CN100359836C - Method and realizing apparatus for interlacing orthogonal transmitting diversity least squares soft decode between coordinates - Google Patents

Abstract

The present invention discloses a method and a realizing device for interlacing orthogonal transmitting diversity least squares soft decode between coordinates, which comprises the following steps: a signal vector transmitted by a transmitting end is received by a receiving end according to an alamouti orthogonal receiving principle, and the left of the signal vector is multiplied by a conjugation transpose of a channel estimation matrix to obtain a conversion vector; virtual parts of each component in the conversion vector are mutually interchanged according to an interchange rule of virtual parts of each component of a signal vector transmitted by the transmitting end to obtain a calculation vector; the channel weighting rotation of the calculation vector is carried out under the least square meaning to obtain constellation point coordinates after judgment. The realizing device comprises a vector orthogonal conversion module, a vector virtual part exchange module and a vector weighting rotation module which are orderly connected. The present invention is the optimal soft decoding method of CIOD under the least square meaning, and can ensure the full transmitting diversity under a plurality of antennas. A code rate is 1. A single symbol decoding can be realized, and the diversity gain of a receive chain is enhanced.

Description

Interlacing orthogonal transmitting diversity least squares soft decode between coordinates method and implement device

Technical field

The present invention relates to orthogonal transmitting diversity least squares soft decode method in the moving communicating field, especially relate to a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates method and implement device.

Background technology

In the mobile communication, physical channel must occupy regular hour, frequency, power and space, and these resources are limited, and is all Channel Sharing.In fdma system, mainly divide different channels with the difference of frequency; In time-division multiple address system, on the basis that frequency is separated, many channels have been marked off again in time; Code division multiple access system is the self-interference system, and its essence is many Channel Sharing power resources, and what distinguish these channels is the PN sign indicating number.Time, frequency and power resource can be regarded as the resource of one dimension, and its separation can represent on a reference axis that industry is very ripe to the utilization of these resources.And space resources is a kind of two dimension or even three-dimensional resource, and the technical difficulty that really makes full use of space resources is also corresponding bigger.

In recent years, along with userbase enlarges and the class of business variation, cellular carrier to the requirement of wireless system voice and high-speed data providing capability also corresponding improved many.In order to satisfy these requirements, must there be new technology to occur and application, farthest to improve the capacity in the existing bandwidth, improve the availability of frequency spectrum.

MIMO (Multiple-Input Multiple-Output) system is proposed in 1908 by Marconi (Marconi) the earliest, and it utilizes many antennas to suppress channel fading.According to the transmitting-receiving two-end antenna amount, with respect to common SISO (Single-Input Single-Output) system, MIMO can also comprise SIMO (Single-Input Multi-ple-Output) system and MISO (Multiple-InputSingle-Output) system.Can prove the linear increase of channel capacity this moment along with the increase of antenna amount.That is to say and can utilize mimo channel to improve the wireless channel capacity exponentially, under the situation that does not increase bandwidth and antenna transmission power, the availability of frequency spectrum can improve exponentially.

Therefore, in recent years, many mechanisms were all in the Space Time Coding technology of research based on the MIMO antenna system.1998, Canada scholar Siavash.M.Alamouti proposes a kind of Space-Time Block Coding form based on two transmit antennas, because it can improve the quality that receives link greatly and realize simply under fading channel, the receiver decoding complexity is low and by 3GPP (3 ^RdGeneration Partnership Project the 3rd generation mobile communication partnership project) and IEEE802.16 agreements such as (Institute ofElectrical and ElectronicsEngineers IEEE) adopt.After this, another one Canada scholar Vahid.Tarokh promotes this method, based on the orthogonal design principle multi-antenna space time block code is proposed, but when number of transmit antennas greater than 2 and when adopting the modulation of complex signal planisphere, the STBC that Tarokh provides (Space-Time Block Code Space-Time Block Coding) can not realize the full rate transmission.Be to solve the full rate transmission problem, some scholars have proposed many quasi-orthogonal STBC methods for designing, but these methods be not lost the part diversity gain, be exactly encoding rate be not 1, perhaps do not realize simple single symbol substitution.Therefore, how to realize that full rate, the full diversity transmission under many antennas condition is one of hot issue in the sky time-code research field, people study respectively from different perspectives.

