CN101909028B - Single carrier demodulation method and system, transmitter and receiver - Google Patents

Single carrier demodulation method and system, transmitter and receiver Download PDF

Info

Publication number
CN101909028B
CN101909028B CN 201010228243 CN201010228243A CN101909028B CN 101909028 B CN101909028 B CN 101909028B CN 201010228243 CN201010228243 CN 201010228243 CN 201010228243 A CN201010228243 A CN 201010228243A CN 101909028 B CN101909028 B CN 101909028B
Authority
CN
China
Prior art keywords
word
matrix
channel
current data
sebolic addressing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN 201010228243
Other languages
Chinese (zh)
Other versions
CN101909028A (en
Inventor
裴玉奎
朱亮
葛宁
陆建华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Original Assignee
Tsinghua University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tsinghua University filed Critical Tsinghua University
Priority to CN 201010228243 priority Critical patent/CN101909028B/en
Publication of CN101909028A publication Critical patent/CN101909028A/en
Application granted granted Critical
Publication of CN101909028B publication Critical patent/CN101909028B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)

Abstract

The invention provides a single carrier demodulation method, a single carrier demodulation system, a transmitter and a receiver. The method comprises: setting a UW word as a new demodulation symbol; performing the QR decomposition of a channel matrix H in the receiver to obtain an orthogonal matrix Q and an upper triangular matrix R; multiplying a received data signal vector Y by QH to obtain an orthonormalized receiving vector Z, wherein the QH is a conjugate transpose matrix of the orthogonal matrix Q; and performing maximum likelihood estimation of the demodulation symbol according to the UW word, the upper triangular matrix R and the receiving vector Z. In the invention, the demodulation complexity can be reduced on the premise of high demodulation performance.

