CN104394105A - TDS-OFDM (Time Domain Synchronous-Orthogonal Frequency Division Multiplexing) channel estimation equalization method and system - Google Patents

Abstract

The invention provides a TDS-OFDM (Time Domain Synchronous-Orthogonal Frequency Division Multiplexing) channel estimation equalization method. The TDS-OFDM channel estimation equalization method comprises receiving and storing information frames and extracting a frame header and the data portion of the current information frame and a frame header of the next information frame; processing the data portion of the current information frame so as to obtain data time domain signals, wherein the trailing interference of the frame header of the current information frame in the data time domain signals is eliminated; performing estimation on a channel which is corresponding to the data portion of the current information frame so as to obtain the frequency domain response of the channel; performing the frequency domain equalization on the time domain signals through the frequency domain response so as to obtain a data symbol sequence. The TDS-OFDM channel estimation equalization method can be applied to the large doppler range and a rapid time-varying channel and is suitable for ultra-high-speed mobile application scenarios. The invention also provides a TDS-OFDM channel estimation equalization system.

Description

TDS-OFDM channel estimation balancing method and system

Technical field

The invention belongs to digital information transmission technical field, particularly a kind of TDS-OFDM channel estimation balancing method and system.

Background technology

Terrestrial wireless TV broadcast transmission channel itself also exists various multipath and fade-out, consider the receiver under situation of movement, then can introduce larger Doppler effect, the frequency selective fading channels become when therefore terrestrial wireless TV broadcast transmission channel is.

Ultrahigh speed Mobile solution scene can correspond to low-altitude navigation wireless channel, and namely there is Doppler's Rice channel model in straight length (LOS) main footpath, the discrete expression of its impulse response is:

$h(i,k)=a\·e^{j2\mathrm{\π}{f}_{D,\mathrm{LOS}}k{T}_S}\·\mathrm{\δ}\left({\mathrm{iT}}_S\right)+c\frac{1}{\sqrt{N}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{e}^{j{\mathrm{\θ}}_n}{e}^{j2\mathrm{\π}{f}_{D,n}{\mathrm{kT}}_S}\mathrm{\δ}({\mathrm{iT}}_S-{\mathrm{\τ}}_n)---\left(1\right)$

Wherein T _sfor the sampling period, τ=iT _sfor time delay, t=kT _s.Rice factor (K=a ²/ c ²) ∈ [2,20] dB.

Multipath phase θ _nobey [0,2 π) be uniformly distributed, multipath Doppler f _d,nobey Jakes spectrum, multipath delay τ _nobeys index distribution.So make u _n~ U (0,1), then have

θ _n＝2πu _n(2)

τ _n＝τ _slopeln(1-u _n) (4)

Wherein with be respectively the maximum incident angle in reflection footpath and minimum incidence angle, representative value is (181.75 °, 178.25 °); τ _slopefor time delay attenuation rate; Maximum doppler frequency f _{d, max}=f _{d, LOS}≈ f _rFv/c, f _rFfor radio carrier frequency, v is sending and receiving end diametrically translational speed, and c is the light velocity (wherein v<<c).

According to above-mentioned theoretical model, multipath phase can think constant constant at short notice, in the reflection maximum incident angle in footpath and the minizone of minimum incidence angle nearly all near 180 °, so can do following simplification to channel model:

θ _n＝0 (5)

f _D,n＝f _D,LOS(6)

${\mathrm{\&tau;}}_n=\frac{n}{N}{\mathrm{\&tau;}}_{\max }---\left(7\right)$

In addition, Doppler effect can cause channel gain to occur rising and falling, and channel impulse response expression formula (1) is above multiplied by fluctuating factor P (t)=1+dAsin (2 π f _at), finally add white Gaussian noise, namely obtain channel simplified model.

