CN101404633A - Carrier wave tracing method for single carrier system based on block transmission - Google Patents

Abstract

The invention provides a carrier wave tracking method of a single carrier wave system based on block transmission. The method includes: step A: carrying out equalization treatment of the data block by phase compensation in the current frame and selecting the corresponding soft judgment data of cyclic prefix from soft judgment element data after equalization and compensation; step B: utilizing the phase estimation algorithm to calculate the soft judgment data and obtaining the phase estimation of the present data block; step C: carrying out accumulated smoothing treatment of the estimation phase and using the phase estimation value of next data block by accumulated smooth output to compensate the next data block; carrying out step A to step C repeatly until the transmission of all data blocks in the current frame is finished. The invention solves the shortcomings of the original calculation and achieves the frequency tracking directly by phase estimation. In the tracking process, the invention carries out the phase compensation and reduces the realization complexity; in addition, the algorithm is not only high in accuracy but also low in complexity.

Description

Carrier wave tracing method based on the single-carrier system of piece transmission

Technical field

The present invention designs communication technique field, relates in particular to a kind of carrier wave tracing method of the single-carrier system based on piece transmission.

Background technology

SC-FDE (single carrier frequency domain equalization) is the method that a kind of up-and-coming anti-multipath disturbs in the broadband wireless transmission, the same branch of taking with OFDM transmits soon, and adopt Cyclic Prefix Cyclic Prefix (CP) mode, so just can be converted into circular convolution to signal in the linear convolution of channel impulse response, and eliminate the interference of the data block that multipath causes.Adopt simple frequency-domain equalization technology just can eliminate intersymbol interference at receiving terminal.

A most crucial problem in the digital communication technology field is exactly a stationary problem, is divided into carrier synchronization technically with regularly synchronously.In the baseband sampling signal that obtains through the receiving terminal behind the wireless mobile channel, exist carrier frequency offset, phase deviation, timing offset, these deviations generally derive from Doppler effect and these several aspects of frequency selectivity in not matching of crystal oscillator between the transmitter and receiver and the wireless mobile channel, even if in the indoor environment of channel relatively flat, also can there be these deviations.

Use the SC-FDE system of piece transmission technology that these synchronous deviations are had higher requirement.The simultaneous techniques of SC-FDE system is divided into carrier synchronization and regularly synchronously, its receiving terminal mainly contain following task synchronously:

Frame arrive to detect, and is used for judging whether detect signal, and this is the synchronous first step, has only and judges that what receive is that useful signal just can carry out the processing of back;

Carrier wave frequency deviation is caught;

Carrier frequency tracking;

The symbol original position is estimated (piece is synchronous), promptly accurately knows original position (initial Fourier transform FFT position) before equilibrium; For obtaining more accurate position, need timing estimation.

The tracking of sampling clock frequency.

The signal of SC-FDE system arrives along the time shaft order, and each data block of SC-FDE signal of piecemeal transmission is made up of data segment and Cyclic Prefix.Because not matching of crystal oscillator can produce very large frequency departure, the error that frequency deviation causes pure single-carrier system is equivalent to cause signal rotation and decay to the SC-FDE radiofrequency signal in separating the process of being transferred to baseband signal; When this frequency deviation appears at the SC-FDE system, also can cause intersymbol interference ISI, inter-carrier interference ICI, have influence on the frequency domain equalization algorithm accuracy of SC-FDE system, therefore the frequency deviation of the baseband sampling signal that obtains after the receiving terminal down-conversion must catch earlier and compensate, the frequency deviation of local carrier is locked in the small range, makes the system can operate as normal.The remaining frequency deviation in compensation back is referred to as residual frequency departure again.As time goes on, the phase shift difference that residual frequency departure causes is along with the accumulation of time, can make angle of receiving symbol rotation, cause can't be correct the symbol that demodulates; Channel estimating and frequency domain equalization that the cumulative departure of while phase place also can lead to errors.In order to keep good synchronization performance, must carry out carrier track.Carrier track is promptly followed the tracks of the variation of carrier phase deviation, prevents the accumulation of phase effect that frequency deviation causes.