Description

A kind of single carrier demodulation method and system, transmitter and receiver
Technical field
The present invention relates to communication technical field, particularly relate to a kind of single carrier demodulation method and system, transmitter and receiver.
Background technology
The multidiameter delay expansion of channels with memory can cause strides symbol-interference (ISI, Inter-SymbolInterference), and long time delay expansion can cause that serious ISI disturbs, and will cause the receiving terminal demodulation of single carrier communication a large amount of error codes to occur, seriously reduces receptivity.Especially for the abominable wireless multipath channel of environment, deep fade can bring serious ISI.
Single carrier communication tackles the multipath channel time delay and generally adopts balancing technique, and it is divided into two classes usually:
1, linear equalization technology;
The algorithm that linear equalization algorithm commonly used is commonly used has ZF (ZF, Zero Forcing) algorithm and least mean-square error (MMSE, Minimum Mean Squared Error) algorithm, the linear equalization algorithm is relatively simple, can not eliminate preferably channel effect when the channel frequency domain has deep fade.
But, when there is the long delay multipath in wireless channel, in the channel frequency domain response a lot of deep fades at zero point can appear.For the ISI that compensates the place at zero point disturbs, the linear equalization technology must be amplified this section frequency spectrum, thereby the noise of this frequency range is significantly strengthened, and greatly reduces the signal to noise ratio of system, causes the serious variation of demodulation performance.
2, nonlinear equalization technology.
Maximum-likelihood sequence estimation (MLSE, Maximum likelihood sequence estimation) be a kind of nonlinear equalization technology, its basic thought is all possible list entries of traversal, therefrom selects with receiving symbol to have the list entries of maximum comparability as estimated result.Because MLSE do not need to use filter, therefore do not have the problem of Noise enhancement after the filtering of linear equalization, so MLSE is a kind of demodulating algorithm of performance the best.
And the problem that MLSE estimates is that complexity is too high, and the Viterbi algorithm changes the judgement convergence of the infinite length of sequence into about 5 times of channel delay length recall, and in the short occasion of channel delay, under the magnetic disc storage channel, the Viterbi algorithm can reduce complexity.But in the wireless channel of long delay, because still being exponential relationship with channel length L, the complexity of Viterbi algorithm (is C L, wherein, C is the number of constellation points of multi-system symbol), thereby Viterbi algorithm complexity under the channel of long delay expansion is high.
In a word, need the urgent technical problem that solves of those skilled in the art to be exactly: how can having reduction demodulated complex degree under the prerequisite of better demodulation performance.
Summary of the invention
Technical problem to be solved by this invention provides a kind of single carrier demodulation method and system, transmitter and receiver, in order to reduce the demodulated complex degree under the prerequisite of better demodulation performance.
In order to address the above problem, the invention discloses a kind of single carrier demodulation method, comprising:
At transmitting terminal, the UW word is added modulation symbol;
At receiving terminal, channel matrix H is carried out QR decompose, obtain orthogonal matrix Q and upper triangular matrix R;
To the data-signal vector Y that receives, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
According to described UW word, upper triangular matrix R and reception vector Z, demodulation symbol is carried out maximal possibility estimation.
Preferably, described demodulation symbol is carried out the step of maximal possibility estimation, comprising:
For the every delegation of the upper triangular matrix R except the UW word, calculate the accumulative total Euclidean distance;
Choose the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
According to the UW word length, judge whether element number satisfies prerequisite in the current data symbol sebolic addressing, if, then adjudicate according to the current data symbol sebolic addressing, obtain demodulation symbol; Otherwise, return the step of carrying out described calculating accumulative total Euclidean distance.
Preferably, described step of adjudicating according to the current data symbol sebolic addressing comprises:
According to the accumulative total Euclidean distance, P current data symbol sebolic addressing carried out hard decision or soft-decision, wherein, p is natural number.
Preferably, the pass between described reception vector Z and the upper triangular matrix R is:
Wherein,
Figure BSA00000194621200022
Be the additivity additive white Gaussian, s is signal vector to be demodulated.
Preferably, obtain channel matrix H by following steps:
Extract the UW word at receiving terminal;
According to described UW word, by calculating the time domain impulse response of channel, obtain channel matrix H.
Preferably, the T that described modulation symbol length is the UW word length times, wherein, 5＜T＜6.
According to another embodiment, the invention also discloses a kind of single carrier demodulation system, comprise transmitter and receiver, wherein, described transmitter comprises:
The UW word adds module, is used for the UW word is added modulation symbol;
Described receiver comprises:
The QR decomposing module is used for that channel matrix H is carried out QR and decomposes, and obtains orthogonal matrix Q and upper triangular matrix R;
The orthogonalization module is used for the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Estimation module is used for according to described UW word, upper triangular matrix R and reception vector Z demodulation symbol being carried out maximal possibility estimation.
Preferably, described estimation module comprises:
Computing unit is used for the every delegation for the upper triangular matrix R except the UW word, calculates the accumulative total Euclidean distance;
Choose the unit, be used for choosing the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Judging unit is used for according to the UW word length, judges whether element number satisfies prerequisite in the current data symbol sebolic addressing;
Decision unit is used for adjudicating according to the current data symbol sebolic addressing when current data symbol sebolic addressing element number satisfies prerequisite, obtains demodulation symbol;
Return the unit, be used for when current data symbol sebolic addressing element number does not satisfy prerequisite, returning described computing unit.
Preferably, described decision unit, concrete being used for carried out hard decision or soft-decision according to the accumulative total Euclidean distance to P current data symbol sebolic addressing, and wherein, p is natural number.
Preferably, described receiver also comprises:
Extraction module is used for extracting the UW word at receiving terminal;
The channel matrix acquisition module is used for by calculating the time domain impulse response of channel, obtaining channel matrix H according to described UW word.
According to another embodiment, the invention also discloses a kind of transmitter, comprising:
The UW word adds module, is used for the UW word is added modulation symbol.
Preferably, the T that described modulation symbol length is the UW word length times, wherein, 5＜T＜6.
According to another embodiment, the invention also discloses a kind of receiver, comprising:
The QR decomposing module is used for that channel matrix H is carried out QR and decomposes, and obtains orthogonal matrix Q and upper triangular matrix R;
The orthogonalization module is used for the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Estimation module is used for according to described UW word, upper triangular matrix R and reception vector Z demodulation symbol being carried out maximal possibility estimation.
Preferably, described estimation module comprises:
Computing unit is used for the every delegation for the upper triangular matrix R except the UW word, calculates the accumulative total Euclidean distance;
Choose the unit, be used for choosing the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Judging unit is used for according to the UW word length, judges whether element number satisfies prerequisite in the current data symbol sebolic addressing;
Decision unit is used for adjudicating according to the current data symbol sebolic addressing when current data symbol sebolic addressing element number satisfies prerequisite, obtains demodulation symbol;
Return the unit, be used for when current data symbol sebolic addressing element number does not satisfy prerequisite, returning described computing unit.
Preferably, described decision unit, concrete being used for carried out hard decision or soft-decision according to the accumulative total Euclidean distance to p current data symbol sebolic addressing, and wherein, p is natural number.
Preferably, described receiver also comprises:
Extraction module is used for extracting the UW word at receiving terminal;
The channel matrix acquisition module is used for by calculating the time domain impulse response of channel, obtaining channel matrix H according to described UW word.
Compared with prior art, the present invention has the following advantages:
At first, demodulated complex degree of the present invention comprises that mainly M algorithm search complexity and QR decompose complexity:
About M algorithm search complexity, owing to choose p data symbol sebolic addressing in the front M algorithm search process as the vector of surviving, therefore, C constellation point symbol for this search, only need to calculate Cp Euclidean distance, thereby M algorithm search complexity of the present invention is Cp;
Decomposing complexity about QR, is circular matrix owing to contain the channel matrix of UW word, therefore adopts the peculiar QR matrix of circular matrix Fast decomposition algorithms, can make multiplication quantity and the n of this QR decomposition algorithm 2Be directly proportional, wherein, length is that to include length in the n modulation signal data block be n uUW block and length be n-n uThe data field piece, like this, for each data field piece, the demodulated complex degree of its corresponding modulation symbol is equivalent to be directly proportional with n; Here the length L of n size and channel is at the same order of magnitude, so the QR of this algorithm decomposition complexity can be regarded L as;
To sum up, demodulated complex degree of the present invention is L+Cp, and in the reality, the value of C is smaller when C=4 (for example qpsk modulation signal) generally, and p≤C generally speaking, so described Cp complexity can ignore, therefore, with respect to the MLSE algorithm of prior art, the present invention can be with the non-linear demodulation complexity from C LReduce to L.
Secondly, because transmitting terminal adds modulation signal with the UW word, the present invention can be lower assisting of UW word known array, the component at zero point of crossing the QR decomposition carries out demodulation, greatly reduce the error propagation probability of initial demodulation, because remaining data to be demodulated have higher power, therefore can substantially reach the demodulation effect of optimal demodulation MLSE.
To sum up, the present invention under the ingenious demodulation structure of the structure of originating data and receiving end, in the time of can significantly improving the long multidiameter delay of channel, the performance of low complex degree equalization device;
Moreover the present invention can also carry out channel measurement based on transmitting terminal, particularly, can utilize in the straight plural number of frequency domain response or the time domain PN sequence of real number, and the impact of measuring channel is corresponding, thereby provides channel information for the demodulation of receiving terminal; And very frequent in the insertion of described channel measurement UW word, be convenient to channel estimation method opposing noise, improve estimated accuracy.
Description of drawings
Fig. 1 is the flow chart of a kind of single carrier demodulation method embodiment of the present invention;
Fig. 2 is the application example of a kind of modulation signal of the present invention;
Fig. 3 is the application example of the present invention in the receiving terminal demodulation;
Fig. 4 is the Another Application example of the present invention in the receiving terminal demodulation;
Fig. 5 is the structure chart of a kind of single carrier demodulation system embodiment of the present invention;
Fig. 6 is the structure chart of a kind of transmitter embodiment of the present invention;
Fig. 7 is the structure chart of a kind of receiver embodiment of the present invention.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
Core idea of the present invention is:
1, decomposes by the QR to channel matrix so that M beta pruning algorithm estimates that restituted signal becomes possibility, to reduce the complexity of maximum-likelihood sequence estimation;
2, in the modulation signal of transmitting terminal, add the UW word, because this UW word can serve as known data in demodulation, thereby can skip channel deep fade at the zero point coefficient in the lower right corner among the upper triangular matrix R, reduce error propagation, and then increase substantially demodulation performance.
With reference to Fig. 1, show the flow chart of a kind of single carrier demodulation method embodiment of the present invention, specifically can comprise:
Step 101, at transmitting terminal, the UW word is added modulation symbol;
The present invention adopts UW-QRM-MLD, and (the UW word is assisted, the QR of maximum likelihood decomposes the M algorithm, Unique Word assisted QR decomposition associated with the M-algorithm tomaximum likelihood detection) carries out single carrier demodulation, wherein, described M algorithm also is M beta pruning algorithm, can be used to reduce the complexity of route searching.
