CN101841359A - Data transmission method and system based on channel Quality to Interference Ratio - Google Patents

Abstract

The invention discloses a limiting feedback beamforming data transmission method and system using channel Quality to Interference Ratio as quantitative standard. The method comprises the following special steps: using discrete Fourier transform for reference, designing a DFTC codebook; also designing a quantitative standard based on channel Quality to Interference Ratio (QIR), wherein by using the standard, the overall quality of the quantified channel and the reduction of the mutual interference of component channels can be considered at the same time; adopting the proportional fairness scheduling (PFS) to dispatch users, adopting the water-filling algorithm to distribute power for the dispatched users; and adopting the optimal codeword fed back by users as beam vector to transmit data to users. The method of the invention overcomes the problems that when the number of users is small, the system performance is reduced under the traditional beamforming mode; when the number of users is large, the system feedback quantity is increased quickly, and the system performance is reduced under low SNR scene. Therefore, the method is a better beamforming scheme.

Description

Data transmission method and system based on the channel quality interference ratio

Technical field

The present invention relates to wireless communication field, particularly a kind of method and system that forms transfer of data based on the channel quality interference ratio for the Limited Feedback wave beam that quantizes criterion.

Background technology

MIMO (Multiple-Input Multiple-Output, multiple-input and multiple-output) communication system is by adopting many to transmit and receive antenna and have gain of higher spatial reuse and diversity gain, and can in multi-user system, obtain significant multi-user diversity gain and multiplexed gain by wave beam formation and user's dispatching technique, therefore it is adopted by the LTE-Advanced standard, also will become the key technology of 4G wireless communication system.In mimo system, can utilize beam-forming technology to obtain certain wave beam and form gain, improve systematic function, but, conventional beam-forming technology systematic function when lacking the user obviously descends, and feedback quantity is excessive when the multi-user, and especially (Signal toNoise Ratio, SNR) systematic function descends obvious especially under the scene at low signal-to-noise ratio.

At present, a kind of employing random wave bundle forms (Random Beamforming, RBF) wireless communication technology of pattern, having wave beam forms simply, the feedback complexity is low, feedback quantity is few and can obtain when the multi-user Yu know channel condition information (Channel State Information fully at transmitter, CSI) and utilize dirty paper code (Dirty-Paper Coding, DPC) therefore the advantages such as throughput of system of identical growth rate during technology extensively come into one's own.But the beam vector that this technology adopts is to generate at random, and especially system can't find dispatched users with its coupling when few user, so systematic function seriously descends.

In order to overcome the above problems, adopt eigen beam to form (Eigen Beamforming, EBF) wireless communication technology of pattern, the complete feedback user CSI of this specification requirement, and a right singular vector of the subscriber channel singular value decomposition that feeds back is as beam vector, even therefore under few user, wave beam also mates fully with subscriber channel, improve the systematic function of system when lacking the user greatly, but along with increasing of number of users, the system feedback amount seriously increases, take valuable system resource, therefore can't extensive use.

In addition, relevant opportunity beam forming forms (Coherent Opportunistic Beamforming, COBF) pattern adopts user's normalization channel vector to form vector as the wave beam of optimum, this pattern can obtain the systematic function near DPC under low SNR scene, but it requires the base station must accurately know user's CSI, in addition, this pattern does not adopt the code book scheme, so is difficult in the reality use.

Summary of the invention

The problem that the embodiment of the invention will solve provides a kind of based on Limited Feedback wave beam formation data transmission method and the system of channel quality interference ratio for the quantification criterion, the defective that the system feedback amount increases severely when forming pattern (RBF, EBF and COBF) in systematic function decline under few user, the multi-user to overcome existing traditional wave beam, and the problem that systematic function descends under low SNR scene.

For reaching above purpose, the invention provides a kind of method of transfer of data, may further comprise the steps: the system model brief introduction; The design of DFTC code book; The design of QIR criterion and corresponding quantitative criterion thereof; New beam forming process; Power system capacity analysis, power division and user's dispatching algorithm.