India scholar B.Sundar Rajan has proposed the orthogonal space time-code of a kind of CIOD (interlacing orthogonal design between Co-ordinate InterleavedOrthogonal Design coordinate), can satisfy full transmit diversity under many antennas, encoding rate is 1 and can realizes single symbol substitution, but Rajan just proves the orthogonal space time-code of CIOD theoretically and can accomplish full transmit diversity, this proof is not a constructive proof, be the implementation method of the Rajan CIOD that just proposed transmitting terminal and the method that receives hard decision, do not provide corresponding soft decoding algorithm, yet hard decision not optimum under the least square meaning.In addition, " the Comment on thePre-coding of STC for 3﹠amp that in standard proposals, mentions from the scholar Wen Tong of Nortel Networks; 4 Transmit Antennas " in, think that this kind method does not gain in whole link, promptly CIOD has gain under codeless situation, and not gain under the situation of coding is being arranged.

Summary of the invention

The present invention provides a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates method and implement device, this soft decoding method is the optimum soft decoding method of CIOD under the least square meaning, can guarantee full transmit diversity under many antennas, and encoding rate is 1, improves the diversity gain that receives link.

In order to solve the problems of the technologies described above, the invention provides a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates method, comprise the steps:

(a) receive principle according to the Alamouti quadrature, receiving terminal receives the signal phasor that transmitting terminal sends, and to the conjugate transpose of this signal phasor premultiplication with channel estimate matrix, obtains transformation vector;

(b) will exchange mutually according to transmitting terminal each component imaginary part exchange regulation through the imaginary part of each component in the transformation vector that obtains after step (a) processing, obtain compute vectors transmitting vector;

(c) will under the least square meaning, carry out channel weights rotation, the constellation point coordinate after obtaining adjudicating through the compute vectors that obtains after step (b) processing.

Further, said method can have following characteristics: the channel estimate matrix in the described step (a) is by pilot tone, and methods such as prefix draw after channel is estimated, and obtains its conjugate transpose through matrixing.

Further, said method can have following characteristics: described step (c) further can be divided into following steps:

(c1) will filter through the noise that obtains in the compute vectors after step (b) processing, obtain the least-squares estimation matrix;

(c2) will carry out channel weights rotation, the constellation point coordinate after obtaining adjudicating through obtaining the least-squares estimation matrix after step (c1) processing.

Further, said method can have following characteristics: receiving terminal is an antenna in the described step (a), transmitting terminal is 4 antennas, an antenna of receiving terminal receives 4 dimensional signal vectors of 4 antenna transmission of transmitting terminal, and the imaginary part of first dimension of this 4 dimensional signal vector and third dimension component exchanges mutually, second dimension exchanges mutually with the imaginary part of fourth dimension component; In the described step (b) in the transformation vector imaginary part of each component exchange mutually, carry out according to the mode that the imaginary part of first dimension and third dimension component imaginary part, second dimension and fourth dimension component exchanges mutually, obtain 4 dimension compute vectors; 4 dimension compute vectors are carried out the channel weights rotation in the described step (c) under the least square meaning, obtain 4 constellation point coordinates and soft information thereof.

In order to solve the problems of the technologies described above, the present invention also provides a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device, this device includes vector orthogonal transform module, vector imaginary part Switching Module and the vector weighting rotary module that is connected successively, wherein, described vector orthogonal transform module is used for the received signal vector, and, obtain transformation vector to the conjugate transpose of this signal phasor premultiplication with channel estimate matrix; Described vector imaginary part Switching Module is used for the imaginary part with described each component of transformation vector, exchanges mutually according to transmitting terminal each component imaginary part exchange regulation to the signal phasor launched, obtains compute vectors; Described vector weighting rotary module is used for described compute vectors is carried out channel weights rotation under the least square meaning, obtains and export constellation point coordinate and soft information thereof after the judgement.

