CN103166891A - Channel estimation method used in amplification limiting orthogonal frequency division multiplexing (OFDM) system based on virtual pilot frequency - Google Patents

Abstract

The invention provides a channel estimation method used in an amplification limiting orthogonal frequency division multiplexing (OFDM) system based on virtual pilot frequency. The channel estimation method includes the steps of: equalizing a channel using estimated initial channel information, judging and obtaining transmitted data symbols; selecting a part of or the whole data symbols as virtual pilot frequency, combining the data symbols with pilot frequency of the amplification limiting OFDM system and forming a blending pilot frequency sequence; estimating the current channel by means of the blending pilot frequency sequence again, obtaining channel response of a sub-carrier of the blending pilot frequency; feeding the channel response of the sub-carrier of the blending pilot frequency back to step one, and using the channel response as basis of next channel equalization and data symbol judgment. The channel estimation method is capable of balancing system performance requirements and computation complexity requirements, solving the problems that performance is poor or computation complexity is high in the current published method, and improving system performance compared with method using pilot frequency sub-carrier only because the channel estimation method estimates the channel using a part of data symbols as supplementary means.

Description

The channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency

Technical field

The invention belongs to communication technical field, relate to a kind of channel estimation methods, be specifically related to a kind of channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency.

Background technology

Compare with traditional wireless sensor networks, wireless multimedia sensor network is more paid close attention to the multimedia collections such as audio frequency, video, image, is processed and transmission.The data transfer mode of low power consuming, two-forty becomes one of key technical problem of restriction wireless multimedia sensor network development.OFDM be a kind of efficiently, the data transfer mode of extensive use, be suitable as wireless multimedia sensor network.The OFDM time-domain signal by multichannel separate, modulated subsignal is formed by stacking, and may produce larger peak-to-average force ratio (PAPR).And large peak-to-average force ratio signal is very responsive to the nonlinear characteristic of power amplifier, can cause the error rate of system to raise, and cause larger out-of-band radiation.Also can cause the energy efficiency of power amplifier to descend and energy dissipation when in addition, peak-to-average force ratio is larger.

In the method for the Inhibitory signal peak-to-average force ratio that has existed at present, limit filtration is the most frequently used, the simplest mode.But limit filtration is a kind of nonlinear transformation, can bring nonlinear distortion.This distortion comprises nonlinear distortion and the out-of-band energy emission in frequency band.Nonlinear distortion in frequency band can cause noise energy to increase, and reduces the signal to noise ratio (SNR) of system.And out-of-band radiation can disturb the transfer of data of adjacent channel.In the ofdm system of honeycomb pilot frequency distribution, pilot sub-carrier and data subcarrier are subject to the impact of nonlinear distortion jointly, thereby cause the performance of channel estimating to reduce.For this problem, there has been the channel estimation methods of some anti-Nonlinear perturbations at present, as: S.Schwartz etc. have proposed based on the joint channel estimation of EM-iteration and signal detection algorithm; B.Wu etc. have studied the theoretical channel gain in the amplitude-limiting OFDM system, and have proposed a channel estimation method based on decision-feedback; This algorithm has utilized all data and pilot frequency carrier wave to carry out channel estimating, and then the nonlinear distortion in the estimation of iteration and elimination data symbol, improve systematic function; J.Sterba has proposed an associating Data Detection that is used for ofdm system and channel algorithm for estimating again; The nonlinear transformations that the estimation transmitting terminal of this algorithm iteration produces, and be used for the nonlinear distortion of offset data and frequency pilot sign, frequency pilot sign after compensation is used for obtaining more accurate channel response in next iteration, the data symbol after compensation is used for obtaining data decision more accurately.Above-mentioned algorithm or be to utilize merely pilot sub-carrier to carry out channel estimating, or be to utilize all data and pilot sub-carrier to carry out channel estimating.Utilize merely pilot sub-carrier to carry out the computation complexity of algorithm of channel estimating relatively low, but its performance is relatively relatively poor; Although and the algorithm that utilizes total data and pilot sub-carrier to carry out channel estimating has significantly improved the performance of channel estimating, also brought very large operand simultaneously.

Summary of the invention

The shortcoming of prior art in view of the above, the object of the present invention is to provide a kind of channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency, in order to select flexibly the quantity of virtual pilot frequency, reach the contradiction of flexible balance system performance and computation complexity.

Reach for achieving the above object other relevant purposes, the invention provides a kind of channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency.

