CN102215184B - Method and system for estimating uplink timing error - Google Patents

Abstract

The invention discloses a method and system for estimating an uplink timing error. The method comprises the following steps of: estimating a channel frequency domain response of a pilot frequency position by using frequency domain data of an orthogonal frequency division multiplexing (OFDM) symbol; converting the estimated channel frequency domain response to a time domain so as to obtain a channel time domain response and carrying out zero resetting a tap determined as a noise item in the channel time domain response; converting the channel time domain response, which is obtained after the tap of the noise item is subjected to zero resetting, to a frequency domain so as to obtain a converted channel frequency domain response; carrying out conjugate correlation on pilot frequency points spaced in the converted channel frequency domain response and carrying out sequence averaging on a conjugate correlation result to obtain a sequence average value; and reconstructing a noise interference item by using rest taps after the tap of the noise item is subjected to zero resetting and estimating a timing error after subtracting the noise interference item from the sequence average value. According to the invention, the estimation precision of the timing error can be effectively improved and the demodulation property of the uplink channel can be assured.

Description

A kind of method of estimation of uplink timing error and system

Technical field

The present invention relates to communication technical field, relate in particular to method of estimation and the system of uplink timing error in a kind of OFDM (Orthogonal FrequencyDivision Multiplexing, OFDM) system.

Background technology

In wireless communication system, the performance of simultaneous techniques is directly connected to the performance of whole communication system, and it is the prerequisite of information transmitting.OFDM (Orthogonal Frequency DivisionMultiplexing, OFDM) system synchronously can be divided into sample-synchronous, carrier synchronization and Timing Synchronization.

In ofdm system, Timing Synchronization comprises timing synchronization and sample value Timing Synchronization.The object of timing synchronization is to make receiving terminal determine the start-stop moment of each OFDM symbol, determines FFT (fast Fourier transform) window position accurately, and further realizes the synchronous or frame synchronization of piece; Sample value Timing Synchronization is in order to make receiving terminal determine the start-stop moment of each sample value symbol.In ofdm system, realize sign synchronization and need to determine symbol time offset and sample value timing offset, as shown in Figure 1, sample value symbol when the first behavior ideal distributes, and the second behavior exists symbol time offset Δ t with respect to ideal _ftime definite sample value symbol distribute, there is sample value timing offset Δ t when with respect to ideal in the third line _ytime definite sample value symbol distribute, there is symbol time offset Δ t when with respect to ideal in fourth line _fwith sample value timing offset Δ t _ytime definite sample value symbol distribute.

For symbol time offset Δ t _f, as long as the initial time of symbol drops in Cyclic Prefix (CyclicPrefix, CP), just can not destroy the orthogonality between each subcarrier, only can cause phase rotating.Otherwise, can destroy the orthogonality between each subcarrier, also can between created symbol, crosstalk simultaneously.Sample value timing offset Δ t _ycan be equivalent to a fixing time delay.Therefore generally by symbol time offset Δ t _fwith sample value timing offset Δ t _yimpact be equivalent to a total timing offset value, by estimating that this value carries out best timing to ofdm system, to ensure that ofdm system obtains a best performance.

At present about estimating that timing offset realizes the method for Timing Synchronization and be mainly divided into following two classes:

1) time domain correlation based on CP

The method is to utilize the redundancy of Cyclic Prefix, carries out auto-correlation carry out tracking symbol in time domain, intercepts CP and OFDM symbol afterbody corresponding part is correlated with, and obtains maximum peak point, can determine the position of synchronization timing deviation.

Adopt the timing offset method of estimation of this kind of mode, due to the impact of channel multi-path, in the front portion of CP, the hangover part of previous OFDM symbol has produced certain interference, that is to say that to utilize the data of CP front portion unreliable, thereby need to obtain the starting point of CP valid data.But when system bandwidth hour, because the number of samples of whole CP is smaller, if remove some sampling points that are disturbed, just still less, just there is the problem that timing offset estimated accuracy is not high in available effective sampling point now again.

