CN102045134B - Full-rate and full-diversity space-time coding/decoding method - Google Patents

Abstract

The invention discloses a full-rate and full-diversity space-time coding/decoding method. In the method, a receiving end determines a precoding parameter of an emission end according to channel information, quantizes the parameter to obtain a quantized value of the parameter and feeds serial number of the quantized value in a quantization codebook back to the emission end; and the emission end carries out precoding on a full-rate and full-diversity space-time code according to the received feedback information. When in decoding, the receiving end respectively decodes a real component and a virtual component of a modulation signal. Compared with a traditional full-rate and full-diversity space-time coding/decoding scheme, by virtue of the method disclosed by the invention, lower decoding complexity is achieved while the same reliability is kept.

Description

The Full-rate full-diversity space-time compilation code method

Technical field

The invention belongs to communication technical field, relate in particular to a kind of Full-rate full-diversity space-time compilation code method of low complex degree.

Background technology

Space-Time Codes can significantly improve reliability of Communication System and validity.Traditional Space Time Coding comprises two kinds of orthogonal space time packet (Orthogonal Space-Time Block Code, OSTBC) and Bell Laboratory vertical layered space-time codes (Vertical Bell Labs LayeredSpace-Time, VBLAST).VBLAST can reach full rate, yet can not realize full-diversity; Though OSTBC can realize full-diversity, can not reach full rate.For this problem, in succession proposed to avoid simultaneously in conjunction with both advantages in recent years the coded system of its shortcoming, the empty time-code of Golden is exactly wherein a kind of, however its decoding complexity is directly proportional with the biquadratic of order of modulation.In order to reduce decoding complexity, there is the scholar to propose a kind of empty time-code (Full Rate Full Diversity Space Time Code of Full-rate full-diversity that is called, FR-FD STC) new coding (is designated hereinafter simply as the Full-rate full-diversity Space Time Coding, perhaps FR-FD STC encodes), square being directly proportional of its decoding complexity and order of modulation, be starkly lower than the empty time-code of Golden, but its decoding complexity is still higher.

The below introduces coding and the interpretation method of FR-FD STC.

S=[s ₁, s ₂, s ₃, s ₄] ^TExpression modulation signal vector, s _i(i=1,2,3,4) are the points on MQAM (QuadratureAmplitude Modulation, the quadrature amplitude modulation) planisphere, and M is order of modulation, () ^TThe expression transposition.S is carried out the Space Time Coding of Full-rate full-diversity, obtaining encoder matrix X can be expressed as

$X=\left[\begin{array}{cc}{\mathrm{as}}_1+{\mathrm{bs}}_3& -\mathrm{<figure-callout\; id="2"\; label="c\; s"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">c</figure-callout>}{s_{}}^{}{{}_2}^{*}-\mathrm{<figure-callout\; id="4"\; label="d\; s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">d</figure-callout>}{s_{}}^{}{{}_4}^{*}\\ {\mathrm{as}}_2+{\mathrm{bs}}_4& \mathrm{<figure-callout\; id="1"\; label="c\; s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">c</figure-callout>}{s_{}}^{}{{}_1}^{*}+\mathrm{<figure-callout\; id="3"\; label="d\; s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">d</figure-callout>}{s_{}}^{}{{}_3}^{*}\end{array}\right]---\left(1\right)$

Wherein, a, b, c and d are plural numbers, and they satisfy | a| ²+ | b| ²=| c| ²+ | d| ²=| a| ²+ | b| ²=| d| ²+ | d| ²=1, () ^*The expression complex conjugate.

If it is 2 * 2 channel matrix that H represents exponent number, receiving signal matrix Y can be expressed as

Y＝HX+W (2)

Wherein, W is noise matrix; The exponent number of Y and W is 2 * 2.