A kind of channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency, the method comprises:

Step 1 utilizes the initial channel information of estimating to carry out channel equalization, and adjudicates the data symbol that obtains transmitting;

Step 2, selected part or total data symbol be as virtual pilot frequency, and form the hybrid pilot sequence together with the pilot tone of amplitude-limiting OFDM system; Utilize described hybrid pilot sequence that current channel is estimated again, obtain the channel response at hybrid pilot subcarrier place;

Step 3 feeds back to step 1 with the channel response at described hybrid pilot subcarrier place, as the foundation of next round channel equalization and data symbol judgement.

As a preferred embodiment of the present invention, the specific implementation process of described step 1 is: receiving terminal carries out Amplitude Compensation to the reception signal from ofdm system; Utilize the signal after Amplitude Compensation to carry out initial channel estimation, obtain the channel estimation value at pilot sub-carrier place according to the pilot tone of known ofdm system, and obtain the channel response of all pilot sub-carriers; Signal according to the channel response of all pilot sub-carriers after to Amplitude Compensation carries out channel equalization; The nonlinear distortion of the signal after elimination channel equalization; Extracted data subcarrier from the signal of eliminating nonlinear distortion, judgement obtains the data symbol of transmission.

As another kind of preferred version of the present invention, the implementation procedure of described initial channel estimation is: utilize the LS algorithm to calculate the channel estimation value at pilot sub-carrier place, then calculate the channel response of all pilot sub-carriers by interpolation algorithm.

As another preferred version of the present invention, described interpolation algorithm is DFT interpolation algorithm, linear interpolation algorithm or polynomial interopolation algorithm.

As another preferred version of the present invention, the specific implementation process of described step 2 is: reconfigure out the OFDM symbol according to described data symbol and pilot frequency sequence; Described OFDM symbol is carried out limit filtration, the signal after the acquisition limit filtration; Selected part or total data symbol are as virtual pilot frequency from described data symbol, the pilot tone of described virtual pilot frequency and described ofdm system consists of the hybrid pilot sequence, obtains the channel estimation value at the channel estimation value at virtual pilot frequency subcarrier place and channel response and pilot frequency sequence subcarrier place; According to the Signal estimation nonlinear distortion information after limit filtration, be used for the nonlinear distortion at the step 1 erasure signal of next round iteration.

As another preferred version of the present invention, the process of described limit filtration is: first described OFDM sign reversing to time domain is carried out limit filtration, and then conversion is returned frequency domain, the frequency-region signal after the acquisition amplitude limit, i.e. signal after described limit filtration.

As another preferred version of the present invention, described nonlinear distortion information is: the frequency-region signal after amplitude limit divided by distortion factor after with the difference of described OFDM symbol; The function of the equal inhibiting rate γ in peak that described distortion factor α is described reception signal, namely $\mathrm{\α}=1-e^{-{\mathrm{\γ}}^2}+\sqrt{\mathrm{\π}}/2\·\mathrm{\γ}\·\mathrm{erfc}\left(\mathrm{\γ}\right).$

As mentioned above, the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention has following beneficial effect:

1, utilize the partial data symbol to carry out channel estimating as virtual pilot frequency, channel estimating performance utilizes merely the method for pilot sub-carrier to be greatly improved with respect to tradition;

2, the computational complexity of the method for the invention is greatly reduced with respect to the algorithm that utilizes all data and pilot sub-carrier, and algorithm performance does not have too many decline simultaneously;

3, along with the increase of the equal inhibiting rate in peak, the performance performance of the method for the invention is more excellent;

4, the method for the invention can according to the requirement of performance and the requirement of complexity, be selected the quantity of virtual pilot frequency, the requirement of balance quality and computation complexity flexibly.

Description of drawings

Fig. 1 is the fundamental block diagram of ofdm system.

Fig. 2 is the structural representation of virtual pilot frequency.

Fig. 3 is the flow chart of the described channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of embodiment two.

Fig. 4 is the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention and the curve comparison schematic diagram of the error rate of traditional channel estimation methods with the signal to noise ratio variation.

Fig. 5 is the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention and the curve comparison schematic diagram of the mean square error of traditional channel estimation methods with the signal to noise ratio variation.

Fig. 6 is the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention and the curve comparison schematic diagram of the error rate of traditional channel estimation methods with the equal inhibiting rate variation in peak.

Embodiment

Below by specific instantiation explanation embodiments of the present invention, those skilled in the art can understand other advantages of the present invention and effect easily by the disclosed content of this specification.The present invention can also be implemented or be used by other different embodiment, and the every details in this specification also can be based on different viewpoints and application, carries out various modifications or change under spirit of the present invention not deviating from.

