU-MIMO user's set in groups

Step S304 confirms that according to dispatching priority the master scheduling user (user who for example, selects to have the highest dispatching priority is as the master scheduling user) among the Ω for example is U ₁, the rank of the code word of its feedback is n, codewords indexes is I ¹With Ω _MUBe updated to Ω _MU={ Ω _MU, U ₁;

Step S306 judges set omega _MUWhether middle number of users equals the predefined MU-MIMO of system number of users threshold value N in groups _MUIf, equate, then change step S316 over to, if not, then change step S308 over to;

Step S308 is when the correlation of calculating in the step 1 is coefficient correlation, at form T _nIn I ¹Row or I ¹Search numerical value in the row less than the predefined correlation coefficient threshold TH of system _rElement, all these elements are at form T _nMiddle corresponding column index or line index constitute set (promptly being called second set); Perhaps, when the correlation of calculating in the step 1 be apart from the time, at form T _nIn I ¹Row or I ¹Search numerical value in the row greater than the predefined distance threshold TH of system _dElement, all these elements are at form T _nMiddle corresponding column index or line index constitute set

(promptly being called second set);

Step S310 judges set

Whether be empty set,, then get into step S316 if be empty, otherwise at set omega={ U ₁, U ₂..., U _MIn seek the user satisfy following condition and constitute set omega ₁(being called first set): user's correspondence code word indexing belongs to set

${\mathrm{<figure-callout\; id="1"\; label="I"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">I</figure-callout>}}^1=\{I_1^1,I_2^1,...,I_{{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">N</figure-callout>}}_1}^1\};$

Step S312 is updated to Ω with set omega ₁: Ω ← Ω ₁

Step S314 judges whether set omega is empty set, if be empty, then get into step S316, otherwise returns step S304;

Step S316, MU-MIMO user is EO in groups.

The corresponding above-mentioned step S204 of above-mentioned steps 3 is to step S208.The user in groups the time; Only need directly from the form of correspondence, to find with master scheduling user's coefficient correlation less than the codewords indexes of predetermined correlation coefficient threshold or directly from corresponding form, find and master scheduling user's code word distance codewords indexes greater than predetermined distance threshold; Thereby the codewords indexes that can in treating the set of user in groups, seek feedback is user's set (being the first above-mentioned set) of the codewords indexes that finds; Gather renewal with this user and treat user's set in groups; Confirm in treating user's set in groups that once more the master scheduling user adds in user's set in groups; Promptly, finally be met user's group of condition (for example, number of users reaches predetermined threshold value in groups) through the process of step S304 to step S314.Like this, only need the once calculating of step S202 get final product, follow-up need are simply searched and relatively can be accomplished the user in groups, can reduce amount of calculation significantly, thus the complexity of reduction system.

Be example with embodiment 1 to embodiment 3 below, come to explain particularly above-mentioned MU-MIMO user flow process in groups.

Embodiment 1

How present embodiment explanation base station side utilizes known code book structure code word coefficient correlation form or code word apart from form; Certainly the computational methods of wherein employed coefficient correlation or code word distance are just wherein a kind of, can also adopt other mode to calculate coefficient correlation or distance in the reality: to establish the base station that known codebook C is used to dispose 4 transmitting antennas.Wherein C comprises the code word of 16 rank=1, the code word of 16 rank=2, and the code word of the code word of 8 rank=3 and 4 rank=4, present embodiment explains that mainly the coefficient correlation that makes up rank=1 and 2 is perhaps apart from form T ₁And T ₂

(1) calculating rank (order) among the code book C for the situation of the coefficient correlation between all code words of n is:

For the situation of rank=1, establish 16 code word column vectors and be respectively

The coefficient correlation of then calculating between any two code words is: I, j=1 ..., 16, obtain being used for the form T of rank=1 like this ₁, shown in form 1:

Form 1T ₁

For rank=2, establish 16 code word column vectors and be respectively

The coefficient correlation of calculating between any two code words is:

I, j=1 ..., 16, wherein Representing matrix

Minimal eigenvalue, obtain being used for the form T of rank=2 like this ₂, shown in form 2:

Form 2T ₂

Similarly, can calculate the coefficient correlation form of rank=3 and rank=4.