If | a|=|c| and employing maximum-likelihood decoding, the decision statistics D (s of receiving terminal ₁, s ₂, s ₃, s ₄) to embody form as follows

$D({\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_2,{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_3,s_4)=\underset{k=1}{\overset{2}{\mathrm{\&Sigma;}}}({\left|y_{k1}\right|}^2+{\left|y_{k2}\right|}^2)+f_1({\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_1,{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_3,s_4)+f_2({\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_2,{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">s</figure-callout>}}_3,s_4)---\left(3\right)$

Wherein,

Wherein, y _KiAnd h _KiThe element that represents respectively the capable i row of k of Y and H, k=1,2, i=1,2.

With

Represent respectively real component and imaginary part component.

If known s ₃And s ₄Value, then can be respectively by f ₁(s ₁, s ₃, s ₄) and f ₂(s ₂, s ₃, s ₄) estimate s ₁And s ₂Value.To sum up, its concrete decoding procedure is:

Step 1 provides set G, its element g _iBe (s ₃, s ₄) all possible combination, g _i=(s ₃, s ₄) i, i=1,2 ..., M ², the gesture of G is M ²

Step 2 is got i the element g of G _i, make (s ₃, s ₄)=g _i, find out respectively and make f ₁(s ₁, s ₃, s ₄) and f ₂(s ₂, s ₃, s ₄) obtain the s of minimum value ₁And s ₂, obtain vector

I=1,2 ..., M ²This step obtains M altogether ²The group vector;

Step 3, the M that obtains from step 2 ²Select in the group vector and make D (s ₁, s ₂, s ₃, s ₄) one group of minimum vector,

Be the vector that decoding obtains

Known by above-mentioned decoding procedure, decode procedure needs traversal (s ₃, s ₄) all possible value, its decoding complexity and M ²Be directly proportional.If order of modulation is higher, the amount of calculation during receiving terminal decoding will be very large.

Summary of the invention

The technical problem that (one) will solve

For the too high problem of the decoding complexity that solves FR-FD STC, the present invention proposes a kind of Full-rate full-diversity space-time compilation code method of low complex degree.

(2) technical scheme

For solving the problems of the technologies described above, the invention provides a kind of Full-rate full-diversity space-time compilation code method, comprise the steps:

A, receiving terminal is determined the precoding parameter according to channel matrix H, quantizes this parameter and obtains quantized value, and the sequence number of this quantized value in quantizing code book fed back to transmitting terminal;

B, transmitting terminal at first carry out the Full-rate full-diversity Space Time Coding to modulation signal, obtain encoder matrix X, and obtain pre-coding matrix F according to the serial number information of feedback, then utilize pre-coding matrix F that X is carried out precoding;

C, receiving terminal calculate equivalent real-valued channel matrix H according to channel matrix H and pre-coding matrix F ^θ

D, receiving terminal process and receive signal, obtain respectively the reception signal Y corresponding to real component of described modulation signal ^θ ₁The reception signal Y corresponding with the imaginary part component ^θ ₂

E, receiving terminal is according to the real-valued channel matrix H of described equivalence ^θFront four row and Y ^θ ₁, decipher out the real component of described modulation signal;

F, receiving terminal is according to the real-valued channel matrix H of described equivalence ^θRear four row and Y ^θ ₂, decipher out the imaginary part component of described modulation signal;

G obtains the estimated value of transmission signal vector according to the decode results of step e and step F.

Wherein, steps A specifically comprises:

A1, receiving terminal calculate the tangent value of precoding parameter θ according to described channel matrix H, and try to achieve θ by its arctan function;

A2, receiving terminal quantizes θ, obtains quantized value and the sequence number i in quantizing code book thereof;

A3, receiving terminal feeds back to transmitting terminal with i.

Wherein, step B specifically comprises:

B1, transmitting terminal carries out the Full-rate full-diversity Space Time Coding to modulation signal, obtains encoder matrix X;

B2, transmitting terminal obtain angle θ ' by quantizing code book, thereby obtain pre-coding matrix F according to the feedback information i that receives, and are expressed as

B3 utilizes F that X is carried out precoding.