See also accompanying drawing.Need to prove, the diagram that provides in the present embodiment only illustrates basic conception of the present invention in a schematic way, satisfy only show in graphic with the present invention in relevant assembly but not component count, shape and size drafting when implementing according to reality, during its actual enforcement, kenel, quantity and the ratio of each assembly can be a kind of random change, and its assembly layout kenel also may be more complicated.

Be subjected to the impact of limit filtration, in the amplitude-limiting OFDM system of honeycomb pilot frequency distribution, data and pilot sub-carrier all can produce nonlinear distortion.And after pilot sub-carrier was subject to nonlinear distortion, channel estimating performance can produce serious floor effect.For head it off, the present invention proposes one based on the channel estimation methods that is used for amplitude-limiting OFDM system of virtual pilot frequency, the nonlinear distortion information in the data estimator symbol of the method iteration, and it is used for data decision; Then general's partial data symbol subsequently as virtual pilot frequency, obtains more accurate channel estimated value again, estimates simultaneously the nonlinear distortion information of more accurate data symbol; At last, with above-mentioned channel again estimated value and more accurate nonlinear distortion information feed back to the reception signal processing unit, re-start channel equalization and symbol judgement.Simulation result shows, only needs suitable increase operand, and the method for the invention can be improved systematic function significantly.In addition, the method for the invention can be selected the quantity of virtual pilot frequency flexibly, reaches the requirement of flexible balance quality and computation complexity.

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in further detail.

Embodiment one

The present embodiment provides a kind of channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency, the nonlinear distortion information in the data estimator symbol of this channel estimation methods iteration, and this nonlinear distortion information is used for data decision; Then general's partial data symbol subsequently as virtual pilot frequency, obtains more accurate channel estimated value again, estimates simultaneously the nonlinear distortion information of more accurate data symbol; At last, with above-mentioned channel again estimated value and more accurate nonlinear distortion information feed back to the reception signal processing unit, re-start channel equalization and symbol judgement.

The specific implementation process of the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention is as follows:

If the time-domain expression of OFDM symbol is:

$x\left(n\right)=1/\sqrt{N}\underset{k=0}{\overset{N-1}{\mathrm{\&Sigma;}}}X\left(k\right)\exp (\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000120003980000041.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;kn}/N),$ 0≤n＜N-1 (1)

The frequency domain presentation of supposing x (n) is X (k), and frequency domain signal X (k) is independent identically distributed stochastic variable, when sub-carrier number N is very large, according to central-limit theorem, approximate multiple Gaussian Profile, its whose amplitude obeys rayleigh distributed of obeying of its time-domain signal x (n).

For suppressing the peak-to-average force ratio of ofdm signal, usually transmitted signal is carried out limit filtration, the limit filtration expression formula is:

$x_g\left(n\right)=g\left(x\left(n\right)\right)=\left\{\begin{array}{cc}{\mathrm{Ae}}^{j\arg \left\{x\left(n\right)\right\}}& \left|x\left(n\right)\right|>A\\ x\left(n\right)& \left|x\left(n\right)\right|\&le;A\end{array}\right.---\left(2\right)$

The equal inhibiting rate in definition peak is

According to the Bussgang theorem, through the stationary Gaussian process of a memoryless nonlinear system, the decay that can be expressed as inputting and incoherent distortion component sum.The output through the OFDM of limit filtration symbol can be expressed as:

x _g(n)＝αx(n)+d(n) (3)

Wherein, α is the constant that depends on the equal inhibiting rate γ in peak, and d (n) is the nonlinear distortion of signal, and x (n) and d (n) are incoherent mutually, wherein

$\mathrm{\&alpha;}=R_{{\mathrm{xx}}_g}{R}_{\mathrm{xx}}^{-1}=1-e^{-{\mathrm{\&gamma;}}^2}+\sqrt{\mathrm{\&pi;}}/2\&CenterDot;\mathrm{\&gamma;}\&CenterDot;\mathrm{erfc}\left(\mathrm{\&gamma;}\right)---\left(4\right)$

Limit filtration comprises interference two parts in out-of-band radiation and band to the impact of useful signal.Out-of-band radiation can pass through the low pass filter filtering; And the impact of disturbing in band is embodied in two aspects: amplitude fading and nonlinear noise disturb.Traditional receiver can't compensate the hydraulic performance decline that causes due to this nonlinear noise.

The fundamental block diagram of ofdm system as shown in Figure 1.Supposes that ofdm system is Time and Frequency Synchronization, and without intersymbol interference (ISI), become when the channel in single OFDM symbol period is non-, and the channel between the OFDM symbol becomes when being, remove Cyclic Prefix, and the reception signal that carries out after the FFT conversion is:

Y＝diag(αX+D)H+Z (5)

Wherein, X is data to be transmitted, and D is the nonlinear distortion that amplitude limit causes, and H is the multipath channel frequency domain response in symbol period, and Z is additive white Gaussian noise, and X and D are incoherent.

