CN100446434C - Simplified block linear equalizer with block space time transmit diverstiy - Google Patents

Abstract

The present invention is a method and system for receiving data transmitted using block space time transmit diversity (BSTTD) in a code division multiple access (CDMA) communication system. The system comprises a transmitter, for transmitting a first data field using a first antenna and a second data field using a second antenna, and a receiver. The receiver includes an antenna for receiving the first and second transmitted data fields, and a BSTTD joint detector which determines symbols of the first and second transmitted data fields using a minimum mean square error block linear equalizer model and an approximated Cholesky decomposition of the model. The receiver is simplified by the use of block banded propagation matrices.

Description

Receive method and the receiver and the wireless communication system of data

(1) technical field

The present invention is relevant for the communication system of a kind of request code division multiple access (CDMA) technology.More particularly, the present invention is relevant a kind of transmit diversity scheme that can be applicable to a CDMA communication.

(2) background technology

Propose spatial diversity and supported high message transmission rate user in the third generation wide band code division multiple access system.When using multiple antenna, this system can realize preferable gain and link quality, and this will cause the power system capacity of increase.Typically, utilized variation via using beam deflection or having made up already via variation.

Recently, realized already realizing the collaborative variation of using via usage space-timing code.The capacity increase is reached be equal to a factor of transmission and reception antenna quantity in the array.The space-time sign indicating number is to produce the upward running of a block input symbols (symbol) that a matrix by antenna and time transfers out.

In the past, the space-time transmission diversity system is its conjugate complex number of synchronous transmission when the continuous code element of transmission.Yet this type system may cause code element to overlap at receiving terminal.The amount that overlaps is to decide according to the length of the pulse reaction of propagating channel.Under variation duplex (TDD) pattern, must in engaging (joint) search receiner, this code element of compensation overlap.This engages transmitted symbol and conjugation thereof that detector will be estimated this overlapping, and this complexity that this joint will be surveyed improves.

Survey the increase of complexity for this joint of releiving, set up out the system that can transmit two data fields similar but inequality.Has one first part D ₁, and one second D partly ₂A data field be by first antenna transmission.One second data field is to produce D by modifying this first data field ₂The conjugation negative value ,-D ₂* be first part of this second data field, and D ₁Conjugation, D ₁* be second part.This second data field is by the second antenna synchronous transmission.

Although this diversified transmission plan will reduce the complexity of receiver, the receiver that is used for this scheme is still very complicated.This receiver is to utilize two to engage sniffer.Each engages sniffer can restore the wherein data field of the independent transmission of an antenna.This utensil is to handle two chiasma interferences of transmitting between the data fields by the transmission of handling each antenna discretely.As a result, each to engage sniffer be that transmission with other antenna is considered as noise.Be to use a decoder in conjunction with deciding by each code element that engages the sniffer recovery

And

It among Fig. 1 a calcspar that shows this system.Receiver in this system will have a high complexity because of using two joint detectors, so that cause higher receiver expense.

Therefore, need a kind of receiver utensil of modification badly.

(3) summary of the invention

The present invention is a kind of in a code division multiple access (CDMA) system, is used for receiving the method and system of the data of using block space transmission diversity (BSTTD) transmission.This system comprises that use one first antenna transmits one first data field and uses one second antenna to transmit a reflector and a receiver of one second data field.This receiver comprises an antenna that is used to receive this first and second transmission data field and uses a least mean-square error block linear equalizer model and approximate Cholesky (Cholesky) decomposition model decides a BSTTD of this first and second transmission data field to engage detector.

(4) description of drawings

Fig. 1 is the calcspar of communication system of a prior art of utilization space-time transmission diversity.

Fig. 2 is the calcspar according to a receiver of preferred embodiment of the present invention.

Fig. 3 is the matrix structure icon according to the approximate block space time transmission diversity (BSTTD) of preferred embodiment.

Fig. 4 is the flow chart that engages detection method according to the block space time transmission diversity of preferred embodiment.

(5) embodiment

The calcspar that Fig. 2 is in the CDMA communication system, be positioned at preferably that a subscriber terminal equipment (UE) locates according to a receiver 10 of preferred embodiment of the present invention.Although preferably make this receiver be positioned at this UE place, receiver 10 can be positioned at base station and operate when the cochain communication.Receiver 10 comprises that a BSTTD engages sniffer (BSTTD JD) 12, one channel estimation device 13 and an antenna 16.The antenna 16 of this UE is to receive various radio frequencies (RF) signal that comprises from first and second communication burst (burst) of a reflector.