Wherein, step C specifically comprises:

C1, receiving terminal calculates according to channel matrix H and pre-coding matrix F

Be expressed as

C2, according to

Obtain equivalent channel matrix

Be expressed as:

${\hat{H}}^{\&prime;}=\left[\begin{array}{cccc}{{\mathrm{ah}}^{\mathrm{\&theta;}}}_{11}& {{\mathrm{ah}}^{\mathrm{\&theta;}}}_{12}& {{\mathrm{bh}}^{\mathrm{\&theta;}}}_{11}& {{\mathrm{bh}}^{\mathrm{\&theta;}}}_{12}\\ c^{*}{{h^{\mathrm{\&theta;}}}_{12}}^{*}& -c^{*}{{h^{\mathrm{\&theta;}}}_{11}}^{*}& d^{*}{{h^{\mathrm{\&theta;}}}_{12}}^{*}& -d^{*}{{h^{\mathrm{\&theta;}}}_{11}}^{*}\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ a{h^{\mathrm{\&theta;}}}_{N1}& a{h^{\mathrm{\&theta;}}}_{N2}& b{h^{\mathrm{\&theta;}}}_{N1}& b{h^{\mathrm{\&theta;}}}_{N2}\\ c^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}2}}^{*}& -c^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}1}}^{*}& d^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}2}}^{*}& -d^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}1}}^{*}\end{array}\right]$

Wherein, h ^θ _IjExpression

The element of the capable j of i row, i=1,2 ..., N, j=1,2, N represent the number of reception antenna; A, b, c and d are plural numbers, and satisfy | a| ²+ | b| ²=| c| ²+ | d| ²=| a| ²+ | c| ²=| b| ²+ | d| ²=1, () ^*The expression complex conjugate;

C3, according to Obtain equivalent real-valued channel matrix H ^θ, be expressed as

Wherein,

With

Represent respectively real component and imaginary part component.

Wherein, step D is specially:

Receiving terminal is at first according to receiving signal matrix Y _θObtain vectorial Y _θ'=[y ^θ ₁₁, y ^θ ₁₂ ^*..., y ^θ _N1, y ^θ _N2 ^*] ^T, y ^θ _IjExpression Y ^θThe element of the capable j of i row, i=1,2 ..., N, j=1,2; Then, by Y _θ' obtain

Order

Y ^θ ₁The reception signal corresponding to real component of modulation signal, Y ^θ ₂The reception signal corresponding to imaginary part component of modulation signal.

Wherein, step e specifically comprises:

E1, receiving terminal adopt the HMLIC method to decipher the real component of described modulation signal: at first travel through

All possible value, even

I=1,2 ..., M, g _iM quadrature amplitude modulation qam constellation point, if i ≠ k, then g _i≠ g _k

Then E2 on the basis of step e 1, carries out the interference of HMLIC interpretation method and eliminates part, eliminates

Right

Interference, estimate respectively

With

Value, obtain vector

This step obtains M group vector altogether;

E3 selects one group that satisfies following formula from the M group vector that step e 2 obtains

${\hat{s}}^{\&prime;}=\arg \underset{{s_i}^{\&prime;},i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,M}{\min }{\left|\right|{Y^{\mathrm{\&theta;}}}_1-\left[{\stackrel{\&OverBar;}{h}}_1^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_2^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_3^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_4^{\mathrm{\&theta;}}\right]{s_i}^{\&prime;}\left|\right|}^2$

Wherein,

It is the real component of deciphering the modulation signal that obtains;

Expression H ^θI row, i=1,2,3,4.

Wherein, step F specifically comprises:

F1, according to the HMLIC interpretation method, traversal

All possible value, even

F2 on the basis of step F 1, carries out the interference of HMLIC interpretation method and eliminates part, namely eliminates

Right Interference, estimate respectively

With

Value, obtain vector

I=1,2 ..., M, this step obtains M group vector altogether;

F3 selects one group that satisfies following formula from the M group vector that step F 2 obtains

${\hat{s}}^{\&prime;\&prime;}=\arg \underset{{s_i}^{\&prime;\&prime;},i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,M}{\min }{\left|\right|{Y^{\mathrm{\&theta;}}}_2-\left[{\stackrel{\&OverBar;}{h}}_5^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_6^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_7^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_8^{\mathrm{\&theta;}}\right]{s_i}^{\&prime;\&prime;}\left|\right|}^2$

Wherein,

Decipher exactly the imaginary part component of the modulation signal that obtains;

Expression H ^θI row, i=5,6,7,8.