OFDM receiving end signal handling process as shown in Figure 1, at first the reception signal after synchronous removes Cyclic Prefix, implements the FFT conversion, then carries out Amplitude Compensation; Then extract pilot frequency information the signal after Amplitude Compensation, carry out initial channel estimation.Receive signal through after channel equalization, enter nonlinear distortion compensation unit elimination nonlinear distortion noise, and send into and carry out symbol judgement in decision device.Data after judgement are reformulated the OFDM symbol, and again experience the limit filtration identical with transmitting terminal after the IFFT conversion.At last, utilize pilot frequency sequence and partial data symbol to carry out the estimation again of channel, estimate simultaneously the nonlinear distortion information of data symbol.Estimate that again the channel information that obtains is used for carrying out channel equalization in next circulation, nonlinear distortion information is used for the nonlinear noise of offset data symbol.Along with the increase of iterations, the channel information of estimation and nonlinear distortion information will be more and more accurate, and data decision also will be more and more reliable.Concrete channel estimation process can be divided into initial estimation and iterative estimate two parts.

Initial channel estimation procedure is as follows:

There is certain amplitude fading in OFDM symbol after limit filtration.Before signal processing and channel estimating, at first to carry out Amplitude Compensation, the reception signal after compensation is:

Y _c＝α ^-1Y＝diag(X+α ^-1D)H+α ^-1Z (6)

In following formula, α is constant, and it is relevant with the equal inhibiting rate in peak.

Initial channel estimation adopts the LS algorithm, then calculates the response of all subchannels by interpolation algorithm.Through the LS algorithm for estimating, the channel response at pilot sub-carrier place is:

${\hat{H}}_p=\mathrm{diag}{\left(X_p\right)}^{-1}{Y}_c=H_p+{\mathrm{\&alpha;}}^{-1}\mathrm{diag}{\left(X_p\right)}^{-1}\mathrm{diag}\left(D_p\right)H_p+{\mathrm{\&alpha;}}^{-1}\mathrm{diag}{\left(X_p\right)}^{-1}{Z}_p---\left(7\right)$

The channel response at expression pilot sub-carrier place, subscript p represents traditional pilot tone, X _pExpression pilot sub-carrier sequence.By interpolation algorithm, the channel estimation value that obtains all subcarriers places is:

$\hat{H}=\mathrm{FLTR}\left({\hat{H}}_p\right)---\left(8\right)$

In following formula, the interpolation algorithm that FLTR () expression is common comprises the DFT interpolation, linear interpolation, polynomial interopolation etc.

The iterative channel estimation process is as follows:

The mean square error of channel estimation value is relevant with the number p of pilot frequency carrier wave.Along with the increase of p, the mean square error of the channel estimation value trend that tapers off.Therefore, the number of suitable increase pilot sub-carrier in amplitude-limiting OFDM system can effectively reduce the evaluated error of channel.But in real system, the number of pilot sub-carrier is limited, has restricted the performance of channel estimating.In order to increase the number of pilot tone, to promote channel estimating performance, the present invention has introduced the concept of virtual pilot frequency.After initial channel estimation was completed, the present invention utilized the initial channel information of estimation to carry out channel equalization, then adjudicated the data symbol that obtains transmitting.And some or all of data symbol will be for the estimation again of current channel.The present invention is called virtual pilot frequency with this data symbol.Traditional pilot tone (actual pilot) number of supposing current system is p, and the virtual pilot frequency number of estimating again for channel is P _v, total number of pilots of system is:

P′＝P+P _v (9)

If traditional pilot frequency sequence of system is X _p, the virtual pilot frequency of estimating again for channel is

The new pilot frequency sequence that is comprised of both is X ' _pIf

For the nonlinear distortion information of hybrid pilot sequence (comprising traditional pilot tone and virtual pilot frequency) of estimation, utilize the LS algorithm to carry out channel and estimate again, the response that obtains hybrid pilot carrier wave place is:

${\hat{H}}_p=\mathrm{diag}{(\mathrm{\&alpha;}{X}_p^{\&prime;}+{\hat{D}}_p)}^{-1}Y=H_p+\mathrm{diag}{(\mathrm{\&alpha;}{X}_p^{\&prime;}+{\hat{D}}_p)}^{-1}(\mathrm{diag}(\mathrm{\&alpha;\&Delta;}{X}_p+\mathrm{\&Delta;}{D}_p)H_p+Z_p)---\left(10\right)$

In formula (10),

Be non-linear residual noise, Δ X _pBe the residual noise of hybrid pilot sequence, it is that error diffusion by virtual pilot frequency causes.Work as P _vDuring=N-P, all data symbols all are used for channel estimating, and channel response only is subjected to the impact of the error diffusion of virtual pilot frequency.And work as P _v=0 o'clock, data symbol did not participate in channel estimating, and channel response only is subject to the impact of the remaining nonlinear noise of pilot frequency sequence.