This first and second communication burst comprises respectively as first and second above-mentioned data field.This first data field comprises partly D2 of the first part D1 and second; This second data field comprises the conjugation negative value-D2* of D2 and the conjugation D1* of D1.One typical communication burst has two parts of this data field that separates by an intermediate code (midamble).This burst has a protection cycle (guardperiod) at its end and has the different times of advent between this burst allowing.Each data field of a certain communication burst is to be encoded into the first data field D1, D2.Each data field of another communication burst is to be encoded into second data field-D2*, D1*.Launch to comprise in each data field and this data field that an intermediate code is to generate this first and second communication burst respectively.Each this communication burst is all by each other first and second antenna, transfer to receiver 10 with radio frequency signals.

This radio frequency signals that receives that will comprise this first and second communication burst is separated modulation and is delivered to channel estimation device 13 and BSTTD JD 12.Usually estimation unit 13 is to handle this to separate the signal of modulation and this channel information is delivered to BSTTD JD 12.

BSTTD JD 12 is the modulating signals of separating that comprise first and second communication burst and channel information that receive from channel estimation device 13.Use the spreading codes of this channel signals and this reflector, BSTTD JD 12 can estimate first and second data field of each communication burst data symbols D1, D2 ,-D2* ,-D1, and in conjunction with D1, D2 ,-D2* ,-D1 to be to restore initial data field D.

According to preferred embodiment of the present invention, BSTTD JD 12 is the data field that each receives is estimated in utilization based on the detector of the least mean-square error block linear equalizer (MMSE-BLE) of simplification data symbols.BSTTD JD 12 is according to following actions.A and B be the channel 1 that is relevant to antenna 1 respectively, with propogator matrix block polymerization (banded) variant that is relevant to the channel 2 of antenna 2.It is rewritten as 2 * 2 following block matrixes.

$A=\left[\begin{array}{cc}{A}_{11}& 0\\ A_{21}& A_{22}\end{array}\right],$ $B=\left[\begin{array}{cc}{B}_{11}& 0\\ B_{21}& B_{22}\end{array}\right].$

Edge is, with the received signal model of equation 1 expression block space time transmission diversity.

$\left[\begin{array}{c}{\stackrel{\&RightArrow;}{r}}_1\\ {{\stackrel{\&RightArrow;}{r}}_2}^{*}\end{array}\right]=\left[\begin{array}{cccc}{A}_{11}& 0& 0& {-B}_{11}\\ {B_{22}}^{*}& {{-\mathrm{<figure-callout\; id="21"\; label="B"\; filenames="C0280452900151.png,C0280452900152.png"\; state="\{\{state\}\}">B</figure-callout>}}_{21}}^{*}& {A_{21}}^{*}& {}_{22}\stackrel{*}{A}\end{array}\right]\left[\begin{array}{c}{\stackrel{\&RightArrow;}{d}}_1\\ {\stackrel{\&RightArrow;}{d}}_2\\ {{\stackrel{\&RightArrow;}{d}}_1}^{*}\\ {{\stackrel{\&RightArrow;}{d}}_2}^{*}\end{array}\right]+\left[\begin{array}{c}{\stackrel{\&RightArrow;}{n}}_1\\ {{\stackrel{\&RightArrow;}{n}}_2}^{*}\end{array}\right]$ Equation 1

Because the length of this block is launched much larger than this channel delay, therefore can ignore the interference between adjacent block A21 and the B21, and this received signal model can be simplified to equation 2:

Equation 2

In order to estimate this data block, can use the MMSE BLE algorithm of BSTTD.Utilize to become plain sound coupling filtration (whitening matched fitering), can be by following equation 3 and 4 these data blocks of expression.

${\hat{d}}_{\mathrm{wmf}1}={A_{11}}^H{\stackrel{\&RightArrow;}{r}}_1+{\left({B_{22}}^H{\stackrel{\&RightArrow;}{r}}_2\right)}^{*}$ Equation 3

${\hat{d}}_{\mathrm{wmf}2}={A_{22}}^H{\stackrel{\&RightArrow;}{r}}_2+{\left({B_{11}}^H{\stackrel{\&RightArrow;}{r}}_1\right)}^{*}$ Equation 4

With equation 5 expression MMSE-BLE outputs.