Wherein, step G obtains the estimated value of transmission signal vector by following formula

Wherein, in steps A 2, adopt the method for uniform quantization to quantize precoding parameter θ.

(3) beneficial effect

The present invention is by FR-FD STC coding is carried out precoding, so that the real component of modulation signal and imaginary part component keep quadrature in transmission course, thereby can decipher out respectively real component and the imaginary part component of modulation signal.Derivation can get

Namely

With

In transmission course, keep quadrature, therefore, during the real component of decoding modulation signal, only need traversal

All possible value, thus the complexity of the real component of decoding modulation signal is directly proportional with M.In like manner as can be known, the complexity of the imaginary part component of decoding modulation signal is directly proportional with M.Therefore, decoding complexity of the present invention is directly proportional with M, compares with existing FR-FD STC decoded mode, and complexity reduces greatly.Method of the present invention is particularly useful for MIMO (Multiple-Input and Multiple-Output, the multiple-input and multiple-output) system of 2 transmitting antenna N root reception antennas.

Description of drawings

Fig. 1 is the method flow diagram of the embodiment of the invention;

Fig. 2 is method and the feedback method flow chart that receiving terminal calculates precoding parameter θ in the method for the embodiment of the invention;

Fig. 3 is the cataloged procedure flow chart of transmitting terminal in the method for the embodiment of the invention;

Fig. 4 is the decode procedure flow chart of receiving terminal in the method for the embodiment of the invention;

The error rate correlation curve of Fig. 5 (a) and 5 (b) prior aries and method of the present invention;

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention will be further described.

Thinking of the present invention is: receiving terminal is determined the precoding parameter θ of transmitting terminal according to channel matrix, quantizes θ and the sequence number of quantized value in quantizing code book fed back to transmitting terminal; Transmitting terminal carries out precoding according to the feedback information that receives to FR-FD STC coding; Receiving terminal is deciphered respectively real component and the imaginary part component of modulation signal.

Such as Fig. 1, and in conjunction with shown in Figure 2, the method for receiving terminal calculating precoding parameter θ and the concrete steps of feedback method are as follows:

Steps A 1, receiving terminal calculate the tangent value of the precoding parameter θ of transmitting terminal according to channel matrix H

The expression formula of g and f is as follows, tries to achieve by arctan function again

Arctan () represents arc tangent.

Steps A 2, receiving terminal quantizes θ, obtains quantized value β (i) and β (i) the sequence number i in quantizing code book.Adopt the method for uniform quantization in this supposition, the quantification code book is β=[0,2 pi/2 ⁿ, 4 pi/2s ⁿ..., 2 π (2 ⁿ-1)/2 ⁿ], n calculates and selection for being used for the bit number of feedback As quantized value, i is 0 to 2 ⁿInteger between-1, β (i) expression quantizes i+1 the element of code book β, 0≤i≤2 ⁿ-1.

Steps A 3, receiving terminal utilize n bit that i is fed back to transmitting terminal.

Fig. 3 and Fig. 4 show respectively the cataloged procedure of transmitting terminal among the present invention and the decode procedure of receiving terminal.In conjunction with Fig. 2, the concrete steps of transmitting terminal cataloged procedure are as follows:

B1, transmitting terminal is to modulation signal vector s=[s ₁, s ₂, s ₃, s ₄] ^TCarry out FR-FD STC coding, obtain encoder matrix

Wherein, a, b, c and d are plural numbers, and they satisfy | a| ²+ | b| ²=| c| ²+ | d| ²=1, | a| ²+ | c| ²=| b| ²+ | d| ²=1.