Suppose that virtual pilot frequency is desirable, namely do not have error diffusion, formula (10) can be reduced to:

${\hat{H}}_p=H_p+\mathrm{diag}{(\mathrm{\&alpha;}{X}_p^{\&prime;}+{\hat{D}}_p)}^{-1}(\mathrm{diag}\left(\mathrm{\&Delta;}{D}_p\right)H_p+Z_p)---\left(11\right)$

But in real system, data symbol is might be misjudged.If in last iterative cycles, the virtual pilot frequency symbol of judgement is wrong, can draw error diffusion, causes the convergence rate of Iterative Method even can't restrain slowly.Especially when system works during at the higher environment of the error rate, algorithm performance reduces even more serious.In order to reduce the impact of error diffusion, accelerate convergence of algorithm speed, the present invention has further improved the channel estimation method based on virtual pilot frequency.As shown in Figure 2, the channel response at virtual pilot frequency subcarrier place is by virtual pilot frequency with traditional pilot tone is common determines.

Limit in the width ofdm system, the nonlinear distortion of traditional pilot frequency sequence is

If utilize merely traditional pilot frequency sequence to carry out channel estimating, the channel response at pilot sub-carrier place is:

${\hat{H}}_p=\mathrm{diag}{\left({\hat{X}}_{g,p}\right)}^{-1}Y=H_p+\mathrm{diag}{(\mathrm{\&alpha;}{X}_p+{\hat{D}}_p)}^{-1}(\mathrm{diag}\left(\mathrm{\&Delta;}{D}_p\right)H_p+Z_p)---\left(12\right)$

Wherein,

Be the traditional pilot frequency sequence after amplitude limit of estimating,

Nonlinear distortion for traditional pilot frequency sequence of estimating.By interpolation algorithm, the channel response that interpolation obtains hybrid pilot subcarrier place is

And the channel response at the hybrid pilot subcarrier place that is derived by virtual pilot frequency By the formula mistake! Do not find Reference source.Determine the channel estimating at the virtual pilot frequency place after improving For:

${\hat{H}}_{p^{\&prime;}}\left(k\right)=c\left(k\right){\hat{H}}_p\left(k\right)+(1-c\left(k\right)){\tilde{H}}_p\left(k\right)---\left(13\right)$

In following formula, c (k) is the confidence level parameter of virtual pilot frequency, and k is the sequence number of pilot tone.When c (k) was 1, virtual pilot frequency was fully believable; And when c (k) was 0, virtual pilot frequency was fully insincere, and channel estimation value depends on traditional pilot tone fully.By adjusting the value of confidence level parameter c (k), the channel estimation value that can be optimized.In the present invention, c (k) can get c (k)=| X _d(k) | ²/ (| X _p(k) | ²+ | X _d(k) | ²), X wherein _d(k) expression data subcarrier.

At last, according to the channel estimation value at hybrid pilot subcarrier place

Interpolation calculation obtains the channel value of all carrier waves, simultaneously, further estimates the nonlinear distortion information of more accurate data symbol.And the nonlinear distortion information of the channel value that estimates and data symbol together feeds back to the reception signal processing unit, is used for obtaining more reliable data decision in next circulation.

At first the method for the invention eliminates the nonlinear noise in data symbol, and carries out the symbol hard decision.And the partial data symbol after judgement is taken as virtual pilot frequency, and channel response is reappraised.While is data estimator symbol nonlinear distortion information further, and feed back to the reception signal processing unit.This process will continue circulation always goes down, until iteration finishes.

The performing step of the described channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present embodiment can be summarized as:

Step 1 utilizes the initial channel information of estimating to carry out channel equalization, and adjudicates the data symbol that obtains transmitting; The further specific implementation process of described step 1 is: receiving terminal carries out Amplitude Compensation to the reception signal from ofdm system; Utilize the signal after Amplitude Compensation to carry out initial channel estimation, obtain the channel estimation value at pilot sub-carrier place according to the pilot tone of known ofdm system, and obtain the channel response of all pilot sub-carriers; Signal according to the channel response of all pilot sub-carriers after to Amplitude Compensation carries out channel equalization; The nonlinear distortion of the signal after elimination channel equalization; Extracted data subcarrier from the signal of eliminating nonlinear distortion, judgement obtains the data symbol of transmission.