Further, above-mentioned interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device can have following characteristics: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module can be realized after the establishment respective code by general CPU (central processing unit).

Further, above-mentioned interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device can have following characteristics: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module can be realized by DSP (Digital Signal Processing digital signal processor), FPGA (FieldProgrammable Gate Array field programmable gate array) or ASIC (Application SpecficIntegrated Circuit application-specific integrated circuit (ASIC)).

Compared with prior art, interlacing orthogonal transmitting diversity least squares soft decode between coordinates method of the present invention and implement device have the following advantages:

A, satisfy optimum soft decoding requirement under the least square meaning through the constellation point coordinate that the present invention obtains, promptly have the suitable exponent number factor, make from the angle of average meaning, the present invention can realize full transmit diversity, is CIOD optimum soft decoding method under the least square meaning;

B, the present invention can guarantee under many antennas full transmit diversity, and encoding rate is 1, can realize single symbol substitution, the diversity gain that receives link is improved, and exports soft information.

Description of drawings

Fig. 1 is the flow chart of interlacing orthogonal transmitting diversity least squares soft decode between coordinates method among the present invention.

Embodiment

For understanding interlacing orthogonal transmitting diversity least squares soft decode between coordinates method of the present invention and implement device in depth, the present invention is described in detail below in conjunction with drawings and the specific embodiments.

Present embodiment is the optimum soft decoding method of CIOD that adopts 4 transmit antennas, and the CIOD least squares soft decode method situation of all the other many antennas is similar to the situation of 4 antennas, and their similar soft decoding method equally should be in protection scope of the present invention.

Before explanation CIOD least squares soft decode method of the present invention, at first introduce the launch scenario of CIOD, need to suppose four symbols of emission to be

x _i＝x _iI+jx _iQ，i＝1，2，3，4 (1.1)

According to the theory of CIOD, at first need angle the symbol of emission rotation θ, obtain s _i, i=1,2,3,4, promptly

s _i＝(x _iI+jx _iQ)e ^jθ＝(x _iI?cosθ-x _iQ?sinθ)+j(x _iQ?cosθ+x _iI?sinθ) i＝1，2，3，4(1.2)

Exchange s then ₁And s ₃Imaginary part, the exchange s ₂And s ₄Imaginary part, obtain

I=1,2,3,4

${\tilde{s}}_1=(x_{1I}\cos \mathrm{\θ}-x_{1Q}\sin \mathrm{\θ})+j(x_{3Q}\cos \mathrm{\θ}+x_{3I}\sin \mathrm{\θ})$

${\tilde{s}}_2=(x_{2I}\cos \mathrm{\θ}-x_{2Q}\sin \mathrm{\θ})+j(x_{4Q}\cos \mathrm{\θ}+x_{4I}\sin \mathrm{\θ})$

${\tilde{s}}_3=(x_{3I}\cos \mathrm{\θ}-x_{3Q}\sin \mathrm{\θ})+j(x_{1Q}\cos \mathrm{\θ}+x_{1I}\sin \mathrm{\θ})$

${\tilde{s}}_4=(x_{4I}\cos \mathrm{\θ}-x_{4Q}\sin \mathrm{\θ})+j(x_{2Q}\cos \mathrm{\θ}+x_{2I}\sin \mathrm{\θ})---\left(1.3\right)$

At antenna 1,2,3 and 4 by following matrix (space-time two-dimensional) form emission respectively, and wherein the line display transmitting antenna is tabulated and shown the symbol time slot,

Represent respectively corresponding to

With

Conjugation:

$\left[\begin{array}{cccc}{\tilde{s}}_1& -{\tilde{s}}_2^{*}& 0& 0\\ {\tilde{s}}_2& {\tilde{s}}_1^{*}& 0& 0\\ 0& 0& {\tilde{s}}_3& -{\tilde{s}}_4^{*}\\ 0& 0& {\tilde{s}}_4& {\tilde{s}}_3^{*}\end{array}\right]---\left(1.4\right)$