With reference to Fig. 2, in a kind of application example of the present invention, described modulation signal can comprise two parts: data field and UW field; The length of supposing modulated signal frames is n, and the UW field length is n u, then data-field length is n-n u
In practice, the UW word can be the time domain PN sequence (pseudo noise sequence, Pseudo-noise Sequence) of plural number or real number, and its conversion at frequency domain has constant amplitude.This UW word can be used to remove the linear convolution hangover between Frame, and effect is equivalent to Cyclic Prefix.Therefore, the length requirement of this UW field is greater than the length of the root mean square multidiameter delay of channel, and the multipath that could eliminate between Frame disturbs.
In general, in order to guarantee data transmission efficiency, the 4-5 that the length of data field is generally the UW word length doubly, if high then process complexity and will increase again.For example, the length of QPSK (Quadrature Phase Shift Keying, Quadrature Phase Shift Keying) modulation signal is generally T times of UW word length, wherein, and 5＜T＜6.
Certainly, the value of above-mentioned T is just as example, and those skilled in the art can arrange other value of T according to actual needs, for example, T=10 etc., the present invention is not limited this.
Step 102, at receiving terminal, channel matrix H is carried out QR decomposes, obtain orthogonal matrix Q and upper triangular matrix R;
For using M beta pruning algorithm that each modulation symbol of above-mentioned transmission is searched for, the present invention at first carries out QR to the channel matrix H of receiving terminal and decomposes.
The method that receiving terminal obtains channel matrix H has a lot, for example, can utilize the characteristic of UW word channel measurement to obtain this channel matrix H, specifically can comprise:
At first extract the UW word at receiving terminal;
Then, according to described UW word, calculate the time domain impulse response of channel by adopting the slip correlation technique, obtain channel matrix H.
Because channel matrix H is a Theory of Circular Matrix, have nonsingular characteristic, therefore, according to the QR decomposition theorem, there are an orthogonal matrix Q and upper triangular matrix R, satisfy H=QR, and the diagonal element of upper triangular matrix R is positive number entirely.Because there is QQ H=Q HQ=I.So following formula can be set up:
Wherein, r I, j=0, for i＞j; Q HAssociate matrix for orthogonal matrix Q; H is the matrix of n * n, and n is the length of modulation signal here.
Step 103, the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
In practice, can be at first show relation between data-signal vector Y and the channel matrix H with following formula table:
Y＝Hs+N
(2)
＝QRs+N
Here, N is the additivity additive white Gaussian, and s is symbolic vector to be demodulated.
Then, to data-signal vector Y premultiplication with Q H, obtain orthogonalized reception vector Z, obtain:
Figure BSA00000194621200082
Here
Figure BSA00000194621200083
Because Q HBe orthogonal matrix, so this matrix multiple can't increase the power of noise.Top formula also can be expressed as:
Step 104, according to described UW word, upper triangular matrix R with receive vector Z, demodulation symbol is carried out maximal possibility estimation.
Common QRM-MLD algorithm (QR of maximum likelihood decomposes the M algorithm) carries out the search of demodulation symbol vector s with the behavior unit of upper triangular matrix R; And, in each search procedure, choose p data symbol sebolic addressing.
In specific implementation, the direction of search is general last column r from the R matrix N, nVector sum s nBegin, proceed at last the first row r of R matrix 1Vector sum symbol s 1
With s nSearch be example, calculate its corresponding Euclidean distance:
E n ( m , j ) = | z n - r n , n s n ( m ) | 2 - - - ( 5 )
Here, m represents demodulation symbol s nThe constellation point sequence number, also, by the traversal s nAll constellation point, obtain corresponding Euclidean distance.
For choosing of the vector of surviving, for p data symbol sebolic addressing choosing successively the Euclidean distance minimum, take p=2 as example, also namely choose minimum and inferior 2 the little data symbol sebolic addressings of Euclidean distance as the vector of surviving; And reservation respective symbol estimated value s nSequence number m.Simultaneously, the j=1 here ..., p represents j minimum Euclidean distance.
Then, search for again r N-1Vectorial and corresponding s N-1The estimated value of symbol, it is as follows to calculate the accumulative total Euclidean distance:
E n - 1 ( m , j ) = | z n - 1 - ( r n - 1 , n - 1 s n - 1 ( m ) + r n - 1 , n s n ( m ) ) | 2 + E n ( m , j ) - - - ( 6 )
Owing to p candidate symbol arranged, s corresponding to each symbol this moment before N-1Have C constellation point hunting zone, so for this
Figure BSA00000194621200093
Have the combination that Cp possible data symbol sequence estimated.This search then from this Cp data symbol sebolic addressing estimate to choose the combination p vectorial as surviving, specifically choose process can for, add up Euclidean distance from Cp
Figure BSA00000194621200094
In choose p minimum, and data symbol sequence corresponding to record.
Above process is sustained, until proceed to s 1Corresponding minimum accumulative total Euclidean distance
Figure BSA00000194621200095
Obtained, can adjudicate this moment.
In the demodulating process of above-mentioned QRM-MLD because multipath channel exists zero point of frequency domain, the computational process of decomposing according to QR as can be known, the R that generates at last is a sorted matrix, r in the matrix N, nPower is minimum and approach zero, and r 1,1Power the highest, surpass the power of original transmission modulation symbol.Carry out like this QRM-MLD demodulation, can cause s nSignal to noise ratio extremely low, fully correct demodulating data symbols also can cause follow-up sequencal estimation search cumulative error to occur simultaneously, causes high error floor.So should be avoided in r N, nThe signal demodulation at place.
For the low situation of above-mentioned demodulation performance, UW-QRM-MLD demodulating algorithm of the present invention can solve the r of above-mentioned QRM-MLD method by the auxiliary receiver signal demodulation of known channel estimating UW word N, nPlace's signal zero problem.
Its basic thought is that the channel matrix H size is n * n, wherein, both can comprise data field, can comprise the UW word again; This means and the s vector is being carried out in the process of demodulation, to known, be included in the n among the s uLong UW word need not demodulation fully, like this, and can be in matrix R, from Corresponding
Figure BSA00000194621200102
The sign-on demodulation.
In practice, step 104 specifically can comprise:
Substep A1, for the every delegation of the upper triangular matrix R except the UW word, calculate the accumulative total Euclidean distance;