(2) in above-mentioned example, if select to calculate the distance between all code words that rank (order) among the code book C be n, then method is accordingly:

For the situation of rank=1, ${\mathrm{<figure-callout\; id="1"\; label="D\; Ij"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">D</figure-callout>}}_{\mathrm{Ij}}^{}=\sqrt{1-{\left|{\left(V_i^1\right)}^{\mathrm{<figure-callout\; id="1"\; label="H\; V\; j"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">H</figure-callout>}}{V}_j^{}{}_1^{}\right|}^2},i,j=1,...,16;$

For the situation of rank=2, ${\mathrm{<figure-callout\; id="2"\; label="D\; Ij"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">D</figure-callout>}}_{\mathrm{Ij}}^{}=\sqrt{1-{\mathrm{\&lambda;}}_{\mathrm{Min}}\left({\left({\left(V_i^2\right)}^H{V}_i^2\right)}^H\left({\left(V_i^2\right)}^H{V}_i^2\right)\right)},i,j=1,...,16.$

Corresponding form T ₁, T ₂Similar with above table 1 and form 2.

Similarly, can calculate rank=3 and rank=4 apart from form.

Embodiment 2

Present embodiment is used for explaining that base station side treating that in groups user's set selects N _MU=4 user pairing are used MU-MIMO.If currently treat in groups that by 20 it is Ω={ U that the user forms set ₁, U ₂... U ₂₀, current MU-MIMO user's set in groups The base station is provided with MU-MIMO user's correlation coefficient threshold TH _r=0.2.

Step 1, base station are selected the master scheduling user at first therein, are made as U ₃, and the rank=1 codewords indexes of feedback is 3 recently, the base station is with set omega _MUBe updated to Ω _MU={ U ₃, then at form T ₁All numerical value of the 3rd row in select less than TH _r=0.2 numerical value is established them and is respectively

Then its corresponding column index constitutes set I ¹={ 2,8,12,14,16} is then at Ω={ U ₁, U ₂... U ₂₀In seek user feedback codewords indexes belong to I ¹={ 2,8,12,14, user's set of 16} is made as Ω ₁={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆, Ω is updated to Ω ₁, i.e. Ω={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆.

Step 2, base station confirm that in Ω the master scheduling user is U ₇, upgrade Ω _MUBe Ω _MU={ U ₃, U ₇, establish U ₇Corresponding codewords indexes is 8, then at form T ₁In eighth row to seek column index be I ¹={ 2,8,12,14, the numerical value of 16} is established in the numerical value that finds and is satisfied numerical value less than TH _rThe column index of=0.2 numerical value is 2,12,14, then upgrades I ¹Be I ¹={ 2,12,14} continues at Ω={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆In seek user feedback codewords indexes belong to I ¹={ 2,12, user's set of 14} is made as Ω ₁={ U ₅, U ₁₀, U ₁₂, U ₁₆, Ω is updated to Ω ₁, i.e. Ω={ U ₅, U ₁₀, U ₁₂, U ₁₆.

Step 3, base station confirm that in Ω the master scheduling user is U ₁₀, upgrade Ω _MUBe Ω _MU={ U ₃, U ₇, U ₁₀, establish U ₁₀Corresponding codewords indexes is 12, then at form T ₁In the 12nd row to seek index be I ¹={ 2,12, the numerical value of 14} is established and is wherein satisfied numerical value less than TH _r=0.2 index is 2, then upgrades I ¹Be I ¹={ 2} continues at Ω={ U ₅, U ₁₀, U ₁₂, U ₁₆In seek user feedback codewords indexes belong to I ¹={ user's set of 2} is made as Ω ₁={ U ₅, U ₁₂, Ω is updated to Ω ₁, i.e. Ω={ U ₅, U ₁₂.

Step 4, base station confirm that in Ω the master scheduling user is U ₁₂, upgrade Ω _MUBe Ω _MU={ U ₃, U ₇, U ₁₀, U ₁₂, this moment Ω _MUElement number be 4, with N _MUEquate, so MU-MIMO user's operation has in groups been accomplished in the base station.The final set omega of user in groups _MU={ U ₃, U ₇, U ₁₀, U ₁₂Form MU-MIMO user's group, at same time transmitting data.