In the general SC-FDE system, data flow is transmitted by piecemeal, the time domain data vector paragraph that it is N that every blocks of data stream certain encoding process of process and constellation mapping form a length afterwards

At data segment

Adding that afterwards length is the Cyclic Prefix of L---UW sequence (the unique word sequence is generally chu, frank, PN sequence) obtains the vector that length is M=L+N

Be called data block.Data block obtains the signal flow corresponding with it by after the Channel Transmission at receiving terminal

Obtain soft-decision data through isostatic compensation again Decode at last.

At present, general carrier tracking method utilizes the information of the Cyclic Prefix in the adjacent data blocks among the SC-FDE to obtain the frequency deviation of current block, or utilize information after soft decision information behind the frequency domain equalization or reconstruct recover to estimate the frequency deviation of current block, reach the effect of frequency tracking with this, and then carry out phase compensation, realize carrier tracking.These solutions mostly are to use for reference from the related synchronization algorithm of OFDM and come.

Carry out this class algorithm of frequency offset tracking for the phase difference that directly utilizes the UW in the receiving terminal adjacent data blocks, because the reception data of algorithm utilization this moment are the information through not passing through isostatic compensation behind the wireless mobile channel, therefore the carrier wave frequency deviation error that obtains is bigger, precision is low, can't reach the purpose of tracking.And soft-decision data after the existing another kind of utilization equilibrium or reconstruct recovered data are carried out the algorithm that frequency deviation is followed the trail of, and utilize the current block frequency deviation that obtains to remove to compensate current data block, and the algorithm complexity implements time-delay greatly.These algorithms all are after finishing the frequency deviation tracking simultaneously, carry out phase compensation again, have further strengthened the complexity that realizes.

Summary of the invention

In view of above-mentioned analysis, the object of the invention is to provide a kind of carrier wave tracing method of the single-carrier system based on the piece transmission, in order to solve the problem that carrier tracking methodical error of the prior art is big, precision is low or complexity is big.

Purpose of the present invention mainly is achieved through the following technical solutions:

The invention provides a kind of carrier wave tracing method of the single-carrier system based on piece transmission, described method comprises:

Steps A: the data block through phase compensation in the present frame is carried out equilibrium treatment, and choose the soft-decision data of Cyclic Prefix correspondence the soft decision symbols data that after isostatic compensation, obtain;

Step B: utilize maximum likelihood phase estimation algorithm that described soft-decision data is calculated, obtain the phase estimation of current data block;

Step C: the estimation phase place that will the obtain smoothing processing that adds up, and the level and smooth output that will add up is used to compensate next data block as the phase estimation value of next data block;

Repeated execution of steps A is to step C, up to all end of transmission blocks of present frame.

Further, described step B specifically comprises:

Obtain the correlation of the known Cyclic Prefix soft-decision data corresponding earlier, then this correlation is asked argument, obtain the estimation phase place of current data block with it;

Perhaps, described step B specifically comprises:

Utilize the decoding judgement output of soft-decision data to carry out relevant estimation phase place of trying to achieve current data block with soft-decision data.

Further, described step C specifically comprises:

By a predefined smoother that the adds up smoothing processing that adds up, the time domain expression formula of the smoother that adds up is with the estimation phase place that obtains: $S(i+1)=\left(\begin{array}{c}S\left(i\right)+\frac{1}{i-1}\left(\underset{m=2}{\overset{i}{\mathrm{\Σ}}}\mathrm{est}\left(i\right)\right)....i\≥2\\ \mathrm{est}\left(i\right)..............................i=1\end{array}\right),$ S (1)=0 wherein, i represents the i blocks of data, est (i) is the smoother input of i piece, i.e. the previous step phase estimation value of trying to achieve

S (i+1) is the level and smooth output of adding up of i piece, and the level and smooth phase estimation value of exporting as next data block that adds up is used to compensate next data block.

Beneficial effect of the present invention is as follows:

The invention solves the shortcoming of original algorithm, directly reach frequency tracking, in the process of following the trail of, carrying out phase compensation simultaneously, reduced the complexity that realizes by phase estimation; The while algorithm is the precision height not only, and complexity is very low, and whole algorithm does not constitute loop structure truly, and it is very little to delay time, and very high practical value is arranged.

Other features and advantages of the present invention will be set forth in the following description, and becoming apparent from specification of part perhaps understood by implementing the present invention.Purpose of the present invention and other advantages can realize and obtain by specifically noted structure in the specification of being write, claims and accompanying drawing.

Description of drawings

Fig. 1 is for being the structural representation of the data block of single-carrier frequency domain equalization system;

Fig. 2 is the schematic flow sheet of the described method of the embodiment of the invention.

Embodiment

Specifically describe preferential embodiment of the present invention below in conjunction with accompanying drawing, wherein, accompanying drawing constitutes the application's part, and is used from explaination principle of the present invention with embodiments of the invention one.

Consider and before carrier tracking, carried out capturing carrier, the frequency deviation of this moment has been locked in the small range, it is not responsive especially to carrier synchronization that while SC-FDE compares OFDM, therefore utilize through Cyclic Prefix in the current data block after the equilibrium, it is UW sequence (unique word sequence, be generally chu sequence, frank-zadaff sequence, the PN sequence of IEEE802.16a standard code) corresponding soft-decision data r, estimate the phase place of current data block according to ML (maximum likelihood, Maximum Likelihood) phase estimation algorithm; Again with the phase place that estimates by the smoother that adds up; To go to compensate next data block through the phase value that obtains after the smoothing processing that adds up, go down, to reach the purpose of carrier tracking with this recursion.

The SC-FDE system is transmitted frame by frame, and every frame is n data block altogether, and the every blocks of data of SC-FDE signal of piecemeal transmission is made up of useful data and Cyclic Prefix, and Cyclic Prefix is taken as Chu sequence (L Baud Length), obtains the vector that length is M=L+N As shown in Figure 1, i data block correspondence Obtain the signal flow corresponding by after the Channel Transmission at receiving terminal with it ${\stackrel{\→}{r}}_i=[r\left(1\right),\·\·\·,r\left(M\right)],$ Obtain the soft decision symbols data through isostatic compensation again ${\stackrel{\→}{z}}_i=[z\left(1\right),\·\·\·,z\left(M\right)]$ Decode at last.

Specifically can may further comprise the steps:

Step 200: phase compensation; The expression formula of phase compensation is: ${\stackrel{\→}{r}}_i={\stackrel{\→}{r}}_i\×e^{-\mathrm{j\φ}\left(i\right)}.$

Step 201: with first valid data piece of present frame ${\stackrel{\→}{r}}_1=[r\left(1\right),\·\·\·,r\left(M\right)]$ By obtaining the soft decision symbols data after the isostatic compensation ${\stackrel{\→}{z}}_1=[z\left(1\right),\·\·\·,z\left(M\right)],$ Choose Chu sequence correspondence soft-decision data [z (N+1) ..., z (M)], use to quantity symbol ${\stackrel{\→}{u}}_1=[u_1\left(1\right),\·\·\·,u_1\left(L\right)]$ Expression [z (N+1) ..., z (M)].Here, the existing in the prior art ripe solution of isostatic compensation repeats no more herein.

Step 202: because fixing frequency deviation causes is the linear phase deviation, the accumulated phase that residual carrier wave frequency deviation causes in a data block is very little, carrier wave frequency deviation is regarded as the fixedly photograph deviation of an accumulation to the phase effect of every blocks of data, promptly can regard carrier wave frequency deviation as a fixed value to the phase effect of each data in the data block is approximate, therefore can utilize one section known Cyclic Prefix in each block data structure of SC-FDE, the ML in the binding data aided algorithm (maximum likelihood) phase estimation algorithm obtains the phase estimation of current data block.Specifically can utilize ML phase estimation algorithm to obtain the phase estimation value of first data block, this algorithm is to obtain known Chu sequence U=[U (1) earlier ..., U (L)] and the correlation of the soft-decision data corresponding with it $\underset{k=0}{\overset{L-1}{\mathrm{\Σ}}}U{\left(k\right)}^{*}\×u_1\left(k\right)$ (k is the sequence number in the sequence, i.e. k data in i data block); Again correlation is asked argument, promptly obtain the estimation phase place of current data block $\widehat{\mathrm{\φ}}\left(1\right)=\arg (\underset{k=0}{\overset{L-1}{\mathrm{\Σ}}}U{\left(k\right)}^{*}\×u_1\left(k\right)).$ ML phase estimation algorithm in the embodiment of the invention also can select to utilize soft-decision data decoding judgement output A=(a (1) ..., a (N)) with soft-decision data z=[z (1) ..., z (N)] be correlated with and try to achieve phase estimation

But under the lower situation of signal to noise ratio, follow the trail of precision like this, equally also can utilize soft-decision data to utilize the auxiliary phase estimation algorithm of free of data, for example V﹠amp because the unreliable meeting of judgement reduces; V (A.J.Viterbi and A.M.Viterbi) algorithm carries out phase estimation, and such precision of estimation result is lower, but it is better to compare a kind of scheme effect under the low situation of signal to noise ratio.

Step 203: the estimation phase place that will obtain

Obtain φ (1) by a predefined smoother that the adds up smoothing processing that adds up, the time domain expression formula of the smoother that adds up is: $S(i+1)=\left(\begin{array}{c}S\left(i\right)+\frac{1}{i-1}\left(\underset{m=2}{\overset{i}{\mathrm{\Σ}}}\mathrm{est}\left(i\right)\right)....i\≥2\\ \mathrm{est}\left(i\right)..............................i=1\end{array}\right),$ S (1)=0 wherein, i represents the i blocks of data, est (i) is the smoother input of i piece, i.e. the previous step phase estimation value of trying to achieve S (i+1) is the level and smooth output φ (1) of adding up of i piece.Because what handled this moment is first blocks of data, what therefore try to achieve is S (2).

Original algorithm all is to estimate frequency deviation to carry out phase compensation again, consider through the residual frequency departure behind the capturing carrier and regard a fixed value as that it is that the linear phase of variable changes that the i.e. phase deviation that forms of residual frequency departure can be regarded as based on data block to the phase effect of each data in the data block is approximate; Therefore the accumulated phase α of i piece in theory _i=α ₁+ k (i-1) (k=2 π fMT is a slope) represents every theoretic accumulation variable.Therefore the accumulation smoother of design designs according to this formula, owing to be to utilize next piece of estimation phase estimation of last, so the phase estimation value of each data block $\mathrm{est}\left(i\right)=\widehat{\mathrm{\φ}}\left(i\right)$ With regard to a measured value of the slope k on the theory of correspondences, pass through smoothing filter $\tilde{k}=\frac{1}{i-1}\left(\underset{m=2}{\overset{i}{\mathrm{\Σ}}}\mathrm{est}\left(i\right)\right)$ Existing measured value asked on average reduce the influence that noise brings, the estimated value of the k that obtains more accurately

Utilize sum filter $S(i+1)=S\left(i\right)+\tilde{k}$ Can reach the purpose of carrier tracking, therefore the time domain expression formula of the smoothing filter that adds up of design is $S(i+1)=\left(\begin{array}{c}S\left(i\right)+\frac{1}{i-1}\left(\underset{m=2}{\overset{i}{\mathrm{\Σ}}}\mathrm{est}\left(i\right)\right)....i\≥2\\ \mathrm{est}\left(i\right)..............................i=1\end{array}\right).$ The accumulation function of smoother of adding up solves the accumulation of phase between the data block that frequency deviation causes, and has reached the purpose that carrier wave frequency deviation is followed the trail of; Smoothing function has solved the effect of the level and smooth accumulation of influence of the phase estimation result's shake that is caused by noise, has further improved the tracking precision.Certainly, the smoother that adds up also can have other forms, must have accumulation function and smoothing function just can achieve the goal, but the smoother performance that adds up of embodiment of the invention design is relatively good.

Forward step 200 to, with the level and smooth output φ (1) that adds up that obtains, promptly the level and smooth output s (2) that adds up in the step 203 compensates the next data block that receives ${\stackrel{\→}{r}}_2=[r\left(1\right),\·\·\·,r\left(M\right)],$ The expression formula of phase compensation is: ${\stackrel{\→}{r}}_i={\stackrel{\→}{r}}_i\×e^{-\mathrm{j\φ}\left(i\right)}.$

Data after the compensation obtain the soft-decision data of second data block correspondence again through the equilibrium treatment of step 201

Try to achieve successively

φ (2) utilizes φ (2) to go to compensate the 3rd data block, with this recursion all end of transmission blocks up to present frame again.

Here, the embodiment of the invention is that example is illustrated with the Chu sequence in the UW sequence only, but those skilled in the art of the present invention will be appreciated that, two kinds of forms of UW sequence (frank-zadaff sequence and PN sequence) go for the embodiment of the invention equally, just the size with length L has relation, be the length of frank-zadaff sequence be necessary for n square, the PN sequence is necessary for 2 n power and subtracts 1.

In sum, the embodiment of the invention provides a kind of carrier wave tracing method of the single-carrier system based on piece transmission, because what algorithm adopted is through the soft-decision data after the isostatic compensation, two characteristics utilizing equalization algorithm can compensate the stationary phase deviation and must carry out the piecemeal processing, the phase estimation result of current block is used for next pending of compensation, whole algorithm does not form real loop processed like this, but reached the effect of loop, reached the purpose of following the trail of, it is big to compare the loop time-delay that original algorithm solved in the common tracing algorithm, poor stability, the problem that precision is low.

And the accumulation of phase between the data block that the frequency deviation that solves smoothing filter that adds up of the embodiment of the invention, the accumulation function of the smoother that adds up causes has reached the purpose that carrier wave frequency deviation is followed the trail of; Smoothing function has solved the effect of the level and smooth accumulation of influence of the phase estimation result's shake that is caused by noise, has further improved the tracking precision.

The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claims.

Claims (3)

1, a kind of carrier wave tracing method of the single-carrier system based on piece transmission is characterized in that described method comprises:

Steps A: the data block through phase compensation in the present frame is carried out equilibrium treatment, and choose the soft-decision data of Cyclic Prefix correspondence the soft decision symbols data that after isostatic compensation, obtain;

Step B: utilize maximum likelihood phase estimation algorithm that described soft-decision data is calculated, obtain the phase estimation of current data block;

Step C: the estimation phase place that will the obtain smoothing processing that adds up, and the level and smooth output that will add up is used to compensate next data block as the phase estimation value of next data block;

Repeated execution of steps A is to step C, up to all end of transmission blocks of present frame.

2, method according to claim 1 is characterized in that, described step B specifically comprises:

Obtain the correlation of the known Cyclic Prefix soft-decision data corresponding earlier, then this correlation is asked argument, obtain the estimation phase place of current data block with it;

Perhaps, described step B specifically comprises:

Utilize the decoding judgement output of soft-decision data to carry out relevant estimation phase place of trying to achieve current data block with soft-decision data.

3, method according to claim 1 and 2 is characterized in that, described step C specifically comprises:

By a predefined smoother that the adds up smoothing processing that adds up, the time domain expression formula of the smoother that adds up is with the estimation phase place that obtains: $S(i+1)=\left(\begin{array}{c}S\left(i\right)+\frac{1}{i-1}\left(\underset{m=2}{\overset{i}{\mathrm{\Σ}}}\mathrm{est}\left(i\right)\right)....i\≥2\\ \mathrm{est}\left(i\right)..............................i=1\end{array}\right),$ S (1)=0 wherein, i represents the i blocks of data, est (i) is the smoother input of i piece, i.e. the previous step phase estimation value of trying to achieve

S (i+1) is the level and smooth output of adding up of i piece, and the level and smooth phase estimation value of exporting as next data block that adds up is used to compensate next data block.

Images