Wherein, the system model brief introduction is as follows:

The present invention supposes M antenna of transmitting terminal configuration, N antenna of receiving terminal configuration, and base station and user are in the environment that enriches scattering object, and the m transmit antennas is h to the channel coefficients of n reception antenna _Nm, they all obey independent Gaussian Profile CN (0,1) again.Suppose that channel is the piece fading channel, promptly in the time of code element transmission for flat fading and remain unchanged, the channel between the different user is separate.K user in the received signal of time slot t is:

Y _k(t)＝H _k(t)X _k(t)+Φ _k(t)

Wherein, Y _k(t) ∈ C ^NExpression user k is in the received signal of time slot t; H _k(t) ∈ C ^{N * M}Expression user k is in the multiple Gaussian channel of time slot t; X _k(t) ∈ C ^MThe expression base station is at the transmission symbol of time slot t; Φ _k(t) ∈ C ^NThe additivity white complex gaussian noise signal that expression user k receives at time slot t, its power is N _ΦSuppose that the total transmitting power in base station is

All save time slot t for convenient the discussion below.

Wherein, the design of DFTC code book is as follows:

The code book set is C={C ⁽⁰⁾..., C ^(G-1), wherein, G=2 ^B, B is a quantization bit.Method for generating codebooks based on DFT is as follows,

Be g pre-coding matrix, guarantee that pre-coding matrix is a unitary matrice, wherein Be m precoding vector in this pre-coding matrix, its each element generates based on Fourier basis, and is as follows:

$c_m^{\left(g\right)}=\frac{1}{\sqrt{M}}{[c_{0m}^{\left(g\right)},...,c_{(M-1)m}^{\left(g\right)}]}^T$

$c_{\mathrm{nm}}^{\left(g\right)}=\exp \left\{j\frac{2\mathrm{\πn}}{M}(m+\frac{g}{G})\right\},m,n,=0,...,M-1,g=1,...,G$

Wherein, the design of QIR criterion and corresponding quantitative criterion thereof specifically comprises:

Receiving terminal can be known oneself CSI by channel estimating, to H _kDo singular value (SVD) and decompose, promptly U _k, V _kBe unitary matrice, Λ _kBe H _kThe diagonal matrix that singular value is formed.In order to consider user k overall channel quantification quality quality simultaneously and quantize orthogonality size between the subchannel that the definition Mass Interference is than (QIR, matter is done ratio) parameter, that is:

${\mathrm{QIR}}_k^{\left(g\right)}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{{\left|\right|V_k^H{C}^{\left(g\right)}\left|\right|}_F^2-\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{\underset{\underset{m\≠n}{m=1,n=1}}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mn}}\right|}^2},k=1,...,K,m=1,...,M$

Wherein,

Representing matrix Diagonal entry,

The matter of expression when user k selects g code book is done than size, and its minute subrepresentation employing wave beam forms back all subchannel quality information of user k, and its denominator represents to adopt wave beam to form mutual interference situation between all subchannels of user k of back.Therefore,

Can comprise mutual interference information between all subchannel quality information of user k and the subchannel simultaneously, Big more, the subchannel quality of expression user k is good more, and mutual interference is weak more between the subchannel, and this is to adopt the advantage place of QIR as quantitative criteria.It is as follows to quantize criterion based on QIR.

Described optimum code word indexing is by formula

${\hat{g}}_k=\arg \underset{1\≤g\≤G}{\max }{\mathrm{QIR}}_k^{\left(g\right)},k=1,...,K$

Wherein,

It is the optimum code word indexing of user k.