Orthogonal frequency division multiplexi (orthogonal frequency division multiplexing, OFDM) is a kind of data transmission technology efficiently, has the advantages such as the availability of frequency spectrum is high, anti-multipath fading ability is strong.General ofdm system is mainly divided into two kinds for the selection of channel estimation balancing method: a kind of is on frequency domain, insert known weight of pilot frequency, utilizes these weight of pilot frequency to calculate and estimate channel parameter at receiving terminal.Another kind is in time domain, insert known training sequence (being generally PN sequence), utilizes known time-domain training sequence to carry out channel estimating and equilibrium at receiving terminal.Adopt the ofdm system of pilot tone usually can carry out pilot tone pattern reasonable in design by some conditions of theory deduction and channel parameter, complete the accurate estimation to channel.Need in varying Channels all to design rational pilot interval in time domain and frequency domain, could the fading characteristic of simultaneous adaptation channel in time domain and frequency domain two dimensions.And adopt the ofdm system of time-domain training sequence to need training sequence length reasonable in design to meet the multipath maximum delay of channel.General time-domain synchronization OFDM digital transmission technology (time domain synchronous OFDM, TDS-OFDM) in system, use time-domain training sequence when carrying out channel estimating, constant when the channel parameter supposing in a subframe is.But, if in practical application scene, there is the relative movement of receiver and transmitter and the Doppler effect produced be can not ignore time, the performance of system can degradation.

OFDM is more responsive for Doppler frequency shift, and traditional ofdm system can produce serious inter-channel crosstalk (ICI) under larger Doppler frequency shift.Existing solution mainly contains: subcarrier from null method by the appropriate design of data subcarrier and coding, realize ICI from eliminating.Information assists anti-Doppler method (IAADO) to be obtained the estimated value of starting two ends relative moving speed by the prior information of external system, and section carries out the most of Doppler frequency shift of process elimination in front of the receiver.But subcarrier needs to use the original subcarrier for transmitting data information of part to encode from null method, reduces band efficiency, and its Doppler spread be suitable for cannot reach the kHz magnitude that described application requires.And general information assists anti-Doppler method can eliminate LOS main footpath Doppler shift when diametrically translational speed is accurately measured in sending and receiving end preferably, but the velocity measurement that real system obtains has error, and except the main footpath of LOS, in the scene required in described application, also there is the interference of multipath.

Summary of the invention

The present invention is intended to solve one of technical problem in correlation technique at least to a certain extent.For this reason, the object of first aspect present invention is to propose a kind ofly have that accuracy is high, structure is simple, be easy to the TDS-OFDM channel estimation balancing method that realizes.

The object of second aspect present invention is to propose a kind of TDS-OFDM channel estimation balancing system.

In order to achieve the above object, the TDS-OFDM channel estimation balancing method of first aspect present invention embodiment, comprises the following steps: A, receives and store information frame, and information frame extracts the frame head of the frame head of current information frame, data division and next information frame; B, processes the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame; C, the channel corresponding to the data division of described current information frame is estimated, to obtain the frequency domain response of described channel; D, utilizes described frequency response to carry out frequency domain equalization, to obtain data symbol sequence to described data time-domain signal.

According to the TDS-OFDM channel estimation balancing method of the embodiment of the present invention, utilize and the frame head of current information frame that be close to of data division in information frame and the frame head of next information frame, achieve the independence of channel and exactly estimation with balanced.The TDS-OFDM channel estimation balancing method of proposition of the present invention can be tackled up to the High-speed mobile Channel condition under 1kHz Doppler effect, channel estimating can adapt to fast time variant feature, and realize simple, solve quick estimation and the equalization problem of ground TV broadcast transmission channel under high-speed mobile.

In some instances, step B comprises: the Received signal strength of the 1st section of training sequence position of the 2nd of the frame head of described current information frame the section of Received signal strength of training sequence position and the frame head of next information frame described and the data division of described current information frame are carried out alignment operation; Linear operation is carried out to the Received signal strength of the 2nd section of training sequence position of the frame head of described current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame described and the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame.