B2, transmitting terminal at first obtain angle beta (i) according to the feedback information i that receives and quantification code book β, make θ '=β (i), thereby obtain pre-coding matrix F, are expressed as

B3 utilizes F that X is carried out precoding, obtains encoder matrix X _θ=XF.In conjunction with Fig. 5, the decode procedure of receiving terminal is as follows: it is Y that order receives signal matrix _θ, Y _θ=HX _θ+ W, wherein, H represents channel matrix, W represents noise matrix, Y _θ, H and W exponent number be N * 2.

C1, receiving terminal at first calculate according to channel matrix H and pre-coding matrix F

With

Exponent number be N * 2,

C2, according to

Obtain equivalent channel matrix

Form as follows

${\hat{H}}^{\&prime;}=\left[\begin{array}{cccc}{{\mathrm{ah}}^{\mathrm{\&theta;}}}_{11}& {{\mathrm{ah}}^{\mathrm{\&theta;}}}_{12}& {{\mathrm{bh}}^{\mathrm{\&theta;}}}_{11}& {{\mathrm{bh}}^{\mathrm{\&theta;}}}_{12}\\ c^{*}{{h^{\mathrm{\&theta;}}}_{12}}^{*}& -c^{*}{{h^{\mathrm{\&theta;}}}_{11}}^{*}& d^{*}{{h^{\mathrm{\&theta;}}}_{12}}^{*}& -d^{*}{{h^{\mathrm{\&theta;}}}_{11}}^{*}\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ a{h^{\mathrm{\&theta;}}}_{N1}& a{h^{\mathrm{\&theta;}}}_{N2}& b{h^{\mathrm{\&theta;}}}_{N1}& b{h^{\mathrm{\&theta;}}}_{N2}\\ c^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}2}}^{*}& -c^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}1}}^{*}& d^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HSA00000372550900022.png,HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}2}}^{*}& -d^{*}{{h^{\mathrm{\&theta;}}}_{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HSA00000372550900032.png"\; state="\{\{state\}\}">N</figure-callout>}1}}^{*}\end{array}\right]$

Wherein, h ^θ _IjExpression

The element of the capable j of i row, i=1,2 ..., N, j=1,2.

C3, according to

Obtain

Exponent number be 2N * 4, H ^θExponent number be 4N * 8.

D, receiving terminal are at first according to receiving signal matrix Y _θObtain vectorial Y _θ'=[y ^θ ₁₁, y ^θ ₁₂ ^*..., y ^θ _N1, y ^θ _N2 ^*] ^T, y ^θ _IjExpression Y _θThe element of the capable j of i row, i=1,2 ..., N, j=1,2; Then, by Y _θ' obtain

Order

Y ^θ ₁The reception signal corresponding to real component of modulation signal, Y ^θ ₂The reception signal corresponding to imaginary part component of modulation signal.

E1, receiving terminal adopt the real component of HMLIC (Hybrid maximum-likelihood interference cancellation, maximum likelihood is in conjunction with disturbing cancellation) method decoding modulation signal.At first order

I=1,2 ..., M, g _iThe MQAM constellation point, if i ≠ k, g _i≠ g _k

Then E2 on the basis of step e 1, carries out the interference of HMLIC interpretation method and eliminates part, eliminates

Right

Interference, estimate respectively

With

Value, obtain vector This step obtains M group vector altogether.

E2 selects one group that satisfies following formula from the M group vector that step e 2 obtains

${\hat{s}}^{\&prime;}=\arg \underset{{s_i}^{\&prime;},i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,M}{\min }{\left|\right|{Y^{\mathrm{\&theta;}}}_1-\left[{\stackrel{\&OverBar;}{h}}_1^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_2^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_3^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_4^{\mathrm{\&theta;}}\right]{s_i}^{\&prime;}\left|\right|}^2$

Wherein, Decipher exactly the real component of the modulation signal that obtains;

Expression H ^θI row, i=1,2,3,4.

F1, receiving terminal adopt the imaginary part component of HMLIC method decoding modulation signal.At first order

I=1,2 ..., M, g _iThe MQAM constellation point, if i ≠ k, g _i≠ g _k

F2 eliminates

Right Interference, estimate respectively

With

Value, obtain vector

This step obtains M group vector altogether.