Situation when above-mentioned formula (1.1) to (1.4) has illustrated the emission of CIOD method, the step of many antennas base band signal process is: the first step is to separate CIOD, promptly to the soft-decision of each transmitting antenna constellation point, this step is that multiaerial system is peculiar, it is exactly will distinguish each transmitting antenna at a time respectively to have launched which constellation point in the planisphere that constellation point is carried out soft-decision, and corresponding soft information is provided; Second step was each layer digital demodulation, i.e. mapping from the constellation point to the bit; The 3rd step was deinterleaving, decoding; The 4th step was a bit decision.Therefore, separate CIOD, promptly quite crucial to the soft-decision of each transmitting antenna constellation point.

The signal of Space Time Coding arrives receiving terminal through the relevant less wireless channel of too much bar, receiving terminal need be known the ideal parameters of each wireless channel usually, this just requires different pilot tone or the prefix sequence of transmitting terminal emission, receiving terminal adopts a large amount of channel estimating computings, just can reach the space and time diversity effect.

The following describes the embodiment of 4 transmitting antenna CIOD least squares soft decode methods of the present invention.

Suppose transmitting antenna 1,2,3 and 4 gains of arriving between the reception antenna (suppose a reception antenna, many reception antennas are similar) are H _i, i=1,2,3,4, suppose that again the signal that reception antenna receives is r _i, i=1,2,3,4.Then the signal that receives of receiving terminal can be written as the form of vector representation:

$r=H\tilde{s}+n---\left(1.5\right)$

Wherein, vector n represents noise, matrix H is represented channel estimate matrix, slightly through putting in order:

$\left[\begin{array}{c}{r}_1\\ {\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">r</figure-callout>}}_2^{*}\\ r_3\\ r_4^{*}\end{array}\right]=\left[\begin{array}{cccc}{H}_1& H_2& 0& 0\\ H_2^{*}& {-H}_1^{*}& 0& 0\\ 0& 0& H_3& H_4\\ 0& 0& H_4^{*}& -H_3^{*}\end{array}\right]\left[\begin{array}{c}{\tilde{s}}_1\\ {\tilde{s}}_2\\ {\tilde{s}}_3\\ {\tilde{s}}_4\end{array}\right]+\left[\begin{array}{c}{n}_1\\ {\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_2^{*}\\ n_3\\ n_4^{*}\end{array}\right]---\left(1.6\right)$

As scheme shown in the l, this CIOD least squares soft decode method comprises the steps:

Step 1, receiving terminal receives the signal phasor r that transmitting terminal sends, and to the conjugate transpose H of this signal phasor r premultiplication with channel estimate matrix ^H, this channel estimate matrix H draws after by pilot tone channel being estimated, obtains transformation vector after matrix operation

Be expressed as follows

$\left[\begin{array}{c}{{\tilde{s}}_1}^{\&prime;}\\ {{\tilde{s}}_2}^{\&prime;}\\ {{\tilde{s}}_3}^{\&prime;}\\ {{\tilde{s}}_4}^{\&prime;}\end{array}\right]=\left[\begin{array}{cccc}{\left|H_1\right|}^2+{\left|H_2\right|}^2& 0& 0& 0\\ 0& {\left|H_1\right|}^2+{\left|H_2\right|}^2& 0& 0\\ 0& 0& {\left|H_3\right|}^2+{\left|H_4\right|}^2& 0\\ 0& 0& 0& {\left|H_3\right|}^2+{\left|H_4\right|}^2\end{array}\right]\left[\begin{array}{c}{\tilde{s}}_1\\ {\tilde{s}}_2\\ {\tilde{s}}_3\\ {\tilde{s}}_4\end{array}\right]+\begin{array}{c}\left[\begin{array}{cccc}{H}_1^{*}& H_2& 0& 0\\ H_2^{*}& -H_1& 0& 0\\ 0& 0& H_3^{*}& H_4\\ 0& 0& H_4^{*}& -H_3\end{array}\right]\left[\begin{array}{c}{n}_1\\ {\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_2^{*}\\ n_3\\ n_4^{*}\end{array}\right]---\left(1.7\right)\end{array}$

