CN103916333A - Channel estimation method and device based on time-domain sliding window - Google Patents
Channel estimation method and device based on time-domain sliding window Download PDFInfo
- Publication number
- CN103916333A CN103916333A CN201210591397.7A CN201210591397A CN103916333A CN 103916333 A CN103916333 A CN 103916333A CN 201210591397 A CN201210591397 A CN 201210591397A CN 103916333 A CN103916333 A CN 103916333A
- Authority
- CN
- China
- Prior art keywords
- sliding window
- reference signal
- symbol
- channel
- estimation
- 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.)
- Pending
Links
Abstract
The invention belongs to the technical field of mobile communication and provides a channel estimation method based on a time-domain sliding window. The method includes the following steps: according to reference signal sign frequency domain estimation values in the time-domain sliding window, calculating time-domain estimation coefficients corresponding to subcarriers on OFDM symbol positions; according to the reference signal symbol frequency domain estimation values and the time-domain estimation coefficients, performing time-domain channel estimation on OFDM symbols without reference signals and outputting a channel estimation value; and updating the reference signal symbol frequency domain estimation values in the sliding window. The invention also provides a channel estimation device based on the time-domain sliding window. The device is capable of obtaining a higher channel estimation precision compared with a linear interpolation estimation technology based on two reference signals; the time-domain sliding window is adopted and in terms of time, channel information in the history or at a present moment is mainly used and a repeated or mixed output mode is adopted so that the estimation performance is improved and at the same time, larger processing waiting or realization difficulties are not brought; and the real-time update of the sliding window ensures self-adaption to a time-varying channel in an estimation process.
Description
Technical field
The invention belongs to mobile communication technology field, specially refer to the channel estimation methods based on time domain sliding window and device in Long Term Evolution (Long Term Evolution, LTE) system.
Background technology
At 3rd Generation Partnership Project (3rd Generation Partnership Project; 3GPP) in LTE system; conventionally can adopt least-squares estimation (Least Square; LS) to reference signal (Reference Signal; RS) channel response value of position carrier wave is estimated, supposes that the frequency-domain expression of reception reference signal is as follows:
Y
p=X
pH
p+N
p （1）
Wherein, X
prepresent to send reference signal, Y
prepresent to receive reference signal, N
prepresent to receive the noise jamming superposeing in reference signal, H
prepresent channel response value.Channel response after least-squares estimation (LS) is as follows:
Wherein,
represent H
plS estimated result.
In LTE system, the two-dimensional reference signal structure that has adopted time-frequency domain to disperse, as shown in Figure 1, each lattice represents a Resource Unit (corresponding to a subcarrier on an orthogonal frequency division multiplex OFDM symbol), on part Resource Unit, comprise reference signal (marking the lattice of R0 position in figure), the carry out channel estimating of symbol that can be by comprising pilot tone (R0) to non-pilot tone (blank space) position.
Channel estimating apparatus and channel estimation methods thereof that Chinese invention patent CN201010102955.X provides a kind of channel estimation methods for orthogonal FDM communication system and channel estimator and Chinese invention patent CN200710179922.3 that a kind of ofdm system is provided, proposed respectively to utilize Wiener filtering to carry out the method for channel estimation in frequency domain.Obtain the channel response value on whole subcarriers by LS estimated result is carried out to Wiener filtering, Wiener filtering criterion is that LS channel estimation results is carried out to a frequency domain interpolation filtering processing, obtains the locational channel estimation value of data carrier, is expressed as:
Wherein, F is Wiener filtering interpolating matrix, F=R
hH(R
hH+ σ I)
-1,
for the channel estimation in frequency domain value that adopts LS channel estimation method to obtain.
By aforementioned formula (1) ~ (3), can complete the channel estimation in frequency domain on reference signal symbol, then according to estimated result, the symbol (as the blank square in Fig. 2) that does not contain reference signal is carried out to time domain channel estimation, thereby obtain the channel estimation results on whole Resource Units, what aforementioned two sections of patent documentations proposed is time domain linear interpolation scheme, mainly utilizes two reference signal symbols adjacent in time domain to carry out time domain linear interpolation and obtains the channel estimation results on non-reference signal symbol;
As utilized the channel estimation results on B, two Resource Units of C to carry out Linear Estimation to A Resource Unit position in Fig. 2, method of estimation is as follows:
Wherein
represent respectively the channel estimation results on A, B, C Resource Unit.α
0, α
1be linear interpolation time domain estimation coefficient, be respectively the ratio of time interval length between time interval length between AC and between AB and BC.
Prior art all adopts two adjacent reference signal symbols to carry out time domain linear interpolation, and its processing delay is less, but its reference signal symbolic number adopting is less, causes performance not ideal enough.
Summary of the invention
For overcoming the above problems, the invention provides a kind of channel estimation methods and device based on time domain sliding window.The present invention is based on completing the channel estimation in frequency domain on reference signal symbol, then according to channel estimation in frequency domain result to not containing carrying out time domain channel estimation on the symbol of reference signal, and delivery channel estimated value, in the time carrying out time domain channel estimation, adopt sliding window internal reference signals symbol frequency domain estimated value to calculate time domain estimation coefficient corresponding to each subcarrier on OFDM character position, when delivery channel estimated value, the feature of considering sliding window adopts the part or all of way of output, promote channel estimating performance, finally slide in time domain, upgrade sliding window internal reference signals symbol frequency domain estimated value.
Channel estimation methods based on time domain sliding window of the present invention, comprising:
According to time domain sliding window internal reference signals symbol frequency domain estimated value, calculate time domain estimation coefficient corresponding to each subcarrier on OFDM character position;
According to described reference signal symbol frequency domain estimated value and described time domain estimation coefficient, the OFDM symbol that does not contain reference signal is carried out to time domain channel estimation, delivery channel estimated value;
Upgrade sliding window internal reference signals symbol frequency domain estimated value, repeat above step.
Channel estimating apparatus based on time domain sliding window of the present invention, comprising:
Sliding window control unit, for loading and renewal reference signal symbol frequency domain estimated value;
Estimation coefficient computing unit, according to reference signal symbol frequency domain estimated value, calculates time domain estimation coefficient corresponding to each subcarrier on OFDM character position;
Channel estimating unit, according to reference signal symbol frequency domain estimated value and time domain estimation coefficient, carries out time domain channel estimation, delivery channel estimated value to the OFDM symbol that does not contain reference signal.
Compared with prior art, the present invention proposes a kind of concrete methods of realizing and device that utilizes multiple reference signals to carry out time domain sliding window channel estimating, the linear interpolation estimation technique based on two reference signals can obtain higher precision of channel estimation; Adopt time domain sliding window, mainly utilize in time the channel information of historical and present moment, take repetition or mix output mode, in improving estimated performance, can't bring larger processing wait for and realize difficulty, the real-time update of sliding window has ensured the self adaptation to time varying channel in estimation procedure.
Brief description of the drawings
Fig. 1 is the time-frequency domain distributed architecture schematic diagram of reference signal in prior art LTE system;
Fig. 2 is the time-frequency domain distributed architecture schematic diagram that the present invention treats time domain interpolation;
Fig. 3 is the channel estimation methods embodiment schematic flow sheet that the present invention is based on time domain sliding window;
Fig. 4 is sliding window of the present invention and Output rusults example structure schematic diagram;
Fig. 5 is sliding window of the present invention and another example structure schematic diagram of Output rusults;
Fig. 6 is the channel estimating apparatus example structure schematic diagram that the present invention is based on time domain sliding window.
Embodiment
For making the present invention clearer, describe below in conjunction with accompanying drawing, the invention is not restricted to down address execution mode shown in the drawings or embodiment, all amendments of making without prejudice to spirit of the present invention and distortion, within all should being included in the scope of the invention.
Research shows, in the time that multiple adjacent reference signal symbols carry out time domain channel estimation in employing time domain, can bring larger performance gain.As shown in Figure 2, can utilize the channel estimation results on five Resource Units of E ~ I to carry out non-linear estimations to D Resource Unit position, method of estimation is as follows:
Wherein, β
0, β
1... β
4it is non-linear interpolation coefficient.
Therefore,, for OFDM symbol l, the locational Resource Unit of subcarrier k, can obtain time domain and estimate that mode is as follows:
In formula (6), M represents the reference signal resource unit number of each time domain estimated service life,
represent the time domain channel estimated result on symbol l, subcarrier k, α
irepresent time domain estimation coefficient,
represent each selected reference signal time domain subscript of estimating.
The invention provides a kind of channel estimation methods based on time domain sliding window, as an embodiment, as shown in Figure 3, comprising:
Step 301, adopt sliding window to load reference signal symbol frequency domain estimated value in time domain;
Describedly adopt sliding window to refer to that sliding window slides in time domain in time domain, wherein, described sliding window is long is M, and M span is 3 ~ 4, and sliding step is N, and the span of N is that N is 1 ~ M.
Described sliding window length refers to and can be used for the adjacent reference signal symbolic number that time domain is estimated, illustrate, as shown in Figure 2, OFDM symbol 0,4,7,11,14,18(shadow positions) comprise reference signal, if selecting sliding window length is 4, available continuous 4 reference signals, 0,4,7,11 symbols carry out interpolation to other sign resources unit of 0 ~ 11 of OFDM symbol.
Step 302, calculate time domain estimation coefficient corresponding to each subcarrier on OFDM character position according to sliding window internal reference signals symbol frequency domain estimated value;
Described according to time domain estimation coefficient corresponding to each subcarrier on reference signal symbol frequency domain estimated value calculating OFDM character position, can adopt in one of mode:
Mode one, employing multinomial criterion generate coefficient
The interpolation formula that multinomial criterion provides is as follows:
Wherein polynomial interpolation coefficient producing method is as follows, obtains according to the time interval of adjacent 3 reference signal symbols:
Wherein, a represents data symbol and the intersymbol time interval of reference signal; α
0, α
1, α
2represent 3 adjacent polynomial interpolation coefficients that reference signal symbol is corresponding in time domain.
Mode two, employing minimum mean square error criterion generate coefficient
Wherein, MMSE Interpolation coefficient producing method is as follows:
1) obtain the intersymbol auto-correlation coefficient of each reference signal in sliding window
cross-correlation coefficient with reference signal symbol and other symbols
2) obtain channel noise power measurement result σ;
3) basis computing formula below, generates the interpolation coefficient that each reference signal symbol is corresponding:
Wherein, α
krepresent in sliding window k the MMSE Interpolation coefficient that reference signal symbol is corresponding.
According to least mean-square estimate criterion, in the estimated result performance of acquisition, there is minimum estimation mean square error, compared with Linear Estimation and Polynomial Estimation, more excellent on estimated performance; Meanwhile, in least mean-square estimate coefficient generative process, need to utilize channel relevancy and channel noise power measurement result, so to thering is adaptivity with time varying channel; Meanwhile, the manner is applicable to the long setting of variable sliding window, and can select the strongest reference signal symbol of correlation calculate according to the channel relevancy obtaining, and can ensure estimated performance optimum.
The present invention adopts with one of upper type, will adopt different estimation coefficients for each sliding window, when specific implementation, and the estimation coefficient of the each sliding window of real-time update.
Special, if sliding window is long and when all identical with reference to the letter time interval, and adopt unified coefficient generating mode, each sliding window can use identical estimation coefficient and not upgrade.Illustrate, for example, when sliding window is fixing, to adopt 4 reference signal symbols, sliding window to upgrade symbolic number be 1 while carrying out channel estimating, as shown in Figure 2, sliding window symbolization 0,4,7,11 for the first time, the time interval of interior each reference signal symbol of sliding window is respectively 3,2,3 symbols for the first time; Sliding window symbolization 4,7,11,14 for the second time, the time interval of interior each reference signal symbol of sliding window is respectively 2,3,2 symbols for the second time; Sliding window symbolization 7,11,14,18 for the third time, the time interval of interior each reference signal symbol of sliding window is respectively 3,2,3 symbols for the third time; Comprise reference signal symbolic number in the sliding window of sliding window and sliding window for the third time for the first time identical, and the time interval of reference signal is also identical, the coefficient generating mode that if sliding window adopts with sliding window is identical for the first time for the third time, sliding window can directly adopt the estimation coefficient of sliding window for the first time for the third time, and need not remove to calculate estimation coefficient again.
Step 303, according to reference signal symbol frequency domain estimated value and time domain estimation coefficient, to not carrying out time domain channel estimation, delivery channel estimated value containing the OFDM symbol of reference signal.
That is to say, calculate the channel response value of all symbols of sliding window according to the channel estimation results of reference signal, also can only select part symbol wherein to calculate; Preferably, according to channel relevancy principle, upper interval of time symbol far away, relevant less; So it is more excellent to calculate channel response value performance on sliding window mid portion character position at every turn.
Carry out time domain channel and estimate to calculate, according to formula (6) interpolation method, use the reference signal in same sub-carrier positions to carry out time domain interpolation to each Resource Unit, no longer describe in detail.
Described delivery channel estimated value, the present invention can adopt one of following output mode:
Mode one: non-repetition output mode
Described non-repetition output mode is, sliding window output and all channel estimation values of different symbols of adjacent sliding window before and after it, illustrate, as shown in Figure 4, suppose that sliding window length is 4, to upgrade symbolic number be 1 to sliding window, in the 1st sliding window, symbol the 0,4,7, the 11st, the reference signal symbol in sliding window, included 0,1,2,3,4 outputs that are the 1st sliding window within the scope of arrow line in figure; In the 2nd sliding window, 4,7,11,14 symbols are the reference signal symbols in sliding window, and within the scope of arrow line, included 5,6,7,8,9 is the output of the 2nd sliding window; In like manner, symbol 10,11,12,13,14,15 is the output of the 3rd sliding window; Can see the output symbol that the 1st, 2,3 sliding windows all do not repeat.
The OFDM symbol at the end that preferably, the symbol of each sliding window output is this sliding window.
The present embodiment could be exported after all sign computation owing to waiting in window, may have larger output time delay.
Mode two: repeat to export output mode
Describedly repeat to export output mode and be, sliding window is exported the channel estimation value of the symbol identical with adjacent sliding window part before and after it, as shown in Figure 5, suppose that sliding window length is 4, to upgrade symbolic number be 1 to sliding window, in the 1st sliding window, symbol the 0,4,7, the 11st, the reference signal symbol in sliding window, included 0,1,2,3,4 outputs that are the 1st sliding window within the scope of arrow line in figure; In the 2nd sliding window, 4,7,11,14 symbols are the reference signal symbols in sliding window, and within the scope of arrow line, included 4,5,6,7,8,9 is the output of the 2nd sliding window; In like manner, symbol 7,8,9,10,11,12 is the output of the 3rd sliding window; Can see, the 1st, 2 sliding windows have all been exported symbol 4 simultaneously; The 2nd sliding window, 3 has all been exported symbol 7,8,9 simultaneously.
The present embodiment repeats output can reduce last time the time of waiting for output, reduces to a certain extent output time delay, but front and back repeat output information and underuse, and output performance is affected.
Mode three: mix output mode
On mode two bases, the Output rusults in two sliding window same-sign is weighted to merging:
ω
0, ω
1represent weight coefficient,
represent respectively the estimated result on the same position symbol of two adjacent windows output.
Before and after the present embodiment utilizes, repeat output information, reduced output time delay, the performance gain that has brought again multiple symbol interpolating to bring.
Step 304, time area update sliding window, load reference signal symbol frequency domain estimated value, repeat above step 302-304;
As each reference signal symbol estimated result in last time sliding window is:
this sliding window internal reference signals channel estimation results after renewal is:
the channel estimation results of N reference signal symbol being above about in sliding window last time removes, and increases the channel estimation results of N subsequent reference signal simultaneously.Load the channel estimation in frequency domain result on reference signal character position, to the channel estimation in frequency domain result of each reference signal
press the input of OFDM symbol sequential order,
represent the channel estimation results on the whole subcarriers on m reference signal symbol.
Described renewal sliding window comprises, upgrades the interior part reference signal symbol frequency domain estimated value of sliding window or upgrades whole reference signal symbol frequency domain estimated values.
In described renewal sliding window, part reference signal symbol frequency domain estimated value comprises: according to reference signal symbol time order, remove one or more reference signal symbols the earliest on the time, add follow-up one or more reference signal symbols simultaneously.As shown in Figure 4: the 1st sliding window and the 2nd sliding window, window length is set to 4, upgrades reference signal symbol numbers is 1, in the 1st sliding window, comprise the reference signal symbol 0 that removes in reference signal symbol 0,4,7,11, the 2 sliding windows on the time the earliest, add follow-up reference signal symbol 14 simultaneously.
The whole reference signal symbol of described renewal frequency domain estimated value comprises: by reference signal time sequencing, remove the whole reference signal symbols in this sliding window, add the reference signal of the long whole sliding windows next time that arrange of follow-up window simultaneously.As shown in Figure 5, for example as for the first time sliding window window length is 4, comprises reference signal symbol 0,4,7,11; Next time, sliding window window length was 4, comprised reference signal symbol 14,18,21,25 after update all.
Preferably, area update sliding window when described, may first upgrade sliding window long, according to the long definite reference signal symbol frequency domain estimated value quantity that needs renewal of new window, loads reference signal symbol frequency domain estimated value.
Complete on the OFDM symbol of selected output of this sliding window time domain scope all channel estimating of Resource Units, upgrade after sliding window, get back to step 302 and carry out the channel estimating in next sliding window.
Compared with prior art, the present invention adopts sliding window to load reference signal symbol frequency domain estimated value, then calculate time domain estimation coefficient corresponding to each subcarrier on OFDM character position according to the reference signal symbol frequency domain estimated value loading, adopt multiple time domain samples points, more only adopted the linear interpolation scheme of two reference signal symbols can obtain better estimation effect; According to reference signal symbol frequency domain estimated value and time domain estimation coefficient, the OFDM symbol that does not contain reference signal is carried out to time domain channel estimation, delivery channel estimated value, not only avoided reducing time delay but also promoting the performance gain that multiple symbol interpolating are brought, after the channel estimation calculation of all symbols in this sliding window completes, upgrade sliding window, slide by sliding step N, slide and load reference signal symbol frequency domain estimated value, repeat above process.
Compared with prior art, the present invention proposes a kind of concrete methods of realizing that utilizes multiple reference signals to carry out time domain sliding window channel estimating, have the linear interpolation estimation technique based on two reference signals can obtain higher precision of channel estimation; Adopt time domain sliding window, mainly utilize in time the channel information of historical and present moment, in improving estimated performance, can't bring larger processing wait for and realize difficulty, the real-time update of sliding window has ensured the self adaptation to time varying channel in estimation procedure.
The present invention is based on the channel estimating apparatus of time domain sliding window, as shown in Figure 6, comprising:
Sliding window control unit, for loading and renewal reference signal symbol frequency domain estimated value;
Describedly adopt sliding window to refer to that sliding window slides in time domain in time domain, wherein, described sliding window is long is M, and M span is 3 ~ 4, and sliding step is N, and the span of N is that N is 1 ~ M.
Described sliding window length refers to and can be used for the adjacent reference signal symbolic number that time domain is estimated, illustrate, as shown in Figure 2, OFDM symbol 0,4,7,11,14,18(shadow positions) comprise reference signal, if selecting sliding window length is 4, available continuous 4 reference signals, 0,4,7,11 symbols carry out interpolation to other sign resources unit of 0 ~ 11 of OFDM symbol.
As each reference signal symbol estimated result in last time sliding window is:
this sliding window internal reference signals channel estimation results after renewal is:
the channel estimation results of N reference signal symbol being above about in sliding window last time removes, and increases the channel estimation results of N subsequent reference signal simultaneously.Load the channel estimation in frequency domain result on reference signal character position, to the channel estimation in frequency domain result of each reference signal
press the input of OFDM symbol sequential order,
represent the channel estimation results on the whole subcarriers on m reference signal symbol.
Described renewal sliding window comprises, upgrades the interior part reference signal symbol frequency domain estimated value of sliding window or upgrades whole reference signal symbol frequency domain estimated values.
In described renewal sliding window, part reference signal symbol frequency domain estimated value comprises: according to reference signal symbol time order, remove one or more reference signal symbols the earliest on the time, add follow-up one or more reference signal symbols simultaneously.As shown in Figure 4: the 1st sliding window and the 2nd sliding window, window length is set to 4, upgrades reference signal symbol numbers is 1, in the 1st sliding window, comprise the reference signal symbol 0 that removes in reference signal symbol 0,4,7,11, the 2 sliding windows on the time the earliest, add follow-up reference signal symbol 14 simultaneously.
The whole reference signal symbol of described renewal frequency domain estimated value comprises: by reference signal time sequencing, remove the whole reference signal symbols in this sliding window, add the reference signal of the long whole sliding windows next time that arrange of follow-up window simultaneously.As shown in Figure 5, for example as for the first time sliding window window length is 4, comprises reference signal symbol 0,4,7,11; Next time, sliding window window length was 4, comprised reference signal symbol 14,18,21,25 after update all.
Preferably, area update sliding window when described, may first upgrade sliding window long, according to the long definite reference signal symbol frequency domain estimated value quantity that needs renewal of new window, loads reference signal symbol frequency domain estimated value.
Complete on the OFDM symbol of selected output of this sliding window time domain scope all channel estimating of Resource Units, upgrade after sliding window, get back to step 302 and carry out the channel estimating in next sliding window.
Estimation coefficient computing unit, according to time domain sliding window internal reference signals symbol frequency domain estimated value, calculates time domain estimation coefficient corresponding to each subcarrier on OFDM character position
Described according to time domain estimation coefficient corresponding to each subcarrier on reference signal symbol frequency domain estimated value calculating OFDM character position, can adopt in one of mode:
Mode one, employing multinomial criterion generate coefficient
The interpolation formula that multinomial criterion provides is as follows:
Wherein polynomial interpolation coefficient producing method is as follows, obtains according to the time interval of adjacent 3 reference signal symbols:
Wherein, a represents data symbol and the intersymbol time interval of reference signal; α
0, α
1, α
2represent 3 adjacent polynomial interpolation coefficients that reference signal symbol is corresponding in time domain; K represents reference signal symbol subscript, and L is the mark space of adjacent two reference signals.
Mode two, employing minimum mean square error criterion generate coefficient
Wherein, MMSE Interpolation coefficient producing method is as follows:
1) obtain the intersymbol auto-correlation coefficient of each reference signal in sliding window
cross-correlation coefficient with reference signal symbol and other symbols
2) obtain channel noise power measurement result σ;
3) basis computing formula below, generates the interpolation coefficient that each reference signal symbol is corresponding:
Wherein, α
krepresent in sliding window k the MMSE Interpolation coefficient that reference signal symbol is corresponding.
According to least mean-square estimate criterion, in the estimated result performance of acquisition, there is minimum estimation mean square error, compared with Linear Estimation and Polynomial Estimation, more excellent on estimated performance; Meanwhile, in least mean-square estimate coefficient generative process, need to utilize channel relevancy and channel noise power measurement result, so to thering is adaptivity with time varying channel; Meanwhile, the manner is applicable to the long setting of variable sliding window, and can select the strongest reference signal symbol of correlation calculate according to the channel relevancy obtaining, and can ensure estimated performance optimum.
The present invention adopts with one of upper type, will adopt different estimation coefficients for each sliding window, when specific implementation, and the estimation coefficient of the each sliding window of real-time update.
Special, if sliding window is long and when all identical with reference to the letter time interval, and adopt unified coefficient generating mode, each sliding window can use identical estimation coefficient and not upgrade.Illustrate, for example, when sliding window is fixing, to adopt 4 reference signal symbols, sliding window to upgrade symbolic number be 1 while carrying out channel estimating, as shown in Figure 2, sliding window symbolization 0,4,7,11 for the first time, the time interval of interior each reference signal symbol of sliding window is respectively 3,2,3 symbols for the first time; Sliding window symbolization 4,7,11,14 for the second time, the time interval of interior each reference signal symbol of sliding window is respectively 2,3,2 symbols for the second time; Sliding window symbolization 7,11,14,18 for the third time, the time interval of interior each reference signal symbol of sliding window is respectively 3,2,3 symbols for the third time; Comprise reference signal symbolic number in the sliding window of sliding window and sliding window for the third time for the first time identical, and the time interval of reference signal is also identical, the coefficient generating mode that if sliding window adopts with sliding window is identical for the first time for the third time, sliding window can directly adopt the estimation coefficient of sliding window for the first time for the third time, and need not remove to calculate estimation coefficient again.
Channel estimating unit, according to reference signal symbol frequency domain estimated value and time domain estimation coefficient, carries out time domain channel estimation, delivery channel estimated value to the OFDM symbol that does not contain reference signal;
That is to say, calculate the channel response value of all symbols of sliding window according to the channel estimation results of reference signal, also can only select part symbol wherein to calculate; Preferably, according to channel relevancy principle, upper interval of time symbol far away, relevant less; So it is more excellent to calculate channel response value performance on sliding window mid portion character position at every turn.
Carry out time domain channel and estimate to calculate, according to formula (6) interpolation method, use the reference signal in same sub-carrier positions to carry out time domain interpolation to each Resource Unit, no longer describe in detail.
Described delivery channel estimated value, the present invention can adopt one of following output mode:
Mode one: non-repetition output mode
Described non-repetition output mode is, sliding window output and all channel estimation values of different symbols of adjacent sliding window before and after it, illustrate, as shown in Figure 4, suppose that sliding window length is 4, to upgrade symbolic number be 1 to sliding window, in the 1st sliding window, symbol the 0,4,7, the 11st, the reference signal symbol in sliding window, included 0,1,2,3,4 outputs that are the 1st sliding window within the scope of arrow line in figure; In the 2nd sliding window, 4,7,11,14 symbols are the reference signal symbols in sliding window, and within the scope of arrow line, included 5,6,7,8,9 is the output of the 2nd sliding window; In like manner, symbol 10,11,12,13,14,15 is the output of the 3rd sliding window; Can see the output symbol that the 1st, 2,3 sliding windows all do not repeat.
The OFDM symbol at the end that preferably, the symbol of each sliding window output is this sliding window.
The present embodiment could be exported after all sign computation owing to waiting in window, may have larger output time delay.
Mode two: repeat to export output mode
Describedly repeat to export output mode and be, sliding window is exported the channel estimation value of the symbol identical with adjacent sliding window part before and after it, as shown in Figure 5, suppose that sliding window length is 4, to upgrade symbolic number be 1 to sliding window, in the 1st sliding window, symbol the 0,4,7, the 11st, the reference signal symbol in sliding window, included 0,1,2,3,4 outputs that are the 1st sliding window within the scope of arrow line in figure; In the 2nd sliding window, 4,7,11,14 symbols are the reference signal symbols in sliding window, and within the scope of arrow line, included 4,5,6,7,8,9 is the output of the 2nd sliding window; In like manner, symbol 7,8,9,10,11,12 is the output of the 3rd sliding window; Can see, the 1st, 2 sliding windows have all been exported symbol 4 simultaneously; The 2nd sliding window, 3 has all been exported symbol 7,8,9 simultaneously.
The present embodiment repeats output can reduce last time the time of waiting for output, reduces to a certain extent output time delay, but front and back repeat output information and underuse, and output performance is affected.
Mode three: mix output mode
On mode two bases, the Output rusults in two sliding window same-sign is weighted to merging:
ω
0, ω
1represent weight coefficient,
represent respectively the estimated result on the same position symbol of two adjacent windows output.Described two sliding windows refer to the previous sliding window of current sliding window and current sliding window.
Before and after the present embodiment utilizes, repeat output information, reduced output time delay, the performance gain that has brought again multiple symbol interpolating to bring.
Compared with prior art, the present invention adopts sliding window control unit to load reference signal symbol frequency domain estimated value, then estimation coefficient computing unit calculates time domain estimation coefficient corresponding to each subcarrier on OFDM character position according to the reference signal symbol frequency domain estimated value loading, adopt multiple time domain samples points, more only adopted the linear interpolation scheme of two reference signal symbols can obtain better estimation effect; Channel estimating unit is according to reference signal symbol frequency domain estimated value and time domain estimation coefficient, the OFDM symbol that does not contain reference signal is carried out to time domain channel estimation, delivery channel estimated value, not only avoided reducing time delay but also promoting the performance gain that multiple symbol interpolating are brought, after the channel estimation calculation of all symbols in this sliding window completes, sliding window control unit upgrades sliding window, slides by sliding step, slides and loads reference signal symbol frequency domain estimated value.
The present invention has carried out further detailed description for execution mode or embodiment to the object, technical solutions and advantages of the present invention; institute is understood that; above lifted execution mode or embodiment are only the preferred embodiment of the present invention; not in order to limit the present invention; all any amendments made for the present invention within the spirit and principles in the present invention, be equal to replacement, improvement etc., within protection scope of the present invention all should be included in.
Claims (17)
1. the channel estimation methods based on time domain sliding window, is characterized in that, comprising:
According to time domain sliding window internal reference signals symbol frequency domain estimated value, calculate time domain estimation coefficient corresponding to each subcarrier on orthogonal frequency division multiplex OFDM character position;
According to described reference signal symbol frequency domain estimated value and described time domain estimation coefficient, the OFDM symbol that does not contain reference signal is carried out to time domain channel estimation, delivery channel estimated value;
Upgrade sliding window internal reference signals symbol frequency domain estimated value, repeat above step.
2. channel estimation methods according to claim 1, it is characterized in that, described according to time domain estimation coefficient corresponding to each subcarrier on time domain sliding window internal reference signals symbol frequency domain estimated value calculating OFDM character position, comprise: adopt multinomial criterion to generate coefficient, obtain according to the time interval of adjacent 3 reference signal symbols:
Wherein, a represents data symbol and the intersymbol time interval of reference signal; α
0, α
1, α
2represent adjacent polynomial interopolation time domain estimation coefficient corresponding to 3 reference signal symbols in time domain.
3. channel estimation methods according to claim 1, it is characterized in that, described according to time domain estimation coefficient corresponding to each subcarrier on time domain sliding window internal reference signals symbol frequency domain estimated value calculating OFDM character position, comprise that employing minimum mean square error criterion generates coefficient, comprising:
1) obtain the intersymbol auto-correlation coefficient of each reference signal in sliding window
cross-correlation coefficient with reference signal symbol and other symbols
2) obtain channel noise power measurement result σ;
3) basis computing formula below, generates the interpolation coefficient that each reference signal symbol is corresponding:
Wherein, α
krepresent in sliding window k the MMSE Interpolation time domain estimation coefficient coefficient that reference signal symbol is corresponding.
4. channel estimation methods according to claim 1, is characterized in that, described delivery channel estimated value, comprising: sliding window output and all channel estimation values of different symbols of its adjacent sliding window in front and back.
5. channel estimation methods according to claim 4, is characterized in that, described sliding window output and adjacent sliding window before and after it all channel estimation value of different symbols are the OFDM symbol at symbol that sliding window the is exported end that is this sliding window.
6. channel estimation methods according to claim 1, is characterized in that, described delivery channel estimated value, comprising: the channel estimation value of the sliding window output symbol identical with its adjacent sliding window part in front and back.
7. channel estimation methods according to claim 6, is characterized in that, described delivery channel estimated value, comprising: the Output rusults in two sliding window same-sign is weighted to merging,
wherein, ω
0, ω
1represent weight coefficient,
represent the estimated result on the same position symbol of two adjacent windows output.
8. channel estimation methods according to claim 1, is characterized in that, described renewal sliding window comprises that first to upgrade sliding window long, according to new window long determine need the reference signal symbol frequency domain estimated value quantity upgraded.
9. according to channel estimation methods described in claim 1 or 8, it is characterized in that, described renewal sliding window comprises some or all of reference signal symbol frequency domain estimated value in renewal sliding window.
10. the channel estimating apparatus based on time domain sliding window, is characterized in that, comprising:
Sliding window control unit, for loading and renewal reference signal symbol frequency domain estimated value;
Estimation coefficient computing unit, according to reference signal symbol frequency domain estimated value, calculates time domain estimation coefficient corresponding to each subcarrier on orthogonal frequency division multiplex OFDM character position;
Channel estimating unit, according to reference signal symbol frequency domain estimated value and time domain estimation coefficient, carries out time domain channel estimation, delivery channel estimated value to the OFDM symbol that does not contain reference signal.
11. channel estimating apparatus according to claim 10, it is characterized in that, described according to time domain estimation coefficient corresponding to each subcarrier on time domain sliding window internal reference signals symbol frequency domain estimated value calculating OFDM character position, comprise: adopt multinomial criterion to generate coefficient, obtain according to the time interval of adjacent 3 reference signal symbols:
Wherein, a represents data symbol and the intersymbol time interval of reference signal; α
0, α
1, α
2represent adjacent polynomial interopolation time domain estimation coefficient coefficient corresponding to 3 reference signal symbols in time domain.
12. channel estimating apparatus according to claim 10, it is characterized in that, described according to time domain estimation coefficient corresponding to each subcarrier on time domain sliding window internal reference signals symbol frequency domain estimated value calculating OFDM character position, comprise that employing minimum mean square error criterion generates coefficient, comprising:
1) obtain the intersymbol auto-correlation coefficient of each reference signal in sliding window
cross-correlation coefficient with reference signal symbol and other symbols
2) obtain channel noise power measurement result σ;
3) basis computing formula below, generates the interpolation coefficient that each reference signal symbol is corresponding:
Wherein, α
krepresent in sliding window k the MMSE Interpolation time domain estimation coefficient coefficient that reference signal symbol is corresponding.
13. channel estimating apparatus according to claim 10, is characterized in that, described delivery channel estimated value, comprising: sliding window output and all channel estimation values of different symbols of its adjacent sliding window in front and back.
14. according to channel estimating apparatus described in claim 13, it is characterized in that, the output of described sliding window and adjacent sliding window before and after it all channel estimation value of different symbols are the OFDM symbol at symbol that each sliding window the is exported end that is this sliding window.
15. channel estimating apparatus according to claim 10, is characterized in that, described delivery channel estimated value, comprising: the channel estimation value of the sliding window output symbol identical with its adjacent sliding window part in front and back.
16. according to channel estimating apparatus described in claim 15, it is characterized in that, described delivery channel estimated value, comprising: the Output rusults in two sliding window same-sign is weighted to merging,
wherein, ω
0, ω
1represent weight coefficient,
represent the estimated result on the same position symbol of two adjacent windows output.
17. channel estimating apparatus according to claim 10, is characterized in that, described renewal sliding window comprises that first to upgrade sliding window long, according to new window long determine need the reference signal symbol frequency domain estimated value quantity upgraded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210591397.7A CN103916333A (en) | 2012-12-30 | 2012-12-30 | Channel estimation method and device based on time-domain sliding window |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210591397.7A CN103916333A (en) | 2012-12-30 | 2012-12-30 | Channel estimation method and device based on time-domain sliding window |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103916333A true CN103916333A (en) | 2014-07-09 |
Family
ID=51041742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210591397.7A Pending CN103916333A (en) | 2012-12-30 | 2012-12-30 | Channel estimation method and device based on time-domain sliding window |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103916333A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104243390A (en) * | 2014-10-15 | 2014-12-24 | 华南师范大学 | Sliding window-based estimation compensation method for CO-OFDM optical phase noise |
| CN108075992A (en) * | 2016-11-09 | 2018-05-25 | 电信科学技术研究院 | A kind of channel estimation methods and device |
| CN110808930A (en) * | 2019-11-06 | 2020-02-18 | 紫光展锐(重庆)科技有限公司 | Channel estimation method, device and storage medium |
| CN113676274A (en) * | 2020-05-15 | 2021-11-19 | 大唐移动通信设备有限公司 | Signal time delay estimation method, device, equipment and storage medium |
| CN114338327A (en) * | 2022-01-05 | 2022-04-12 | 电子科技大学 | Synchronous capture method of short-wave multi-carrier signal |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271102A (en) * | 2008-05-06 | 2008-09-24 | 青岛理工大学 | Cement base material direct pulling stress corrosion instrument |
| CN101827057A (en) * | 2010-01-28 | 2010-09-08 | 北京天碁科技有限公司 | Channel estimation method and channel estimator for orthogonal frequency division multiplexing (OFDM ) communication system |
-
2012
- 2012-12-30 CN CN201210591397.7A patent/CN103916333A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271102A (en) * | 2008-05-06 | 2008-09-24 | 青岛理工大学 | Cement base material direct pulling stress corrosion instrument |
| CN101827057A (en) * | 2010-01-28 | 2010-09-08 | 北京天碁科技有限公司 | Channel estimation method and channel estimator for orthogonal frequency division multiplexing (OFDM ) communication system |
Non-Patent Citations (1)
| Title |
|---|
| 江丽娟: "MIMO-OFDM系统信道估计技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104243390A (en) * | 2014-10-15 | 2014-12-24 | 华南师范大学 | Sliding window-based estimation compensation method for CO-OFDM optical phase noise |
| CN104243390B (en) * | 2014-10-15 | 2017-11-03 | 华南师范大学 | A kind of CO OFDM light phase noise estimation compensation process based on sliding window |
| CN108075992A (en) * | 2016-11-09 | 2018-05-25 | 电信科学技术研究院 | A kind of channel estimation methods and device |
| CN108075992B (en) * | 2016-11-09 | 2020-05-15 | 电信科学技术研究院 | Channel estimation method and device |
| CN110808930A (en) * | 2019-11-06 | 2020-02-18 | 紫光展锐(重庆)科技有限公司 | Channel estimation method, device and storage medium |
| CN110808930B (en) * | 2019-11-06 | 2022-06-07 | 紫光展锐(重庆)科技有限公司 | Channel estimation method, device and storage medium |
| CN113676274A (en) * | 2020-05-15 | 2021-11-19 | 大唐移动通信设备有限公司 | Signal time delay estimation method, device, equipment and storage medium |
| CN113676274B (en) * | 2020-05-15 | 2022-09-13 | 大唐移动通信设备有限公司 | Signal time delay estimation method, device, equipment and storage medium |
| CN114338327A (en) * | 2022-01-05 | 2022-04-12 | 电子科技大学 | Synchronous capture method of short-wave multi-carrier signal |
| CN114338327B (en) * | 2022-01-05 | 2023-03-14 | 电子科技大学 | Synchronous capture method of short-wave multi-carrier signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102724147B (en) | A kind of channel estimation methods of underwater sound OFDM | |
| US8548101B2 (en) | Demodulation method and demodulator for orthogonal frequency multiplexing—multiple input multiple output system | |
| CN103873397B (en) | A kind of new joint time domain and frequency domain OFDM receive channel estimation methods | |
| CN103916333A (en) | Channel estimation method and device based on time-domain sliding window | |
| CN102123127B (en) | Coefficient calculation method and coefficient calculation device of time-domain channel estimation filter | |
| CN101815042B (en) | Orthogonal frequency division multiplexing (OFDM) system channel estimation method and device | |
| CN100521665C (en) | Iterative decomposition method for fixed training sequence stuffing modulation system | |
| CN102571647B (en) | Frequency offset estimation method and device | |
| CN101827057A (en) | Channel estimation method and channel estimator for orthogonal frequency division multiplexing (OFDM ) communication system | |
| CN102185803B (en) | Channel estimation method under high-speed mobile environment | |
| Attar et al. | Efficient equalisers for OFDM and DFrFT-OCDM multicarrier systems in mobile E-health video broadcasting with machine learning perspectives | |
| CN103051571A (en) | Doppler shift estimation method for TD-LTE (Time Division Long-Term Evolution) system | |
| CN103595679A (en) | Method for reducing peak-to-average ratio of LTE uplink single-carrier frequency-division multiple access signal | |
| CN103220242B (en) | Based on the channel estimation methods of pilot blocks in single-carrier frequency domain equalization system | |
| CN103338168A (en) | Iteration time domain MMSE (minimum mean square error) equilibrium method based on weighted-type fractional Fourier transform (WFRFT) in doubly dispersive channel | |
| CN103281272B (en) | Based on the ofdm system signal detecting method of BEM under Cyclic Prefix disappearance | |
| CN100477651C (en) | High-performance OFDM channel estimation method based on combined pilot | |
| CN102413080B (en) | Method for estimating channel in high-speed moving TDD-LTE (time division duplex-long time evolution) uplink | |
| CN102780656A (en) | Method and device for eliminating multi-symbol subcarrier jamming and performing channel estimation jointly | |
| CN101299735A (en) | Method and system for estimating carrier frequency migration | |
| CN104243366A (en) | Frequency offset estimation method and device on condition of multiuser access | |
| CN104639472B (en) | Uplink multi-users mimo channel method of estimation | |
| CN103188053A (en) | Signal detection method for space time block code (STBC)-orthogonal frequency division multiplexing (OFDM) system under condition of lack of cyclic prefix | |
| CN107968760A (en) | Filter a kind of receiving algorithm based on iterative channel estimation in multitone modulating system | |
| CN101510858A (en) | Channel long-range forecast method based on slope correction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140709 |
|
| RJ01 | Rejection of invention patent application after publication |