2) the spectrum correlation method based on pilot tone

This algorithm is to utilize the pilot signal that receives in signal and local pilot signal to carry out frequency domain correlation to realize symbol tracking, reception data and local pilot frequency sequence by pilot frequency locations on same OFDM symbol multiply each other, eliminate the impact of local pilot frequency sequence, also obtained the channel estimation in frequency domain at pilot tone place, calculate afterwards the conjugation correlation of adjacent pilot frequencies channel estimating, the phase place of this conjugation correlation is introduced by timing offset, thereby can obtain timing offset value.

Adopt the timing offset method of estimation of this kind of mode, when the bandwidth taking is less, when signal to noise ratio is lower simultaneously, the estimated accuracy of timing offset is lower.Conventionally the frequency-region signal of eliminating local pilot frequency sequence is transformed to time domain, that the signal value of noise sets to 0 by determining in time domain, afterwards again by DFT (Direct FourietTransformer, discrete Fourier transform) transform to frequency domain, then carry out the calculating of the conjugation correlation of adjacent pilot frequencies channel estimating below.The uplink channel estimation that transforms from the time domain to frequency domain adopts DFT (DirectFouriet Transformer, discrete Fourier transform) algorithm, getting window in time domain carries out noise suppression and improves the precision of channel estimating, but there is very large impact to the estimation of timing offset in this operation, even cannot follow the tracks of, estimate timing offset in the time of low signal-to-noise ratio.

Adopt existing timing offset method of estimation due to the accurate timing offset of estimating OFDM system, in the time of ofdm system life period deviation, the impact bringing is mainly reflected in: increase the sensitivity of ofdm system to time delay expansion, and cause that thereby phase deviation causes demodulation performance to decline, error rate increase.

Summary of the invention

The invention provides a kind of uplink timing error method of estimation and system, for effectively improving the estimated accuracy of timing offset, ensured the demodulation performance of up channel.

The method of estimation that the invention provides a kind of uplink timing error, comprising:

Utilize the frequency domain data of orthogonal frequency division multiplex OFDM symbol and the channel frequency domain response of local pilot data estimating pilot frequency position that receive;

The channel frequency domain response of estimation is transformed to time domain and obtain channel time domain response, and will in described channel time domain response, determine it is the tap zero setting of noise item;

The channel time domain response transform obtaining after noise item tap zero setting, to frequency domain, is obtained converting rear channel frequency domain response;

The pilot tone point at interval in channel frequency domain response after described conversion is carried out to conjugation and be correlated with, and ask sequence average to obtain serial mean to the relevant result of conjugation;

After utilizing noise item tap zero setting, remaining tap reconstructs noise jamming item, estimates timing offset after deducting described noise jamming item in described serial mean;

Described utilize noise item tap zero setting after remaining tap reconstruct noise jamming item, be specially by noise jamming item described in following formula reconstruct:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\frac{2\mathrm{\π}\·n\·\mathrm{\Δk}}{N}}$

Wherein I is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

The present invention also provides a kind of estimating system of uplink timing error, comprising:

The first channel frequency domain response estimation unit, for utilizing the frequency domain data of orthogonal frequency division multiplex OFDM symbol and the channel frequency domain response of local pilot data estimating pilot frequency position of reception;

The first channel time domain response unit, obtains channel time domain response for the channel frequency domain response of estimation being transformed to time domain;

Tap zero setting unit, for determining it is the tap zero setting of noise item by described channel time domain response;

Second channel frequency domain response unit, for the channel time domain response transform obtaining after noise item tap zero setting is arrived to frequency domain, obtains converting rear channel frequency domain response;

Conjugation correlation unit, carries out conjugation for the pilot tone point to converting rear channel frequency domain response interval and is correlated with;

All value cells, for asking sequence average to obtain serial mean to the relevant result of conjugation;

Interference reconstruction unit, reconstructs noise jamming item for remaining tap after utilizing noise item tap zero setting;