Embodiment 3

Present embodiment is used for explaining that the base station treating that in groups user's set selects N _MU=4 user pairing are used MU-MIMO.If currently treat in groups that by 20 it is Ω={ U that the user forms set ₁, U ₂... U ₂₀, current MU-MIMO user's set in groups

The base station is provided with MU-MIMO user's correlation threshold T H _r=0.2.

Step 1, base station are at first at Ω={ U ₁, U ₂... U ₂₀Middle selection master scheduling user, be made as U ₃, and the rank=1 codewords indexes of feedback is 3 recently, the base station is with set omega _MUBe updated to Ω _MU={ U ₃, then at form T ₁All numerical value of the 3rd row in select less than TH _r=0.2 numerical value is established them and is done

Then its corresponding column index constitutes set I ¹={ 2,8,12,14,16} is then at Ω={ U ₁, U ₂... U ₂₀In seek the user codewords indexes belong to I ¹={ 2,8,12,14, user's set of 16} is made as Ω ₁={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆, Ω is updated to Ω ₁, i.e. Ω={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆.

Step 2, base station confirm that in Ω the master scheduling user is U ₇, upgrade Ω _MUBe Ω _MU={ U ₃, U ₇, establishing the corresponding codewords indexes of U7 is 8, then at form T ₁Eighth row to seek index be I ¹={ 2,8,12,14, the numerical value of 16} is established and is wherein satisfied numerical value less than TH _r=0.2 index is 2,12,14, then upgrades I ¹Be I ¹={ 2,12,14} continues at Ω={ U ₁, U ₄, U ₅, U ₇, U ₁₀, U ₁₂, U ₁₆In seek user feedback codewords indexes belong to I ¹={ 2,12, user's set of 14} is made as Ω ₁={ U ₅, U ₁₀, U ₁₂, U ₁₆, Ω is updated to Ω ₁, i.e. Ω={ U ₅, U ₁₀, U ₁₂, U ₁₆.

Step 3, base station confirm that in Ω the master scheduling user is U ₁₀, upgrade Ω _MUBe Ω _MU={ U ₃, U ₇, U ₁₀, establish U ₁₀Corresponding codewords indexes is 12, then at form T ₁The 12nd row in to seek column index be I ¹={ 2,12, the numerical value of 14} is established and is wherein satisfied numerical value less than TH _r=0.2 index is an empty set, promptly

So this moment MU-MIMO user's EO in groups, the final set omega of user in groups _MU={ U ₃, U ₇, U ₁₀Form MU-MIMO user's group, at same time transmitting data.

Preferably, above-mentioned processing method (promptly accomplish user in groups after) after step S208 also comprises: calculating in groups, the user gathers corresponding pre-coding matrix; The pre-coding matrix that use calculates carries out precoding to all users' during the user gathers in groups transmission data; Wherein, this pre-coding matrix is: P=[W ₁W ₂W _N], wherein, N is the number of users in user's set in groups, W _i=orth ((I-W (W ^HW) ^-1W ^H) V _i), wherein I is N _t* N _tUnit matrix, N _tBe constant, orth (.) expression is carried out orthonormalization operation, V to the variable of input _iBe the user U during the user gathers in groups _iThe code word that the codewords indexes of feedback is corresponding, i=1 ..., N.

In the correlation technique because when calculating the MU-MIMO pre-coding matrix of user's group; Be that the code word that the codewords indexes of all user feedbacks in the user group is corresponding is placed in the matrix and carries out computing; Because the dimension of computational complexity and matrix is an exponent function relation; Therefore when the number of users more (user more than three is arranged) in user's group, computation complexity is very high, and the practicality in system is relatively poor.Pre-coding matrix P=[W in the present embodiment ₁W ₂W _N] promptly can the computational process of P be divided into and calculate W ₁, W ₂..., W _NN step carry out (substep computing), the dimension of matrix can be very not big when calculating at every turn, computation complexity is lower; And, calculate W _iNeed use a preceding result calculated W _I-1, like this, realize that at hardware the time ratio is easier to realize.