$\left[\begin{array}{c}{\stackrel{\&RightArrow;}{d}}_{\mathrm{mmse}1}\\ {\stackrel{\&RightArrow;}{d}}_{\mathrm{mmse}2}\end{array}\right]={(E^{H}E+{\mathrm{\&sigma;}}^{2}I)}^{-1}\left[\begin{array}{c}{\hat{d}}_{\mathrm{wmf}1}\\ {\hat{d}}_{\mathrm{wmf}2}^{*}\end{array}\right]$ Equation 5

E shows in equation 1.σ 2 is that average noise variation and I are unit matrixs.

In single antenna BLE, the complexity of block STTD mainly is to be caused by inverse matrix, and it is preferably with an approximate Cholesky and decomposes enforcement.

$D\&equiv;E^{H}E+{\mathrm{\&sigma;}}^{2}I)=\left[\begin{array}{cc}{D}_{11}& {D_{21}}^H\\ D_{21}& D_{22}\end{array}\right],$ Equation 6

The block matrix notation of the correlation matrix that Cholesky decomposes can be write as equation 6.

D11, D22 and D21 are respectively equation 7,8 and 9.

D ₁₁=A ₁₁ ^HA ₁₁+ (B ₂₂ ^HB ₂₂) ^*+ σ ²I equation 7

D ₂₂=B ₁₁ ^HB ₁₁+ (A ₂₂ ^HA ₂₂) ^*+ σ ²I equation 8

D ₂₁=(A ₂₂ ^HB ₂₂) ^*-B ₁₁ ^HA ₁₁Equation 9

Triangular matrices are to be written as equation 10 under Cholesky decomposing D=GGH.

$G=\left[\begin{array}{cc}{G}_{11}& 0\\ G_{21}& G_{22}\end{array}\right]$ Equation 10

Equation 11,12 and 13 is the relations between G11, G21, G22, D11, D21 and the D22.

G ₁₁G ₁₁ ^H=D ₁₁Equation 11

G ₂₁G ₁₁ ^H=D ₂₁ Equation 12

G ₂₂G ₂₂ ^H=D ₂₂-G ₂₁G ₂₁ ^H Equation 13

Separate the triangular system of following equation 14,15,16 and 17, the sequence of symhols that can obtain estimating.

$G_{11}{\stackrel{\&RightArrow;}{m}}_1={\hat{d}}_{\mathrm{wmf}1}$ Equation 14

$G_{22}{\stackrel{\&RightArrow;}{m}}_2={{\hat{d}}_{\mathrm{wmf}2}}^{*}-G_{21}{\stackrel{\&RightArrow;}{m}}_1$ Equation 15

${G_{22}}^H{{\hat{d}}_{\mathrm{mmse}2}}^{*}={\stackrel{\&RightArrow;}{m}}_2$ Equation 16

${G_{11}}^H{\hat{d}}_{\mathrm{mmse}1}={\stackrel{\&RightArrow;}{m}}_1-{G_{21}}^H{{\hat{d}}_{\mathrm{mmse}2}}^{*}$ Equation 17

In a single antenna system, need a Cholesky to decompose.Use a diversified antenna to decompose the complexity that (equation 11 and 13) and a forward direction substitution (equation 12) increase decoded symbol (symbol) because of needs two Cholesky.This will make the complexity of a BSTTD system increase above twice than a single antenna system person.In addition, the BSTTD decoder in this system can't be eliminated the interference of the first sub-block to the second sub-block, and this will cause more multiple error when surveying.

Below explanation will further reduce complexity.By the structure of transmission matrix, but block matrix notation A22 and the B22 of mat A11 and B11 are as follows.

$A_{22}=\left[\frac{A_{11}}{\begin{array}{cc}0& A_3\end{array}}\right]$ And $B_{22}=\left[\frac{B_{11}}{\begin{array}{cc}0& B_3\end{array}}\right]$

Equation 18,19 and 20 is relations of expression A11, A22, B11 and B22.