Step 2, selected part or total data symbol be as virtual pilot frequency, and form the hybrid pilot sequence together with the pilot tone of amplitude-limiting OFDM system; Utilize described hybrid pilot sequence that current channel is estimated again, obtain the channel response at hybrid pilot subcarrier place; The further specific implementation process of described step 2 is: reconfigure out the OFDM symbol according to described data symbol and pilot frequency sequence; Described OFDM symbol is carried out limit filtration, the signal after the acquisition limit filtration; According to the Signal estimation nonlinear distortion information after limit filtration; Selected part or total data symbol are as virtual pilot frequency from described data symbol, the pilot tone of described virtual pilot frequency and described ofdm system consists of the hybrid pilot sequence, obtains the channel estimation value at the channel estimation value at virtual pilot frequency subcarrier place and channel response and pilot frequency sequence subcarrier place.

Step 3 feeds back to step 1 with the channel response at described hybrid pilot subcarrier place, as the foundation of next round channel equalization and data symbol judgement.

The performing step of the described channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present embodiment can be in detail:

A, Amplitude Compensation.The amplitude fading that compensation causes due to limit filtration.

B, initial channel estimation.Known pilot sequence is X _p, utilize the LS algorithm to calculate the channel estimation value at pilot sub-carrier place, then calculate the channel response of all subcarriers by interpolation algorithm.

C, channel equalization.According to the channel response of latest estimated, the decline of the multipath Rayleigh of equalizing signal.

D eliminates nonlinear distortion.Deduct nonlinear distortion information signal after equilibrium Described nonlinear distortion information

If current is initial channel estimation, its value is 0; Otherwise, the described nonlinear distortion information of step H when this information is a upper circulation channel estimating.

E, symbol judgement.Extracted data subcarrier from the described signal of step D, and adjudicate the data symbol that obtains transmitting.

F, restructuring OFDM symbol.Data symbol and the pilot frequency sequence of step e judgement are reformulated symbol

G, limit filtration.The described OFDM sign reversing of step F to time domain, and is carried out limit filtration, then frequency domain is returned in it conversion, the frequency-region signal that obtains after amplitude limit is

Described limit filtration is consistent with the transmitting terminal amplitude limiting processing.

H estimates nonlinear distortion information,

I is based on the channel estimating of virtual pilot frequency.From the described signal of step G Middle selection P _vThen individual data symbol utilizes the LS algorithm to calculate the channel estimation value at virtual pilot frequency subcarrier place as virtual pilot frequency

The quantity P of described virtual pilot frequency _v, its value is 0≤P _v≤ N-P, wherein N is the ofdm system carrier number, P is the number of traditional pilot tone.The suitable P that chooses _vValue, described algorithm can reach the balance of systematic function and computation complexity.

J is based on the channel estimating of traditional pilot tone, from the described signal of step G

The traditional pilot frequency sequence of middle extraction utilizes the LS algorithm to calculate the channel estimation value at traditional pilot tone place.Then utilize interpolation algorithm to calculate the channel response at virtual pilot frequency place

Described interpolation algorithm comprises the DFT interpolation algorithm, linear interpolation and polynomial interopolation etc.

K, improved virtual pilot frequency is estimated.With the described channel response of step I

And the described channel response of step J

Be weighted, the virtual pilot frequency that is improved is estimated.

L carries out the channel interpolation.Then the channel response at the channel response at the traditional pilot tone place that reconfigures that step J obtains and step K obtain virtual pilot frequency place utilizes interpolation algorithm to calculate the channel response of all carrier waves.

The channel response that M, step L calculate feeds back to step C, re-starts channel equalization and symbol judgement.Then further estimate nonlinear distortion and the channel response of more accurate data symbol.

The channel estimating new method that is used for amplitude-limiting OFDM system based on virtual pilot frequency of the present invention is an iterative process, it can weigh performance requirements and computation complexity requirement, solves the problem that the current performance of publish method is too poor or computational complexity is too high.The present invention has utilized the partial data symbol to carry out channel estimating as supplementary means, makes systematic function that certain improvement arranged with respect to the method for utilizing merely pilot sub-carrier; While the method has overcome utilizes all carrier waves to carry out the too high shortcoming of computation complexity of channel estimation methods.

Embodiment two

The present embodiment provides a kind of specific implementation process of the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency, and the flow process of this process supposes that iterations is ITER_NUM as shown in Figure 3, and iteration count is i.Specific algorithm is described below:

1) Amplitude Compensation.The amplitude fading that compensation causes due to limit filtration according to formula (6), the reception signal after compensation is Y _c

2) initiation parameter. I=0, wherein

Be nonlinear distortion information.

