CN102035764A - Channel estimation device and channel estimation method of nonideal synchronous single-carrier frequency division multiple access system - Google Patents

Abstract

The invention discloses a channel estimation device and a channel estimation method of a nonideal synchronous single-carrier frequency division multiple access system. The channel estimation method comprises the following steps of: converting a received time domain base band signal into frequency domain data by a time domain signal processing unit, separating data of each user from the frequency domain data and outputting; acquiring a pilot frequency character in the user data from the user data by a pilot frequency position processing unit, and estimating the channel at a pilot frequency position by utilizing the pilot frequency character in the user data to obtain a channel estimation value of the pilot frequency position; carrying out time bias estimation compensation, noise reduction, filtering and time bias recovery on the channel estimation value of the pilot frequency position, then carrying out frequency bias estimation and initial phase difference compensations and outputting to the channel estimation unit of the data position; and estimating the channel estimation value of the user data by a data position processing unit according to the channel estimation value of the pilot frequency position, wherein the channel estimation value of the pilot frequency position is received from the pilot frequency position processing unit. The invention overcomes the defects of high complexity and sensitivity to time frequency bias of the traditional channel estimation algorithm.

Description

The channel estimating apparatus of imperfect synchronous single carrier-frequency division multiple access system and method

Technical field

The present invention relates to communication technical field, relate in particular to a kind of channel estimating apparatus and method of imperfect synchronous single carrier-frequency division multiple access system.

Background technology

SC-FDMA (Single Carrier-Frequency Division Multiplex Access, single-carrier frequency division multiple access) technology has lower PAPR (Peak Average Power Ratio, peak-to-average power compares) and reach and OFDM (Orthogonal Frequency Division Multiplexing, OFDM) the suitable performance of technology, so in 3G LTE (Long Term Evolution, Long Term Evolution) system, obtained application.Yet the raising of the systematic function that wants to realize that fully the SC-FDMA technology is brought also needs to carry out the research of related key technical, and channel estimation technique is exactly one of them.

In the existing relevant patent documentation of channel estimating, major part all is the method that provides channel estimating from the algorithm angle, the complexity height, and operand is big, and the influence that the existence of frequency deviation brings channel estimating when not considering.And in practical communication system, because the influence of non-ideal factor, inclined to one side and frequency deviation when receiver exists inevitably is so traditional channel estimation methods often exists performance flat in real system.

Summary of the invention

In view of above-mentioned analysis, the present invention aims to provide a kind of channel estimating apparatus and method of imperfect synchronous single carrier-frequency division multiple access system, in order to solving the channel estimation method complexity that exists in the prior art, to the time frequency deviation sensitive issue.

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

The invention provides a kind of channel estimating apparatus of imperfect synchronous single carrier-frequency division multiple access system, comprising: time domain baseband signal processing unit, pilot frequency locations processing unit and Data Position processing unit, wherein,

Described time domain baseband signal processing unit, the time domain baseband signal that is used for it is received is transformed to frequency domain data, and isolates each user data and export from described frequency domain data;

Described pilot frequency locations processing unit and described Data Position processing unit carry out following processing respectively to each user data:

Described pilot frequency locations processing unit is used for obtaining from described time domain baseband signal processing unit the frequency pilot sign of this user data, utilizes the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated to obtain pilot frequency locations place channel estimation value; Inclined to one side estimation compensation when described pilot frequency locations place channel estimation value is carried out, noise reduction filtering and the time recover partially, carry out then exporting to described Data Position processing unit after the compensation of frequency offset estimating and frequency pilot sign start-up phase potential difference;

Described Data Position processing unit is used for estimating according to the pilot frequency locations channel estimation value that receives from described pilot frequency locations processing unit the channel estimation value of this user data.

Further, described time domain baseband signal processing unit specifically comprises: fast Fourier transform module and frequency domain data separation module, wherein,

Described fast Fourier transform module, the time domain baseband signal that is used for that it is received are carried out the fast Fourier transform processing, and will export to described frequency domain data separation module through the frequency domain data that fast Fourier transform obtains;

Described frequency domain data separation module is used to receive the frequency domain data that described fast Fourier transform module is imported, and isolates each user data and output from this frequency domain data, and wherein the frequency pilot sign in the user data is exported to described pilot frequency locations processing unit.

Further, described pilot frequency locations processing unit specifically comprises: pilot frequency locations place channel estimation module, frequency deviation estimating modules and frequency pilot sign start-up phase potential difference compensating module, wherein,

Described pilot frequency locations place channel estimation module, be used for after getting access to the frequency pilot sign of user data from described time domain baseband signal processing unit, utilize the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated, and inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain carried out, noise reduction filtering and the time export to described frequency deviation estimating modules and described frequency pilot sign start-up phase potential difference compensating module after recovering partially;

Described frequency deviation estimating modules, be used for carrying out obtaining the frequency offset estimating value after the residual frequency deviation value is estimated, and the frequency offset estimating value that obtains is exported to described frequency pilot sign start-up phase potential difference compensating module and described Data Position processing unit according to the pilot frequency locations place channel estimation value of described pilot frequency locations place channel estimation module output;

Described frequency pilot sign start-up phase potential difference compensating module, the pilot frequency locations place channel estimation value that is used to utilize described frequency offset estimating value that described pilot frequency locations place channel estimation module is exported carries out the compensation of frequency pilot sign start-up phase potential difference, and the pilot frequency locations place channel estimation value after the compensation of output process frequency pilot sign start-up phase potential difference is to described Data Position processing unit.

Further, described pilot frequency locations place channel estimation module specifically comprises: pilot frequency locations place least square channel estimation module, the time inclined to one side estimation module, partial compensation for the time module, noise reduction process module and the time recover module partially, wherein,

Described pilot frequency locations place least square channel estimation module, be used for estimating by the least square channel estimation method that output pilot frequency locations place least square channel estimation value is given inclined to one side estimation module and described partial compensation for the time module when described according to the frequency pilot sign of the user data that receives and this user's local frequency pilot sign;

Inclined to one side estimation module when described is estimated when being used for carrying out according to the described pilot frequency locations place least square channel estimation value that receives partially, and the time inclined to one side estimated value that estimates is transferred to described partial compensation for the time module and recovers module when described partially;

Described partial compensation for the time module, be used for according to receive described the time inclined to one side estimated value described pilot frequency locations place least square channel estimation value is carried out partial compensation for the time, the pilot frequency locations place channel estimation value after the output compensation;

Described noise reduction process module is used for the noise of the pilot frequency locations place channel estimation value after the described compensation being suppressed the pilot frequency locations place channel estimation value behind the noise that is inhibited;

Partially recover module when described, be used for recovering the pilot frequency locations place channel estimation value after output recovers partially when when the described inclined to one side estimated value pilot frequency locations place channel estimation value after to described inhibition noise carries out.

Further, described noise reduction process module specifically is used for, and the noise in the pilot frequency locations place channel estimation value after adopting limited long impulse response filtering denoise processing method to described compensation suppresses the pilot frequency locations place channel estimation value behind the noise that is inhibited.

Further, described Data Position processing unit specifically comprises: interpolating module and phase difference recover module, wherein,

Described interpolating module, be used to utilize interpolation algorithm that described pilot frequency locations place channel estimation value through the compensation of frequency pilot sign start-up phase potential difference is estimated to obtain the channel estimation value of this user data, and will estimate that the channel estimation value of this user data of obtaining exports to described phase difference recovery module through interpolation;

Described phase difference recovers module, is used to utilize described frequency offset estimating value that the channel estimation value of this user data of described interpolating module output is carried out phase difference and recovers, and obtains the channel estimation value of this user data of recovering through phase difference.

The present invention also provides a kind of channel estimation methods of imperfect synchronous single carrier-frequency division multiple access system, and described method comprises:

Steps A: will carry out obtaining frequency domain data after fast Fourier transform is handled from the time domain baseband signal that receiver receives, and from this frequency domain data, isolate each user data;

Each user data is carried out the processing of step B and step C respectively:

Step B: according to the frequency pilot sign in this user data that obtains the channel at pilot frequency locations place is estimated, and inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain carried out, noise reduction filtering, the time recovery partially, frequency offset estimating and the compensation deals of frequency pilot sign start-up phase potential difference;

Step C: utilize described through out-of-date inclined to one side estimation compensation, noise reduction filtering, the time pilot frequency locations place channel estimation value after recovery partially, frequency offset estimating and the compensation deals of frequency pilot sign start-up phase potential difference estimate the channel estimation value of this user data.

Further, described step B specifically comprises:

Step B1: utilize the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated, inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain is carried out, noise reduction filtering and the time recover partially, obtain through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially;

Step B2: according to through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially carry out the residual frequency deviation value and estimate, obtain the frequency offset estimating value;

Step B3: utilize described frequency offset estimating value to described through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially carry out the compensation of frequency pilot sign start-up phase potential difference, obtain through the pilot frequency locations place channel estimation value after the compensation of frequency pilot sign start-up phase potential difference.

Further, described step B1 specifically comprises:

Step B11: obtain pilot frequency locations place least square channel estimation value after utilizing frequency pilot sign in this user data and this user's local frequency pilot sign to estimate by the least square channel estimation method, and estimate inclined to one side estimated value when obtaining partially when described pilot frequency locations place least square channel estimation value carried out;

Step B12: inclined to one side estimated value is carried out partial compensation for the time to described pilot frequency locations place least square channel estimation value, the pilot frequency locations place channel estimation value after being compensated when described;

Step B13: the noise in the pilot frequency locations place channel estimation value after the described compensation is suppressed the pilot frequency locations place channel estimation value behind the noise that is inhibited;

Step B14: recover partially when the pilot frequency locations place channel estimation value of inclined to one side estimated value after to described inhibition noise carries out when described.

Further, described step B13 specifically comprises:

Noise in the pilot frequency locations place channel estimation value after adopting limited long impulse response filtering denoise processing method to described compensation suppresses, the pilot frequency locations place channel estimation value behind the noise that is inhibited.

Further, described step C specifically comprises:

Step C1: utilize interpolation algorithm to estimating through the pilot frequency locations place channel estimation value after the compensation of frequency pilot sign start-up phase potential difference among the described step B3, obtain channel estimation value through this user after the interpolation estimation;

Step C2: utilize described frequency offset estimating value that described channel estimation value through this user after the interpolation estimation is carried out phase difference and recover, obtain channel estimation value through this user data of phase difference recovery.

Beneficial effect of the present invention is as follows:

The present invention has overcome existing channel algorithm for estimating complexity height, to the time frequency deviation sensitivity shortcoming.

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

Description of drawings

Fig. 1 is in the embodiment of the invention, the frame structure schematic diagram of 3G LTE system;

Fig. 2 is the structural representation of the channel estimating apparatus of the described imperfect synchronous SC-FDMA of embodiment of the invention system;

Fig. 3 is in the embodiment of the invention, the structural representation of described pilot frequency locations place channel estimation module;

Fig. 4 is in the embodiment of the invention, the structural representation of described Data Position processing unit;

Fig. 5 is the schematic flow sheet of the channel estimation methods of the described imperfect synchronous SC-FDMA of embodiment of the invention system.

Embodiment

The objective of the invention is to, from the angle of system, the influence that frequency deviation causes channel estimating when having taken into full account realizes that a kind of realization is simple, flexible design and to the time the insensitive channel estimation methods of frequency deviation and device.

Specifically describe preferential embodiment of the present invention below in conjunction with accompanying drawing, wherein, accompanying drawing constitutes the application's part, and is used from explaination principle of the present invention with embodiments of the invention one.For clear and simplification purpose, when it may make theme of the present invention smudgy, with specifying in detail of known function and structure in the omission device described herein.

1 be elaborated at first in conjunction with the accompanying drawings to 4 pairs of described devices of the embodiment of the invention of accompanying drawing.

Suppose that the SC-FDMA system is a 3G LTE system, its frame structure as shown in Figure 1, system adopts block pilot configuration, every seven adjacent SC_FDMA symbols comprise a frequency pilot sign and six data symbols, and the channel estimating at the channel estimating at pilot frequency locations place and Data Position place will be finished by the information that this frequency pilot sign provides.

As shown in Figure 2, Fig. 2 is the structural representation of the channel estimating apparatus of imperfect synchronous SC-FDMA system, specifically can comprise: time domain baseband signal processing unit, pilot frequency locations processing unit and Data Position processing unit; Below will be described in detail respectively each functional unit.

(1) time domain baseband signal processing unit, the time domain baseband signal that is used for it is received is transformed to frequency domain data, from this frequency domain data, isolate each user data (comprising frequency pilot sign and data symbol) and output, wherein, the frequency pilot sign in the user data is exported to the pilot frequency locations processing unit.

The time domain baseband signal processing unit specifically can comprise: FFT (Fast Fourier Transform, fast Fourier transform) module and frequency domain data separation module, wherein,

(1) FFT module, when receiver was input to the FFT module with its time domain baseband signal that receives, the FFT module carried out obtaining frequency domain data after FFT handles to the time domain baseband signal, then this frequency domain data is exported to the frequency domain data separation module.

(2) frequency domain data separation module is responsible for receiving the frequency domain data that the FFT module is imported, and isolates each user data from this frequency domain data, and respectively the frequency pilot sign in each user data is exported to the pilot frequency locations processing unit; Owing to be frequency domain data, thus only need promptly can realize the separation of each user data according to the data that the shared resource location of each user extracts response, and the separating of data symbol in the user data and frequency pilot sign.

(2) pilot frequency locations processing unit, after obtaining the frequency pilot sign of certain user data, utilize the frequency pilot sign in this user data that the channel at this user's pilot frequency locations place is estimated to obtain pilot frequency locations place channel estimation value from described time domain baseband signal processing unit; Inclined to one side estimation compensation when pilot frequency locations place channel estimation value is carried out, noise reduction filtering and and the time recover partially, carry out then exporting to described Data Position processing unit after the compensation of frequency offset estimating and frequency pilot sign start-up phase potential difference.

The pilot frequency locations processing unit specifically can comprise: pilot frequency locations place channel estimation module, frequency deviation estimating modules and frequency pilot sign start-up phase potential difference compensating module, wherein,

(1) pilot frequency locations place channel estimation module, behind the frequency pilot sign that obtains from described time domain baseband signal processing unit certain user data, utilize the frequency pilot sign in this user data that the channel at this user's pilot frequency locations place is estimated, and when the pilot frequency locations place channel estimation value of estimating to obtain carried out inclined to one side estimation compensation noise reduction filtering handle and the time export to described frequency deviation estimating modules after the recovery partially; The concrete structure of pilot frequency locations place channel estimation module is referring to Fig. 3.

As shown in Figure 3, Fig. 3 is the structural representation of pilot frequency locations place channel estimation module, mainly comprises: the LS of pilot frequency locations place channel estimation module, the time inclined to one side estimation module, partial compensation for the time module, noise reduction process module and the time recover module partially; Below will be described in detail respectively each module.

The LS of pilot frequency locations place (least square) estimation module utilizes the frequency pilot sign of certain user data that receives and this user's local frequency pilot sign to estimate by the LS algorithm for estimating, the output pilot frequency locations LS of place channel estimation value

Inclined to one side estimation module and partial compensation for the time module when giving;

Be exactly specifically, when supposing the system exists under the frequency deviation condition certain user data be:

$y^{\left(i\right)}=e^{j2\mathrm{\π}\·\mathrm{\Δ}{f}^{\left(i\right)}\·t^{\left(i\right)}}.\left(\mathrm{diag}\right\{h^{\left(i\right)}\}\·\mathrm{\Λ}\left({\mathrm{\Δt}}^{\left(i\right)}\right)).(F\·\mathrm{\Λ}\left({\mathrm{\Δf}}^{\left(i\right)}\right)\·F^H)\·s^{\left(i\right)}+n^{\left(i\right)}---\left(1\right);$

In the formula, diag{} represents { } is converted to diagonal matrix, and i represents to constitute SC_FDMA symbol sequence number (comprising data symbol and the frequency pilot sign) i=0 of user data, and 1,2 ..., 13, Δ t ⁽ⁱ⁾, Δ f ⁽ⁱ⁾And t ⁽ⁱ⁾When representing the originator system in i the symbol duration respectively, frequency deviation and the initial moment (part except that CP), s ⁽ⁱ⁾Be illustrated in i signal vector (except that CP) that SC_FDMA equivalence baseband sampling sequence constitutes under the own system clock of making a start, h ⁽ⁱ⁾The channel frequency response value of representing i SC_FDMA symbol correspondence, n ⁽ⁱ⁾The interchannel noise sequence of representing i SC_FDMA symbol correspondence is done following notation convention simultaneously in the formula

(2)；

This user data pilot frequency locations LS of place channel estimation value then

For,

${\hat{h}}_{\mathrm{ls}}^{\left(l\right)}={\mathrm{\Λ}}^{*}\left(s\right)\·y^{\left(l\right)}=e^{j2\mathrm{\π}\·\mathrm{\Δ}{f}^{\left(l\right)}\·t^{\left(l\right)}}\·\mathrm{\Λ}\left({\mathrm{\Δt}}^{\left(l\right)}\right)\·h^{\left(l\right)}+{\mathrm{\Λ}}^{*}\left(s\right)\·n^{\left(l\right)}---\left(3\right),$

Wherein, l represents the frequency pilot sign in this user data, l=3,10.

The time inclined to one side estimation module, utilize the LS of pilot frequency locations place channel estimation value

Estimate inclined to one side estimated values theta during output when carrying out partially ₁, and be transferred to the partial compensation for the time module and the time partially recover module;

Be exactly specifically, at the LS of pilot frequency locations place channel estimation value

The basis on, be respectively to sue for peace after the conjugate multiplication successively at interval with δ on the frequency domain according to each pilot tone.The sub-carrier number of supposing CU is N _c, l has for frequency pilot sign

${\mathrm{\β}}_1=\underset{m=1}{\overset{N_c-6}{\mathrm{\Σ}}}[{\hat{h}}_{\mathrm{ls}}^{\left(l_1\right)}\left(m\right)\·\mathrm{conj}\left({\hat{h}}_{\mathrm{ls}}^{\left(l_1\right)}(m+\mathrm{\δ})\right)+{\hat{h}}_{\mathrm{ls}}^{\left(l_2\right)}\left(m\right)\·\mathrm{conj}\left({\hat{h}}_{\mathrm{ls}}^{\left(l_2\right)}(m+\mathrm{\δ})\right)]---\left(4\right);$

(annotating: when having between slot frequency hopping, need carry out phase difference to each slot and ask angle)

To β ₁Ask and remove δ behind the phase angle, can obtain by the time phase deviation θ that causes partially ₁

θ ₁＝angle(β ₁)/δ (5)；

Wherein, phase angle function (radian) is got in angle () expression.

The partial compensation for the time module, inclined to one side estimated values theta when utilizing ₁To the LS of pilot frequency locations place channel estimation value

Carry out partial compensation for the time, the pilot frequency locations place channel estimation value after the output compensation

Be exactly specifically, with this phase deviation θ ₁The LS channel estimation value is given in compensation

Pilot frequency locations place channel estimation value after being compensated

That is,

${\tilde{h}}_{\mathrm{ls}}^{\left(l\right)}=\mathrm{\Λ}\left({\mathrm{\θ}}_1\right)\·{\hat{h}}_{\mathrm{ls}}^{\left(l\right)}---\left(6\right);$

Wherein, $\mathrm{\Λ}\left({\mathrm{\θ}}_1\right)\stackrel{\mathrm{\Δ}}{=}\mathrm{diag}\left\{\right[e^{j{\mathrm{\θ}}_1\·0}\·\·\·e^{j{\mathrm{\θ}}_1\·(N-1)}\left]\right\}.$

The noise reduction process module is to the pilot frequency locations place channel estimation value after the compensation

In noise suppress the pilot frequency locations place channel estimation value behind the noise that is inhibited Wherein, Method is FIR (limited long impulse response) filtering denoise processing method in embodiments of the present invention;

Be exactly specifically, adopt FIR filtering denoise processing method in the embodiment of the invention, the pilot frequency locations place channel estimation value that suppresses behind the noise is,

${\tilde{h}}_{\mathrm{FIR}}^{\left(l\right)}=\mathrm{Toeplitz}\left(f_{\mathrm{half}}\right)*{\tilde{h}}_{\mathrm{ls}}^{\left(l\right)}---\left(7\right);$

Wherein, The vector that expression is made up of P rank FIR filter segment tap coefficient, and the tap coefficient vector of P rank FIR filter is as the formula (23).

The tap coefficient of FIR filter obtains by the following method:

A, determine the bandwidth of low-pass FIR filter: because wireless multidiameter delay channel can be regarded sinusoidal wave superimposed the forming by many low frequencies as, therefore, wireless multi channel frequency domain response can be regarded a low frequency signal as.The peak frequency of this signal is decided by the maximum delay of wireless channel.Its normalization bandwidth can be decided by formula (24):

$f_{\mathrm{Normal}}=\frac{{\mathrm{\τ}}_{\max }}{T_{\mathrm{sym}}}---\left(9\right);$

In the formula, τ _MaxThe maximum delay of expression wireless channel, for specific system works environment, its span can be known in advance.T _SymRepresent the time that a symbol continues, determine by system parameters.Like this, the normalization bandwidth of wireless channel just can pre-determine.And the bandwidth of low-pass FIR filter can be according to f _NormalDecide the normalization bandwidth f of FIR filter _{FIR, Normal}As long as satisfy following relation:

f _FIR，Normal＞f _Normal (10)；

The design of B, low pass filter: in the embodiment of the invention, do not have fixing restriction, can choose any one low pass filter, as long as filter satisfies the condition among the A according to the needs of system for FIR Filter Design method;

C: the decision of low-pass FIR filter exponent number: the exponent number of FIR filter has determined the performance of filter on large program very, in general, the exponent number of filter is high more, and the performance of filter is good more.But the increase of filter order will increase computational complexity, and can cause bigger rim effect, thus the exponent number of filter comprehensively the factor of each side freely determine;

In real system, if maximum delay that can estimating system, filter storage that then can pre-designed different bandwidth is moved, then according to real system estimate the time partially value select corresponding filter as the noise reduction process filter, realize the purpose of adaptive-filtering.

The time partially recover module, inclined to one side estimated values theta when utilizing ₁To the pilot frequency locations place channel estimation value behind the inhibition noise

Recover output when carrying out partially

Be exactly specifically, inclined to one side in the time of need recovering the channel estimation results after the noise reduction process after finishing noise reduction process, the pilot frequency locations place channel estimation value after promptly noise reduction in time recovers partially

${\hat{h}}^{\left(l\right)}=\mathrm{\Λ}(-{\mathrm{\θ}}_1)\·{\tilde{h}}^{\left(l\right)}---\left(11\right).$

(2) frequency deviation estimating modules is recovered the pilot frequency locations place channel estimation value of module output partially when utilizing Carry out user's residual frequency deviation value and estimate to obtain the frequency offset estimating value

Be exactly to suppose specifically

l ₁And l ₂Represent a TTI (Transmission Time Interval respectively, time slot) Nei two RS (Reference Signals, pilot tone) sequence number of symbol is utilized the frequency deviation that differs estimating system of the channel estimation value on the subcarrier of the same frequency location on the different RS.Because adjacent two RS differ 0.5ms in time, consider " coiling " factor of phase place, the scope of this frequency deviation valuation will [1000,1000), promptly have

$\mathrm{\Δ}\hat{f}=\frac{\mathrm{angle}({\left({\hat{h}}^{\left(l_1\right)}\right)}^H\·{\hat{h}}^{\left(l_2\right)})}{2\mathrm{\π}\·(t^{\left(l_2\right)}-t^{\left(l_1\right)})}\left(\mathrm{Hz}\right)---\left(12\right).$

(3) frequency pilot sign start-up phase potential difference compensating module utilizes the frequency offset estimating value

To pilot frequency locations place channel estimation value

Carry out start-up phase potential difference compensation between frequency pilot sign, and the pilot frequency locations place channel estimation value after the compensation of output process frequency pilot sign start-up phase potential difference

That is,

${\hat{h}}^{\left(l\right)}=e^{j{\mathrm{\θ}}_2\·t^{\left(l\right)}}\·{\hat{h}}^{\left(l\right)}---\left(13\right).$

(3) Data Position processing unit, according to through out-of-date inclined to one side estimation compensation, noise reduction filtering, the time recovery partially, frequency offset estimating and the compensation of frequency pilot sign start-up phase potential difference pilot frequency locations place channel estimation value

Estimate, estimate the channel estimation value of this user data

(comprising Data Position channel estimation value and pilot frequency locations place channel estimation value); Be exactly after through frequency offset estimating and the compensation of frequency pilot sign start-up phase potential difference, to have obtained the frequency offset estimating value of system specifically

And the pilot frequency locations place channel estimation value of process frequency pilot sign start-up phase potential difference compensation

Utilize interpolation algorithm and phase difference recovery algorithms just can estimate the channel estimation value of this final user data

As shown in Figure 4, Fig. 4 is the structural representation of Data Position processing unit, mainly comprises: interpolating module and phase difference recover module and form; Wherein,

Interpolating module is according to the pilot frequency locations channel estimation value that obtains from frequency pilot sign start-up phase potential difference compensating module

Utilize interpolation algorithm to carry out the channel estimation value that interpolation estimates this user data

Wherein, the value of i is 0～13, when the value of i is 3 and 10,

Be pilot frequency locations place channel estimation value, when i gets other values,

Be Data Position place channel estimation value, that is,

${\tilde{h}}^{\left(i\right)}=(1-\frac{(l_2-i)}{l_2-l_1}){\tilde{h}}^{\left(l_2\right)}+\frac{(l_2-i)}{l_2-l_1}{\tilde{h}}^{\left(l_1\right)},i\∈\{\mathrm{0,1},\·\·\·,13\}---\left(14\right);$

Phase difference recovers module, the frequency offset estimating value of utilizing formula (30) and obtaining from frequency deviation estimating modules

For the channel estimation value (comprising pilot frequency locations place channel estimation value and Data Position place channel estimation value) of this user data all adds the start-up phase parallactic angle, obtain the channel estimation value of this user data of recovering through phase difference

Be the channel estimation value of our final conceivable this user data, that is,

${\hat{h}}^{\left(i\right)}=e^{j2\mathrm{\π\Δ}\hat{f}{t}^{\left(i\right)}}{\tilde{h}}^{\left(i\right)}---\left(15\right).$

Channel estimation methods below in conjunction with the described imperfect synchronous SC-FDMA of 5 pairs of embodiment of the invention of accompanying drawing system is elaborated.

As shown in Figure 5, Fig. 5 is the schematic flow sheet of the channel estimation methods of the described imperfect synchronous SC-FDMA of embodiment of the invention system, specifically can comprise the steps:

Step 501: when supposing the system exists under the frequency deviation condition certain user data be:

$y^{\left(i\right)}=e^{j2\mathrm{\π}\·\mathrm{\Δ}{f}^{\left(i\right)}\·t^{\left(i\right)}}.\left(\mathrm{diag}\right\{h^{\left(i\right)}\}\·\mathrm{\Λ}\left({\mathrm{\Δt}}^{\left(i\right)}\right)).(F\·\mathrm{\Λ}\left({\mathrm{\Δf}}^{\left(i\right)}\right)\·F^H)\·s^{\left(i\right)}+n^{\left(i\right)}---\left(16\right)$

(1) in the formula, diag{} represents { } is converted to diagonal matrix, Δ t ⁽ⁱ⁾, Δ f ⁽ⁱ⁾And t ⁽ⁱ⁾When representing the originator system in i the SC_FDMA symbol duration respectively, frequency deviation and the initial moment (part except that CP), s ⁽ⁱ⁾Be illustrated in i signal vector (except that CP) that SC_FDMA equivalence baseband sampling sequence constitutes under the own system clock of making a start.Do following notation convention in the formula simultaneously

Step 502: the LS of pilot frequency locations place channel estimating;

Frequency position LS channel estimation value

For,

${\hat{h}}_{\mathrm{ls}}^{\left(l\right)}={\mathrm{\Λ}}^{*}\left(s\right)\·y^{\left(l\right)}=e^{j2\mathrm{\π}\·\mathrm{\Δ}{f}^{\left(l\right)}\·t^{\left(l\right)}}\·\mathrm{\Λ}\left({\mathrm{\Δt}}^{\left(l\right)}\right)\·h^{\left(l\right)}+{\mathrm{\Λ}}^{*}\left(s\right)\·n^{\left(l\right)}---\left(18\right)$

Step 503: in time, estimated and compensation partially

At the LS of pilot frequency locations place channel estimation value

The basis on, be respectively to sue for peace after the conjugate multiplication successively at interval with δ on the frequency domain according to each pilot tone.The sub-carrier number of supposing CU is N _c, l has for symbol

${\mathrm{\β}}_1=\underset{m=1}{\overset{N_c-6}{\mathrm{\Σ}}}[{\hat{h}}_{\mathrm{ls}}^{\left(l_1\right)}\left(m\right)\·\mathrm{conj}\left({\hat{h}}_{\mathrm{ls}}^{\left(l_1\right)}(m+\mathrm{\δ})\right)+{\hat{h}}_{\mathrm{ls}}^{\left(l_2\right)}\left(m\right)\·\mathrm{conj}\left({\hat{h}}_{\mathrm{ls}}^{\left(l_2\right)}(m+\mathrm{\δ})\right)]---\left(19\right)$

(annotating: when having between slot frequency hopping, need carry out phase difference to each slot and ask angle)

To β ₁Ask and remove δ behind the phase angle, can obtain by the time phase deviation θ that causes partially ₁

θ ₁＝angle(β ₁)/δ (20)

Wherein phase angle function (radian) is got in angle () expression, and this phase deviation is compensated the channel estimation value to LS

${\tilde{h}}_{\mathrm{ls}}^{\left(l\right)}=\mathrm{\Λ}\left({\mathrm{\θ}}_1\right)\·{\hat{h}}_{\mathrm{ls}}^{\left(l\right)}---\left(21\right)$

Wherein, $\mathrm{\Λ}\left({\mathrm{\θ}}_1\right)\stackrel{\mathrm{\Δ}}{=}\mathrm{diag}\left\{\right[e^{j{\mathrm{\θ}}_1\·0}\·\·\·e^{j{\mathrm{\θ}}_1\·(N-1)}\left]\right\}.$

Step 504:FIR filtering noise reduction process;

${\tilde{h}}_{\mathrm{FIR}}^{\left(l\right)}=\mathrm{Toeplitz}\left(f\right)*{\tilde{h}}_{\mathrm{ls}}^{\left(l\right)}---\left(22\right)$

Wherein,

The vector that expression is made up of the tap coefficient of p rank FIR filter.The tap coefficient vector of p rank FIR filter as the formula (23).

Wherein, the tap coefficient of FIR filter obtains by the following method:

A, determine the bandwidth of low-pass FIR filter; Because wireless multidiameter delay channel can be regarded sinusoidal wave superimposed the forming by many low frequencies as, therefore, wireless multi channel frequency domain response can be regarded a low frequency signal as.The peak frequency of this signal is decided by the maximum delay of wireless channel.Its normalization bandwidth can be decided by formula (24):

$f_{\mathrm{Normal}}=\frac{{\mathrm{\τ}}_{\max }}{T_{\mathrm{sym}}}---\left(24\right)$

In the formula, τ _MaxThe maximum delay of expression wireless channel, for specific system works environment, its span can be known in advance.T _SymRepresent the time that a symbol continues, determine by system parameters.Like this, the normalization bandwidth of wireless channel just can pre-determine.And the bandwidth of low-pass FIR filter can be according to f _NormalDecide the normalization bandwidth f of FIR filter _{FIR, Normal}As long as satisfy following relation:

f _FIR，Normal＞f _Normal (25)

The method for designing of B, low-pass FIR filter; Here, FIR Filter Design method does not have fixing restriction, can choose any one low pass filter according to the needs of system.As long as filter satisfies the condition among the A.

The decision of C, low-pass FIR filter exponent number; The exponent number of FIR filter has determined the performance of filter on large program very, in general, the exponent number of filter is high more, and the performance of filter is good more.But the increase of filter order will increase computational complexity, and can cause bigger rim effect, thus the exponent number of filter comprehensively the factor of each side freely determine.

In real system, if maximum delay that can estimating system, filter storage that then can pre-designed different bandwidth is moved, then according to real system estimate the time partially value select corresponding filter as the noise reduction process filter, realize the purpose of adaptive-filtering.

Step 505: in time, recover partially;

After finishing noise reduction process, need the time partially recover to go back.Inclined to one side in the time of can recovering the channel estimation results after the noise reduction process.

${\hat{h}}^{\left(l\right)}=\mathrm{\Λ}(-{\mathrm{\θ}}_1)\·{\tilde{h}}^{\left(l\right)}---\left(26\right)$

Step 506: frequency offset estimating;

Suppose

l ₁And l ₂The sequence number of representing two RS symbols in the TTI is respectively utilized the frequency deviation that differs estimating system of the channel estimation value on the subcarrier of the same frequency location on the different RS.Because adjacent two RS differ 0.5ms in time, consider " coiling " factor of phase place, the scope of this frequency deviation valuation will [1000,1000), promptly have

$\mathrm{\Δ}\hat{f}=\frac{\mathrm{angle}({\left({\hat{h}}^{\left(l_1\right)}\right)}^H\·{\hat{h}}^{\left(l_2\right)})}{2\mathrm{\π}\·(t^{\left(l_2\right)}-t^{\left(l_1\right)})}\left(\mathrm{Hz}\right)---\left(27\right)$

Step 507: frequency pilot sign start-up phase potential difference compensation;

${\hat{h}}^{\left(l\right)}=e^{j{\mathrm{\θ}}_2\·t^{\left(l\right)}}\·{\hat{h}}^{\left(l\right)}---\left(28\right)$

Step 508: after through frequency offset estimating and the compensation of frequency pilot sign start-up phase potential difference, obtained the frequency deviation value of system And the pilot frequency locations place channel estimation value of process phase difference compensation The channel estimation value of this user data so

(comprising Data Position place channel estimation value and pilot frequency locations place channel estimation value) can obtain by following interpolation algorithm:

${\tilde{h}}^{\left(i\right)}=(1-\frac{(l_2-i)}{l_2-l_1}){\tilde{h}}^{\left(l_2\right)}+\frac{(l_2-i)}{l_2-l_1}{\tilde{h}}^{\left(l_1\right)},i\∈\{\mathrm{0,1},\·\·\·,13\}---\left(29\right)$

Step 509: utilize formula (30) to be the channel estimation value of this user data at last

Add the start-up phase parallactic angle, carry out phase difference and recover, get the channel estimation value of our final this user data that needs

${\hat{h}}^{\left(i\right)}=e^{j2\mathrm{\π\Δ}\hat{f}{t}^{\left(i\right)}}{\tilde{h}}^{\left(i\right)}---\left(30\right).$

The above is that example has been advanced explanation with the 3G LTE system among the SC-FDMA only, but it will be recognized by those skilled in the art that the embodiment of the invention is equally applicable to the other system among the SC-FDMA that processing procedure is identical, herein explanation no longer one by one.

In sum, the embodiment of the invention provides a kind of channel estimating apparatus and method of imperfect synchronous single carrier-frequency division multiple access system, has overcome existing channel algorithm for estimating complexity height, to the time frequency deviation sensitivity shortcoming.

The present invention only needs less cost can realize the function of data transaction by simple FFT conversion; Before noise reduction, add the processing of fashionable inclined to one side estimation compensation, overcome the conventional channel algorithm for estimating to the time responsive partially shortcoming; By in noise reduction process, adopting the filtering noise-reduction method in the signal processing theory, and provide the parameter of Design of Filter, made the noise reduction process simple and flexible; Before linear interpolation, carry out simultaneously the compensation of frequency offset estimating and pilot symbol phases difference, overcome the shortcoming of conventional channel algorithm for estimating, improved the channel estimation method anti-jamming capacity the frequency deviation sensitivity.

There is not the problem of Model Matching in this method and apparatus, and only needs less priori, so have stronger practicality.

Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with the general calculation device, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation element forms, optionally, they can be realized with the executable program code of calculation element, thereby they are stored in the storage device, carry out by calculation element.Perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.The invention is not restricted to any specific hardware and software combination like this.

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

Claims (11)

1. the channel estimating apparatus of an imperfect synchronous single carrier-frequency division multiple access system is characterized in that, comprising: time domain baseband signal processing unit, pilot frequency locations processing unit and Data Position processing unit, wherein,

Described time domain baseband signal processing unit, the time domain baseband signal that is used for it is received is transformed to frequency domain data, and isolates each user data and export from described frequency domain data;

Described pilot frequency locations processing unit and described Data Position processing unit carry out following processing respectively to each user data:

Described pilot frequency locations processing unit is used for obtaining from described time domain baseband signal processing unit the frequency pilot sign of this user data, utilizes the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated to obtain pilot frequency locations place channel estimation value; Inclined to one side estimation compensation when described pilot frequency locations place channel estimation value is carried out, noise reduction filtering and the time recover partially, carry out then exporting to described Data Position processing unit after the compensation of frequency offset estimating and frequency pilot sign start-up phase potential difference;

Described Data Position processing unit is used for estimating according to the pilot frequency locations channel estimation value that receives from described pilot frequency locations processing unit the channel estimation value of this user data.

2. device according to claim 1 is characterized in that, described time domain baseband signal processing unit specifically comprises: fast Fourier transform module and frequency domain data separation module, wherein,

Described fast Fourier transform module, the time domain baseband signal that is used for that it is received are carried out the fast Fourier transform processing, and will export to described frequency domain data separation module through the frequency domain data that fast Fourier transform obtains;

Described frequency domain data separation module is used to receive the frequency domain data that described fast Fourier transform module is imported, and isolates each user data and output from this frequency domain data, and wherein the frequency pilot sign in the user data is exported to described pilot frequency locations processing unit.

3. device according to claim 1 is characterized in that, described pilot frequency locations processing unit specifically comprises: pilot frequency locations place channel estimation module, frequency deviation estimating modules and frequency pilot sign start-up phase potential difference compensating module, wherein,

Described pilot frequency locations place channel estimation module, be used for after getting access to the frequency pilot sign of user data from described time domain baseband signal processing unit, utilize the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated, and inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain carried out, noise reduction filtering and the time export to described frequency deviation estimating modules and described frequency pilot sign start-up phase potential difference compensating module after recovering partially;

Described frequency deviation estimating modules, be used for carrying out obtaining the frequency offset estimating value after the residual frequency deviation value is estimated, and the frequency offset estimating value that obtains is exported to described frequency pilot sign start-up phase potential difference compensating module and described Data Position processing unit according to the pilot frequency locations place channel estimation value of described pilot frequency locations place channel estimation module output;

Described frequency pilot sign start-up phase potential difference compensating module, the pilot frequency locations place channel estimation value that is used to utilize described frequency offset estimating value that described pilot frequency locations place channel estimation module is exported carries out the compensation of frequency pilot sign start-up phase potential difference, and the pilot frequency locations place channel estimation value after the compensation of output process frequency pilot sign start-up phase potential difference is to described Data Position processing unit.

4. device according to claim 3, it is characterized in that, described pilot frequency locations place channel estimation module specifically comprises: pilot frequency locations place least square channel estimation module, the time inclined to one side estimation module, partial compensation for the time module, noise reduction process module and the time recover module partially, wherein

Described pilot frequency locations place least square channel estimation module, be used for estimating by the least square channel estimation method that output pilot frequency locations place least square channel estimation value is given inclined to one side estimation module and described partial compensation for the time module when described according to the frequency pilot sign of the user data that receives and this user's local frequency pilot sign;

Inclined to one side estimation module when described is estimated when being used for carrying out according to the described pilot frequency locations place least square channel estimation value that receives partially, and the time inclined to one side estimated value that estimates is transferred to described partial compensation for the time module and recovers module when described partially;

Described partial compensation for the time module, be used for according to receive described the time inclined to one side estimated value described pilot frequency locations place least square channel estimation value is carried out partial compensation for the time, the pilot frequency locations place channel estimation value after the output compensation;

Described noise reduction process module is used for the noise of the pilot frequency locations place channel estimation value after the described compensation being suppressed the pilot frequency locations place channel estimation value behind the noise that is inhibited;

Partially recover module when described, be used for recovering the pilot frequency locations place channel estimation value after output recovers partially when when the described inclined to one side estimated value pilot frequency locations place channel estimation value after to described inhibition noise carries out.

5. device according to claim 4, it is characterized in that, described noise reduction process module specifically is used for, noise in the pilot frequency locations place channel estimation value after adopting limited long impulse response filtering denoise processing method to described compensation suppresses, the pilot frequency locations place channel estimation value behind the noise that is inhibited.

6. according to any described device in the claim 3 to 5, it is characterized in that described Data Position processing unit specifically comprises: interpolating module and phase difference recover module, wherein,

Described interpolating module, be used to utilize interpolation algorithm that described pilot frequency locations place channel estimation value through the compensation of frequency pilot sign start-up phase potential difference is carried out the channel estimation value that interpolation estimates to obtain this user data, and will estimate that the channel estimation value of this user data of obtaining exports to described phase difference recovery module through interpolation;

Described phase difference recovers module, is used to utilize described frequency offset estimating value that the channel estimation value of this user data of described interpolating module output is carried out phase difference and recovers, and obtains the channel estimation value of this user data of recovering through phase difference.

7. the channel estimation methods of an imperfect synchronous single carrier-frequency division multiple access system is characterized in that, described method comprises:

Steps A: will carry out obtaining frequency domain data after fast Fourier transform is handled from the time domain baseband signal that receiver receives, and from this frequency domain data, isolate each user data;

Each user data is carried out the processing of step B and step C respectively:

Step B: according to the frequency pilot sign in this user data that obtains the channel at pilot frequency locations place is estimated, and inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain carried out, noise reduction filtering, the time recovery partially, frequency offset estimating and the compensation deals of frequency pilot sign start-up phase potential difference;

Step C: utilize described through out-of-date inclined to one side estimation compensation, noise reduction filtering, the time pilot frequency locations place channel estimation value after recovery partially, frequency offset estimating and the compensation deals of frequency pilot sign start-up phase potential difference estimate the channel estimation value of this user data.

8. method according to claim 7 is characterized in that, described step B specifically comprises:

Step B1: utilize the frequency pilot sign in this user data that the channel at pilot frequency locations place is estimated, inclined to one side estimation compensation when the pilot frequency locations place channel estimation value of estimating to obtain is carried out, noise reduction filtering and the time recover partially, obtain through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially;

Step B2: according to through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially carry out the residual frequency deviation value and estimate, obtain the frequency offset estimating value;

Step B3: utilize described frequency offset estimating value to described through out-of-date inclined to one side estimation compensation, noise reduction filtering and the time pilot frequency locations place channel estimation value that recovers partially carry out the compensation of frequency pilot sign start-up phase potential difference, obtain through the pilot frequency locations place channel estimation value after the compensation of frequency pilot sign start-up phase potential difference.

9. method according to claim 8 is characterized in that, described step B1 specifically comprises:

Step B11: obtain pilot frequency locations place least square channel estimation value after utilizing frequency pilot sign in this user data and this user's local frequency pilot sign to estimate by the least square channel estimation method, and estimate inclined to one side estimated value when obtaining partially when described pilot frequency locations place least square channel estimation value carried out;

Step B12: inclined to one side estimated value is carried out partial compensation for the time to described pilot frequency locations place least square channel estimation value, the pilot frequency locations place channel estimation value after being compensated when described;

Step B13: the noise in the pilot frequency locations place channel estimation value after the described compensation is suppressed the pilot frequency locations place channel estimation value behind the noise that is inhibited;

Step B14: recover partially when the pilot frequency locations place channel estimation value of inclined to one side estimated value after to described inhibition noise carries out when described.

10. method according to claim 8 is characterized in that, described step B13 specifically comprises:

Noise in the pilot frequency locations place channel estimation value after adopting limited long impulse response filtering denoise processing method to described compensation suppresses, the pilot frequency locations place channel estimation value behind the noise that is inhibited.

11. any described method in 10 according to Claim 8 is characterized in that described step C specifically comprises:

Step C1: utilize interpolation algorithm to estimating through the pilot frequency locations place channel estimation value after the compensation of frequency pilot sign start-up phase potential difference among the described step B3, obtain channel estimation value through this user after the interpolation estimation;

Step C2: utilize described frequency offset estimating value that described channel estimation value through this user after the interpolation estimation is carried out phase difference and recover, obtain channel estimation value through this user data of phase difference recovery.

Images