${A_{22}}^H{A}_{22}={A_{11}}^H{A}_{11}+\left[\begin{array}{cc}0& 0\\ 0& {A_3}^H{A}_3\end{array}\right]$ Equation 18

${B_{22}}^H{B}_{22}={B_{11}}^H{B}_{11}+\left[\begin{array}{cc}0& 0\\ 0& {B_3}^H{B}_3\end{array}\right]$ Equation 19

${A_{22}}^H{B}_{22}={A_{11}}^H{B}_{11}+\left[\begin{array}{cc}0& 0\\ 0& {A_3}^H{B}_3\end{array}\right]$ Equation 20

Person skilled in the art person can understand, A ₂₂ ^HA ₂₂, B ₂₂ ^HB ₂₂And A ₂₂ ^HB ₂₂Be block Toeplitz matrix, but A ₁₁ ^HA ₁₁, B ₁₁ ^HB ₁₁And A ₁₁ ^HB ₁₁Be because of the sub-block in the lower right corner of equation 18,19 and 20 in last really not so.

Behind the substitution equation 18, equation 4 will become equation 21.

$D_{11}={A_{22}}^H{A}_{22}+{\left({B_{22}}^H{B}_{22}\right)}^{*}+{\mathrm{\&sigma;}}^{2}I-\left[\begin{array}{cc}0& 0\\ 0& {A_3}^H{A}_3\end{array}\right]$ Equation 21

Equation 21 is block hermitian (Hermitian).But mat is ignored last Cholesky decomposition repetition variant and is come separating of approximate equation formula 7, and promptly equation 22.

${\hat{G}}_{11}{{\hat{G}}_{11}}^H={\hat{D}}_{11}$ Equation 22

D11 is according to equation 23.

D ₁₁=A ₂₂ ^HA ₂₂+ (B ₂₂ ^HB ₂₂) ^*+ σ ²I equation 23

Equation 22 is block Toeplitz approximate matrixs.Its complexity is to equate with approximate decomposition under the single antenna situation.Person skilled in the art person will recognize that aforesaid equation will cause one of G11 to be similar to, and lower the complexity of BSTTD JD 12.

Can in G22 approximate, find the further complexity of attenuating BSTTD JD 12.By equation 11 and 12, equation will become equation 24.

G ₂₂G ₂₂ ^H=D ₂₂-D ₂₁D ₁₁ ^-1D ₂₁ ^HEquation 24

Suppose norm (D ₂₂Norm (the D of)＞＞ ₂₁D ₁₁ ^-1D ₂₁ ^H), equation 24 will become equation 25.G ₂₂G ₂₂ ^H≈ D22 equation 25

In addition, by equation 8,9 and 12, will realize equation 26.

$D_{22}={{\hat{D}}_{11}}^{*}-\left[\begin{array}{cc}0& 0\\ 0& {B_3}^H{B}_3\end{array}\right]$ Equation 26

Approximate similar in appearance to above-mentioned G11, but mat is ignored last Cholesky and is decomposed and repeat variant and be similar to above-mentioned separating and become equation 27.

${\hat{G}}_{22}={{\hat{G}}_{11}}^{*}$ Equation 27

Approximate by this, G22, and therefore D22 (equation 8 and 13) need not obvious calculating.Be that the complexity that the Cholesky with, BSTTD decomposes will be identical with single antenna system person.

BSTTD is that the matrix G21 of equation 12,15 and 17 is relevant than the main complexity of single antenna complexity.Complex operation number of times in the equation 15 and 17 is identical with the nonzero element quantity of G21.Less nonzero element will reduce the complexity of equation 15 and 17.The method that lowers complexity is a hypothesis=0.Yet this being similar to will be introduced an error to equational separating, and not wish on the typical case so.

Therefore, the other method of minimizing complexity is to be similar to according to following person By equation 9 and 12, will cause equation 28.

${\hat{G}}_{21}{{\hat{G}}_{11}}^H={\hat{D}}_{21}$ Equation 28

Be according to equation 29.

D ₂₁=(A ₂₂ ^HB ₂₂) ^*-B ₂₂ ^HA ₂₂Equation 29

Equation 29 will cause a block Toeplitz matrix.Yet its general solution will be separated too complexity for multiple forward direction triangular system because of it, can't carry out easily.But, can use following character to simplify.

Character 1: matrix

Be antisymmetry block Toeplitz, promptly ${\hat{D}}_{21}=-{{\hat{D}}_{21}}^T.$

The diagonal angle item must be zero.

Character 2:

All items (entry) be all zero, the element in sub-block matrix last column or last row (seeing also Fig. 2).

Character 3:

Have a block Toeplitz structure.