In some instances, step C comprises: carry out linear interpolation arithmetic to the Received signal strength of the Received signal strength of the 2nd section of training sequence position of the frame head of described current information frame and the 2nd section of training sequence position of the frame head of next information frame described, with the time-domain signal of channel estimating corresponding to the data division obtaining described current information frame; FFT conversion is carried out with the frequency domain symbol sequence obtaining described time-domain signal to described time-domain signal; Extract the frequency domain PN sequence that described frequency domain symbol sequence pair is answered; The pre-estimation frequency domain response of described channel is obtained according to described frequency domain symbol sequence and described frequency domain PN sequence; IFFT conversion is carried out to described pre-estimation frequency domain response, to obtain the time-domain response of described channel; Filtering is carried out to described time-domain response, to obtain the first time-domain response; Described first time-domain response is processed, to obtain the second time-domain response and to carry out FFT conversion to described second time-domain response, to obtain the frequency domain response of described channel.

In some instances, the value of described frequency domain PN sequence is 1 or-1.

In some instances, carry out filtering to described time-domain response to comprise multipath component power in described time-domain response is set to 0, to obtain the first time-domain response lower than the part of the power predetermined threshold in the main footpath of straight length.

In some instances, carry out process to described first time-domain response to comprise: to described first time-domain response zero padding to make the length of described first time-domain response equal with the length of the data division of described current information frame.

The TDS-OFDM channel estimation balancing system of second aspect present invention embodiment, comprising: data acquisition module, data processing module, channel estimation module and balance module.Receiver module is used for receiving and storing information frame, extracts the frame head of the frame head of current information frame, data division and next information frame.Data processing module is used for processing the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame.Channel estimation module is used for the channel corresponding to the data division of described current information frame and estimates, to obtain the frequency domain response of described channel.Balance module is used for utilizing described frequency response to carry out frequency domain equalization, to obtain data symbol sequence to described data time-domain signal.

According to the TDS-OFDM channel estimation balancing system of the embodiment of the present invention, utilize and the frame head of current information frame that be close to of data division in information frame and the frame head of next information frame, achieve the independence of channel and exactly estimation with balanced.The TDS-OFDM channel estimation balancing method of proposition of the present invention can be tackled up to the High-speed mobile Channel condition under 1kHz Doppler effect, channel estimating can adapt to fast time variant feature, and realize simple, solve quick estimation and the equalization problem of ground TV broadcast transmission channel under high-speed mobile.

In some instances, described data processing module comprises: pretreatment unit and go interference units.Pretreatment unit is for carrying out alignment operation by the Received signal strength of the 1st section of training sequence position of the 2nd of the frame head of described current information frame the section of Received signal strength of training sequence position and the frame head of next information frame described and the data division of described current information frame.Interference units is gone to carry out linear operation for the Received signal strength of the 2nd section of training sequence position of the frame head to described current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame described and the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame.

In some instances, described channel estimation module comprises: time domain interpolation unit, the first FFT unit, PN retrieval unit, sign bit arithmetic element, an IFFT unit, filter unit and the second FFT unit.Time domain interpolation unit is used for carrying out linear interpolation arithmetic to the Received signal strength of the Received signal strength of the 2nd section of training sequence position of the frame head of described current information frame and the 2nd section of training sequence position of the frame head of next information frame described, with the time-domain signal of channel estimating corresponding to the data division obtaining described current information frame.First FFT unit is used for carrying out FFT conversion with the frequency domain symbol sequence obtaining described time-domain signal to described time-domain signal.The frequency domain PN sequence that PN retrieval unit is answered for extracting described frequency domain symbol sequence pair.Sign bit arithmetic element is used for the pre-estimation frequency domain response obtaining described channel according to described frequency domain symbol sequence and described frequency domain PN sequence.One IFFT unit is used for carrying out IFFT conversion to pre-estimation frequency domain response, to obtain the time-domain response of described channel.Filter unit is used for carrying out filtering to described time-domain response, to obtain the first time-domain response.Second FFT unit is used for processing described first time-domain response, to obtain the second time-domain response and to carry out FFT conversion to described second time-domain response, to obtain the frequency domain response of described channel.