In the following formula $\left[\begin{array}{c}{{\tilde{s}}_1}^{\&prime;}\\ {{\tilde{s}}_2}^{\&prime;}\\ {{\tilde{s}}_3}^{\&prime;}\\ {{\tilde{s}}_4}^{\&prime;}\end{array}\right]=\left[\begin{array}{cccc}{H}_1^{*}& H_2& 0& 0\\ H_2^{*}& -H_1& 0& 0\\ 0& 0& H_3^{*}& H_4\\ 0& 0& H_4^{*}& -H_3\end{array}\right]\left[\begin{array}{c}{r}_1\\ {\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">r</figure-callout>}}_2^{*}\\ r_3\\ r_4^{*}\end{array}\right]---\left(1.8\right)$

If | H ₁| ²+ | H ₂| ²=| H ₁₂| ², | H ₃| ²+ | H ₄| ²=| H ₃₄| ², then above-mentioned matrix form can be write as:

${{\tilde{s}}_1}^{\&prime;}={\left|H_{12}\right|}^2(x_{1I}\cos \mathrm{\&theta;}-x_{1Q}\sin \mathrm{\&theta;})+j{\left|H_{12}\right|}^2(x_{3Q}\cos \mathrm{\&theta;}+x_{3I}\sin \mathrm{\&theta;})+H_1^{*}{n}_1+H_2{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_2^{*}---\left(1.9\right)$

${{\tilde{s}}_2}^{\&prime;}={\left|H_{12}\right|}^2(x_{2I}\cos \mathrm{\&theta;}-x_{2Q}\sin \mathrm{\&theta;})+j{\left|H_{12}\right|}^2(x_{4Q}\cos \mathrm{\&theta;}+x_{4I}\sin \mathrm{\&theta;})+H_2^{*}{n}_1+H_1{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_2^{*}---(1.10)$

${{\tilde{s}}_3}^{\&prime;}={\left|H_{34}\right|}^2(x_{3I}\cos \mathrm{\&theta;}-x_{3Q}\sin \mathrm{\&theta;})+j{\left|H_{34}\right|}^2(x_{1Q}\cos \mathrm{\&theta;}+x_{1I}\sin \mathrm{\&theta;})+H_3^{*}{\mathrm{<figure-callout\; id="3"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_3+H_4{n}_4^{*}---(1.11)$

${{\tilde{s}}_4}^{\&prime;}={\left|H_{34}\right|}^2(x_{4I}\cos \mathrm{\&theta;}-x_{4Q}\sin \mathrm{\&theta;})+j{\left|H_{34}\right|}^2(x_{2Q}\cos \mathrm{\&theta;}+x_{2I}\sin \mathrm{\&theta;})+H_4^{*}{\mathrm{<figure-callout\; id="3"\; label="n"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">n</figure-callout>}}_3+H_3{n}_4^{*}---(1.12)$

Step 2, the transformation vector that will obtain after will handling through step 1

In each component

Imaginary part, exchange mutually according to transmitting terminal each component imaginary part exchange regulation transmitting vector, obtain compute vectors s ^/, promptly exchange component

With

Imaginary part, again the exchange With

Imaginary part, obtain s ^/ _iI=1,2,3,4, notice that the noise between formula (1.9) and the formula (1.11) is independently, the noise between formula (1.10) and the formula (1.12) is also independently, notice simultaneously that the noise between formula (1.9) and the formula (1.10) is not independent, the noise between formula (1.11) and the formula (1.12) is not independent yet yet.So be independently between the real part of each component and the imaginary part noise after the exchange imaginary part, be made as n _IIAnd n _IQ, i=1,2,3,4, i.e. n _IIAnd n _IQBetween separate.Exchange back compute vectors s ^/Each component can be written as:

s ₁′＝|H ₁₂| ²(x _1I?cosθ-x _1Q?sinθ)+n _1I+j(|H ₃₄| ²(x _1Q?cosθ+x _1I?sinθ)+n _1Q)(1.13)

s ₂′＝|H ₁₂| ²(X _2I?cosθ-x _2Q?sinθ)+n _2I+j(|H ₃₄| ²(x _2Q?cosθ+x _2I?sinθ)+n _2Q)(1.14)

s ₃′＝|H ₃₄| ²(x _3I?cosθ-x _3Q?sinθ)+n _3I+j(|H ₁₂| ²(x _3Q?cosθ+x _3I?sinθ)+n _3Q)(1.15)

s ₄′＝|H ₃₄| ²(x _4I?cosθ-x _4Q?sinθ)+n _4I+j(|H ₁₂| ²(x _4Q?cosθ+x _4I?sinθ)+n _4Q)(1.16)

Can see from above four equatioies, expect x _i=x _II+ jx _IQ, as long as be concerned about i equation, i=1,2,3,4.

Step 3 will obtain compute vectors s after handling through step 1, step 2 ^/In noise filter, obtain the least-squares estimation matrix; Subsequently this least-squares estimation matrix is carried out the channel weights rotation, the constellation point coordinate after demodulation obtains adjudicating.

Below with demodulation x ₁=x _1I+ jx _1QBe example, the rest may be inferred by analogy for it.Clearly, this moment, first equation (1.13) equivalence can be write as two equations:

Re(s ₁′)＝|H ₁₂| ²(x _1I?cosθ-x _iQ?sinθ)+n _1I (1.17)

Im(s ₁′)＝|H ₃₄| ²(x _1Q?cosθ+x _1I?sinθ)+n _1Q (1.18)

Write formula (1.17) and (1.18) form of matrix as, had

$\left[\begin{array}{c}\mathrm{Re}\left({s_1}^{\&prime;}\right)\\ \mathrm{Im}\left({s_1}^{\&prime;}\right)\end{array}\right]=\left[\begin{array}{cc}{\left|H_{12}\right|}^2\cos \mathrm{\&theta;}& -{\left|H_{12}\right|}^2\sin \mathrm{\&theta;}\\ {\left|H_{34}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{34}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}{x}_{1I}\\ x_{1}Q\end{array}\right]+\left[\begin{array}{c}{n}_{1I}\\ n_{1Q}\end{array}\right]---\left(1.19\right)$

Because n _1IAnd n _1QBe independent white noise, so this moment is to x ₁=x _1I+ jx _1QLeast-squares estimation be:

$\left[\begin{array}{c}{x}_{1I}\\ x_{1Q}\end{array}\right]\&ap;{\left[\begin{array}{cc}{\left|H_{12}\right|}^2\cos \mathrm{\&theta;}& -{\left|H_{12}\right|}^2\sin \mathrm{\&theta;}\\ {\left|H_{34}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{34}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]}^{-1}\left[\begin{array}{c}\mathrm{Re}\left({s_1}^{\&prime;}\right)\\ \mathrm{Im}\left({s_1}^{\&prime;}\right)\end{array}\right]$

$=\frac{1}{{\left|H_{12}\right|}^2{\left|H_{34}\right|}^2}\left[\begin{array}{cc}{{|H}_{34}|}^2\cos \mathrm{\&theta;}& {\left|H_{12}\right|}^2\sin \mathrm{\&theta;}\\ -{\left|H_{34}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{12}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}\mathrm{Re}\left({s_1}^{\&prime;}\right)\\ \mathrm{Im}\left({s_1}^{\&prime;}\right)\end{array}\right]---\left(1.20\right)$

So have:

${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{1I}\\ x_{1Q}\end{array}\right]\&ap;\left[\begin{array}{cc}{\left|H_{34}\right|}^2\cos \mathrm{\&theta;}& {\left|H_{12}\right|}^2\sin \mathrm{\&theta;}\\ -{\left|H_{34}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{12}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}\mathrm{Re}\left({s_1}^{\&prime;}\right)\\ \mathrm{Im}\left({s_1}^{\&prime;}\right)\end{array}\right]---\left(1.21\right)$

