CN106301515A - A kind of multiple-input and multiple-output detection method and system - Google Patents

Abstract

The present invention provides a kind of multiple-input and multiple-output detection method, including: create transmission channel matrix H；Fix a transmitting signal x in i-th layer of transport layer_i, its value chooses a constellation point in the constellation point determined according to preset number modulation system；By described transmitting signal x_iRemove from launching signal concentration, update and form transmitting signal update collection X_-i, the transmission channel row simultaneously removing correspondence form transmission channel renewal matrix H_-i, update and form reception signal update collection R_-i；The sequence QR decomposition algorithm prestored is used to obtain X_-iIn the likelihood probability of each transmitting signal, by firmly sentencing the constellation point position obtaining each transmitting signal, by X_-iSame x_iMerge and obtained for x_iTransmitting signals all in i-th layer of transport layer and all transport layers are performed above step acquisition judgement transmitting signal collection and therefrom calculate the transmitting signal meeting maximum likelihood standard by one group of judgement under the constellation point chosen.It is excellent that the present invention detects performance, and complexity is lower than the complexity of existing detection algorithm.

Description

A kind of multiple-input and multiple-output detection method and system

Technical field

The invention belongs to the calculating field that communicates, especially belong to the downlink multiple-input and multiple-output detection of tdd mode in 4G system Technical field, particularly relates to a kind of multiple-input and multiple-output detection method and system.

Background technology

MIMO technology can produce independent parallel channel in space, it is achieved transmits while multichannel data, effectively improves The transfer rate of system and the availability of frequency spectrum, owing to mimo system all uses multiple antennas to improve system communication speed in sending and receiving end Rate, so the signal of many reception antenna receptions is from the superposition of gained signal after respective channels of all transmitting aerial signals, The signal that every reception antenna receives can interfere, and therefore, how to recover original transmitted antenna from many reception antennas Signal becomes key issue.MIMO detection technique is exactly that the channel matrix utilizing channel estimation module to obtain receives with receiving side Receive signal, signal detection will be sent out.Design high-performance, the MIMO detection algorithm of low complex degree becomes and ground in recent years The emphasis studied carefully.Many researchers has carried out substantial amounts of research work to MIMO detection technique, it is proposed that some detection algorithms.

The following is some the MIMO detection algorithms having pointed out at present.Several typical algorithm is described below, and it is described not simultaneously In place of foot.

The first, maximum likelihood (Maximum Likelihood, ML) checks algorithm

Maximum likelihood algorithm [1] is that numerous MIMO detection calculates the detection algorithm sending out middle best performance, owing to this algorithm considers Sending all constellation point in vector space, its main thought is in launching signal space, search out and receive signal phasor away from From minimum transmitting signal.Assume that receiving channel state information (CSI) is known, i.e. H is known, and ML algorithm can be used Following formula represents:

${\hat{x}}_{\mathrm{ML}}=\underset{x\∈\mathrm{\Ω}}{\arg \min }\left\{{\left|\right|y-\mathrm{Hx}\left|\right|}^2\right\}$

Wherein Ω represents all N_tThe constellation point space that individual transmitting antenna is corresponding, if Nt transmitting antenna all have employed order of modulation For the modulation system of M, then to calculateSecondary norm computing, the most also will beFind in individual unordered result of calculation Minima.So complexity of result of calculation can be in exponential increase along with the increase of transmission antenna number, especially is using more In the case of system modulation, computation complexity can be the highest.

The second, ZF (Zero Forcing, ZF) checks algorithm

Squeeze theorem algorithm [2] is mainly based upon the weighted least-squares criterion of Guass-Markov theorem, and its design object is to make to send out Penetrate the estimated value of signal vector sUnder conditions of by channel matrix H with original transmitted signal actual value error sum of squares Little.It is the one of linear detection algorithm, directly the signal vector received is multiplied by weighting matrix at receiving terminal thus is estimated Meter vector.The advantage of ZF detection algorithm is that complexity is low, and main amount of calculation is matrix inversion, and its complexity level is O (N³ _t), Though realizing simple, but not accounting for effect of noise during realizing, noise robustness is poor, therefore in the feelings that signal to noise ratio is the highest Preferable performance just can be had under condition.