In some instances, the value of described frequency domain PN sequence is 1 or-1.

In some instances, described filter unit is also for being set to 0, to obtain the first time-domain response by multipath component power in described time-domain response lower than the part of the power predetermined threshold in the main footpath of straight length.

In some instances, described second FFT unit also for: to described first time-domain response zero padding to make the length of described first time-domain response equal with the length of the data division of described current information frame.

The aspect that the present invention adds and advantage will part provide in the following description, and part will become obvious from the following description, or be recognized by practice of the present invention.

Accompanying drawing explanation

Fig. 1 is the flow chart of TDS-OFDM channel estimation balancing method according to an embodiment of the invention;

Fig. 2 is two PN frame structure schematic diagrames of one embodiment of the invention;

Fig. 3 is the filter implementation structure schematic diagram of the channel of one embodiment of the invention;

Fig. 4 is the filter effect schematic diagram of one embodiment of the invention;

Fig. 5 is the baseband transmission system structured flowchart of one embodiment of the invention;

Fig. 6 is the emulation testing BER-SNR curve chart of one embodiment of the invention; With

Fig. 7 is the structured flowchart of TDS-OFDM channel estimation balancing system according to an embodiment of the invention.

Embodiment

In describing the invention, it will be appreciated that, term " " center ", " longitudinal direction ", " transverse direction ", " length ", " width ", " thickness ", " on ", D score, " front ", " afterwards ", " left side ", " right side ", " vertically ", " level ", " top ", " end " " interior ", " outward ", " clockwise ", " counterclockwise ", " axis ", " radial direction ", orientation or the position relationship of the instruction such as " circumference " are based on orientation shown in the drawings or position relationship, only the present invention for convenience of description and simplified characterization, instead of indicate or imply that the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore limitation of the present invention can not be interpreted as.

In addition, term " first ", " second " only for describing object, and can not be interpreted as instruction or hint relative importance or imply the quantity indicating indicated technical characteristic.Thus, be limited with " first ", the feature of " second " can express or impliedly comprise at least one this feature.In describing the invention, the implication of " multiple " is at least two, such as two, three etc., unless otherwise expressly limited specifically.

In the present invention, unless otherwise clearly defined and limited, the term such as term " installation ", " being connected ", " connection ", " fixing " should be interpreted broadly, and such as, can be fixedly connected with, also can be removably connect, or integral; Can be mechanical connection, also can be electrical connection; Can be directly be connected, also indirectly can be connected by intermediary, can be the connection of two element internals or the interaction relationship of two elements, unless otherwise clear and definite restriction.For the ordinary skill in the art, above-mentioned term concrete meaning in the present invention can be understood as the case may be.

In the present invention, unless otherwise clearly defined and limited, fisrt feature second feature " on " or D score can be that the first and second features directly contact, or the first and second features are by intermediary indirect contact.And, fisrt feature second feature " on ", " top " and " above " but fisrt feature directly over second feature or oblique upper, or only represent that fisrt feature level height is higher than second feature.Fisrt feature second feature " under ", " below " and " below " can be fisrt feature immediately below second feature or tiltedly below, or only represent that fisrt feature level height is less than second feature.

Be described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Be exemplary below by the embodiment be described with reference to the drawings, be intended to for explaining the present invention, and can not limitation of the present invention be interpreted as.

In describing the invention, mark before step or action, such as " steps A " ~ " step D " or (1)-(4) are only for describing the object of the TDS-OFDM channel estimation balancing method of the embodiment of the present invention, and instruction or hint relative ranks relation can not be interpreted as, therefore can not be interpreted as limitation of the present invention.

See Fig. 1, the TDS-OFDM channel estimation balancing method of first aspect present invention embodiment, comprises the following steps: A, receives and store information frame, and information frame comprises the frame head of the frame head of current information frame, data division and next information frame; B, processes the data division of current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame; C, the channel corresponding to the data division of current information frame is estimated, to obtain the frequency domain response of channel; D, utilizes frequency response to carry out frequency domain equalization to data time-domain signal, to obtain data symbol sequence.Specific implementation process is as follows:

Steps A, receives and stores information frame, extracts the frame head of the frame head of current information frame, data division and next information frame.