Substep A2, choose the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Substep A3, according to the UW word length, judge whether element number satisfies prerequisite in the current data symbol sebolic addressing, if, then adjudicate according to the current data symbol sebolic addressing, obtain demodulation symbol; Otherwise, return the step of carrying out described calculating accumulative total Euclidean distance.
For example, substep A1 at first for
Figure BSA00000194621200103
The row correspondence
Figure BSA00000194621200104
Sign computation accumulative total Euclidean distance; Substep A2 is to C
Figure BSA00000194621200105
The accumulative total Euclidean distance of the correspondence of symbol is carried out ordering from small to large, and chooses p as the current data symbol sebolic addressing of survival from C according to described ranking results, also be the current data symbol sebolic addressing can for
Figure BSA00000194621200106
Here { } represents sequence symbol; The prerequisite of substep A3 then is that whether element number is more than or equal to n-n in the current data symbol sebolic addressing uIf,, represent the current data symbol sebolic addressing and be
Figure BSA00000194621200107
Then judge according to p described current data symbol sebolic addressing and corresponding accumulative total Euclidean distance, otherwise the search procedure of substep A1 and substep A2 is carried out in circulation.
In practice, described judgement can be hard decision, for example, and directly with minimum
Figure BSA00000194621200108
Corresponding modulating data sequencal estimation value is originating data; Also can be soft-decision, for example calculate log-likelihood ratio according to the accumulative total Euclidean distance of the survival estimate vector of afterbody, and provide soft information to channel decoder, improve the performance of channel decoding; The present invention is not limited concrete decision method.
Because n uLong UW field has surpassed the length of channel multi-path time delay, so the relevant data of deep fade at zero point need not demodulation fully under known UW word is auxiliary.And because it is Partition of Unity that QR decomposes, the conservation of energy.So remaining data unknown to be demodulated
Figure BSA00000194621200111
Average power exceed original transmitting terminal symbol power so that the demodulation performance of this patent algorithm can reach the optimal demodulation performance of existing MLSE technology substantially.
The present invention has following advantage:
At first, demodulated complex degree of the present invention comprises that mainly M algorithm search complexity and QR decompose complexity:
About M algorithm search complexity, owing to choose p data symbol sebolic addressing in the front M algorithm search process as the vector of surviving, therefore, C constellation point symbol for this search, only need to calculate Cp Euclidean distance, thereby M algorithm search complexity of the present invention is Cp;
Decomposing complexity about QR, is circular matrix owing to contain the channel matrix of UW word, therefore adopts the peculiar QR matrix of circular matrix Fast decomposition algorithms, can make complexity and the n of this algorithm 2Be directly proportional (n is modulation signal length), to n-n altogether uIndividual the transmission of data, the demodulated complex degree of each modulation symbol is equivalent to be directly proportional with n; And the size of n and channel length L be at an order of magnitude, so the QR of this algorithm decomposes complexity and can be considered as L;
To sum up, demodulated complex degree of the present invention is L+Cp, and in the reality, the value of C is smaller (for example 4) generally, and p≤C generally speaking, so described Cp complexity can be ignored, therefore, with respect to the MLSE algorithm of prior art, the present invention can be with the non-linear demodulation complexity from C LReduce to L;
Secondly, because transmitting terminal adds modulation signal with the UW word, the present invention can be lower assisting of UW word known array, the component at zero point of crossing the QR decomposition carries out demodulation, and data to be demodulated have higher power, therefore the demodulation effect of optimal demodulation MLSE can be substantially reached, the performance of low complex degree equalization device can be significantly improved under the long channel multidiameter delay.
Moreover, the present invention can also carry out channel measurement based on transmitting terminal, particularly, can utilize in the straight plural number of frequency domain response or the time domain PN sequence of real number, it is corresponding at the channel of frequency domain or time domain to measure channel, thereby provides channel information for the demodulation of receiving terminal; And very frequent in the insertion of described channel measurement UW word, be convenient to channel estimation method and improve estimated accuracy.
For making those skilled in the art understand better the present invention, below provide the present invention the application example in the receiving terminal demodulation:
Example 1,
With reference to Fig. 3, specifically can comprise:
Step 301, for the radiofrequency signal from antenna, carry out radio-frequency front-end and amplify;
Step 302, after down-converted, radiofrequency signal is carried out the A/D conversion, and estimates to obtain channel matrix H;
Step 303, channel matrix H is carried out QR decompose, obtain orthogonal matrix Q and upper triangular matrix R;
Step 304, the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Step 305, capable for the upper triangular matrix R except the UW word is calculated the accumulative total Euclidean distance, and wherein, described UW word adds at transmitting terminal;
Step 306, choose the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Step 307, judge that whether element number is more than or equal to n-n in the current data symbol sebolic addressing uIf,, then execution in step 308; Otherwise, return step 305, wherein, n is modulation signal length, n uBe the UW word length;
Step 308, carry out the hard decision of minimum accumulative total Euclidean distance according to described current data symbol sebolic addressing and corresponding accumulative total Euclidean distance, obtain restituted signal, perhaps, soft-decision calculates the log-likelihood ratio of the current data symbol sebolic addressing of surviving.
In this example, the log-likelihood ratio of described restituted signal or calculating can be input in the channel decoder.
Example 2,
With reference to Fig. 4, specifically can comprise:
Step 401, for the radiofrequency signal from antenna, carry out radio-frequency front-end and amplify;
Step 402, after down-converted, radiofrequency signal is carried out the A/D conversion;
Step 403, for the radiofrequency signal after the A/D conversion, the UW word according to transmitting terminal adds calculates the time domain impulse response of channel by adopting the slip correlation technique, obtains channel matrix H;
Step 404, channel matrix H is carried out QR decompose, obtain orthogonal matrix Q and upper triangular matrix R;