The third, SQRD (Sort QR) checks algorithm

The error propagation in QR catabolic process, document [3]-[4] is overcome to propose one based on changing in order to obtain the detection ordering of optimum The Gram-Schmidt orthogonalized sequence QR entered decomposes, and this algorithm, before first step orthogonalization procedure, needs at channel The N of matrix H_TIndividual column vector is found out the row of 2 Norm minimums, if when prostatitis is not minimum 2 norm row, then needed N_TIndividual row The norm value of vector, inside Q, R, P, corresponding row all swap, and obtain new Q matrix and R matrix, do orthogonal the most again Change；Before second step orthogonalization, need the N to Q_T-1 column vector is reformed aforesaid operations, the like, until having considered After all of row, then carry out signal detection.Gram-Schmidt algorithm owing to improving is calculating the suitable of R diagonal of a matrix Sequence is from r₁₁To r_NtNt, but the detection ordering of optimum is from r_NtNtTo r₁₁, so that signal detection to kth walk (k=NT ..., 1) Time, SNR_kMaximum, needs to introduce a switching matrix P, is used for recording exchanged position.Comprising the following steps that of algorithm, Whole algorithm is decomposed by the QR sorted and signal detection two parts form:

SQRD algorithm:

1) R=0, Q=H, P=eye (N_T)

2) for i=1 ..., N_T

3)k_i=argmin | | q_l||²L=i ..., N_T

4) exchange Q, Ki Yu the i row of P, R

5)R_i,i=| | q_i||

6)q_i=q_i/R_k,k

7) forl=i+1 ..., N_T

8)R_i,l=q_i ^Hq_l

9)q_l=q_l-R_i,lq_i

10)end

11)end

Signal detection:

12) y=Q^Hr

13) for k=N_T,...,1

14) $d_k=\underset{i=k+1}{\overset{N_T}{\mathrm{\Σ}}}{R}_{k,i}{x}_i$

15)z_k=y_k-d_k

16) ${\hat{x}}_k=Q[z_k/R_{k,k}]$

17)end

18) $x=P\·{\hat{x}}_k$

Although being sorted QR overhaul algorithm, but the detection performance comparing SQRD algorithm with ML is the most very poor.

In 4G system down link, MIMO checks the improvement of algorithm in tdd mode.When the transmitting antenna number of system is N_t, adjust When exponent number processed is M, owing to ML algorithm considers whole constellation point, therefore the transmission assemble of symbol that ML algorithm considers has Planting probability, this patent is aiming at ML algorithm complex height and proposes, and its thinking is by ML algorithm and SQRD algorithm phase In conjunction with, each layer of probability sending symbol is 2^MKind, the remaining system SQRD algorithm after removing this symbols influence is examined Survey, therefore the transmission assemble of symbol number that innovatory algorithm considers is 2^M×N_tPlant probability.4G system down link under tdd mode The maximum antenna supported is configured to 8 × 8, if selecting the modulation system of 64QAM, then the complexity in ML algorithm search space It is 2^6×8, the complexity of innovatory algorithm search volume is 2⁶× 8, the complexity comparing innovatory algorithm with ML algorithm has clearly Reduction.

Therefore, how to provide a kind of multiple-input and multiple-output detection method and system, to solve MIMO of the prior art detection skill Art scheme detection poor performance, the many disadvantages such as complexity is high, become the technical problem that practitioner in the art is urgently to be resolved hurrily in fact.

Summary of the invention

The shortcoming of prior art in view of the above, it is an object of the invention to provide a kind of multiple-input and multiple-output detection method and is System, is used for solving MIMO detection technique scheme detection poor performance, the problem that complexity is high in prior art.