Wherein, beam forming process is summarized as follows, specifically comprises:

The optimum code word indexing of all online user's feedbacks is collected in the base station, finds out the CSI quantized value of corresponding code word as this user according to the codewords indexes of user feedback from code book, and forms vector as its wave beam, so the wave beam of user k formation vector is:

$B_k=C^{\left({\hat{g}}_k\right)}$

Give transmission signals X _kPremultiplication pre-coding matrix B _k, then transmitting on antenna is B _kX _kAt receiving terminal, received signal Y _kPremultiplication

Then the received signal of user k is:

${\hat{Y}}_k=U_k^H\left(H_k{B}_k{X}_k\right)+U_k^H{\mathrm{\Φ}}_k={\mathrm{\Λ}}_k{V}_k^H{B}_k{X}_k+{\widehat{\mathrm{\Φ}}}_k$

Wherein, power system capacity analysis, power division and user's dispatching algorithm are as follows, specifically comprise:

At first analytical system capacity and power distribution algorithm; Order

User k is at r son

The effective signal-to-noise ratio of channel (Effective SNR ESNR) is:

${\mathrm{\γ}}_k^r=\frac{{\mathrm{\λ}}_k^r{\mathrm{\ψ}}_{\mathrm{rr}}}{{\mathrm{\λ}}_k^r\underset{j,r\≠j}{\mathrm{\Σ}}{\mathrm{\ψ}}_{\mathrm{rj}}+{\mathrm{\σ}}_r^2}$

Wherein, ψ _RjBe matrix ψ _kThe capable j column element of r value, Be k subscriber channel matrix H _kR singular value,

It is the noise power of r subchannel.The base station is according to the feedback of scheduled user k Value, available water-filling algorithm maximized system capacity, throughput is by formula behind the employing water-filling algorithm

Determine, wherein

(0, A), R represents effective subchannel number to expression max.

The equitable proportion of next employing classics (Proportional Fairness Scheduling, PFS) user's dispatching algorithm is summarized as follows, the described user who is scheduled is by formula

$k^{*}\left(t\right)=\arg \underset{1\≤k\≤K}{\max }\frac{R_k\left(t\right)}{C_k\left(t\right)}$

Determine.Wherein user k speed more new formula be:

$C_k(t+1)=\left\{\begin{array}{cc}(1-\frac{1}{t_c})C_k\left(t\right)+\frac{1}{t_c}{R}_k\left(t\right),& k=k^{*}\\ (1-\frac{1}{t_c})C_k\left(t\right)& k\≠k^{*}\end{array}\right.$

R wherein _k(t) be the demand rate of user, t at time slot t _cBe time constant.

The present invention also provides a kind of system of transfer of data, comprising: the base station end adopts DFTC code book C, and utilizes this code book C to send a plurality of different training sequence symbols to subscriber equipment; Subscriber equipment adopts based on QIR to be quantification criterion quantized channel state information, and to the required optimum code word indexing of each user of base station feedback; Utilize the ESNR criterion to calculate its value then; The base station end according to the codewords indexes and the ESNR value of each user feedback, calculates its throughput, and selects the optimal scheduling user in conjunction with the PFS algorithm.Simultaneously, adopt water-filling algorithm to transmit data to it for this user's each subchannel dynamic assignment power and employing beam-forming technology.

Wherein, described base station end comprises:

The DFTC code book is adopted in the base station; Collect the optimum code word indexing and the effective signal-to-noise ratio value of all user feedbacks; And adopt the PFS algorithm to find out the optimal scheduling user of system based on above feedback parameter; Adopt water-filling algorithm dynamic assignment power and its optimum DFTC codeword transmission data of use for the optimal scheduling user.

Wherein, described subscriber equipment comprises:

Optimum code word acquiring unit is used to obtain each user's optimum code word; All subchannel effective signal-to-noise ratio acquiring units are used for the effective signal-to-noise ratio value that the user obtains its all subchannels;

Form (RBF, EBF and COBF) isotype with existing wave beam and compare, technical scheme of the present invention has following advantage:

The present invention both had been fit to few user's scene, also was fit to multi-user's scene and low SNR scene, reduced the system feedback amount simultaneously, improved systematic function.