F3 selects one group that satisfies following formula from the M group vector that step F 2 obtains

${\hat{s}}^{\&prime;\&prime;}=\arg \underset{{s_i}^{\&prime;\&prime;},i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,M}{\min }{\left|\right|{Y^{\mathrm{\&theta;}}}_2-\left[{\stackrel{\&OverBar;}{h}}_5^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_6^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_7^{\mathrm{\&theta;}}{\stackrel{\&OverBar;}{h}}_8^{\mathrm{\&theta;}}\right]{s_i}^{\&prime;\&prime;}\left|\right|}^2$

Wherein,

It is the imaginary part component of deciphering the modulation signal that obtains;

Expression H ^θI row, i=5,6,7,8.

G is according to the decode results of step e 2

And the decode results of step F 3 Obtain the estimated value of transmission signal vector

Known by above-mentioned decoding procedure, only need respectively traversal in the decode procedure

All possible value and

All possible value, its decoding complexity is directly proportional with M, and therefore, decoding complexity of the present invention is lower.

In order to further specify useful result of the present invention, table 1 has provided the decoding amount of calculation of the inventive method at this; Table 2 compares the decoding amount of calculation of the inventive method and traditional scheme, the unit of amount of calculation is per second flops (Floating-point Operations PerSecond, referred to as flops), namely once plural multiplier computing needs 6 flops operations, and a complex addition computing needs 2 flops operations.Relatively can get, decoding amount of calculation of the present invention approximately is 7/ (6M) of tradition decoding amount of calculation.When modulation system was respectively QPSK, 16QAM and 64QAM, the former decoding amount of calculation only was the latter's 29.17%, 7.29% and 1.82%.

Table 1 decoding amount of calculation of the present invention

Step	Amount of calculation (flops)
		Step 1	12(N+1)
Step 3	M(56N+3)
		Step 4	28MN+8M
Step 5	M(56N+3)
		Step 6	28MN+8M
Step
	7	4
Amount to			168MN+22M+12N+16

The comparison of the decoding amount of calculation of table 2 two schemes

Scheme	Amount of calculation (flops)
		The present invention	168MN+22M+12N+16
Traditional scheme	M 2(146N+22)+12N+16

[0113]Fig. 5 (a) and 6 (b) have provided the error rate (BER) curve (abscissa is signal to noise ratio snr) of coding and decoding scheme of the present invention, and compare with traditional FR-FD STC, modulation system is QPSK (Quadrature Phrase Shift Key, Quadrature Phase Shift Keying).Wherein Fig. 5 (a) be two kinds of encoding schemes in 2 * 2 the system reliability ratio, Fig. 5 (b) be two kinds of encoding schemes in 4 * 2 the system reliability ratio.As can be seen from Figure, if there is not quantization error, the reliability of two kinds of encoding schemes is suitable; During n=7, quantization error can be ignored substantially on the impact of systematic function, and therefore, coding and decoding scheme of the present invention has exchanged the very big reduction of decoding complexity for the feedback quantity of several bits.

Above execution mode only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; in the situation that do not break away from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.

Claims (9)

1. a Full-rate full-diversity space-time compilation code method is characterized in that, comprises the steps:

A, receiving terminal is determined the precoding parameter according to channel matrix H, quantizes this parameter and obtains quantized value, and the sequence number of this quantized value in quantizing code book fed back to transmitting terminal;

B, transmitting terminal at first carry out the Full-rate full-diversity Space Time Coding to modulation signal, obtain encoder matrix x, and obtain pre-coding matrix F according to the serial number information of feedback, then utilize pre-coding matrix F that x is carried out precoding;

C, receiving terminal calculate equivalent real-valued channel matrix H according to channel matrix H and pre-coding matrix F ^θ, wherein θ is the precoding parameter;

D, receiving terminal process and receive signal, obtain respectively the reception signal Y corresponding to real component of described modulation signal ^θ ₁The reception signal Y corresponding with the imaginary part component ^θ ₂

E, receiving terminal is according to the real-valued channel matrix H of described equivalence ^θFront four row and Y ^θ ₁, decipher out the real component of described modulation signal;

F, receiving terminal is according to the real-valued channel matrix H of described equivalence ^θRear four row and Y ^θ ₂, decipher out the imaginary part component of described modulation signal;

G obtains the estimated value of transmission signal vector according to the decode results of step e and step F.