Step 405, the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Step 406, capable for the upper triangular matrix R except the UW word is calculated the accumulative total Euclidean distance;
Step 407, choose the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Step 408, judge that whether element number is more than or equal to n-n in the current data symbol sebolic addressing uIf,, then execution in step 409; Otherwise, return step 406, wherein, n is modulation signal length, n uBe the UW word length;
Step 409, carry out the hard decision of minimum accumulative total Euclidean distance according to described current data symbol sebolic addressing and corresponding accumulative total Euclidean distance, obtain restituted signal, perhaps, soft-decision calculates the log-likelihood ratio of the current data symbol sebolic addressing of surviving.
In this example, described restituted signal is input in the channel decoder and goes.
Difference with example 1 in this example is, obtain channel matrix H by the UW word, because the ratio of UW word in restituted signal can reach 10～20%, if channel variation is slower, can be to the summation that adds up of many groups UW word, can reduce the channel additive noise, greatly improve the precision of channel estimation of UW word.
Need to prove that the UW field can also be synchronous for the system clock of radio communication, the present invention is not limited the concrete application of UW word.
In addition, those skilled in the art can compositive complexity and amount of information select the value of p, for example, when tending to little complexity, the value of p can be for being less than or equal to the number of constellation points C of modulated data symbol; When tending to high amount of information, the value of p can be greater than C; The present invention is not limited this.
Embodiment is corresponding with preceding method, and the present invention gives a kind of single carrier demodulation system, and with reference to Fig. 5, described system specifically can comprise transmitter 501 and receiver 502, and wherein, described transmitter 501 specifically can comprise:
The UW word adds module 511, is used for the UW word is added modulation symbol;
Described receiver 502 specifically can comprise:
QR decomposing module 521 is used for that channel matrix H is carried out QR and decomposes, and obtains orthogonal matrix Q and upper triangular matrix R;
Orthogonalization module 522 is used for the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Estimation module 523 is used for according to described UW word, upper triangular matrix R and reception vector Z demodulation symbol being carried out maximal possibility estimation.
In specific implementation, described estimation module 523 specifically can comprise:
Computing unit B1 is used for the every delegation for the upper triangular matrix R except the UW word, calculates the accumulative total Euclidean distance;
Choose unit B 2, be used for choosing the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Judging unit B3 is used for according to the UW word length, judges whether element number satisfies prerequisite in the current data symbol sebolic addressing;
Decision unit B4 is used for adjudicating according to the current data symbol sebolic addressing when current data symbol sebolic addressing element number satisfies prerequisite, obtains demodulation symbol;
Return unit B 5, be used for when current data symbol sebolic addressing element number does not satisfy prerequisite, returning described computing unit.
Preferably, described decision unit B4 can specifically be used for according to the accumulative total Euclidean distance, and p current data symbol sebolic addressing carried out hard decision or soft-decision, and wherein, p is natural number.
In a preferred embodiment of the present invention, can utilize the characteristic of UW word channel measurement to obtain this channel matrix H, at this moment, described receiver 502 can also comprise:
Extraction module C1 is used for extracting the UW word at receiving terminal;
Channel matrix acquisition module C2 is used for by calculating the time domain impulse response of channel, obtaining channel matrix H according to described UW word.
With reference to Fig. 6, show the structure chart of a kind of transmitter embodiment of the present invention, specifically can comprise:
The UW word adds module 601, is used for the UW word is added modulation symbol.
In practice, those skilled in the art can be as required, and the T that described modulation symbol length is set is the UW word length doubly, and is preferred, 5＜T＜6.
With reference to Fig. 7, show the structure chart of a kind of receiver embodiment of the present invention, specifically can comprise:
QR decomposing module 701 is used for that channel matrix H is carried out QR and decomposes, and obtains orthogonal matrix Q and upper triangular matrix R;
Orthogonalization module 702 is used for the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Estimation module 703 is used for according to described UW word, upper triangular matrix R and reception vector Z demodulation symbol being carried out maximal possibility estimation.
Concrete, the structure of described estimation module 703 can comprise:
Computing unit D1 is used for the every delegation for the upper triangular matrix R except the UW word, calculates the accumulative total Euclidean distance;
Choose cells D 2, be used for choosing the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Judging unit D3 is used for according to the UW word length, judges whether element number satisfies prerequisite in the current data symbol sebolic addressing;
Decision unit D4 is used for adjudicating according to the current data symbol sebolic addressing when current data symbol sebolic addressing element number satisfies prerequisite, obtains demodulation symbol;
Return cells D 5, be used for when current data symbol sebolic addressing element number does not satisfy prerequisite, returning described computing unit.
In practice, described decision unit D4 can specifically be used for according to the accumulative total Euclidean distance, and p current data symbol sebolic addressing carried out hard decision or soft-decision, and p is natural number.
In a preferred embodiment of the present invention, can utilize the characteristic of UW word channel measurement to obtain this channel matrix H, at this moment, described receiver can also comprise:
Extraction module E1 is used for extracting the UW word at receiving terminal;
Channel matrix acquisition module E2 is used for by calculating the time domain impulse response of channel, obtaining channel matrix H according to described UW word.
Each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For system embodiment because itself and embodiment of the method basic simlarity, so describe fairly simple, relevant part gets final product referring to the part explanation of embodiment of the method.
More than to a kind of single carrier demodulation method provided by the present invention and system, transmitter and receiver, be described in detail, used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.