As shown in Figure 4, the process of calculating the pre-coding matrix of MU-MIMO user's group may further comprise the steps:

Step S402 is to step S408, through the definite set omega of user in groups of above-mentioned steps _MUIn N user arranged, and the corresponding code word of codewords indexes of the last feedback of this N user is respectively V ₁, V ₂..., V _N, user U then _i=, i=1 ..., the pre-coding matrix that N is corresponding, the base station uses following computational methods to obtain the corresponding pre-coding matrix of user:

W _i=orth ((I-W (W ^HW) ^-1W ^H) V _i), wherein

I representes N _t* N _tThe unit matrix of dimension, promptly Orth (.) expression is carried out the orthonormalization operation to the variable of input, even each column vector mould value that input variable comprises is 1, and quadrature each other.

Step S410, the corresponding pre-coding matrix of each user that utilizes step S402 to calculate to step S408 forms final pre-coding matrix and is: P=[W ₁W ₂W _N], then PS is mapped to N respectively _tOn the transmitting antenna, wherein S _i, i=1 ..., N representes to belong to user U _iData flow.

Embodiment 4

Present embodiment is used to explain how the base station utilizes the corresponding code word of MU-MIMO user to calculate final this user and organize corresponding pre-coding matrix and data transmission method for uplink.

If base station configuration has 4 descending transmitting antennas; In current MU-MIMO user's group, 4 users 1 are arranged; 2; 3,4 (promptly one has 4 users in user's set in groups), the corresponding code word of codewords indexes of their the last feedbacks is respectively

${\mathrm{<figure-callout\; id="2"\; label="V"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">V</figure-callout>}}_2=\left[\begin{array}{c}0.3082\\ 0.0104+0.3151i\\ 0.4077+0.4887i\\ -0.4783+0.4145i\end{array}\right],$ $V_3=\left[\begin{array}{c}0.5587\\ 0.2735-0.3361i\\ 0.1135+0.5471i\\ 0.2735-0.3361i\end{array}\right],$ $V_4=\left[\begin{array}{c}0.5587\\ -0.1492-0.2737i\\ -0.2735-0.3361i\\ -0.6245+0.1132i\end{array}\right].$

Step 1 makes W ₁=V ₁, promptly ${\mathrm{<figure-callout\; id="1"\; label="W"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">W</figure-callout>}}_1=\left[\begin{array}{c}-0.5000\\ -0.5000\\ 0.5000\\ 0.5000\end{array}\right];$

Step 2 makes W=W ₁, calculate W ₂=orth ((I ₄-W (W ^HW) ^-1W ^H) V ₂), obtain

${\mathrm{<figure-callout\; id="2"\; label="W"\; filenames="HSA00000177659700031.png"\; state="\{\{state\}\}">W</figure-callout>}}_2=\left[\begin{array}{c}0.2254+0.1571i\\ -0.0929+0.4938i\\ 0.5397+0.3651i\\ -0.4071+0.2858i\end{array}\right];$

Step 3 makes W=[W ₁W ₂], calculate W ₃=orth ((I ₄-W (W ^HW) ^-1W ^H) V ₃), obtain $W_3=\left[\begin{array}{c}0.5056+0.1201i\\ 0.2380-0.1964i\\ 0.2968+0.3767i\\ 0.4468-0.4530i\end{array}\right];$

Step 4 makes W=[W ₁W ₂W ₃], calculate W ₄=orth ((I ₄-W (W ^HW) ^-1W ^H) V ₄), obtain

So finally the corresponding MU-MIMO pre-coding matrix of these 4 users can be write as:

$P=\left[\begin{array}{cccc}{W}_1& W_2& W_3& W_4\end{array}\right]=\left[\begin{array}{cccc}-0.5000& 0.2254+0.1571i& 0.5056+0.1201i& 0.5176+0.3696i\\ -0.5000& -0.0929+0.4938i& 0.2380-0.1964i& -0.5180-0.3660i\\ 0.5000& 0.5397+03651i& 0.2968+0.3767i& -0.0198-0.3084i\\ 0.5000& -0.4071+0.2858i& 0.4468-0.4530i& 0.0194+0.3120i\end{array}\right].$

Obviously, the precoding vector in the said process is asked in proper order and can be changed according to actual conditions.Simultaneously, when the corresponding code word of user is matrix, utilize similar step can obtain final pre-coding matrix.