For achieving the above object and other relevant purposes, one aspect of the present invention provides a kind of multiple-input and multiple-output detection method, application In including for sending the multiple-input and multiple-output transmitter launching signal and for receiving the multi-input multi-output receiver of signal Communication system, it is S=[s that described multiple-input and multiple-output transmitter has L layer for transmitting signal sequence₁,s₂,…,s_L]^T's Launch the transport layer of signal, there is N_TIndividual transmitting antenna, described multi-input multi-output receiver has N_RIndividual reception antenna, sends out Penetrate signal collectionReceive signal collectionDescribed multiple-input and multiple-output detection method bag Include following steps: based on N_TIndividual transmitting antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L]； A transmitting signal x in i-th layer of transport layer is fixed according to preset number modulation system_i, its value is being modulated according to preset number The constellation point that mode determines is chosen a constellation point；Wherein, i is the integer less than or equal to L more than or equal to 1；By described transmitting Signal x_iRemove from launching signal concentration, update and launch signal collection, formed and launch signal update collectionFrom described transmission channel matrix H, remove described i-th layer of transport layer corresponding simultaneously Transmission channel row h_iForm transmission channel and update matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L]；More newly received signal collection is formed and receives Signal update collection R_-i；The sequence QR decomposition algorithm prestored is used to obtain in the matrix updated and renewal collectionIn the likelihood probability of each transmitting signal, by firmly sentencing the star obtaining each transmitting signal Seat point position, and by X_-iSame x_iMerge and obtained for x_iOne group of judgement under the constellation point chosen, in i-th layer of transport layer All transmitting signals perform above step, all transport layers perform above step afterwards and launches signal collection from falling into a trap to obtain judgement Operator closes the transmitting signal of maximum likelihood standard.

Alternatively, the vector expression of described reception signal collection is: R=HS+N=HWX+N；Wherein, R is for receiving letter Number collection, H is transmission channel matrix, and W is that every layer of transport layer launches the signal sequence mapping matrix to transmitting signal, and N is N_RIndividual The noise variance of each reception antenna in reception antenna.

Alternatively, signal data=h that the reception antenna that i-th layer of transport layer is corresponding receives_i×x_i。

Alternatively, a kind of implementation of maximum likelihood calculation | | r-H × x | | that is arg min{²}。

Alternatively, preset number modulation system includes ASK, PSK, FSK, QAM, MSK, GMSK or OFDM.

Another aspect of the present invention also provides for a kind of multiple-input and multiple-output detecting system, is applied to include for sending transmitting signal many Input multi output transmitter and the communication system of the multi-input multi-output receiver for reception signal, how defeated described multi input is Going out to launch facility has L layer to be S=[s for transmitting signal sequence₁,s₂,…,s_L]^TLaunch signal transport layer, there is N_TIndividual Launching antenna, described multi-input multi-output receiver has N_RIndividual reception antenna, launches signal collection Receive signal collectionDescribed multiple-input and multiple-output detecting system includes: creation module, for based on N_TIndividual Launch antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L]；The originally determined module of constellation, uses In the transmitting signal x fixed according to preset number modulation system in i-th layer of transport layer_i, and determine that this layer launches the star of signal Seat point position；Wherein, i is the integer less than or equal to L more than or equal to 1；First chooses module, for from launching signal x_iValue A constellation point is chosen in the constellation point determined according to preset number modulation system；Remove module, for by described transmitting signal x_i Remove from launching signal concentration, update and launch signal collection, formed and launch signal update collection From described transmission channel matrix H, remove the transmission channel row h that described i-th layer of transport layer is corresponding simultaneously_iFormation transmission channel updates Matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L], and more newly received signal collection formation reception signal update collection R_-i；Processing module, For using the sequence QR decomposition algorithm prestored to obtain in the matrix updated and renewal collection In the likelihood probability of each transmitting signal, by firmly sentencing the constellation point position obtaining each transmitting signal, and by X_-iSame x_iMerge Obtain for x_iOne group of judgement under the constellation point chosen；Recursive call module, for all transmittings in i-th layer of transport layer Signal and to all above modules of transport layer recursive call with obtain judgement launch signal collection；Computing module, for from institute State judgement transmitting signal centralized calculation and meet the transmitting signal of maximum likelihood standard.

Alternatively, described creation module is additionally operable to create received signal vector expression formula, i.e. R=HS+N=HWX+N；Its In, R is for receiving signal collection, and H is transmission channel matrix, and W is that every layer of transport layer launches signal sequence to the mapping launching signal Matrix, N is N_RThe noise variance of each reception antenna in individual reception antenna.

Alternatively, | | the r-H × x | | that is arg min{ of a kind of maximum likelihood calculation in described computing module²}。

As it has been described above, the multiple-input and multiple-output detection method of the present invention and system, have the advantages that

Multiple-input and multiple-output detection method of the present invention and system detection performance are better than SQR, ZR detection algorithm, and complicated Spend lower than the complexity of existing detection algorithm, and improve the traffic rate of MIMO communication system.

Accompanying drawing explanation

Fig. 1 is shown as the multiple-input-multiple-output communication system theory structure schematic diagram of the present invention.