Timing offset estimation unit is estimated timing offset after deducting described noise jamming item in described serial mean;

Described interference reconstruction unit is specifically for by noise jamming item described in following formula reconstruct:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\frac{2\mathrm{\π}\·n\·\mathrm{\Δk}}{N}}$

Wherein I is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

Utilize method of estimation and the system of uplink timing error provided by the invention, there is following beneficial effect: frequency domain conjugation sequence of correlation values average packet Noise distracter after getting window is eliminated, efficiently solve and utilize the frequency domain channel value of getting after window in the time of low signal-to-noise ratio, cannot carry out the problem of reliable timing estimation, effectively improve the precision that timing offset is estimated, ensured the demodulation performance of up channel.

Brief description of the drawings

Fig. 1 is the schematic diagram that ofdm system exists uplink timing error;

Fig. 2 is the flow chart that the existing spectrum correlation method based on pilot tone is estimated timing offset;

Fig. 3 is the method for estimation flow chart of uplink timing error in the embodiment of the present invention;

Fig. 4 is the estimating system structured flowchart of uplink timing error in the embodiment of the present invention.

Embodiment

Method of estimation and the system of uplink timing error the present invention being proposed below in conjunction with drawings and Examples illustrate in greater detail.

In ofdm system, suppose that the frequency domain reception data of an a certain moment OFDM symbol are:

R _k＝H _kS _k+W _kk＝0，1，…，N-1 (1)

Wherein k represents OFDM symbol subcarrier sequence number, R _kthe frequency domain that represents pilot tone correspondence position on OFDM symbol receives data, H _krepresent the frequency domain channel impulse response of pilot frequency locations, S _krepresent the pilot data of transmitting terminal, W _krepresent additive white Gaussian noise, N represents useful subcarrier number, the i.e. number of OFDM symbol subcarrier in ofdm system bandwidth.

The spectrum correlation method based on pilot tone of providing below can be estimated the principle of uplink timing error.According to the existing spectrum correlation method based on pilot tone, as shown in Figure 2, carry out following flow process:

Step S201, carries out conjugate multiplication by the frequency domain data of the orthogonal frequency division multiplex OFDM symbol of reception and local pilot data, thereby eliminates the impact of pilot frequency sequence, has also obtained the channel frequency domain response of pilot sub-carrier position

$\hat{H}=R_k{S}_k^{*}=H_k{S}_k{S}_k^{*}+W_k{S}_k^{*}=P_s{H}_k+W_k^{\′}---\left(2\right)$

Suppose that signal carries out power normalization,

Step S202, to channel frequency domain response

carry out Inverse Discrete Fourier Transform IDFT, obtain the channel time domain response of estimating

that is:

${\hat{h}}_n=\mathrm{IDFT}{\left(\hat{H}\right)}_n=h_n+w_n$ n＝0，1，…，N-1 (3)

Step S203, will determine it is the tap zero setting of noise item in channel time domain response in step S202, obtains channel time domain response

corresponding sample point when the tap here refers to that n gets different value.

Above-mentioned definite channel time domain response

the existing method of the concrete employing of middle noise item, no longer describes in detail here.

${\tilde{h}}_n=\left\{\begin{array}{cc}{\hat{h}}_n& n\∈\mathrm{\Ω}\\ 0& \mathrm{others}\end{array}\right.---\left(4\right)$

Wherein, the tap position set that Ω retains after for tap zero setting.