In like manner, we can obtain:

${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{2I}\\ x_{2Q}\end{array}\right]\&ap;\left[\begin{array}{cc}{\left|H_{34}\right|}^2\cos \mathrm{\&theta;}& {\left|H_{12}\right|}^2\sin \mathrm{\&theta;}\\ -{\left|H_{34}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{12}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}\mathrm{Re}\left({{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}^{\&prime;}\right)\\ \mathrm{Im}\left({{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}^{\&prime;}\right)\end{array}\right]---\left(1.22\right)$

${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{3I}\\ x_{3Q}\end{array}\right]\&ap;\left[\begin{array}{cc}{\left|H_{12}\right|}^2\cos \mathrm{\&theta;}& {\left|H_{34}\right|}^2\sin \mathrm{\&theta;}\\ -{\left|H_{12}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{34}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}\mathrm{Re}\left({{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">s</figure-callout>}}_3}^{\&prime;}\right)\\ \mathrm{Im}\left({{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="C20041008673300131.png"\; state="\{\{state\}\}">s</figure-callout>}}_3}^{\&prime;}\right)\end{array}\right]---\left(1.23\right)$

${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{4I}\\ x_{4Q}\end{array}\right]\&ap;\left[\begin{array}{cc}{\left|H_{12}\right|}^2\cos \mathrm{\&theta;}& {\left|H_{34}\right|}^2\sin \mathrm{\&theta;}\\ -{\left|H_{12}\right|}^2\sin \mathrm{\&theta;}& {\left|H_{34}\right|}^2\cos \mathrm{\&theta;}\end{array}\right]\left[\begin{array}{c}\mathrm{Re}\left({s_4}^{\&prime;}\right)\\ \mathrm{Im}\left({s_4}^{\&prime;}\right)\end{array}\right]---\left(1.24\right)$

Wherein | H ₁| ²+ | H ₂| ²=| H ₁₂| ², | H ₃| ²+ | H ₄| ²=| H ₃₄| ²

Just obtain formula (1.21)～(1.24), their result through after the step 3 ${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{1I}\\ x_{1Q}\end{array}\right],$ ${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{2I}\\ x_{2Q}\end{array}\right],$ ${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{3I}\\ x_{3Q}\end{array}\right]$ With ${\left|H_{12}\right|}^2{\left|H_{34}\right|}^2\left[\begin{array}{c}{x}_{4I}\\ x_{4Q}\end{array}\right]$ As the constellation point coordinate after the judgement, can deliver to the demodulation that following step is carried out the planisphere soft-decision.

Because the constellation point coordinate after the judgement has | H ₁₂| ²| H ₃₄| ²The factor, the effect of this factor can have been regarded as | H ₁₂| ²| H ₃₄| ²Transmit antennas is launched prosign simultaneously, and receiving terminal aligns in time, so this factor is just represented the transmit diversity exponent number.If from average angle, to this factor ask average after, be easy to obtain E[|H ₁₂| ²| H ₃₄| ²]=4 are so from the angle of average meaning, the present invention has realized full transmit diversity, obtain to constructivity CIOD optimum soft decoding method under the least square meaning, and encoding rate is 1, the diversity gain that receives link is improved, and has exported needed soft information.

The present invention also provides a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device, this device includes vector orthogonal transform module, vector imaginary part Switching Module and the vector weighting rotary module that is connected successively, wherein, described vector orthogonal transform module is used for the received signal vector, and, obtain transformation vector to the conjugate transpose of this signal phasor premultiplication with channel estimate matrix; Described vector imaginary part Switching Module is used for the imaginary part with described each component of transformation vector, exchanges mutually according to transmitting terminal each component imaginary part exchange regulation to the signal phasor launched, obtains compute vectors; Described vector weighting rotary module is used for described compute vectors is carried out channel weights rotation under the least square meaning, obtains and export constellation point coordinate and soft information thereof after the judgement.