Fig. 2 is shown as the multiple-input and multiple-output detection method schematic flow sheet of the present invention.

Fig. 3 is shown as the multiple-input and multiple-output detecting system theory structure schematic diagram of the present invention.

Element numbers explanation

1 multiple-input-multiple-output communication system

11 multiple-input and multiple-output transmitters

12 multi-input multi-output receivers

2 multiple-input and multiple-output detecting systems

21 creation modules

The 22 originally determined modules of constellation

23 first choose module

24 remove module

25 more new module

26 processing modules

27 recursive call modules

28 computing modules

29 second choose module

S1～S6 step

Detailed description of the invention

Below by way of specific instantiation, embodiments of the present invention being described, those skilled in the art can be by disclosed by this specification Content understand other advantages and effect of the present invention easily.The present invention can also be added by the most different detailed description of the invention To implement or application, the every details in this specification can also be based on different viewpoints and application, in the essence without departing from the present invention Various modification or change is carried out under god.It should be noted that, the feature in the case of not conflicting, in following example and embodiment Can be mutually combined.

It should be noted that the diagram provided in following example illustrates the basic conception of the present invention the most in a schematic way, then scheme Component count, shape and size when only showing the assembly relevant with the present invention rather than implement according to reality in formula are drawn, in fact When border is implemented, the kenel of each assembly, quantity and ratio can be a kind of random change, and its assembly layout kenel is likely to the most multiple Miscellaneous.

The present invention provides a kind of multiple-input and multiple-output detection method, and basic inventive principle is as follows:

In tdd mode in the mimo system downlink of 4G, in order to make down-link reception side MIMO detection algorithm While performance improves, complexity reduces, and the present invention is directed to ML (Maximum Likelihood, maximum likelihood probability) algorithm complicated Degree height, it is proposed that the MIMO detection algorithm of a kind of improvement.Its main thought is to reduce search space, and considering as much as possible can Can transmission glossary of symbols rather than all.First, the estimated value of first signal to be detected is entered as in certain modulation system A kind of symbol.Then, from original system, remove the interference of this symbol, update system model, the system after updating is SQR Decompose, disposably detect the estimated value remaining signal to be checked, by itself and the estimated value combination of first signal to be detected, obtain To one group of Signal estimation value to be detected of system, when having considered all of symbol of ground floor, consider all of layer successively, can obtain To a Signal estimation value set to be detected.Finally, the assemble of symbol obtained is utilized the processing mode identical with ML algorithm, I.e. find minimum Eustachian distance, thus obtain the transmission estimated value of system.Through emulation and analysis of complexity, the performance of innovatory algorithm It is better than ZF algorithm, is less than ML algorithm close to ML algorithm and complexity.

Embodiment one

The present embodiment provides a kind of multiple-input and multiple-output detection method, is applied to include that how defeated the multi input for sending transmitting signal is Go out transmitter 11 and lead to for receiving the communication system of the multi-input multi-output receiver 12 of signal, i.e. multiple-input and multiple-output In communication system 1, as Fig. 1 shows multiple-input-multiple-output communication system theory structure schematic diagram, described multiple-input and multiple-output transmitter Having L layer for transmitting signal sequence is S=[s₁,s₂,…,s_L]^TLaunch signal transport layer, there is N_TIndividual transmitting sky Line, described multi-input multi-output receiver has N_RIndividual reception antenna, launches signal collectionReceive letter Number collectionRefer to Fig. 2, be shown as multiple-input and multiple-output detection method schematic flow sheet.As in figure 2 it is shown, Described multiple-input and multiple-output detection method comprises the following steps:

S1, based on N_TIndividual transmitting antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L].At this In embodiment, step S1 also includes creating a reception signal collection vector expression model.The vector expression of described reception signal collection is: R=HS+N=HWX+N；Wherein, R is for receiving signal collection, and H is transmission channel matrix, and W is that every layer of transport layer is launched Signal sequence is to the mapping matrix of transmitting signal, and N is N_RThe noise variance of each reception antenna in individual reception antenna.Described reception The matrix form of the vector expression of signal collection is expressed as:

$\left[\begin{array}{c}{r}_1\\ r_2\\ ...\\ r_{N_R}\end{array}\right]=\left[\begin{array}{cccc}{h}_{11}& h_{12}& ...& h_{1l}\\ h_{21}& h_{22}& ...& h_{2l}\\ ...& ...& ...& ...\\ h_{N_{R}1}& h_{N_{R}2}& ...& h_{N_{R}l}\end{array}\right]\×\left[\begin{array}{c}{x}_1\\ x_2\\ ...\\ x_l\end{array}\right]+\left[\begin{array}{c}{n}_1\\ n_2\\ ...\\ n_{N_R}\end{array}\right]$ Formula (1)

In transmission channel matrix H, each element representation is h_jk, h_jkThe antenna transmission channel to jth root reception antenna is launched for kth root.

S2, fixes a transmitting signal x in i-th layer of transport layer according to preset number modulation system_i, its value is according to present count The constellation point that word modulation system determines is chosen a constellation point；Wherein, i is the integer less than or equal to L more than or equal to 1.Described Preset number modulation system includes ASK, PSK, FSK, QAM, MSK, GMSK or OFDM.In the present embodiment, Described preset number modulation system chooses the QPSK in PSK, and so, this layer is launched the constellation point of signal and be $\frac{\sqrt{2}}{2}+i\frac{\sqrt{2}}{2},\frac{\sqrt{2}}{2}-i\frac{\sqrt{2}}{2},-\frac{\sqrt{2}}{2}+i\frac{\sqrt{2}}{2},-\frac{\sqrt{2}}{2}-i\frac{\sqrt{2}}{2}.$

S3, by described transmitting signal x_iRemove from launching signal concentration, update and launch signal collection, formed and launch signal update collectionFrom described transmission channel matrix H, remove described i-th layer of transport layer corresponding simultaneously Transmission channel row h_iForm transmission channel and update matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L].The most also include that renewal connects A collection of letters number collection vector expression model.By formula (1) basisWith H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L], and more newly received signal collection formation reception signal update collection R_-i。

$\left[\begin{array}{c}{r}_1\\ r_2\\ ...\\ r_{N_R}\end{array}\right]-\left[\begin{array}{c}{h}_{1i}\\ h_{2i}\\ ...\\ h_{N_{R}i}\end{array}\right]\×x_i=\left[\begin{array}{cccccc}{h}_{11}& ...& h_{1,i-1}& h_{1,i+1}& ...& h_{1N_T}\\ h_{21}& ...& h_{2,i-1}& h_{2,i+1}& ...& h_{2N_T}\\ ...& ...& ...& ...& ...& ...\\ h_{N_{R}1}& ...& h_{N_R,i-1}& h_{N_R,i+1}& ...& h_{N_{R}l}\end{array}\right]\×\left[\begin{array}{c}{x}_1\\ x_2\\ ...\\ x_l\end{array}\right]+\left[\begin{array}{c}{n}_1\\ n_2\\ ...\\ n_{N_R}\end{array}\right]$ Formula (2)

Order $R_{-i}=\left[\begin{array}{c}{r}_1\\ r_2\\ ..\\ r_{N_R}\end{array}\right]-\left[\begin{array}{c}{h}_{1i}\\ h_{2i}\\ ...\\ h_{N_{R}i}\end{array}\right]\×x_i$

The matrix form of the vector expression of the described reception signal collection after renewal is expressed as:

R_-i=H_-i×X_-i+ n formula (3)

S4, uses the sequence QR prestored to decompose (SQRD) algorithm in the matrix updated and renewal collection and obtains transmitting signal update collectionIn the likelihood probability of each transmitting signal, by firmly sentencing the star obtaining each transmitting signal Seat point position, and by X_-iWith x_iMerge and obtain for x_iOne group of judgement under the constellation point chosen.In this step, by X_-i With x_iCombination, forms complete string and launches the valued combinations of signal.Thus obtain a kind of situation launching signal.

S5, returns step S2, until all possible transmitting signal is all chosen complete in i-th layer of transport layer.

Till L layer transport layer is carried out above step.

S6, when L layer transport layer is carried out above step, can obtain judgement and launch signal collection, launch signal from this judgement and concentrate Find out and meet maximum likelihood standard, i.e. calculate the transmitting signal meeting maximum likelihood standard.In the present embodiment, described maximum is seemingly So the implementation of expression is:

arg min{||r-H×x||²Formula (4)

In this step, classify according to 0/1 value by bit, find out European relative to received signal vector under the combination of every class Minimum euclidean distance in Ju Li, chooses optimal transmitting signal collection.