Fig. 5 is according to the sketch map of the base station of the embodiment of the invention, comprising: correlation calculations module 10 is used for calculating the correlation between all identical code words of local code book order of preserving; Confirm to add module 20, be used for confirming the master scheduling user, and add the master scheduling user to user's set in groups from treating that the user gathers in groups; Update module 30; Be used for according to the correlation between the codewords indexes of master scheduling user feedback all code words identical with precalculated order; From treat the set of user in groups, confirm with the master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with first set and to treat that the user gathers in groups; Control module 40 is used to control and confirm adds module 20 and repeat operation separately with update module 30, and the number of users in user's set in groups reaches in groups the number of users threshold value or treats that user's set is empty set in groups.

Preferably, computing module 10 comprises: computing module, and being used in above-mentioned correlation is under the situation of coefficient correlation, calculates order in the code book and be the coefficient correlation between all code words of n; Be under the situation of distance perhaps, calculate order in the code book and be the distance between all code words of n in above-mentioned correlation; Wherein, order is the quantity of the column vector that comprises of the code word in the code book, n=1, and 2 ..., r _H, r _HBe constant; Memory is used for form T _nOrder that preserve to calculate of form be coefficient correlation or distance between all code words of n, wherein, form T _nIn i element capable and j row place be: in order is in all code words of n, and index is the coefficient R between the code word of i and the code word that index is j ⁿ _IjPerhaps distance

Preferably, above-mentioned base station also comprises: initialization module 50, be used for initialization and treat that in groups user's set comprises all users to be scheduled, and initialization in groups user's set be empty set

Preferably, confirm to add module 20 and comprise: determination module is used for from treating that user's set omega in groups confirms that a user with the highest dispatching priority is as the master scheduling user; Add module, be used for having with what confirm that the user of high dispatching priority adds user's set omega in groups to _MU

What preferably, calculate when above-mentioned said computing module is that above-mentioned predetermined threshold was a correlation coefficient threshold, confirms to add the master scheduling user U that module 20 is confirmed when order was the coefficient correlation between all code words of n in the code book _kThe codewords indexes of feedback is I ^k, said codewords indexes ^IkThe order of corresponding code word is m, and m ∈ [1, r _H]; Update module 30 comprises: judge module is used to judge user's set omega in groups _MUIn number of users whether surpass number of users threshold value in groups; Search module, be used for the judged result of above-mentioned judge module for the situation that does not surpass number of users threshold value in groups under, at form T _mI ^kOK/and being listed as and searching numerical value less than correlation coefficient threshold, the row/row at the numerical value that finds place constitutes second set; Composition module, be used for second set for empty situation under, treating to confirm that the codewords indexes of feedback belongs to the user of second set and constitutes first set in user's set omega in groups, and upgrading with first set and to treat user's set omega in groups.

Perhaps, preferably, when above-mentioned said computing module calculate be in the code book order be between all code words of n apart from the time, above-mentioned predetermined threshold is a distance threshold, confirms to add the master scheduling user U that module 20 is confirmed _kThe codewords indexes of feedback is I ^k, said codewords indexes I ^kThe order of corresponding code word is m, and m ∈ [1, r _H]; Update module 30 comprises: judge module is used to judge user's set omega in groups _MUIn number of users whether surpass number of users threshold value in groups; Search module, be used for the judged result of above-mentioned judge module for the situation that does not surpass number of users threshold value in groups under, at form T _mI ^kOK/and being listed as and searching numerical value greater than distance threshold, the row/row at the numerical value that finds place constitutes second set; Composition module, be used for second set for empty situation under, treating to confirm that the codewords indexes of feedback belongs to the user of second set and constitutes first set in user's set omega in groups, and upgrading with first set and to treat user's set omega in groups.

Preferably; Control module 40 comprises: return module; Be used under treating that user's set omega is not for empty situation in groups; Control and confirm to add module 20 execution definite master scheduling user from treat user's set in groups, and add the master scheduling user to the step (being the step S202 among the said method embodiment) of user's set in groups.

Preferably, this base station also comprises: the pre-coding matrix computing module, and being used to calculate in groups, the user gathers corresponding pre-coding matrix; Precoding module is used for using pre-coding matrix that above-mentioned pre-coding matrix computing module calculates that all users' of user's set transmission data are in groups carried out precoding; Wherein, this pre-coding matrix is: P=[W ₁W ₂W _N], wherein, N is the number of users in user's set in groups, W _i=orth ((I-W (W ^HW) ^-1W ^H) V _i),

Wherein

I is N _t* N _tUnit matrix, N _tBe constant, orth (.) expression is carried out orthonormalization operation, V to the variable of input _iBe the user U during the user gathers in groups _iThe code word that the codewords indexes of feedback is corresponding, i=1 ..., N.