Step S204, to the result of step S203

carry out leaf transformation DFT in N point discrete Fourier, obtain the channel frequency domain response estimation value on all subcarriers

${\tilde{H}}_k=\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{\tilde{h}}_n{e}^{-j\·\frac{2\mathrm{\π}}{N}\·n\·k}=\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}\tilde{h}{e}^{-j\·\frac{2\mathrm{\π}}{N}\·n\·k}=H_k+\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{w}_n{e}^{-j\·\frac{2\mathrm{\π}}{N}\·n\·k}=H_k+{\tilde{W}}_k---\left(5\right)$

Wherein: k=0,1 ..., N-1

Step S205, to being spaced apart two pilot tone point k of Δ k ₁, k ₂do conjugation relevant:

${\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*}=(H_{k_1}+{\tilde{W}}_{k_1}){(H_{k_2}+{\tilde{W}}_{k_2})}^{*}$

$=H_{k_1}\·H_{k_2}^{*}+\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{w}_n{w}_m^{*}{e}^{j\·\frac{2\mathrm{\π}}{N}\·(m{k}_2-n{k}_1)}+H_{k_1}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{w}_n{e}^{j\·\frac{2\mathrm{\π}}{N}\·n\·k_2}+H_{k_2}^{*}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{w}_n{e}^{-j\·\frac{2\mathrm{\π}}{N}\·n\·k_1}---\left(6\right)$

Wherein, k ₂=k ₁+ Δ k, k ₁=0 ..., N-1-Δ k

Step S206, the sequence after conjugation is correlated with is asked serial mean:

$E({\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*})=E(H_{k_1}\·H_{k_2}^{*})+\frac{1}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{\left|\right|w_n\left|\right|}^2{e}^{j\·\frac{2\mathrm{\π}}{N}\·\mathrm{n\Δk}}=1+\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\·\frac{2\mathrm{\π}}{N}\·n\·\mathrm{\Δk}}---\left(7\right)$

Obtaining formula (7) by formula (6) is based on following hypothesis:

1) relevant adjacent pilot frequencies point channel estimation value is approximate identical:

, have: $E(H_{k_1}\·H_{k_2}^{*})=1;$

2) w _nfor additive white Gaussian noise, and || w _n|| ²=σ ², σ ²for the power of additive white Gaussian noise;

3) signal and noise are uncorrelated;

Step S201～step S206 is the derivation based on not there is not timing offset, supposes that ofdm system exists timing offset Δ n:

Formula (1) should be modified to:

k=0,1 ..., N-1 (1 ')

According to the channel frequency domain response of pilot sub-carrier position in step S201

should be:

$\hat{H}=R_k{S}_k^{*}=H_k{e}^{\frac{j2\mathrm{\π\Δnk}}{N}}+W_k^{\′}---\left(2^{,}\right)$

According to the derivation of above-mentioned steps S202～step S205, the serial mean in can determining step S206 should be:

$E({\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*})=W_N^{\mathrm{\Δn\Δk}}+\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{W}_N^{\mathrm{n\Δk}}---\left(7^{,}\right)$

Wherein $W_N^{\mathrm{nk}}=e^{-j\·\frac{2\mathrm{\π}\·n\·k}{N}}.$

Step S207, the sequence of utilizing step S206 to obtain asks serial mean to estimate timing offset:

$\mathrm{\Δn}=\frac{\mathrm{angle}\left(Q\right)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}---\left(8^{,}\right)$

Wherein, angle is for asking phase bit arithmetic, N _fFTfor carrying out counting of fast Fourier transform FFT window that OFDM symbolic blocks uses when synchronous, Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

Get window if do not carry out time domain when time domain transforms to the channel estimating of frequency domain after denoising in step S204, the Section 2 in formula (7 ') equals 0, otherwise this is not 0, can estimate have a certain impact to synchronization timing.Can find out that through getting after window, the serial mean that frequency domain adjacent pilot frequencies point conjugation is relevant not only comprises the phase rotation coefficient that timing offset brings, and also comprises noise jamming item simultaneously.This is that in the relevant serial mean of frequency domain adjacent pilot frequencies point conjugation, playing main contributions will be noise jamming, thereby utilize existing method cannot carry out timing offset estimation if the noise power retaining is excessive.

Get the problem of curtain heading tape for channel estimating time domain, the method for estimation of uplink timing error proposed by the invention, first revises the relevant serial mean of frequency domain adjacent pilot frequencies point conjugation, eliminates the distracter that noise path is brought.In embodiments of the invention one, a kind of method of estimation of uplink timing error is provided, as shown in Figure 3, the method comprises the following steps:

Step S301, the frequency domain data R of the orthogonal frequency division multiplex OFDM symbol that utilization receives _kchannel frequency domain response with local pilot data estimating pilot frequency position.

Preferably, according to formula (2), the frequency domain of pilot frequency locations on same OFDM symbol is received to data R _kmultiply each other with local pilot frequency sequence, obtain the channel frequency domain response H of the pilot frequency locations of estimating.

Step is 302, by the channel frequency domain response of estimating

transform to time domain and obtain channel time domain response

Preferably, by Inverse Discrete Fourier Transform IDFT, according to formula (3) by estimate channel frequency domain response

transform to time domain and obtain channel time domain response

Step S303, by described channel time domain response

in determine be the tap zero setting of noise item;

Wherein determine channel time domain response

the existing method of the concrete employing of middle noise item, no longer describes in detail here.

${\tilde{h}}_n=\left\{\begin{array}{cc}{\tilde{h}}_n& n\∈\mathrm{\Ω}\\ 0& \mathrm{others}\end{array}\right.$

Wherein, the tap position set that Ω retains after for tap zero setting.

Step S304, by the channel time domain response obtaining after noise item tap zero setting transform to frequency domain, obtain converting rear channel frequency domain response

In the present embodiment, the channel time domain obtaining after noise item tap zero setting is responded

carry out leaf transformation DFT in N point discrete Fourier according to formula (5) and transform to frequency domain, described N is the number of OFDM symbol subcarrier.

Step S305, to converting rear channel frequency domain response

the pilot tone point at middle interval carries out conjugation and is correlated with;

Specifically can according to formula (6), to being spaced apart the pilot tone point of Δ k, to carry out conjugation relevant, preferably, to converting rear channel frequency domain response the pilot tone point at middle interval carries out conjugation is correlated with, and is specially converting rear channel frequency domain response

in adjacent pilot tone point to carry out conjugation relevant

Step S306, in step S305

the result that conjugation is relevant asks sequence average to obtain serial mean;

Specifically ask serial mean according to formula (7 ') $E({\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*})=W_N^{\mathrm{\Δn\Δk}}+\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{W}_N^{\mathrm{n\Δk}}.$

Step S307, utilizes channel time domain response after noise item tap zero setting

remaining tap reconstructs noise jamming item.

Specifically can be according to the noise item of tap zero setting in step S303, determine channel time domain response the position of stick signal tap and number, retain the noise jamming item (being the Section 2 in formula (7)) of noise tap introducing and deduct this distracter from serial mean thereby reconstruct, and obtains amended serial mean:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{W}_N^{\mathrm{n\Δk}}$

$Q=E({\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*})-I=W_N^{\mathrm{\Δn\Δk}}$

Wherein

i is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

By this step, the conjugation sequence of correlation values of adjacent pilot frequencies point is average just only relevant with timing offset, thereby can estimate accurately timing offset.

Step S308, utilizes amended serial mean to estimate timing offset Δ n:

$\mathrm{\Δn}=\frac{\mathrm{andle}\left(Q\right)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}=\frac{\mathrm{angle}(E({\tilde{H}}_{k_1}\·{\tilde{H}}_{k_2}^{*})-I)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}$

Wherein, angle is for asking phase bit arithmetic, N _fFTfor carrying out counting of fast Fourier transform FFT window that OFDM symbolic blocks uses when synchronous, be the subcarrier number existing in ofdm system bandwidth, Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

Visible, the present invention eliminates frequency domain conjugation sequence of correlation values average packet Noise distracter after getting window, efficiently solve and utilize the frequency domain channel value of getting after window in the time of low signal-to-noise ratio, cannot carry out the problem of reliable timing estimation, effectively improve the precision that timing offset is estimated, ensured the demodulation performance of up channel.

In embodiments of the invention two, a kind of estimating system of uplink timing error is provided, as shown in Figure 4, this system comprises: the first channel frequency domain response estimation unit 401, for utilizing the frequency domain data of orthogonal frequency division multiplex OFDM symbol and the channel frequency domain response of local pilot data estimating pilot frequency position of reception; The first channel time domain response unit 402, obtains channel time domain response for the channel frequency domain response of estimation being transformed to time domain; Tap zero setting unit 403, for determining it is the tap zero setting of noise item by described channel time domain response; Second channel frequency domain response unit 404, for the channel time domain response transform obtaining after noise item tap zero setting is arrived to frequency domain, obtains converting rear channel frequency domain response; Conjugation correlation unit 405, carries out conjugation for the pilot tone point to converting rear channel frequency domain response interval and is correlated with; All value cells 406, for asking sequence average to obtain serial mean to the relevant result of conjugation; Interference reconstruction unit 407, reconstructs noise jamming item for remaining tap after utilizing noise item tap zero setting; Timing offset estimation unit 408 is estimated timing offset after deducting described noise jamming item in described serial mean.

Preferably, the first channel frequency domain response estimation unit 401, specifically for frequency domain reception data and the local pilot frequency sequence of pilot frequency locations on same OFDM symbol are multiplied each other, obtains the channel frequency domain response of the pilot frequency locations of estimating.The first channel time domain response unit 402, specifically for by Inverse Discrete Fourier Transform IDFT, transforms to time domain by the channel frequency domain response of estimation and obtains channel time domain response.Second channel frequency domain response unit 404 transforms to frequency domain specifically for the channel time domain response obtaining after noise item tap zero setting is carried out to leaf transformation DFT in N point discrete Fourier, and described N is the number of OFDM symbol subcarrier.Specifically to converting pilot tone point adjacent in rear channel frequency domain response, to carry out conjugation relevant for conjugation correlation unit 405.

Interference reconstruction unit 407 in the present embodiment is specifically for by noise jamming item described in following formula reconstruct:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\frac{2\mathrm{\π}\·n\·\mathrm{\Δk}}{N}}$

Wherein I is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

In the present embodiment, timing offset estimation unit 408 specifically comprises: amendment unit, for deduct the noise jamming item reconstructing in serial mean, obtains revised serial mean Q; Timing offset computing unit, for utilizing revised serial mean Q to estimate timing offset Δ n by following formula:

$\mathrm{\Δn}=\frac{\mathrm{angle}\left(Q\right)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}$

Wherein, angle is for asking phase bit arithmetic, N _fFTfor carrying out counting of fast Fourier transform FFT window that OFDM symbolic blocks uses when synchronous, Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

Obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if these amendments of the present invention and within modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (8)

1. a method of estimation for uplink timing error, comprising: utilize the frequency domain data of orthogonal frequency division multiplex OFDM symbol and the channel frequency domain response of local pilot data estimating pilot frequency position that receive; The channel frequency domain response of estimation is transformed to time domain and obtain channel time domain response, and will in described channel time domain response, determine it is the tap zero setting of noise item; The channel time domain response transform obtaining after noise item tap zero setting, to frequency domain, is obtained converting rear channel frequency domain response; The pilot tone point at interval in channel frequency domain response after described conversion is carried out to conjugation and be correlated with, and ask sequence average to obtain serial mean to the relevant result of conjugation; It is characterized in that, also comprise:

After utilizing noise item tap zero setting, remaining tap reconstructs noise jamming item, estimates timing offset after deducting described noise jamming item in described serial mean;

Described utilize noise item tap zero setting after remaining tap reconstruct noise jamming item, be specially by noise jamming item described in following formula reconstruct:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\frac{2\mathrm{\π}\·n\·\mathrm{\Δk}}{N}}$

Wherein I is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

2. the method for claim 1, is characterized in that, describedly estimates timing offset deduct described noise jamming item in serial mean after, specifically comprises:

In serial mean, deduct the noise jamming item reconstructing, obtain revised serial mean Q;

Utilize revised serial mean Q to estimate timing offset Δ n by following formula:

$\mathrm{\Δn}=\frac{\mathrm{angle}\left(Q\right)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}$

Wherein, angle is for asking phase bit arithmetic, N _fFTfor carrying out counting of fast Fourier transform FFT window that OFDM symbolic blocks uses when synchronous, Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

3. the method for claim 1, is characterized in that, the channel frequency domain response of described estimating pilot frequency position, specifically comprises:

Frequency domain reception data and the local pilot frequency sequence of pilot frequency locations on same OFDM symbol are multiplied each other, obtain the channel frequency domain response of the pilot frequency locations of estimating.

4. the method for claim 1, is characterized in that, the described channel frequency domain response by estimation transforms to time domain and obtains channel time domain response, specifically comprises:

By Inverse Discrete Fourier Transform IDFT, the channel frequency domain response of estimation is transformed to time domain and obtain channel time domain response.

5. the method for claim 1, is characterized in that, the channel time domain response transform obtaining after noise item tap zero setting, to frequency domain, is specifically comprised:

The channel time domain response obtaining after noise item tap zero setting is carried out to leaf transformation DFT in N point discrete Fourier and transform to frequency domain, described N is the number of OFDM symbol subcarrier.

6. the method for claim 1, is characterized in that, described to converting the pilot tone point at interval in rear channel frequency domain response, to carry out conjugation relevant, specifically comprises:

To converting pilot tone point adjacent in rear channel frequency domain response, to carry out conjugation relevant.

7. an estimating system for uplink timing error, comprising: the first channel frequency domain response estimation unit, for utilizing the frequency domain data of orthogonal frequency division multiplex OFDM symbol and the channel frequency domain response of local pilot data estimating pilot frequency position of reception; The first channel time domain response unit, obtains channel time domain response for the channel frequency domain response of estimation being transformed to time domain; Tap zero setting unit, for determining it is the tap zero setting of noise item by described channel time domain response; Second channel frequency domain response unit, for the channel time domain response transform obtaining after noise item tap zero setting is arrived to frequency domain, obtains converting rear channel frequency domain response; Conjugation correlation unit, carries out conjugation for the pilot tone point to converting rear channel frequency domain response interval and is correlated with; All value cells, for asking sequence average to obtain serial mean to the relevant result of conjugation; It is characterized in that, also comprise:

Interference reconstruction unit, reconstructs noise jamming item for remaining tap after utilizing noise item tap zero setting;

Timing offset estimation unit is estimated timing offset after deducting described noise jamming item in described serial mean;

Described interference reconstruction unit is specifically for by noise jamming item described in following formula reconstruct:

$I=\frac{{\mathrm{\σ}}^2}{N}\underset{n\∈\mathrm{\Ω}}{\mathrm{\Σ}}{e}^{-j\frac{2\mathrm{\π}\·n\·\mathrm{\Δk}}{N}}$

Wherein I is noise jamming item, σ ²for the power of additive white Gaussian noise, N is the number of OFDM symbol subcarrier, the tap position set that Ω retains after for tap zero setting, and Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

8. system as claimed in claim 7, is characterized in that, described timing offset estimation unit specifically comprises:

Amendment unit, for deduct the noise jamming item reconstructing in serial mean, obtains revised serial mean Q;

Timing offset computing unit, for utilizing revised serial mean Q to estimate timing offset Δ n by following formula:

$\mathrm{\Δn}=\frac{\mathrm{angle}\left(Q\right)*N_{\mathrm{FFT}}}{2\mathrm{\π\Δk}}$

Wherein, angle is for asking phase bit arithmetic, N _fFTfor carrying out counting of fast Fourier transform FFT window that OFDM symbolic blocks uses when synchronous, Δ k is the interval of carrying out between pilot tone point that conjugation is relevant.

Images