Multiple-input and multiple-output detection method detection performance described in the present embodiment is better than SQR, ZR detection algorithm, and complexity ratio The complexity of existing detection algorithm is low, and improves the traffic rate of MIMO communication system.

Embodiment two

The present embodiment improves a kind of multiple-input and multiple-output detecting system 2, is applied to include that the multi input for sending transmitting signal is many Output transmitter and the communication system of the multi-input multi-output receiver for reception signal, described multiple-input and multiple-output is launched Facility have L layer to be S=[s for transmitting signal sequence₁,s₂,…,s_L]^TLaunch signal transport layer, there is N_TIndividual transmitting Antenna, described multi-input multi-output receiver has N_RIndividual reception antenna, launches signal collectionReceive Signal collectionRefer to Fig. 3, be shown as multiple-input and multiple-output detecting system theory structure schematic diagram.Such as figure Shown in 3, described multiple-input and multiple-output detecting system includes: the originally determined module of creation module 21, constellation 22, first chooses mould Block 23, remove module 24, more new module 25, processing module 26, recursive call module 27, computing module 28 and the second choosing Delivery block 29.

Described creation module 21 is for based on N_TIndividual transmitting antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L].In the present embodiment, step S1 also includes creating a reception signal collection vector expression model.Described connect A number vector expression for collection of collecting mail is: R=HS+N=HWX+N；Wherein, R is for receiving signal collection, and H is transmission channel Matrix, W is that every layer of transport layer launches the signal sequence mapping matrix to transmitting signal, and N is N_REach in individual reception antenna connect Receive the noise variance of antenna.The matrix form of the vector expression of described reception signal collection is expressed as:

$\left[\begin{array}{c}{r}_1\\ r_2\\ ...\\ r_{N_R}\end{array}\right]=\left[\begin{array}{cccc}{h}_{11}& h_{12}& ...& h_{1l}\\ h_{21}& h_{22}& ...& h_{2l}\\ ...& ...& ...& ...\\ h_{N_{R}1}& h_{N_{R}2}& ...& h_{N_{R}l}\end{array}\right]\×\left[\begin{array}{c}{x}_1\\ x_2\\ ...\\ x_l\end{array}\right]+\left[\begin{array}{c}{n}_1\\ n_2\\ ...\\ n_{N_R}\end{array}\right]$ Formula (1)

In transmission channel matrix H, each element representation is h_jk, h_jkThe antenna transmission channel to jth root reception antenna is launched for kth root.

The originally determined module of constellation 22 being connected with described creation module 21 is for fixing i-th layer according to preset number modulation system A transmitting signal in transport layer, and determine that this layer launches the constellation point position of signal；Wherein, i is more than or equal to 1 for being less than Integer in L.Described preset number modulation system includes ASK, PSK, FSK, QAM, MSK, GMSK or OFDM. In the present embodiment, described preset number modulation system chooses the QPSK in PSK, and so, this layer launches the constellation point position of signal It is set to $\frac{\sqrt{2}}{2}+i\frac{\sqrt{2}}{2},\frac{\sqrt{2}}{2}-i\frac{\sqrt{2}}{2},-\frac{\sqrt{2}}{2}+i\frac{\sqrt{2}}{2},-\frac{\sqrt{2}}{2}-i\frac{\sqrt{2}}{2}.$

What module 22 originally determined with described constellation was connected chooses module 23 for from launching signal x_iValue is according to preset number The constellation point that modulation system determines is chosen a constellation point.

With described choose that module 23 is connected remove module 24 for by described transmitting signal x_iRemove, more from launching signal concentration New signal collection of launching, signal update collection is launched in formationSimultaneously from described transmission channel square Battle array H removes the transmission channel row h that described i-th layer of transport layer is corresponding_iForm transmission channel and update matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L]。

With described to remove the more new module 25 that module 24 is connected for more newly received signal collection vector expression model and more newly received Signal collection is formed and receives signal update collection R_-i.By formula (1) basisWith H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L]