Character 4:

Be that frequency range is equal to the below block of (L.Ka-1).(seeing also Fig. 2).L is the non-zero block quantity at the first column or row block place.It is equal to interferes length to add 1 between code element, that is L=Lisi+1, and wherein Lisi=ceil (W/SF), W are channel length and are the smallest positive integral of indication greater than x.Ka is the sum of sign indicating number (physical channels) initiatively, such as Ka=K+1 and wherein have K DCH in the BCH time slot.

Use above-mentioned characteristic and be approximated to and have D in the characteristic of being same as 2 ₂₁The block polymer matrix of sub-block structure can significantly simplify complexity.Promptly show this approximate construction among Fig. 3 (c).Fig. 3 (d) shows to have the definite of the collar size (collar scale) that is different from Fig. 3 (b) By above characteristic and following approximate the simplification

Calculating.

Approximate 1:

To have frequency range (L? Ka-1) reach the lower area polymer matrix on.

Approximate 2: Have and identical structure.

By approximate 1, simplification

Can be expressed as:

$\left[\begin{array}{cccccccc}{f}_{11}& {\mathrm{<figure-callout\; id="12"\; label="f"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">f</figure-callout>}}_{12}& ...& f_{1\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="C0280452900154.png"\; state="\{\{state\}\}">L</figure-callout>}}& 0& ...& ...& 0\\ f_{21}& f_{11}& {\mathrm{<figure-callout\; id="12"\; label="f"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">f</figure-callout>}}_{12}& ...& f_{1\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="C0280452900154.png"\; state="\{\{state\}\}">L</figure-callout>}}& 0& & ...\\ & f_{21}& f_{11}& & & ...& ...& ...\\ f_{L1}& & & & & & ...& 0\\ 0& f_{L1}& & & & & & f_{1L}\\ ...& 0& ...& ...& & & & ...\\ ...& & ...& ...& & & f_{11}& {\mathrm{<figure-callout\; id="12"\; label="f"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">f</figure-callout>}}_{12}\\ 0& ...& ...& 0& f_{L1}& & f_{21}& f_{11}\end{array}\right]$

The block matrix notation of correlation matrix and below triangular matrices can be written as equation 30 and 31.

${\hat{G}}_{11}=\left[\begin{array}{cccccccc}{g}_{11}& 0& ...& & ...& ...& 0& \\ g_{21}& g_{22}& 0& ...& & & ...& \\ ...& g_{32}& g_{33}& & & & ...& \\ g_{L1}& ...& & ...& & & 0& \\ 0& ...& & & & & & \\ ...& 0& g_{N,N-L+1}& & & g_{N,N}& & \\ ...& & & & & & & \\ 0& ...& 0& g_{N,N-L+1}& & & & g_{N,N}\end{array}\right]$ Equation 30

$\left[\begin{array}{cccccccc}{d}_{11}& {\mathrm{<figure-callout\; id="12"\; label="d"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">d</figure-callout>}}_{12}& ...& d_{1\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="C0280452900154.png"\; state="\{\{state\}\}">L</figure-callout>}}& 0& ...& ...& 0\\ -d_{12}^T& d_{11}& {\mathrm{<figure-callout\; id="12"\; label="d"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">d</figure-callout>}}_{12}& ...& d_{1\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="C0280452900154.png"\; state="\{\{state\}\}">L</figure-callout>}}& 0& & ...\\ ...& d_{12}^T& d_{11}& & & & & ...\\ -d_{1L}^T& & & ...& & & & 0\\ 0& -d_{1L}^T& & & & & & d_{1L}\\ ...& 0& & & & & & \\ ...& ...& ...& & & d_{11}& {\mathrm{<figure-callout\; id="12"\; label="d"\; filenames="C0280452900142.png"\; state="\{\{state\}\}">d</figure-callout>}}_{12}& \\ 0& ...& ...& 0& -d_{12}^T& ....& -d_{12}^T& d_{11}\end{array}\right]$ Equation 31

d ₁₁Be equal to-d ₁₁ ^T: d _IjAnd f _IjAccording to character 1 and approximate 2 and have a following structure.