Described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module can be realized after the establishment respective code by general CPU; Or by DSP, FPGA or ASIC realization.Above-mentioned each module all is comparatively general matrix operation module and numerical transformation modules.

Be description above to preferred embodiment of the present invention, the method goes for 2～4 antennas of transmitting terminal, receiving terminal is more than or equal to the situation of 1 antenna, so those skilled in the art that should be understood that the various corrections of embodiments of the invention and change and should drop in design of the present invention and the claims limited range.

Claims (9)

1, a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates method is characterized in that, comprises the steps:

(a) receive principle according to the Alamouti quadrature, receiving terminal receives the signal phasor that transmitting terminal sends, and to the conjugate transpose of this signal phasor premultiplication with channel estimate matrix, obtains transformation vector;

(b) will exchange mutually according to transmitting terminal each component imaginary part exchange regulation through the imaginary part of each component in the transformation vector that obtains after step (a) processing, obtain compute vectors transmitting vector;

(c) will under the least square meaning, carry out channel weights rotation, the constellation point coordinate after obtaining adjudicating through the compute vectors that obtains after step (b) processing.

2, interlacing orthogonal transmitting diversity least squares soft decode between coordinates method according to claim 1, it is characterized in that: the channel estimate matrix in the described step (a) is to draw after by pilot tone or prefix channel being estimated, and obtains its conjugate transpose through matrixing.

3, interlacing orthogonal transmitting diversity least squares soft decode between coordinates method according to claim 1 is characterized in that: described step (c) is further divided into following steps:

(c1) will filter through the noise that obtains in the compute vectors after step (b) processing, obtain the least-squares estimation matrix;

(c2) will carry out channel weights rotation, the constellation point coordinate after obtaining adjudicating through obtaining the least-squares estimation matrix after step (c1) processing.

4, interlacing orthogonal transmitting diversity least squares soft decode between coordinates method according to claim 2, it is characterized in that: receiving terminal is an antenna in the described step (a), transmitting terminal is 4 antennas, an antenna of receiving terminal receives 4 dimensional signal vectors of 4 antenna transmission of transmitting terminal, and the imaginary part of first dimension of this 4 dimensional signal vector and third dimension component exchanges mutually, second dimension exchanges mutually with the imaginary part of fourth dimension component; In the described step (b) in the transformation vector imaginary part of each component exchange mutually, carry out according to the mode that the imaginary part of first dimension and third dimension component imaginary part, second dimension and fourth dimension component exchanges mutually, obtain 4 dimension compute vectors; 4 dimension compute vectors are carried out channel weights rotation, 4 constellation point coordinates and soft information thereof after obtaining adjudicating in the described step (c) under the least square meaning.

5, a kind of interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device, it is characterized in that: this device includes vector orthogonal transform module, vector imaginary part Switching Module and the vector weighting rotary module that is connected successively, wherein, described vector orthogonal transform module is used for the received signal vector, and, obtain transformation vector to the conjugate transpose of this signal phasor premultiplication with channel estimate matrix; Described vector imaginary part Switching Module is used for the imaginary part with described each component of transformation vector, exchanges mutually according to transmitting terminal each component imaginary part exchange regulation to the signal phasor launched, obtains compute vectors; Described vector weighting rotary module is used for described compute vectors is carried out channel weights rotation under the least square meaning, obtains and export constellation point coordinate and soft information thereof after the judgement.

6, interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device according to claim 5 is characterized in that: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module are realized after the establishment respective code by general CPU.

7, interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device according to claim 5 is characterized in that: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module are realized by DSP.

8, interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device according to claim 5 is characterized in that: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module are realized by FPGA.

9, interlacing orthogonal transmitting diversity least squares soft decode between coordinates implement device according to claim 5 is characterized in that: described vector orthogonal transform module, vector imaginary part Switching Module and vector weighting rotary module are realized by ASIC.

Images