$\left[\begin{array}{c}{r}_1\\ r_2\\ ...\\ r_{N_R}\end{array}\right]-\left[\begin{array}{c}{h}_{1i}\\ h_{2i}\\ ...\\ h_{N_{R}i}\end{array}\right]\×x_i=\left[\begin{array}{cccccc}{h}_{11}& ...& h_{1,i-1}& h_{1,i+1}& ...& h_{1N_T}\\ h_{21}& ...& h_{2,i-1}& h_{2,i+1}& ...& h_{2N_T}\\ ...& ...& ...& ...& ...& ...\\ h_{N_{R}1}& ...& h_{N_R,i-1}& h_{N_R,i+1}& ...& h_{N_{R}l}\end{array}\right]\×\left[\begin{array}{c}{x}_1\\ x_2\\ ...\\ x_l\end{array}\right]+\left[\begin{array}{c}{n}_1\\ n_2\\ ...\\ n_{N_R}\end{array}\right]$ Formula (2)

Order $R_{-i}=\left[\begin{array}{c}{r}_1\\ r_2\\ ..\\ r_{N_R}\end{array}\right]-\left[\begin{array}{c}{h}_{1i}\\ h_{2i}\\ ...\\ h_{N_{R}i}\end{array}\right]\×x_i$

The matrix form of the vector expression of the described reception signal collection after renewal is expressed as:

R_-i=H_-i×X_-i+ n formula (3)

Remove module 24 with described and the processing module 26 that more new module 25 is connected is pre-for the matrix updated and renewal are collected employing The sequence QR deposited decomposes (SQRD) algorithm and obtainsIn the likelihood of each transmitting signal Probability, by firmly sentencing the constellation point position obtaining each transmitting signal, and by X_-iWith x_iMerge and obtain for x_iThe constellation chosen One group of judgement under Dian.In this step, by X_-iWith x_iCombination, forms complete string and launches the valued combinations of signal.Thus Obtain a kind of situation launching signal.

With described creation module 21, the originally determined module of constellation 22, choose module 23, remove module 24, more new module 25, The described recursive call module 27 that processing module 26 connects is for transmitting signals all on this layer and circulating all transport layers Call all above module and launch signal collection to obtain judgement.

The computing module 28 being connected with described recursive call module 27 is for having performed the transmission of L layer in described recursive call module 27 When having performed all functions of all above module on layer, obtain judgement and launch signal collection, launch signal collection from described judgement and fall into a trap Operator closes the transmitting signal of maximum likelihood standard, i.e. calculates the transmitting signal meeting maximum likelihood standard.In the present embodiment, institute The implementation stating maximum likelihood expression is:

arg min{||r-H×x||²Formula (4)

Described computing module 28 is classified according to 0/1 value by bit, finds out the Europe relative to received signal vector under the combination of every class Minimum euclidean distance in formula distance, makes described second to choose module 29 and chooses optimal transmitting signal.

Multiple-input and multiple-output detection method of the present invention and system detection performance are better than SQR, ZR detection algorithm, and complicated Spend lower than the complexity of existing detection algorithm, and improve the traffic rate of MIMO communication system.

So, the present invention effectively overcomes various shortcoming of the prior art and has high industrial utilization.

The principle of above-described embodiment only illustrative present invention and effect thereof, not for limiting the present invention.Any it is familiar with this skill Above-described embodiment all can be modified under the spirit and the scope of the present invention or change by the personage of art.Therefore, such as All that in art, tool usually intellectual is completed under without departing from disclosed spirit and technological thought etc. Effect is modified or changes, and must be contained by the claim of the present invention.

Claims (8)

1. a multiple-input and multiple-output detection method, be applied to include for send the multiple-input and multiple-output transmitter launching signal and for Receive the communication system of multi-input multi-output receiver of signal, described multiple-input and multiple-output transmitter have L layer for Transmitting signal sequence is S=[s₁,s₂,…,s_L]^TLaunch signal transport layer, there is N_TIndividual transmitting antenna is described many Input multi output receiver has N_RIndividual reception antenna, launches signal collectionReceive signal collectionIt is characterized in that, described multiple-input and multiple-output detection method comprises the following steps:

Based on N_TIndividual transmitting antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L]；

A transmitting signal x in i-th layer of transport layer is fixed according to preset number modulation system_i, its value is according to present count The constellation point that word modulation system determines is chosen a constellation point,；Wherein, i is the integer less than or equal to L more than or equal to 1；