$d_{\mathrm{ij}}=\left[\begin{array}{ccccc}0& 0& ...& 0& \&times;\\ 0& 0& & 0& \&times;\\ ...& & & ...& \&times;\\ 0& 0& ...& 0& \&times;\\ \&times;& \&times;& \&times;& \&times;& \&times;\end{array}\right]$ And $f_{\mathrm{ij}}=\left[\begin{array}{cccccc}& 0& 0& ...& 0& a_{1\mathrm{<figure-callout\; id="0"\; label="Ka"\; filenames="C0280452900154.png"\; state="\{\{state\}\}">Ka</figure-callout>}}\\ & 0& 0& & 0& a_{2\mathrm{Ka}}\\ ...& ...& & ...& & \&times;\\ & 0& 0& ...& 0& a_{\mathrm{Ka}-1,\mathrm{Ka}}\\ & d_{\mathrm{Ka}-1}& d_{\mathrm{Ka}-2}& \&times;& \&times;& d_{\mathrm{Ka},\mathrm{Ka}}\end{array}\right]$

$F{\hat{G}}_{11}^H={\hat{D}}_{21}$

To separate be to be got by first row block and the first row block of calculation equation 32 and 33.

$f_{n1}{g}_{11}^H=-d_{1n}^T$ N=1,2 .L., equation 32

$f_{1n}{g}_{\mathrm{nn}}^H=d_{1n}-\underset{i=1}{\overset{n-1}{\mathrm{\&Sigma;}}}{f}_{1i}{g}_{\mathrm{ni}}^H,n=2,.L,$ Equation 33

A=[a with above-mentioned matrix structure _Ij] ^Ka _{I, j=1}And D=[d _Ij] ^Ka _{I, j=1}, and lower triangular matrix G=[g _Ij] ^Ka _{I, j=1}Can satisfy matrix equation formula AG ^H=D.K _dBe the quantity and the K of dedicated channel (DCH) _a=K _dThe+1st, the physical channels sum in broadcasting channel (BCH) time slot.The one K of last column vector _dElement is to be got divided by real number value by complex values, as equation 34.

$a_{\mathrm{nKa}}=\frac{d_{\mathrm{nKa}}}{g_{\mathrm{KaKa}}},n=\mathrm{1,2}.K_d,-1$ Equation 34

A forward direction substitution of mat Ka can get last column matrix of matrix A, by equation 35 expressions.

$G\&CenterDot;\left[\begin{array}{c}{a_{\mathrm{Ka}1}}^{*}\\ {a_{\mathrm{Ka}2}}^{*}\\ ...\\ {a_{\mathrm{KaKa}}}^{*}\end{array}\right]=\left[\begin{array}{c}{d_{\mathrm{Ka}1}}^{*}\\ {d_{\mathrm{Ka}2}}^{*}\\ ...\\ {d_{\mathrm{KaKa}}}^{*}\end{array}\right]$ Equation 35

In addition, the right-hand side of equation 33 comprises matrix multiplication.Each matrix multiplication can be considered as K because of neutral element _d+ (K _d+ 1) ²Plural multiplier.

Can use above-mentioned being similar to simplify the BSTTD algorithm, as follows:

Machine equation

Matching filter: (3), (4)

Correlation computations: (23), (29)

Cholesky decomposes (22), (32), (33)

The forward direction substitution is according to equation 36 and 37:

${\hat{G}}_{11}{\stackrel{\&RightArrow;}{m}}_1={\hat{d}}_{\mathrm{wmf}1}$ Equation 36

${\hat{G}}_{11}{\stackrel{\&RightArrow;}{m}}_2={\hat{d}}_{\mathrm{wmf}2}-{\left({\hat{G}}_{21}{\stackrel{\&RightArrow;}{m}}_1\right)}^{*}$ Equation 37

Oppositely substitution is according to equation 38 and 39:

${{\hat{G}}_{11}}^H{\hat{d}}_{\mathrm{mmse}2}={\stackrel{\&RightArrow;}{m}}_2$ Equation 38

${{\hat{G}}_{11}}^H{\hat{d}}_{\mathrm{mmse}1}={\stackrel{\&RightArrow;}{m}}_1-{{\hat{G}}_{21}}^H{{\hat{d}}_{\mathrm{mmse}2}}^{*}$ Equation 39

Flow chart in conjunction with Fig. 3 illustrates preferred embodiment.Can come this signal that receives of emulation by the interference of ignoring between this data block, for example according to equation 2 (step 401).This vector that receives is for example to become plain sound coupling according to equation 3 and 4 (steps 402) to filter.The one Cholesky factor of decision equation 10 patterns is with separate (step 403) as MMSE BLE.According to equation 22, the Cholesky factor of mat calculating (equation 7) submatrix D, D11 is calculated the submatrix (step 404) of G, G11 again.According to equation 26, conjugate complex number, the G11* of use G11 calculates another approximate (step 405) of G, G22 one submatrix.User's formula 31 and 32 another submatrixs with G, G21 are approximated to a top and below block polymer matrix (step 406).Use forward direction and reverse substitution, mat equation 35,36,37 and 38 solves code element and (step 407) of two data fields.Then re-use decoder 15, original transmission data (step 408) are deciphered and decided to mat.