Description of drawings

The structured flowchart of a kind of data transmission system of Fig. 1 embodiment of the invention;

The flow chart of a kind of data transmission system of Fig. 2 embodiment of the invention;

The simulation result that Fig. 3 the present invention and traditional wave beam form the throughput of system (RBF, EBF) compares numerical value schematic diagram (the present invention is called for short: QBF, Quantization Beamforming);

The simulation result of Fig. 4 the present invention throughput of system under different SNR is the numerical value schematic diagram relatively;

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.

Embodiment 1

A kind of structured flowchart that forms data transmission method and system for the Limited Feedback wave beam that quantizes criterion based on the channel quality interference ratio comprises the following steps: as shown in Figure 1

Step s201, at the base station end, the generation of DFTC, detailed process is:

The code book set is C={C ⁽⁰⁾..., C ^(G-1), it is as follows based on the method for generating codebooks of DFT,

Be g pre-coding matrix, guarantee that pre-coding matrix is a unitary matrice, wherein

Be m precoding vector in this pre-coding matrix, wherein each element generates based on Fourier basis, and is as follows:

$c_m^{\left(g\right)}=\frac{1}{\sqrt{M}}{[c_{0m}^{\left(g\right)},...,c_{(M-1)m}^{\left(g\right)}]}^T$

$c_{\mathrm{nm}}^{\left(g\right)}=\exp \left\{j\frac{2\mathrm{\πn}}{M}(m+\frac{g}{G})\right\},m,n=0,...,M-1,g=1,...,G$

Step s202, the base station utilizes the DFTC code book to send a plurality of different training sequence symbols to subscriber equipment.

Step s203, subscriber equipment is to base station feedback optimum quantization codewords indexes and corresponding ESNR value.Described optimum quantization codewords indexes is by formula

${\hat{g}}_k=\arg \underset{1\≤g\≤G}{\max }{\mathrm{QIR}}_k^{\left(g\right)},k=1,...,K$

Determine that wherein QIR is channel quality interference ratio (QIR) numerical value of user k under different code words, by formula

${\mathrm{QIR}}_k^{\left(g\right)}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{{\left|\right|V_k^H{C}^{\left(g\right)}\left|\right|}_F^2-\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{\underset{\underset{m\≠n}{m=1,n=1}}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mn}}\right|}^2},k=1,...,K,m=1,...,M$

Determine;

Described ESNR is by formula

${\mathrm{\γ}}_k^r=\frac{{\mathrm{\λ}}_k^r{\mathrm{\ψ}}_{\mathrm{rr}}}{{\mathrm{\λ}}_k^r\underset{j,r\≠j}{\mathrm{\Σ}}{\mathrm{\ψ}}_{\mathrm{rj}}+{\mathrm{\σ}}_r^2},k=1,...,K,r=1,...,R$

Determine.

Step s204, the base station scheduling user also adopts beam-forming technology to transmit data for corresponding user.Detailed process is: all active users of collection sub-district, base station

And calculate its capacity, adopt Proportional Fair (PFS) algorithmic dispatching user; The subscriber equipment formula of described scheduling

$k^{*}\left(t\right)=\arg \underset{1\≤k\≤K}{\max }\frac{R_k\left(t\right)}{C_k\left(t\right)}$

Determine;

To the scheduling the user according to it

Adopt water-filling algorithm dynamic assignment power, and adopt beam-forming technology transmission data for the user of scheduling;

Wherein, the systematic function of embodiment of the invention systematic function and traditional wave beam formation pattern compares the simulation result schematic diagram as shown in Figure 3; Embodiment of the invention system average throughput and SNR relation relatively simulation result respectively as shown in Figure 4 (M=2 wherein, N=2).