3) initial channel estimation.According to formula (7) and formula (8), obtain channel estimation value

4) channel equalization.According to the channel response of estimating

Carry out to the received signal zero forcing equalization, the signal after equilibrium is:

${\hat{Y}}_e^i=\mathrm{diag}{\left({\hat{H}}^i\right)}^{-1}{Y}_c---\left(14\right)$

5) compensating non-linear distortion.Signal after the equilibrium

Middle elimination nonlinear distortion obtains

6) symbol judgement.From

Middle extracted data carrier wave

And select and

The shortest symbol of Euclidean distance is as optimal judgement, and wherein k is the subcarrier sequence number.

7) iterations of judgement iteration count record.If i＞ITER_NUM, iteration finishes, and goes to step 17); Otherwise go to step 8).

8) OFDM symbol restructuring.Insert pilot frequency sequence X in the data symbol of judgement _p, reformulate the OFDM symbol.

9) IFFT conversion.The OFDM symbol is carried out the IFFT conversion, obtain OFDM time domain baseband signal

10) limit filtration.With the time domain baseband signal

Send in the limit filtration device, experience the nonlinear transformation identical with transmitting terminal.

11) FFT conversion.Time domain baseband signal after amplitude limit is converted into frequency domain

12) estimate nonlinear noise.

${\hat{D}}^i={\mathrm{\&alpha;}}^{-1}{\hat{X}}_g^i-{\hat{X}}^i---\left(15\right)$

13) based on the channel estimating of virtual pilot frequency.Carrier wave symbol after the amplitude limit In, choose P _vIndividual data symbol is as virtual pilot frequency.Then according to formula (10), calculate the channel response at virtual pilot frequency subcarrier place

14) based on the channel estimating of traditional pilot tone.From

Traditional pilot tone after middle extraction amplitude limit And calculate the channel response at traditional pilot tone place according to formula (12).Then the interpolation algorithm of through type (8) calculates the channel response at virtual pilot frequency place

15) channel estimating at improved virtual pilot frequency place.According to the channel response at the virtual pilot frequency place after formula (13) computed improved, then calculate the channel estimation value at all pilot tones (comprising virtual pilot frequency and traditional pilot tone) carrier wave place by interpolation algorithm.

16) upgrade iterations i.Iterations increases i=i+1.Go to step 4).

17) iterative algorithm finishes.

The beneficial effect that produces in order to set forth more fully the present invention below in conjunction with simulation analysis and result, further describes validity of the present invention.

The present invention is according to the ofdm system parameter of table 1, the performance of the method for the invention of having utilized monte carlo method emulation.This ofdm system has 512 subcarriers, and wherein 32 subcarriers are pilot sub-carrier.These pilot sub-carriers are evenly distributed in all carrier waves, and its average power is identical with the average power of data symbol.Be the inhibition zone external radiation, adopt over-sampling DFT transform method to carry out limit filtration in emulation, over-sampling rate is 2.The interpolation algorithm that adopts in emulation is the DFT algorithm.The all results of the present invention have all been carried out the inferior emulation of 2000 frames, to guarantee the accurate of result.

Table 1: the main simulation parameter table of system

Fig. 4 has described that equal inhibiting rate is 3dB at the peak, and Doppler frequency shift is in the situation of 50Hz, the error rate of the method for the invention and traditional algorithm (Bit Error Rate) performance comparison.In Fig. 4, Method A represents to adopt merely pilot sub-carrier to carry out the traditional algorithm of channel estimating; Method B represents the method for the invention.In Fig. 4, P '=128, P '=256 and P '=512 represent that respectively the hybrid pilot number is respectively 128,256 and 512.As can be seen from Figure 4, the method for the invention has better performance than traditional algorithm.Simultaneously, along with the increase of virtual pilot frequency number, the error rate of system also decreases.This is due to along with the increasing of virtual pilot frequency, and Iterative Method can be obtained more useful information.Particularly, when P '=256, the method for the invention has promoted 0.2dB than the bit error rate performance of traditional algorithm; When P '=512, performance boost 0.3dB.

Fig. 5 has described that equal inhibiting rate is 3dB at the peak, and Doppler frequency shift is in the situation of 50Hz, the channel estimating mean square error of the method for the invention and traditional algorithm (The MES of channel estimation) contrast.Be 10 in mean square error ^-3Situation under, when P '=256, the method for the invention than the mean square error performance boost of traditional algorithm 2dB; When P '=512, this performance boost 2.3dB.