Although with regard to preferred embodiment the present invention is described,, person skilled in the art person belongs to other interior modification of the scope of the invention that following patent claim is summarized yet can knowing discovery.

Claims (15)

1. one kind is used to receive the method for using the data that block space time transmission diversity transmitted, one block space time transmission diversity reflector is to use one first antenna transmission, one first data field and uses one second antenna transmission, one second data field, this second data field is to be produced by the block of rearranging this first data field, and this method comprises:

Reception comprise this first and second the transmission data field both one of the reception vector;

Become plain sound coupling and filter this reception vector;

Utilize a least mean-square error block linear equalizer model to decide the code element of this first and second data field, this model is the interference of ignoring between this data block; And

The code element that use an approximate Cholesky of this model to decompose, forward direction and reverse substitution decides this first and second transmission data field.

2. the method for claim 1 is characterized in that, deciphers this first and second data field transmitted symbol to estimate the data of this first data field.

3. the method for claim 1, it is characterized in that, a Cholesky factor that is used for this approximate Cholesky decomposition comprises four block matrixes, and one first block of these four block matrixes is complex conjugates that are approximately one second block of these four block matrixes.

4. method as claimed in claim 3 is characterized in that, one the 3rd block of these four block matrixes is to be approximately a top and below block polymer matrix.

5. method as claimed in claim 4 is characterized in that, the 3rd block of these four block matrixes comprises being zero element entirely.

6. one kind is used to receive from a block space time transmission diversity reflector and uses the receiver of the data of block space time transmission diversity transmission, this block space time transmission diversity reflector is to use one first antenna transmission, one first data field and uses one second antenna transmission, one second data field, this second data field is to be produced by the block of rearranging this first data field, and this receiver comprises:

One antenna, receive comprise this first and second transmission data field both one of vector;

One block space time transmission diversity joint detectors, the code element that utilize an approximate Cholesky of a least mean-square error block linear equalizer model, this model to decompose, forward direction and reverse substitution decides this first and second transmission data field; And

This model is the interference of ignoring between this data block.

7. receiver as claimed in claim 6 is characterized in that, this joint detectors is that decoding this first and second is transmitted the code element of data field to estimate the data of this first data field.

8. receiver as claimed in claim 6, it is characterized in that, a Cholesky factor that is used for this approximate Cholesky decomposition comprises four block matrixes, and one first block of these four block matrixes is complex conjugates that are approximately one second block of these four block matrixes.

9. receiver as claimed in claim 8 is characterized in that, one the 3rd block of these four block matrixes is to be approximately a top and below block polymer matrix.

10. receiver as claimed in claim 9 is characterized in that, the 3rd block of these four block matrixes comprises being zero element entirely.

11. a code division multiple access communication system comprises:

One block space time transmission diversity reflector, it utilizes block space time transmission diversity, with one first antenna transmission, one first data field and use one second antenna transmission, one second data field, this second data field is to be produced by the block of rearranging this first data field; And

One receiver is used to receive the data of using block space time transmission diversity to be transmitted, comprising:

One antenna receives and to comprise this first and second both vectors of transmission data field;

One block space time transmission diversity joint detectors, the code element that utilize an approximate Cholesky of a least mean-square error block linear equalizer model, this model to decompose, forward direction and reverse substitution decides this first and second transmission data field; And

This model is the interference of ignoring between this data block.

12. system as claimed in claim 11 is characterized in that, this joint detectors is that decoding this first and second is transmitted the code element of data field to estimate the data of this first data field.

13. system as claimed in claim 11, it is characterized in that, a Cholesky factor that is used for this approximate Cholesky decomposition comprises four block matrixes, and one first block of these four block matrixes is complex conjugates that are approximately one second block of these four block matrixes.

14. system as claimed in claim 13 is characterized in that, one the 3rd block of these four block matrixes is to be approximately a top and below block polymer matrix.

15. system as claimed in claim 14 is characterized in that, the 3rd block of these four block matrixes comprises being zero element entirely.

Images