The system of a kind of transfer of data of the embodiment of the invention: the base station end adopts DFTC code book C, and utilizes this code book C to send a plurality of different training sequence symbols to subscriber equipment; Subscriber equipment adopts based on QIR to be quantification criterion quantized channel state information, and to the required optimum code word indexing of each user of base station feedback; Utilize the ESNR criterion to calculate its value then; The base station according to the codewords indexes and the ESNR value of each user feedback, is calculated its throughput, and is selected the optimal scheduling user in conjunction with the PFS algorithm; Simultaneously, adopt water-filling algorithm to distribute power and employing beam-forming technology transmission data for this user's subchannel.

The base station end comprises that the DFTC code book is adopted in the base station; Collect the optimum code word indexing and the effective signal-to-noise ratio value (ESNR) of all user feedbacks; And adopt the PFS algorithm to find out the optimal scheduling user of system based on above feedback parameter; Adopt water-filling algorithm to distribute power for the optimal scheduling user and use its optimum DFTC code word to do wave beam and form the transmission data.

Subscriber equipment comprises optimum code word acquiring unit, is used to obtain each user's optimum code word; All subchannel effective signal-to-noise ratio acquiring units are used for the effect snr value that the user obtains its all subchannels;

Main feature of the present invention is the quantification criterion of designing based on QIR, it can consider simultaneously that overall system quantizes the mutual interference between quality and the user's subchannel, based on this criterion, the user who guarantees scheduling has mutual interference minimum between equivalent channel quality maximum and the subchannel simultaneously, thereby improve systematic function, this is that all quantize the performance that criterion does not possess at present, and the present invention also adopts the DFTC code book in addition, further reduce the system feedback amount, save system resource overhead.

Should be understood that, for those of ordinary skills, can be improved according to the above description or conversion, and all these improvement and conversion all should belong to the protection range of claims of the present invention.

Claims (10)

1. the data transmission method based on the channel quality interference ratio is characterized in that, comprises following steps: adopt direct computation of DFT phyllopodium code book C, as beam forming matrix; Utilize described beam forming matrix C to send a plurality of different training symbols to user terminal; The described training symbol that described user terminal basis receives is estimated channel information separately, and is quantification criterion quantized channel information based on channel quality interference ratio QIR, selects the optimum code word of oneself from code book C, and its codewords indexes is fed back to the base station; Simultaneously, user terminal calculates the effective signal-to-noise ratio ESNR of subchannel according to above-mentioned optimum code word of selecting and described channel information, and described ESNR is fed back to the base station; The base station is according to the ESNR that feeds back scheduling optimal service user, and adopts the optimum code word that feeds back to do wave beam to form this user is carried out transfer of data.

2. the method for claim 1 is characterized in that, the generative process of described direct computation of DFT phyllopodium code book is as follows: the code book set is C={C ⁽⁰⁾.., C ^(G-1), wherein, G=2 ^B, B is a quantization bit; Wherein,

Be g pre-coding matrix, and guarantee that pre-coding matrix is a unitary matrice, wherein

Be m precoding vector in this pre-coding matrix, wherein each element generates based on Fourier basis, and is as follows:

$c_m^{\left(g\right)}=\frac{1}{\sqrt{M}}{[c_{0m}^{\left(g\right)},...,c_{(M-1)m}^{\left(g\right)}]}^T$

$c_{\mathrm{nm}}^{\left(g\right)}=\exp \left\{j\frac{2\mathrm{\πn}}{M}(m+\frac{g}{G})\right\},m,n=0,...,M-1,g=1,...,G;$

Being called for short the DFT-based code book is DFTC.

3. as the arbitrary described method of claim 1 to 2, it is characterized in that described channel quality interference ratio (QIR) criterion is as follows: to channel matrix H _kDo singular value (SVD) and decompose, promptly

U _kAnd V _kBe unitary matrice, Λ _kBe H _kThe diagonal matrix that singular value is formed.