From above description, can find out that the present invention has realized following technique effect:

(1) can realize the user fast and efficiently in groups, reduce the complexity of system;

(2) owing to being divided into, the computational process of the user being organized corresponding pre-coding matrix P calculates W ₁, W ₂..., W _NN step carry out (substep computing), the dimension of matrix can be very not big when calculating at every turn, computation complexity is lower; And, calculate W _iNeed use a preceding result calculated W _I-1, like this, realize that at hardware the time ratio is easier to realize.

Obviously, it is apparent to those skilled in the art that above-mentioned each module of the present invention or each step can realize with the general calculation device; They can concentrate on the single calculation element; Perhaps be distributed on the network that a plurality of calculation element forms, alternatively, they can be realized with the executable program code of calculation element; Thereby; Can they be stored in the storage device and carry out, and in some cases, can carry out step shown or that describe with the order that is different from here by calculation element; Perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.

The above is merely the preferred embodiments of the present invention, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.All within spirit of the present invention and principle, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (12)

1. the processing method of multi-user's multiple-input and multiple-output MU-MIMO system is characterized in that, comprising:

Step S202, the correlation between all code words that order is identical in the local code book of preserving is calculated in the base station;

Step S204 confirms the master scheduling user from treat in groups user's set, and adds said master scheduling user to user's set in groups;

Step S206; Codewords indexes and precalculated said correlation according to said master scheduling user feedback; From said treat to confirm the set of user in groups with said master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with said first set and saidly to treat that the user gathers in groups;

Repeating said steps S204 and said step S206, the number of users in said user in groups set reach in groups the number of users threshold value or saidly treat that user's set is empty set in groups.

2. method according to claim 1 is characterized in that, from treat in groups user's set, is confirming the master scheduling user, and said master scheduling user is added to before user's set in groups, also comprises:

The initialization of said base station is said treats that user's set comprises all users to be scheduled in groups, and the said set of user in groups of initialization is empty set

3. method according to claim 1 is characterized in that, from treat in groups user's set, confirms the master scheduling user, and adds said master scheduling user in groups that user's set comprises to:

Said base station treats to confirm the set of user in groups that from said a user with the highest dispatching priority is as said master scheduling user and add the said set of user in groups to.

4. method according to claim 1 is characterized in that, said correlation is a coefficient correlation, and the correlation that calculate between all code words that order is identical in the local code book of preserving the base station comprises:

The coefficient correlation between all code words that order in the said code book is n is calculated in said base station, and with form T _nForm preserve;

Wherein, said order is the quantity of the column vector that comprises of the code word in the said code book, n=1, and 2 ..., r _H, r _HBe constant, said form T _nIn i element capable and j row place be: in order is in all code words of n, and index is the coefficient correlation between the code word of i and the code word that index is j

5. method according to claim 4 is characterized in that, said predetermined threshold is a correlation coefficient threshold, said master scheduling user U _kThe codewords indexes of feedback is I ^k, said codewords indexes I ^kThe order of corresponding code word is m, and m ∈ [1, r _H]; Codewords indexes and precalculated said correlation according to said master scheduling user feedback; From said treat to confirm the set of user in groups with said master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with said first set and saidly to treat that the user gathers in groups and comprise:

The said user in groups of said base station judges set omega _MUIn number of users whether surpass the said threshold value of number of users in groups;

If do not surpass the said threshold value of number of users in groups, then at said form T _mI ^kOK/and being listed as and searching numerical value less than said correlation coefficient threshold, the index of the row/row at the numerical value that finds place constitutes second set;

Under said second set be empty situation, confirm that the codewords indexes of feedback belongs to the user of said second set and constitutes said first set in user's set omega in groups said treating, and with the said user's set omega in groups of treating of the said first set renewal.