It is 12dB that Fig. 6 has described in signal to noise ratio, and Doppler frequency shift is in the situation of 50Hz, and the error rate of the method for the invention and traditional algorithm (Bit Error Rate) performance is with the situation of change of the equal inhibiting rate in peak (Clipping ratio).As seen from Figure 6, the quantity of virtual pilot frequency is more, and the performance of the method for the invention is better.Simultaneously, along with the increase of the equal inhibiting rate in peak, the performance boost of the method for the invention is more obvious.This is due to the equal inhibiting rate in peak when larger, and the reliability of virtual pilot frequency is higher, thereby can obtain more accurate channel estimating and more outstanding bit error rate performance.Particularly, be 3 * 10 in the error rate ^-2Situation under, the energy force rate traditional algorithm of the anti-peak equal inhibiting rate of the method for the invention in P '=512 o'clock has promoted 0.3dB.Be 2 * 10 in the error rate ^-2Situation under, the capability improving of the equal inhibiting rate in anti-peak 0.6dB.

So the present invention has effectively overcome various shortcoming of the prior art and the tool high industrial utilization.

Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not is used for restriction the present invention.Any person skilled in the art scholar all can under spirit of the present invention and category, modify or change above-described embodiment.Therefore, have in technical field under such as and know that usually the knowledgeable modifies or changes not breaking away from all equivalences of completing under disclosed spirit and technological thought, must be contained by claim of the present invention.

Claims (7)

1. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency, is characterized in that, the described channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency comprises:

Step 1 utilizes the initial channel information of estimating to carry out channel equalization, and adjudicates the data symbol that obtains transmitting;

Step 2, selected part or total data symbol be as virtual pilot frequency, and form the hybrid pilot sequence together with the pilot tone of amplitude-limiting OFDM system; Utilize described hybrid pilot sequence that current channel is estimated again, obtain the channel response at hybrid pilot subcarrier place;

Step 3 feeds back to step 1 with the channel response at described hybrid pilot subcarrier place, as the foundation of next round channel equalization and data symbol judgement.

2. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 1, it is characterized in that: the specific implementation process of described step 1 is:

Receiving terminal carries out Amplitude Compensation to the reception signal from ofdm system;

Utilize the signal after Amplitude Compensation to carry out initial channel estimation, obtain the channel estimation value at pilot sub-carrier place according to the pilot tone of known ofdm system, and obtain the channel response of all pilot sub-carriers;

Signal according to the channel response of all pilot sub-carriers after to Amplitude Compensation carries out channel equalization;

The nonlinear distortion of the signal after elimination channel equalization;

Extracted data subcarrier from the signal of eliminating nonlinear distortion, judgement obtains the data symbol of transmission.

3. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 2, it is characterized in that: the implementation procedure of described initial channel estimation is: utilize the LS algorithm to calculate the channel estimation value at pilot sub-carrier place, then calculate the channel response of all pilot sub-carriers by interpolation algorithm.

4. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 3, it is characterized in that: described interpolation algorithm is DFT interpolation algorithm, linear interpolation algorithm or polynomial interopolation algorithm.

5. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 1, it is characterized in that: the specific implementation process of described step 2 is:

Reconfigure out the OFDM symbol according to described data symbol and pilot frequency sequence;

Described OFDM symbol is carried out limit filtration, the signal after the acquisition limit filtration;

Selected part or total data symbol are as virtual pilot frequency from described data symbol, the pilot tone of described virtual pilot frequency and described ofdm system consists of the hybrid pilot sequence, obtains the channel estimation value at the channel estimation value at virtual pilot frequency subcarrier place and channel response and pilot frequency sequence subcarrier place;

According to the Signal estimation nonlinear distortion information after limit filtration, be used for the nonlinear distortion at the step 1 erasure signal of next round iteration;

6. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 5, it is characterized in that: the process of described limit filtration is: first described OFDM sign reversing to time domain is carried out limit filtration, and then frequency domain is returned in conversion, frequency-region signal after the acquisition amplitude limit, i.e. signal after described limit filtration.

7. the channel estimation methods that is used for amplitude-limiting OFDM system based on virtual pilot frequency according to claim 6, it is characterized in that: described nonlinear distortion information is: the frequency-region signal after amplitude limit divided by distortion factor after with the difference of described OFDM symbol; The function of the equal inhibiting rate γ in peak that described distortion factor α is described reception signal, namely

$\mathrm{\&alpha;}=1-e^{-{\mathrm{\&gamma;}}^2}+\sqrt{\mathrm{\&pi;}}/2\&CenterDot;\mathrm{\&gamma;}\&CenterDot;\mathrm{erfc}\left(\mathrm{\&gamma;}\right).$

Images