By described transmitting signal x_iRemove from launching signal concentration, update and launch signal collection, formed and launch signal update collectionFrom described transmission channel matrix H, remove described i-th layer of transport layer correspondence simultaneously Transmission channel row h_iForm transmission channel and update matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L]；

More newly received signal collection is formed and receives signal update collection R_-i；

The sequence QR decomposition algorithm prestored is used to obtain in the matrix updated and renewal collectionIn the likelihood probability of each transmitting signal, obtain each transmitting signal by firmly sentencing Constellation point position, and by X_-iSame x_iMerge and obtained for x_iOne group of judgement under the constellation point chosen,

All transmitting signals in i-th layer of transport layer are performed above step, afterwards all transport layers is performed above step to obtain Take judgement transmitting signal collection and therefrom calculate the transmitting signal meeting maximum likelihood standard.

Multiple-input and multiple-output detection method the most according to claim 1, it is characterised in that: the vector expression of described reception signal collection Formula is: R=HS+N=HWX+N；

Wherein, R is for receiving signal collection, and H is transmission channel matrix, and W is that every layer of transport layer launches signal sequence to launching signal Mapping matrix, N is N_RThe noise variance of each reception antenna in individual reception antenna.

Multiple-input and multiple-output detection method the most according to claim 1, it is characterised in that: the reception sky that i-th layer of transport layer is corresponding Signal data=the h received on line_i×x_i。

Multiple-input and multiple-output detection method the most according to claim 1, it is characterised in that: a kind of realization side of maximum likelihood standard Formula is argmin{ ‖ r-H × x ‖²}。

Multiple-input and multiple-output detection method the most according to claim 1, it is characterised in that: preset number modulation system include ASK, PSK, FSK, QAM, MSK, GMSK or OFDM.

6. a multiple-input and multiple-output detecting system, be applied to include for send the multiple-input and multiple-output transmitter launching signal and for Receive the communication system of multi-input multi-output receiver of signal, described multiple-input and multiple-output transmitter have L layer for Transmitting signal sequence is S=[s₁,s₂,…,s_L]^TLaunch signal transport layer, there is N_TIndividual transmitting antenna is described many Input multi output receiver has N_RIndividual reception antenna, launches signal collectionReceive signal collectionIt is characterized in that, described multiple-input and multiple-output detecting system includes:

Creation module, for based on N_TIndividual transmitting antenna and N_RIndividual reception antenna, creates a transmission channel matrix H=[h₁,h₂,…,h_L]；

The originally determined module of constellation, for fixing a transmitting signal in i-th layer of transport layer according to preset number modulation system x_i, and determine that this layer launches the constellation point position of signal；Wherein, i is the integer less than or equal to L more than or equal to 1；

First chooses module, for choosing the constellation point determined according to preset number modulation system from transmitting signal xi value One constellation point；

Remove module, for by described transmitting signal x_iRemove from launching signal concentration, update and launch signal collection, formed and launch Signal update collectionFrom described transmission channel matrix H, remove described i-th layer simultaneously The transmission channel row h that transport layer is corresponding_iForm transmission channel and update matrix H_-i=[h₁,h₂,…h_i-1,h_i+1,…,h_L], and update Receive signal collection and form reception signal update collection R_-i；

Processing module, for using the sequence QR decomposition algorithm prestored to obtain in the matrix updated and renewal collectionIn the likelihood probability of each transmitting signal, obtain each transmitting signal by firmly sentencing Constellation point position, and by X_-iSame x_iMerge and obtain for x_iOne group of judgement under the constellation point chosen；

Recursive call module, for transmitting signals all in i-th layer of transport layer and to all transport layer recursive calls more than All modules launch signal collection to obtain judgement；

Computing module, meets the transmitting signal of maximum likelihood standard for launching signal centralized calculation from described judgement.

Multiple-input and multiple-output detecting system the most according to claim 6, it is characterised in that: described creation module is additionally operable to establishment and connects Collection of letters vector expression, i.e. R=HS+N=HWX+N；

Wherein, R is for receiving signal collection, and H is transmission channel matrix, and W is that every layer of transport layer launches signal sequence to launching signal Mapping matrix, N is N_RThe noise variance of each reception antenna in individual reception antenna.

Multiple-input and multiple-output detecting system the most according to claim 6, it is characterised in that: a kind of maximum in described computing module Likelihood standard is argmin{ ‖ r-H × x ‖²}。