6. method according to claim 1 is characterized in that, said correlation is a distance, and the correlation that calculate between all code words that order is identical in the local code book of preserving the base station comprises:

The distance between all code words that order in the said code book is n is calculated in said base station, and with form T _nForm preserve;

Wherein, said order is the quantity of the column vector that comprises of the code word in the said code book, n=1, and 2 ..., r _H, r _HBe constant, said form T _nIn i element capable and j row place be: in order is in all code words of n, and index is the distance between the code word of i and the code word that index is j

7. method according to claim 6 is characterized in that, said predetermined threshold is a distance threshold, said master scheduling user U _kThe codewords indexes of feedback is I ^k, said codewords indexes I ^kThe order of corresponding code word is m, and m ∈ [1, r _H]; Codewords indexes and precalculated said correlation according to said master scheduling user feedback; From said treat to confirm the set of user in groups with said master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with said first set and saidly to treat that the user gathers in groups and comprise:

The said user in groups of said base station judges set omega _MUIn number of users whether surpass the said threshold value of number of users in groups;

If do not surpass the said threshold value of number of users in groups, then at said form T _mI ^kOK/and being listed as and searching numerical value greater than said distance threshold, the index of the row/row at the numerical value that finds place constitutes second set;

Under said second set be empty situation, confirm that the codewords indexes of feedback belongs to the user of said second set and constitutes said first set in user's set omega in groups said treating, and with the said user's set omega in groups of treating of the said first set renewal.

8. according to claim 5 or 7 described methods; It is characterized in that; Repeating said steps S204 and said step S206; Number of users in said user in groups set reaches in groups the number of users threshold value or saidly treats that user's set comprises for empty set in groups: under the situation for sky, return said step S202 in the said user's set omega of treating in groups.

9. method according to claim 1; It is characterized in that; At repeating said steps S204 and said step S206, the number of users in said user in groups set reaches in groups the number of users threshold value or saidly treats that user's set also comprises for after the empty set in groups:

Calculate said user in groups and gather corresponding pre-coding matrix;

Use said pre-coding matrix that all users' in the said set of user in groups transmission data are carried out precoding;

Wherein, said pre-coding matrix is: P=[W ₁W ₂W _N], wherein, N does

Number of users in the said set of user in groups, W _i=orth ((I-W (W ^HW) ^-1W ^H) V _i),

Wherein I is N _t* N _tUnit matrix, N _tBe constant, orth (.) expression is carried out orthonormalization operation, V to the variable of input _iBe the user U in the said set of user in groups _iThe code word that the codewords indexes of feedback is corresponding, i=1 ..., N.

10. a base station is characterized in that, comprising:

The correlation calculations module is used for calculating the correlation between all identical code words of local code book order of preserving;

Confirm to add module, be used for confirming the master scheduling user, and add said master scheduling user to user's set in groups from treating that the user gathers in groups;

Update module; Be used for codewords indexes and precalculated said correlation according to said master scheduling user feedback; From said treat to confirm the set of user in groups with said master scheduling user between correlation satisfy first set with the magnitude relationship of predetermined threshold, and upgrade with said first set and saidly to treat that the user gathers in groups;

Control module is used to control said definite interpolation module and said update module and repeats operation separately, and the number of users in said user in groups set reaches in groups the number of users threshold value or saidly treats that user's set is empty set in groups.

11. base station according to claim 10 is characterized in that, said correlation calculations

Module comprises:

Computing module, being used in said correlation is under the situation of coefficient correlation, calculates order in the said code book and be the coefficient correlation between all code words of n; Be under the situation of distance perhaps, calculate order in the said code book and be the distance between all code words of n in said correlation; Wherein, said order is the quantity of the column vector that comprises of the code word in the said code book, n=1, and 2 ..., r _H, r _HBe constant;

Memory is used for form T _nSaid order that preserve to calculate of form be coefficient correlation or distance between all code words of n, wherein, said form T _nIn i element capable and j row place be: in order is in all code words of n, and index is the coefficient correlation between the code word of i and the code word that index is j

Perhaps distance

12. base station according to claim 11 is characterized in that, said definite interpolation module comprises:

Determination module is used for confirming that from the said user's set omega in groups treated a user with the highest dispatching priority is as said master scheduling user;

Add module, be used for saidly having that the user of high dispatching priority adds the said set omega of user in groups to what confirm _MU, wherein, said have a user U of high dispatching priority _kThe codewords indexes of feedback is I ^k, said codewords indexes I ^kThe order of corresponding code word is m, and m ∈ [1, r _H].

Images