Claims (6)

1. a single carrier demodulation method is characterized in that, comprising:
At transmitting terminal, special word UW (Unique Word) is added modulation symbol;
At receiving terminal, channel matrix H is carried out QR decompose, obtain orthogonal matrix Q and upper triangular matrix R;
The concrete steps of obtaining channel matrix H comprise: extract the UW word at receiving terminal; According to described UW word, by calculating the time domain impulse response of channel, obtain channel matrix H;
To the data-signal vector Y that receives, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
According to described UW word, upper triangular matrix R and reception vector Z, demodulation symbol is carried out maximal possibility estimation; Specifically comprise:
For the every delegation of the upper triangular matrix R except the UW word, calculate the accumulative total Euclidean distance;
Choose the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
According to the UW word length, judge whether element number satisfies prerequisite in the current data symbol sebolic addressing, if, then adjudicate according to the current data symbol sebolic addressing, obtain demodulation symbol; Otherwise, return the step of carrying out described calculating accumulative total Euclidean distance;
The length of described UW field is greater than the length of the root mean square multidiameter delay of channel.
2. the method for claim 1 is characterized in that, described step of adjudicating according to the current data symbol sebolic addressing comprises:
According to the accumulative total Euclidean distance, p current data symbol sebolic addressing carried out hard decision or soft-decision, wherein, p is natural number.
3. such as each described method in the claim 1 to 2, it is characterized in that the pass between described reception vector Z and the upper triangular matrix R is:
Figure FDA00002375140800011
Wherein,
Figure FDA00002375140800012
Be the additivity additive white Gaussian, s is signal vector to be demodulated.
4. the method for claim 1 is characterized in that, the T that described modulation symbol length is the UW word length times, wherein, 5＜T＜6.
5. a single carrier demodulation system is characterized in that, comprises transmitter and receiver, and wherein, described transmitter comprises:
Special word UW (Unique Word) adds module, is used for special word UW (Unique Word) is added modulation symbol;
Described receiver comprises:
The QR decomposing module is used for that channel matrix H is carried out QR and decomposes, and obtains orthogonal matrix Q and upper triangular matrix R;
Extraction module is used for extracting the UW word at receiving terminal;
The channel matrix acquisition module is used for by calculating the time domain impulse response of channel, obtaining channel matrix H according to described UW word.
The orthogonalization module is used for the data-signal vector Y to receiving, premultiplication Q H, obtain orthogonalized reception vector Z, wherein, Q HAssociate matrix for orthogonal matrix Q;
Estimation module is used for according to described UW word, upper triangular matrix R and reception vector Z demodulation symbol being carried out maximal possibility estimation; Specifically comprise:
Computing unit is used for the every delegation for the upper triangular matrix R except the UW word, calculates the accumulative total Euclidean distance;
Choose the unit, be used for choosing the current data symbol sebolic addressing according to described accumulative total Euclidean distance;
Judging unit is used for according to the UW word length, judges whether element number satisfies prerequisite in the current data symbol sebolic addressing; The length of described UW field is greater than the length of the root mean square multidiameter delay of channel;
Decision unit is used for adjudicating according to the current data symbol sebolic addressing when current data symbol sebolic addressing element number satisfies prerequisite, obtains demodulation symbol;
Return the unit, be used for when current data symbol sebolic addressing element number does not satisfy prerequisite, returning described computing unit.
6. system as claimed in claim 5 is characterized in that, described decision unit, and concrete being used for carried out hard decision or soft-decision according to the accumulative total Euclidean distance to p current data symbol sebolic addressing, and wherein, p is natural number.
CN 201010228243 2010-07-08 2010-07-08 Single carrier demodulation method and system, transmitter and receiver Active CN101909028B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201010228243 CN101909028B (en) 2010-07-08 2010-07-08 Single carrier demodulation method and system, transmitter and receiver