2. method according to claim 1 is characterized in that, wherein, steps A specifically comprises:

A1, receiving terminal calculate the tangent value of precoding parameter θ according to described channel matrix H, and try to achieve θ by its arctan function;

A2, receiving terminal quantizes θ, obtains quantized value and the sequence number i in quantizing code book thereof;

A3, receiving terminal feeds back to transmitting terminal with i.

3. method according to claim 2 is characterized in that, wherein, step B specifically comprises:

B1, transmitting terminal carries out the Full-rate full-diversity Space Time Coding to modulation signal, obtains encoder matrix x;

B2, transmitting terminal obtain angle θ ' by quantizing code book, thereby obtain pre-coding matrix F according to the feedback information i that receives, and are expressed as

B3 utilizes F that x is carried out precoding.

4. method according to claim 2 is characterized in that, wherein, step C specifically comprises:

C1, receiving terminal calculates according to channel matrix H and pre-coding matrix F

, be expressed as

C2, according to

Obtain equivalent channel matrix

, be expressed as:

Wherein, h ^θ _IjExpression

The element of the capable j of i row, i=1,2 ..., N, j=1,2, N represent the number of reception antenna; A, b, c and d are plural numbers, and satisfy | a| ²+ | b| ²=| c| ²+ | d| ²=| a| ²+ | c| ²=| b| ²+ | d| ²=1, () ^*The expression complex conjugate;

C3, according to

Obtain equivalent real-valued channel matrix H ^θ, be expressed as

Wherein, With

Represent respectively real component and imaginary part component.

5. method according to claim 4 is characterized in that, wherein, step D is specially:

Receiving terminal is at first according to receiving signal matrix Y _θObtain vector

y ^θ _IjExpression Y _θThe element of the capable j of i row, i=1,2 ..., N, j=1,2; Then, by

Obtain

Order

Y ^θ ₁The reception signal corresponding to real component of modulation signal, Y ^θ ₂The reception signal corresponding to imaginary part component of modulation signal.

6. method according to claim 5 is characterized in that, wherein, step e specifically comprises:

E1, receiving terminal adopt maximum likelihood to decipher the real component of described modulation signal in conjunction with the method for disturbing cancellation HMLIC: at first travel through

All possible value, even

I=1,2 ..., M, g _iM quadrature amplitude modulation qam constellation point, if i ≠ k, then g _i≠ g _k

Then E2 on the basis of step e 1, carries out the interference of HMLIC interpretation method and eliminates part, eliminates

Right

Interference, estimate respectively

With

Value, obtain vector

This step obtains M group vector altogether, wherein, and s ₁, s ₂, s ₃, s ₄It is modulation signal;

E3 selects one group that satisfies following formula from the M group vector that step e 2 obtains

Wherein,

It is the real component of deciphering the modulation signal that obtains;

Expression H ^θI row, i=1,2,3,4.

7. method according to claim 6 is characterized in that, wherein, step F specifically comprises:

F1, according to the HMLIC interpretation method, traversal

All possible value, even

F2 on the basis of step F 1, carries out the interference of HMLIC interpretation method and eliminates part, namely eliminates

Right

Interference, estimate respectively

With

Value, obtain vector

I=1,2 ..., M, this step obtains M group vector altogether;

F3 selects one group that satisfies following formula from the M group vector that step F 2 obtains

Wherein,

Decipher exactly the imaginary part component of the modulation signal that obtains;

Expression H ^θI row, i=5,6,7,8.

8. method according to claim 7 is characterized in that, wherein, step G obtains the estimated value of transmission signal vector by following formula

9. each described method is characterized in that according to claim 2～8, wherein, in steps A 2, adopts the method for uniform quantization to quantize precoding parameter θ.

Images