4. as the arbitrary described method of claim 1 to 2, it is characterized in that described optimum code word indexing is by formula

K=1 ..., K determines that wherein Mass Interference compares parameter

Calculate according to following formula:

${\mathrm{QIR}}_k^{\left(g\right)}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{{\left|\right|V_k^H{C}^{\left(g\right)}\left|\right|}_F^2-\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}=\frac{\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mm}}\right|}^2}{\underset{m\≠n}{\overset{M}{\underset{m=1,n=1}{\mathrm{\Σ}}}}{\left|{\left[V_k^H{C}^{\left(g\right)}\right]}_{\mathrm{mn}}\right|}^2},k=1,...,K,m=1,...,M$

Wherein,

Representing matrix

Diagonal entry.

5. as the arbitrary described method of claim 1 to 2, it is characterized in that, the corresponding own optimum code word of described user terminal k

By formula

${\mathrm{\γ}}_k^r=\frac{{\mathrm{\λ}}_k^r{\mathrm{\ψ}}_{\mathrm{rr}}}{{\mathrm{\λ}}_k^r\underset{j,r\≠j}{\mathrm{\Σ}}{\mathrm{\ψ}}_{\mathrm{rj}}+{\mathrm{\σ}}_r^2},k=1,...,K,r=1,...,R$

Determine, wherein Ψ _RjBe Ψ _kThe capable j column element of r,

It is user k channel matrix H _kR singular value,

Be the noise power of r subchannel,

Be the SNR of user k at r subchannel.

6. as the arbitrary described method of claim 1 to 2, it is characterized in that according to the ESNR that is scheduled and feeds back, adopt the water-filling algorithm optimized system capacity in described base station, operation below concrete the execution: the base station is according to the feedback of scheduled user k

Value, adopt behind the water-filling algorithm throughput of system by formula:

Determine, wherein

Expression max (0, A).

7. as the arbitrary described method of claim 1 to 2, it is characterized in that adopt the Proportional Fair algorithm to select optimum subscriber equipment in described base station from the sub-district and carry out transfer of data, operation below concrete the execution: the subscriber equipment terminal of described selection is by formula

Determine; Wherein user k speed more new formula be:

$C_k(t+1)=\left\{\begin{array}{cc}(1-\frac{1}{t_c})C_k\left(t\right)+\frac{1}{t_c}{R}_k\left(t\right),& k=k^{*}\\ (1-\frac{1}{t_c})C_k\left(t\right)& k\≠k^{*}\end{array}\right.$

R wherein _k(t) be the demand rate of user, t at time slot t _cBe time constant.

8. one kind is the transmission system that quantizes criterion based on the channel quality interference ratio, it is characterized in that, comprising: user terminal and base station end, and described base station end adopts the DFTC code book, and utilizes this code book C to send a plurality of different training sequence symbols to described user terminal; Described user terminal adopts based on QIR to quantizing criterion quantized channel state information and calculating

And feed back the required optimum code word indexing of each described user terminal and its to described base station end

Value; Described base station end is according to the described optimum code word indexing of each user feedback and described

Value is calculated its throughput, and is selected the optimal scheduling user in conjunction with the PFS algorithm; Adopt water-filling algorithm to distribute power and adopt beam-forming technology transmission data to described optimal scheduling user's subchannel.

9. data transmission system as claimed in claim 8 is characterized in that, described base station end comprises: DFTC code book generation unit is used to generate code book; The system parameters collector unit is used to collect the optimum code word indexing and the effective signal-to-noise ratio value of all user feedbacks

User's scheduling unit is used for finding out sub-district optimal scheduling user; Power distributing unit is used for adopting water-filling algorithm to distribute power to the scheduled user; Wave beam forms the unit, is used for carrying out beam-forming technology transmission data to dispatched users.

10. as data transmission system as described in the claim 8, it is characterized in that described user terminal comprises: optimum code word acquiring unit is used to obtain the optimum code word of each described user terminal; All subchannel effective signal-to-noise ratio acquiring units are used for the effective signal-to-noise ratio value that the user obtains its all subchannels.

Images