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201010228243 CN101909028B (en) 2010-07-08 2010-07-08 Single carrier demodulation method and system, transmitter and receiver

Publications (2)

Publication Number Publication Date
CN101909028A CN101909028A (en) 2010-12-08
CN101909028B true CN101909028B (en) 2013-02-20

Family

ID=43264351

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201010228243 Active CN101909028B (en) 2010-07-08 2010-07-08 Single carrier demodulation method and system, transmitter and receiver

Country Status (1)

Country Link
CN (1) CN101909028B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105071842B (en) * 2015-07-07 2018-05-15 江苏中兴微通信息科技有限公司 First-class triantennary or three four antenna space diversity receiving/transmission methods of stream and device in SC-MIMO systems
CN105871763A (en) * 2016-05-30 2016-08-17 武汉邮电科学研究院 Channel equalization method and system based on QR decomposition
CN112637092B (en) * 2019-09-24 2023-10-27 中兴通讯股份有限公司 BP equalization method, BP equalization device, communication equipment and storage medium
CN111478867B (en) * 2020-03-23 2023-01-31 比科奇微电子(杭州)有限公司 Signal processing method, signal processing device, storage medium, processor, and electronic device
CN111614583B (en) * 2020-05-18 2023-04-28 Oppo广东移动通信有限公司 Signal demodulation method, electronic equipment and computer readable storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1838556A (en) * 2005-03-24 2006-09-27 松下电器产业株式会社 Downlink multi-user space-time packet pre-coding method
CN101150346A (en) * 2000-10-02 2008-03-26 三菱电机株式会社 Communication system and its transmitting channel estimation method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007228029A (en) * 2006-02-21 2007-09-06 Fujitsu Ltd Wireless communication system and receiving device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101150346A (en) * 2000-10-02 2008-03-26 三菱电机株式会社 Communication system and its transmitting channel estimation method
CN1838556A (en) * 2005-03-24 2006-09-27 松下电器产业株式会社 Downlink multi-user space-time packet pre-coding method

Also Published As

Publication number Publication date
CN101909028A (en) 2010-12-08

Similar Documents

Publication Publication Date Title
RU2624101C2 (en) Receiver and method of radio signal processing using soft pilot symbols
EP2245810B1 (en) A method and apparatus for block-based signal demodulation
CN101027848A (en) Reduced complexity soft value generation for MIMO JDGRAKE receivers
CN101909028B (en) Single carrier demodulation method and system, transmitter and receiver
CN102484624A (en) Symbol estimation methods and apparatuses
CN1771671B (en) Joint detection method and system in CDMA communications system
CN101359926A (en) Receiver, channel estimation method and apparatus
US8139669B2 (en) Space domain filter detecting method in a multi-antenna wireless communication system
CN102281129A (en) Information processing method in MIMO two-way relay system
EP1759506B1 (en) Reception with least-squares estimation of interference
CN101133562A (en) Method for restraining user signal interference from neighbor subdistrict
CN100486236C (en) Receiver and method for ID-SCDMA system
EP2834920B1 (en) Soft-bit generation and channel estimation for interference cancellation using decoder output
EP2745405B1 (en) Method and apparatus for soft symbol processing in a communication receiver
CN101964667B (en) High-efficiency multi-antenna detection method for long term evolution scheme
CN111614583B (en) Signal demodulation method, electronic equipment and computer readable storage medium
CN109981151A (en) Improved Gauss tree approximation message transmission detection algorithm in extensive mimo system
EP2680520B1 (en) Method and apparatus for efficient MIMO reception with reduced complexity
CN105406909A (en) MIMO (Multiple Input Multiple Output) detection method and system based on internal iterative interference cancellation
CN100358324C (en) Data equalization method for burst communication
CN101552622A (en) A method and device to estimate the frequency deviation in TD-CDMA access system
CN101292485A (en) Adaptive 8PSK demodulating method suitable for EDGE system
CN101170529A (en) A joint detection method and device
Li et al. A decision feedback channel estimation approach of the satellite mobile communication system
CN102545990B (en) Method and device for demodulating received signal

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant