CN102447655B - The channel estimation methods of up PUSC and base station - Google Patents

Abstract

The invention discloses channel estimation methods and the base station of a kind of up PUSC, wherein, this channel estimation methods comprises: the channel estimation value of the pilot sub-carrier in the sheet (Tile) in up PUSC is determined in base station; Base station obtains the filtering matrix of Tile; Base station is used the channel estimation value of pilot sub-carrier and the filtering matrix of Tile, obtains the channel estimation value of the data subcarrier in Tile. The present invention has reduced amount of calculation, and has kept higher channel estimating performance, and the amount of calculation that has solved prior art existence is larger, the higher problem of difficulty of realization.

Description

The channel estimation methods of up PUSC and base station

Technical field

The present invention relates to wireless communication field, use subcarrier (Partially in particular to a kind of ascenderUsedSub-Carrier, PUSC) channel estimation methods and base station.

Background technology

In wireless communication system, wireless channel also can be predicted due to fixing unlike wire message way, therefore, and channel radioCommunication system can be subject to the shadow fading of wireless channel and the impact of frequency selective fading, and wireless OFDM wherein(OrthogonalFrequencyDivisionMultiplexing, OFDM) communication system is subject to above-mentioned two kinds to declineThe impact falling is more serious, thereby must take certain method to reduce the impact of wireless channel as far as possible. At present, the main letter that adoptsRoad estimation technique reduces the impact of wireless channel, and this has just proposed very large to the channel estimation technique of wireless OFDM communication systemChallenge, the quality of channel estimating will directly have influence on the performance quality of whole system.

At present, channel estimation methods mainly contains MMSE (MinimumMeanSquaredError, least mean-square error) letterChannel estimation method. Although MMSE channel estimation methods can obtain reasonable channel estimating effect,, because the method needsKnow channel coefficients and the signal to noise ratio of intercarrier, so just cause amount of calculation larger, increased the difficulty realizing.

Summary of the invention

Main purpose of the present invention is to provide channel estimation methods and the base station of a kind of up PUSC, at least to solveThe amount of calculation of stating is larger, the higher problem of difficulty of realization.

According to an aspect of the present invention, provide the channel estimation methods of a kind of up PUSC, having comprised: on base station is determinedThe channel estimation value of the pilot sub-carrier in the sheet (Tile) in row PUSC; Base station obtains the filtering matrix of Tile; Base station is usedThe channel estimation value of pilot sub-carrier and the filtering matrix of Tile, obtain the channel estimation value of the data subcarrier in Tile.

Further, the filtering matrix that base station obtains Tile comprises: the time correlation between pilot sub-carrier is calculated in base stationCoefficient and frequency correlation coefficient; Base station coefficient correlation service time and frequency correlation coefficient obtain maximum doppler frequency and all squareTime delay; Base station is used the signal to noise ratio, maximum doppler frequency of Tile and equal square time delays, obtains filtering matrix.

Further, base station coefficient correlation service time and frequency correlation coefficient obtain maximum doppler frequency and equal whens sideProlong and comprise: obtain maximum doppler frequency v according to following formula_max：R_H(Δn)＝J₀(2πN_sΔnv_max); Wherein, R_H(Δn)For the time correlation coefficient between the pilot sub-carrier of the Δ n symbol of being separated by, N_sFor the time interval between adjacent symbol, ΔN is positive integer, J₀(.) is first kind zeroth order Bezier Bessel function.

Further, base station coefficient correlation service time and frequency correlation coefficient obtain maximum doppler frequency and equal whens sideAlso prolong and comprise: obtain all square time delay σ according to following formula_τ：Wherein, R_H(Δ is l) the Δ l of being separated byFrequency correlation coefficient between the pilot sub-carrier of individual carrier wave, L is the frequency interval between adjacent carrier wave, Δ l is positive integer.

Further, base station is used the signal to noise ratio, maximum doppler frequency of Tile and equal square time delays, obtains filtering matrix bagDraw together: calculate filtering matrix W according to following formula:

Wherein, R_hpBetween data subcarrier and pilot sub-carrier in TileCorrelation matrix, R_SiPjFor the coefficient correlation between data subcarrier i and pilot sub-carrier j, R_ppFor pilot tone in Tile carriesCorrelation matrix between ripple and pilot sub-carrier, R_PkPmFor the coefficient correlation between pilot sub-carrier k and pilot sub-carrier m, i,J, k, m is natural number, R_H(Δ n, Δ is l) for pilot sub-carrier and pilot tone of be separated by a Δ n symbol and the Δ l carrier wave of being separated by carryCoefficient correlation between ripple or between pilot sub-carrier and data subcarrier,For the inverse of the signal to noise ratio of Tile.

Further, base station is used the signal to noise ratio, maximum doppler frequency of Tile and equal square time delays, obtains filtering matrix bagDraw together: base station, from local mapping table, finds and signal to noise ratio, maximum doppler frequency and all corresponding filters of square time delayRipple matrix, wherein, corresponding relation table record signal to noise ratio, maximum doppler frequency and equal square time delay three and filtering matrix rightShould be related to.

Further, base station is used the channel estimation value of pilot sub-carrier and the filtering matrix of Tile, obtains in TileThe channel estimation value of data subcarrier comprise: the channel estimation value that obtains the data subcarrier in Tile according to following formula:Wherein,For the matrix that the channel estimation value of the pilot sub-carrier in Tile forms, the filtering square that W is TileBattle array, H_lmmseFor the matrix of the channel estimation value composition of the data subcarrier in Tile.

According to a further aspect in the invention, provide a kind of base station, having comprised: pilot sub-carrier channel estimation module, forDetermine the channel estimation value of the pilot sub-carrier in the Tile in up PUSC; Filtering matrix module, for obtaining the filter of TileRipple matrix; Data subcarrier module, for using the channel estimation value of pilot sub-carrier and the filtering matrix of Tile, obtainsThe channel estimation value of the data subcarrier in Tile.

Further, filtering matrix module comprises: time correlation coefficient and frequency correlation coefficient acquisition module, and for obtainingTime correlation coefficient and frequency correlation coefficient between pilot sub-carrier; Maximum doppler frequency and all square time delay acquisition module,For coefficient correlation and frequency correlation coefficient service time, obtain maximum doppler frequency and equal square time delays; Filtering matrix obtainsModule, for using the signal to noise ratio, maximum doppler frequency of Tile and equal square time delays, obtains filtering matrix.

Further, maximum doppler frequency and all square time delay acquisition module obtain maximum Doppler frequently according to following formulaMove v_max：R_H(Δn)＝J₀(2πN_sΔnv_max); Wherein, R_H(Δ n) between the pilot sub-carrier of a Δ n symbol of being separated by timeBetween coefficient correlation, N_sFor the time interval between adjacent symbol, Δ n is positive integer, J₀(.) is first kind zeroth order BezierBessel function.

By the present invention, adopt the channel estimation methods of a kind of new up PUSC, first determine the pilot tone in TileThe channel estimation value of subcarrier and the filtering matrix of this Tile, then obtain in this Tile according to these two kinds of parameters of determiningThe channel estimation value of data subcarrier, thereby reduced amount of calculation, and kept higher channel estimating performance, solved existingThe amount of calculation that has technology to exist is larger, the higher problem of difficulty of realization.

Brief description of the drawings

Accompanying drawing described herein is used to provide a further understanding of the present invention, forms the application's a part, thisBright schematic description and description is used for explaining the present invention, does not form inappropriate limitation of the present invention. In the accompanying drawings:

Fig. 1 is according to the flow chart of the channel estimation methods of the up PUSC of the embodiment of the present invention;

Fig. 2 is according to the structural representation of the embodiment of the present invention Tile;

Fig. 3 is according to the schematic diagram of the base station of the embodiment of the present invention.

Detailed description of the invention

Hereinafter also describe the present invention in detail with reference to accompanying drawing in conjunction with the embodiments. It should be noted that, what do not conflictIn situation, the feature in embodiment and embodiment in the application can combine mutually.

Fig. 1 is according to the flow chart of the channel estimation methods of the up PUSC of the embodiment of the present invention, comprises the following steps:

Step S102, the channel estimation value of the pilot sub-carrier in the Tile (sheet) in up PUSC is determined in base station;

Regulation according to agreement a: Tile comprises 4 pilot sub-carriers and 8 data subcarriers. A time slot(slot) there are 6 Tile.

For example, the structure of a Tile as shown in Figure 2, by this Tile comprise 4 pilot sub-carrier: P0, P1, P2 andP3, and 8 data subcarrier: S1, S2, S3, S4, S5, S6, S7 and S8. In step S102, base station is according to the letter receivingNumber, in the up PUSC in the signal that utilization receives, 4 pilot sub-carrier P0, P1, P2 and the P3 of a Tile, calculateThe channel estimation value (can utilize existing LS channel estimation methods to obtain) of these four pilot sub-carriers of P0, P1, P2 and P3.The matrix of the channel estimation value composition of these four pilot sub-carriers can be usedRepresent.

Step S104, base station obtains the filtering matrix W of this Tile;

Step S108, base station is used the channel estimation value of the pilot sub-carrier in this Tile and the filtering matrix of this TileW, obtains the channel estimation value of the data subcarrier in this Tile.

The present embodiment has adopted the channel estimation methods of a kind of new up PUSC, first determines that pilot tone in Tile carriesThe channel estimation value of ripple and the filtering matrix of this Tile, then obtain the number in this Tile according to these two kinds of parameters of determiningAccording to the channel estimation value of subcarrier, thereby reduce amount of calculation and reduced computation complexity, and having kept higher channel to estimateMeter performance, the amount of calculation that has solved prior art existence is larger, the higher problem of difficulty of realization.

In actual applications, above-mentioned steps S104 can comprise the following steps (step 1 is to step 3):

Step 1, time correlation coefficient and the frequency correlation coefficient between the pilot sub-carrier in this Tile calculated in base station;

For example, for the example of Tile as shown in Figure 2, utilize these four pilot sub-carriers of P0, P1, P2 and P3, calculateGo out P0, P1 and P2, the frequency correlation coefficient (being frequency dependence) between P3, and P0, P2 and P1, the time correlation between P3Coefficient (being temporal correlation).

Step 2, base station is used the time correlation coefficient and the frequency correlation coefficient that in step 1, calculate to obtain the most generalStrangle frequency displacement and equal square time delays;

For example, can obtain above-mentioned maximum doppler frequency v according to following formula (4)_max：

R_H(Δn)＝J₀(2πN_sΔnv_max)(4)

Wherein, R_H(Δ is n) the time correlation coefficient between the pilot sub-carrier of the Δ n symbol of being separated by, N_sFor adjacent symbolIn the time interval between number, Δ n is positive integer, J₀(.) is first kind zeroth order Bezier (Bessel) function.

The time correlation coefficient of being separated by between the data subcarrier of a Δ n symbol equally, also meets above-mentioned formula (4). ,R_H(Δ is n) also the time correlation coefficient between the data subcarrier of the Δ n symbol of being separated by.

Can also calculate above-mentioned all square time delay σ according to following formula (8)_τ：

$R_H\left(\mathrm{\Δl}\right)=\frac{1}{1+j2\mathrm{\π\Δl}{\mathrm{\σ}}_{\mathrm{\τ}}L}---\left(8\right)$

Wherein, R_H(Δ is l) frequency correlation coefficient between the pilot sub-carrier of the Δ l carrier wave of being separated by, and L is adjacent carryingFrequency interval between ripple, Δ l is positive integer.

Equally, the frequency correlation coefficient between the data subcarrier of the Δ l carrier wave of being separated by also meets above-mentioned formula (8), that is,R_H(Δ is l) also the frequency correlation coefficient between the data subcarrier of the Δ l carrier wave of being separated by.

Introduce in detail the derivation of above-mentioned formula (4) and formula (8) below. In derivation, need to do channelSome are reasonably supposed. And whether hypothesis is reasonable, can verify by emulation and test. Two hypothesis are mainly done: oneThat wireless channel meets Jakes (Ya Keshi) model; The 2nd, multipath power attenuation is obeyed negative exponent and is distributed.

According to the Jakes model of wireless channel, suppose that Doppler's power spectrum can use formula (1) approximate representation:

Wherein, v_maxFor maximum doppler frequency. To S_H(v) carry out anti-Fourier (Fourier) conversion, can obtain letterRoad time domain coefficient R_H(Δt)：

$R_H\left(\mathrm{\Δt}\right)={\∫}_{-\∞}^{\∞}{S}_H\left(v\right)\exp \left(j2\mathrm{\πv\Δt}\right)\mathrm{d\τ}---\left(2\right)$

Above-mentioned formula (3) is arranged and can obtain following formula (3):

R_H(Δt)＝J₀(2πΔtv_max)(3)

The time correlation coefficients R of being separated by between pilot sub-carrier or the data subcarrier of a Δ n symbol so,_H(Δn)For:

R_H(Δn)＝J₀(2πN_sΔnv_max)(4)

Wherein, N_sFor the time interval between adjacent symbol (can be OFDM symbol).

Suppose that multipath power attenuation obedience negative exponent distributes, power delay profile can be approximate with following formula (5) soRepresent:

$S_H\left(\mathrm{\τ}\right)=\left\{\begin{array}{cc}\frac{1}{{\mathrm{\σ}}_{\mathrm{\τ}}}\exp (-\frac{\mathrm{\τ}}{{\mathrm{\σ}}_{\mathrm{\τ}}})& 0\≤\mathrm{\τ}\\ 0& 0>\mathrm{\τ}\end{array}\right.---\left(5\right)$

Wherein, σ_τFor equal square time delays. To S_H(τ) carry out Fourier (Fourier) conversion, can obtain channel spectrum correlationFunction R_H(Δf)：

$R_H\left(\mathrm{\Δf}\right)=\frac{1}{{\mathrm{\σ}}_{\mathrm{\τ}}}{\∫}_0^{\∞}\exp (-\frac{\mathrm{\τ}}{{\mathrm{\σ}}_{\mathrm{\τ}}})\exp (-j2\mathrm{\π\Δf\τ})\mathrm{d\τ}---\left(6\right)$

Above-mentioned formula (6) is arranged and can obtain following formula (7):

$R_H\left(\mathrm{\Δf}\right)=\frac{1}{1+j2\mathrm{\π\Δf}{\mathrm{\σ}}_{\mathrm{\τ}}}---\left(7\right)$

The frequency correlation coefficient R of being separated by between pilot sub-carrier or the data subcarrier of a Δ l carrier wave so,_H(Δl)For:

$R_H\left(\mathrm{\Δl}\right)=\frac{1}{1+j2\mathrm{\π\Δl}{\mathrm{\σ}}_{\mathrm{\τ}}L}---\left(8\right)$

Wherein, L is the frequency interval between adjacent carrier wave.

Step 3, base station is used in the signal to noise ratio (SignalNoiseRatio, SNR), step 2 of this Tile to be estimated to obtainMaximum doppler frequency and equal square time delays, obtain the filtering matrix W of this Tile.

Particularly, in actual applications, can calculate according to following formula (9) the filtering matrix W of this Tile:

$\left\{\begin{array}{c}W=R_{\mathrm{hp}}{(R_{\mathrm{pp}}+{\mathrm{\σ}}_n^2{\left({\mathrm{XX}}^H\right)}^{-1})}^{-1}\\ R_{\mathrm{hp}}=\left\{R_{\mathrm{SiPj}}\right\}\\ R_{\mathrm{pp}}=\left\{R_{\mathrm{PkPm}}\right\}\end{array}\right.---\left(9\right)$

Wherein, R_hpFor the correlation matrix between data subcarrier and pilot sub-carrier in this Tile, R_SiPjFor dataCoefficient correlation between carrier wave i and pilot sub-carrier j, R_ppFor the phase between pilot sub-carrier and pilot sub-carrier in this TileClose matrix, R_PkPmFor the coefficient correlation between pilot sub-carrier k and pilot sub-carrier m, i, j, k, m is natural number,For the inverse of the signal to noise ratio of this Tile. Particularly,For noise, the channel estimating that X is pilot sub-carrier, XX^HFor pilot tone carriesThe autocorrelation matrix of ripple.

Wherein, when the number of pilot sub-carrier is 4 and the number of data subcarrier while being 8, R_ppBe 4 × 4 matrixes, represent 4Coefficient correlation in individual pilot sub-carrier between every two pilot sub-carriers; R_hpBe 8 × 4 matrixes, represent 8 data subcarriers and 4Coefficient correlation between individual pilot sub-carrier; XX^HBe 4 × 4 diagonal matrix, represent the size of the amplitude of 4 pilot sub-carriers.Now, i=1,2,3,4,5,6,7,8, j=0,1,2,3; K=0,1,2,3, m=0,1,2,3.

Introduce in detail the R in above-mentioned formula (9) below_ppAnd R_hpDerivation. If suppose that channel time domain and frequency domain areUncorrelated mutually, so, by above-mentioned formula (4) and formula (8) the leading of a Δ n symbol and the Δ l carrier wave of being separated by that can obtain being separated byFrequently the coefficient correlation between subcarrier and pilot sub-carrier or between pilot sub-carrier and data subcarrier can be expressed as followingFormula (10):

R_H(Δn，Δl)＝R_H(Δn)R_H(Δl)(10)

Because the filtering matrix that calculates Tile according to formula (being above-mentioned formula (9)) is when the practical application, possible amount of calculationLarge, calculate more complicatedly, therefore, in order further to reduce amount of calculation and computation complexity, in the time of practical application, preferably canPrecompute W and SNR, v_maxAnd σ_τThese three values that parameter is corresponding, and be saved in a mapping table. Carrying out letterWhen road is estimated, obtain SNR, the maximum doppler frequency v of this Tile in base station_maxAll square time delay σ_τThe value of these three parametersAfterwards, base station only need be from this mapping table of setting up in advance of this locality, and the concrete value that finds these three parameters is relativeThe filtering matrix of answering. Obviously, signal to noise ratio, maximum doppler frequency and equal square time delay threes in this mapping table, have been recordedCorresponding relation with filtering matrix.

Obtained by step S102 this Tile pilot sub-carrier channel estimation value and obtain by step S104After having arrived the filtering matrix of this Tile, in step S108, can obtain the data in this Tile according to following formula (11)The channel estimation value of subcarrier:

$H_{\mathrm{lmmse}}=W{\hat{H}}_{\mathrm{ls}}---\left(11\right)$

Wherein,For the matrix that the channel estimation value of the pilot sub-carrier in this Tile forms, the filtering square that W is this TileBattle array (its formula is as shown in formula (9)), H_lmmseFor the matrix of the channel estimation value composition of the data subcarrier in this Tile. ItsIn, in the time that the number of pilot sub-carrier is 4,Be 4 × 1 matrixes, represent the channel estimation value of 4 pilot sub-carriers. Wherein, whenThe number of data subcarrier is 8 o'clock, H_lmmseBe 8 × 1 matrixes, represent the channel estimating of 8 subcarriers in a Tile.

In example as shown in Figure 2, finally can obtain 8 data subcarrier S1, S2 in this Tile, S3, S4,The channel estimation value of S5, S6, S7 and S8.

Introduce the computational process of the above embodiment of the present invention below with a concrete calculating embodiment:

In the step S104 of embodiment as shown in Figure 1, want to solve R_ppAnd R_hp, just must know v_maxAnd σ_τ, belowProvide v_maxAnd σ_τComputational process.

With a Tile, 2 antennas are that example illustrates v_maxAnd σ_τComputational methods:

The time correlation coefficients R of the pilot sub-carrier in a tile_H(2) can be expressed as:

$R_H\left(2\right)=\frac{A}{B}$

$A=(H_{00}{H^{*}}_{20}+H_{10}{H^{*}}_{30}+H_{01}{H^{*}}_{21}+H_{11}{H^{*}}_{31})/4$

$B=(H_{00}{H^{*}}_{00}+H_{10}{H^{*}}_{10}+H_{20}{H^{*}}_{20}+H_{30}{H^{*}}_{30}+H_{01}{H^{*}}_{01}+H_{11}{H^{*}}_{11}+H_{21}{H^{*}}_{21}+H_{31}{H^{*}}_{31})/8-{\mathrm{\δ}}^2---\left(12\right)$

Wherein, H₀₀The channel estimating of the 0th pilot sub-carrier of representative antennas 0, H₁₀The 1st pilot sub-carrier of representative antennas 0Channel estimating; H₂₀The channel estimating of the 2nd pilot sub-carrier of representative antennas 0, H₃₀The 3rd pilot sub-carrier of representative antennas 0Channel estimating; H₀₁The channel estimating of the 0th pilot sub-carrier of representative antennas 1, H₁₁The letter of the 1st pilot sub-carrier of representative antennas 1Road is estimated; H₂₁The channel estimating of the 2nd pilot sub-carrier of representative antennas 1, H₃₁The channel of the 3rd pilot sub-carrier of representative antennas 1Estimate.

Therefore, can calculate σ by formula (12) and formula (8)_τ。

In like manner, can calculate v_max, the frequency correlation coefficient R of the pilot sub-carrier in a tile_H(3) can representFor:

$R_H\left(3\right)=\frac{A}{B}$

$A=(H_{00}{H^{*}}_{10}+H_{20}{H^{*}}_{30}+H_{01}{H^{*}}_{11}+H_{21}{H^{*}}_{31})/4$

$B=(H_{00}{H^{*}}_{00}+H_{10}{H^{*}}_{10}+H_{20}{H^{*}}_{20}+H_{30}{H^{*}}_{30}+H_{01}{H^{*}}_{01}+H_{11}{H^{*}}_{11}+H_{21}{H^{*}}_{21}+H_{31}{H^{*}}_{31})/8-{\mathrm{\δ}}^2---\left(13\right)$

Therefore, can calculate v by formula (13) and formula (4)_max。

Can know W and R by formula (9)_pp、R_hpWithRelevant, and R_pp、R_hpAgain with parameter v_maxAnd σ_τIt is relevant,Be exactly the inverse of the SNR of this Tile, therefore, W and three relating to parameters: SNR, v_max、σ_τRelevant, with regard to the calculating of WRealizing calculating for DSP (DigitalSignalProcessing, Digital Signal Processing) is more complicated, amount of calculationLarger, it is very difficult that DSP directly realizes formula (9). Therefore, can construct the matrix table of these three variablees (above-mentionedMapping table), precompute W and these three values that parameter is corresponding, in the time that DSP realizes, only need to calculate SNR, v_maxAnd σ_τValue, then table look-up and just can know the value of W, amount of calculation just can be lacked much like this, is conducive to the realization of DSP.

Fig. 3 is according to the schematic diagram of the base station of the embodiment of the present invention, and this base station comprises with lower module: pilot sub-carrier channelEstimation module 10, for the channel estimation value of the pilot sub-carrier in the Tile of definite up PUSC; Filtering matrix module 20,For obtaining the filtering matrix W of this Tile; Data subcarrier module 30, for use pilot sub-carrier channel estimation value andThe filtering matrix W of this Tile, obtains the channel estimation value of the data subcarrier in this Tile.

Wherein, filtering matrix module 20 can comprise: time correlation coefficient and frequency correlation coefficient acquisition module, and for obtainingGet time correlation coefficient and frequency correlation coefficient between pilot sub-carrier; Maximum doppler frequency and all square time delay are obtained mouldPiece, for using above-mentioned time correlation coefficient and frequency correlation coefficient, obtains maximum doppler frequency v_maxAll square time delaysσ_τ; And filtering matrix acquisition module, for using signal to noise ratio, the maximum doppler frequency v of this Tile_maxAll square time delay σ_τ，Obtain filtering matrix W.

In the time of practical application, maximum doppler frequency and all square time delay acquisition module are according to obtaining by following formula (4)Maximum doppler frequency v_max：

R_H(Δn)＝J₀(2πN_sΔnv_max), wherein, R_H(Δ is n) between the pilot sub-carrier of a Δ n symbol of being separated byTime correlation coefficient, N_sFor the time interval between adjacent symbol, Δ n is positive integer, J₀(.) is first kind zeroth order Bezier(Bessel) function.

Maximum doppler frequency and all square time delay acquisition module can also obtain all square time delay σ according to following formula (8)_τ：

Wherein, R_H(Δ is l) frequency between the pilot sub-carrier of the Δ l carrier wave of being separated byCoefficient correlation, L is the frequency interval between adjacent carrier wave, Δ l is positive integer.

Wherein, filtering matrix acquisition module can calculate above-mentioned filtering matrix W according to following formula (9):

Wherein, R_hpFor the data subcarrier in described Tile and pilot sub-carrierBetween correlation matrix, R_SiPjFor the coefficient correlation between data subcarrier i and pilot sub-carrier j, R_ppFor in described TileCorrelation matrix between pilot sub-carrier and pilot sub-carrier, R_PkPmFor being correlated with between pilot sub-carrier k and pilot sub-carrier mCoefficient, i, j, k, m is natural number, R_H(Δ n, Δ l) for be separated by a Δ n symbol and be separated by a Δ l carrier wave pilot sub-carrier andCoefficient correlation between pilot sub-carrier or between pilot sub-carrier and data subcarrier,For the letter of described TileMake an uproar than inverse.

In order to reduce the larger amount of calculation that may bring according to formula (9) calculation of filtered matrix W and higher calculatingComplexity, in the time of practical application, filtering matrix acquisition module can also from base station in advance set up mapping table,Find with signal to noise ratio, the estimation module of this Tile and estimate the maximum doppler frequency that obtains and all corresponding filtering of square time delayMatrix, wherein, this corresponding relation table record signal to noise ratio, maximum doppler frequency and equal square time delay three and filtering matrix rightShould be related to.

In actual implementation process, data subcarrier module 30 can obtain the number in this Tile according to following formula (11)Channel estimation value according to subcarrier:

Wherein,For the matrix that the channel estimation value of the pilot sub-carrier in this Tile forms, W is for being somebody's turn to doThe filtering matrix of Tile, H_lmmseFor the matrix of the channel estimation value composition of the data subcarrier in this Tile.

As can be seen from the above description, the present invention has realized following technique effect: the invention provides up PUSCThe implementation method of channel estimating, first the method according to the signal receiving, utilizes pilot sub-carrier to carry out LS channel estimating, soThe rear channel estimation value that utilizes calculates the coefficient correlation between carrier wave, estimates maximum doppler frequency and equal square time delays, follows rootAccording to signal to noise ratio, Doppler frequency shift and all three parameters of square time delay table look-up and obtain filtering matrix, finally according to LS channel estimating and filterRipple matrix computations goes out the channel estimating of data subcarrier, thereby can reduce amount of calculation and reduce computation complexity, and canKeep higher channel estimating performance.

Obviously, it is apparent to those skilled in the art that above-mentioned of the present invention each module or each step can be with generalCalculation element realize, they can concentrate on single calculation element, or are distributed in multiple calculation element and formNetwork on, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be storedIn storage device, carried out by calculation element, and in some cases, can be to be different from shown in order execution hereinThe step that goes out or describe, or they are made into respectively to integrated circuit modules, or by multiple modules or step in themBeing made into single integrated circuit module realizes. Like this, the present invention is not restricted to any specific hardware and software combination.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for the skill of this areaArt personnel, the present invention can have various modifications and variations. Within the spirit and principles in the present invention all, to do any repairingProtection scope of the present invention changes, be equal to replacement, improvement etc., within all should be included in.

Claims (8)

1. ascender uses a channel estimation methods of subcarrier PUSC, it is characterized in that, comprising:

The channel estimation value of the pilot sub-carrier in the sheet Tile in up PUSC is determined in base station;

Described base station obtains the filtering matrix of described Tile;

Described base station is used the channel estimation value of described pilot sub-carrier and the filtering matrix of described Tile, obtains described TileIn the channel estimation value of data subcarrier;

Wherein, the filtering matrix that described base station obtains described Tile comprises: calculate between described pilot sub-carrier described base stationTime correlation coefficient and frequency correlation coefficient; Described base station is used described time correlation coefficient and described frequency correlation coefficient to obtainMaximum doppler frequency and all square time delay; Described base station is used the signal to noise ratio of described Tile, described maximum doppler frequency and instituteState all square time delays, obtain described filtering matrix.

2. method according to claim 1, is characterized in that, described base station is used described time correlation coefficient and described frequencyRate coefficient correlation obtain maximum doppler frequency and all square time delay comprise: obtain described maximum doppler frequency according to following formulav_max：

R_H(△n)＝J₀(2πN_s△nv_max)；

Wherein, R_H(△ is n) the time correlation coefficient between the pilot sub-carrier of the △ n symbol of being separated by, N_sFor adjacent symbol itBetween the time interval, △ n is positive integer, J₀(.) is first kind zeroth order Bezier Bessel function.

3. method according to claim 2, is characterized in that, described base station is used described time correlation coefficient and described frequencyRate coefficient correlation obtain maximum doppler frequency and all square time delay also comprise: obtain described all square time delay σ according to following formula_τ： $R_H\left(\mathrm{\Δ}l\right)=\frac{1}{1+j2{\mathrm{\π\Δl\σ}}_{\mathrm{\τ}}L};$

Wherein, R_H(△ is l) frequency correlation coefficient between the pilot sub-carrier of a △ l carrier wave of being separated by, L be adjacent carrier wave itBetween frequency interval, △ l is positive integer.

4. method according to claim 3, is characterized in that, described base station is used the signal to noise ratio, described of described TileLarge Doppler frequency shift and described all square time delays, obtain described filtering matrix and comprise: calculate described filtering according to following formulaMatrix W:

$\left\{\begin{array}{c}W=R_{hp}{(R_{pp}+{\mathrm{\σ}}_n^2{\left({\mathrm{XX}}^H\right)}^{-1})}^{-1}\\ R_{hp}=\left\{R_{SiPj}\right\}\\ R_{pp}=\left\{R_{PkPm}\right\}\\ R_H(\mathrm{\Δ}n,\mathrm{\Δ}l)=R_H\left(\mathrm{\Δ}n\right)R_H\left(\mathrm{\Δ}l\right)\end{array}\right.;$

Wherein, R_hpFor the correlation matrix between data subcarrier and pilot sub-carrier in described Tile, R_SiPjFor data carriesCoefficient correlation between ripple i and pilot sub-carrier j, R_ppFor the phase between pilot sub-carrier and pilot sub-carrier in described TileClose matrix, R_PkPmFor the coefficient correlation between pilot sub-carrier k and pilot sub-carrier m, i, j, k, m is natural number, R_H(△n,△L) for being separated by between the pilot sub-carrier of a △ n symbol and the △ l carrier wave of being separated by and pilot sub-carrier or pilot sub-carrier and numberAccording to the coefficient correlation between subcarrier,For the inverse of the signal to noise ratio of described Tile.

5. method according to claim 1, is characterized in that, described base station is used the signal to noise ratio, described of described TileLarge Doppler frequency shift and described all square time delays, obtain described filtering matrix and comprise:

Described base station from local mapping table, find with described signal to noise ratio, described maximum doppler frequency and described inAll corresponding described filtering matrixs of square time delay, wherein, described corresponding relation table record signal to noise ratio, maximum doppler frequency andAll corresponding relations of square time delay three and filtering matrix.

6. according to the method described in any one in claim 1 to 5, it is characterized in that, described base station is used described pilot tone to carryThe channel estimation value of ripple and the filtering matrix of described Tile, obtain the channel estimation value bag of the data subcarrier in described TileDraw together: the channel estimation value that obtains the described data subcarrier in described Tile according to following formula:

$H_{lmmse}=W{\hat{H}}_{ls};$

Wherein,For the matrix that the channel estimation value of the described pilot sub-carrier in described Tile forms, the filter that W is described TileRipple matrix, H_lmmseFor the matrix of the channel estimation value composition of the described data subcarrier in described Tile.

7. a base station, is characterized in that, comprising:

Pilot sub-carrier channel estimation module, for pilot tone in the sheet Tile of definite ascender use subcarrier PUSCThe channel estimation value of carrier wave;

Filtering matrix module, for obtaining the filtering matrix of described Tile;

Data subcarrier module, for using the channel estimation value of described pilot sub-carrier and the filtering matrix of described Tile,Obtain the channel estimation value of the data subcarrier in described Tile,

Wherein, described filtering matrix module comprises: time correlation coefficient and frequency correlation coefficient acquisition module, and described in obtainingTime correlation coefficient and frequency correlation coefficient between pilot sub-carrier; Maximum doppler frequency and all square time delay acquisition module,Be used for using described time correlation coefficient and described frequency correlation coefficient, obtain maximum doppler frequency and equal square time delays; FilteringMatrix acquisition module, for using the signal to noise ratio of described Tile, described maximum doppler frequency and described all square time delays, obtains instituteState filtering matrix.

8. base station according to claim 7, is characterized in that, described maximum doppler frequency and all square time delay acquisition moduleObtain described maximum doppler frequency v according to following formula_max：

R_H(△n)＝J₀(2πN_s△nv_max)；

Wherein, R_H(△ is n) the time correlation coefficient between the pilot sub-carrier of the △ n symbol of being separated by, N_sFor adjacent symbol itBetween the time interval, △ n is positive integer, J₀(.) is first kind zeroth order Bezier Bessel function.