As shown in Figure 2, K information frame comprises the frame head (PN1, PN2 as after DATA in 2 figure) of the frame head (PN1, PN2 as before DATA in Fig. 2) of current information frame, data division and next information frame.Like this, two sections of PN training sequences of the frame head of the current information frame of K information frame and the data division of K information frame constitute the information frame of two PN frame structure.

Step B, processes the data division of current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame.

Concrete, in one embodiment of the invention, step B comprises: S11, and the Received signal strength of the 1st section of training sequence position of the 2nd of the frame head of current information frame the section of Received signal strength of training sequence position and the frame head of next information frame and the data division of current information frame are carried out alignment operation; S12, linear operation is carried out to the 2nd section of Received signal strength of training sequence position of the frame head of current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame and the data division of current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame.

(1) step S11, by the Received signal strength r of the 2nd of the frame head of current information frame the section of training sequence position _{k, PN2}the Received signal strength r of the 1st section of training sequence position of the frame head of (n) and next information frame _{k+1, PN1}the data division r of (n) and current information frame _{k, DATA}n () carries out alignment operation, by r _{k, PN2}(n), r _{k+1, PN1}(n) and r _{k, DATA}n () carries out alignment operation, make the figure place of three identical.

(2) step S12, to the Received signal strength r of the 2nd section of training sequence position of the frame head of current information frame _{k, PN2}the Received signal strength r of the 1st section of training sequence position of the frame head of (n), next information frame _{k+1, PN1}the data division r of (n) and current information frame _{k, DATA}n () carries out linear operation, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame data time-domain signal computing formula be expressed as follows:

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=r_{k,\mathrm{DATA}}\left(n\right)+r_{k+1,\mathrm{PN}1}\left(n\right){-r}_{k,\mathrm{PN}2}\left(n\right).$

Step C, the channel corresponding to the data division of current information frame is estimated, to obtain the frequency domain response of channel.

Particularly, in one embodiment of the invention, step C comprises: S21, linear interpolation arithmetic is carried out to the Received signal strength of the 2nd section of Received signal strength of training sequence position of the frame head of current information frame and the 2nd section of training sequence position of the frame head of next information frame, with the time-domain signal of channel estimating corresponding to the data division obtaining current information frame; S22, carries out FFT conversion with the frequency domain symbol sequence obtaining time-domain signal to time-domain signal; S23, extracts the frequency domain PN sequence that frequency domain symbol sequence pair is answered; S24, obtains the pre-estimation frequency domain response of channel according to frequency domain symbol sequence and frequency domain PN sequence; S25, carries out IFFT conversion to pre-estimation frequency domain response, to obtain the time-domain response of channel; S26, carries out filtering to time-domain response, to obtain the first time-domain response; S27, processes the first time-domain response, obtains the second time-domain response and carries out FFT conversion to the second time-domain response, to obtain the frequency domain response of channel.

(1) step S21, linear interpolation arithmetic is carried out to the Received signal strength of the 2nd section of Received signal strength of training sequence position of the frame head of current information frame and the 2nd section of training sequence position of the frame head of next information frame, with the time-domain signal of channel estimating corresponding to the data division obtaining current information frame.

From the information frame that time domain receives, kth section receives information frame r _kn () can be regarded as and to be transmitted x by the kth section of correspondence _k(n) response y after channel _k(n), kth-1 hangover z of segment signal _k-1(n) and noise w _kn () is formed.That is,

r _k(n)＝y _k(n)+z _k-1(n)+w _k(n),n＝1,...,N _k

Wherein, y _k(n)=x _k(n) * h (n), n=1 ..., N _k, z _k(n-N _k)=x _k(n) * h (n), n > N _k.

In order to accurate reconstruction data symbol sequence, it is desirable to the time-domain signal received to be built into its own transmission sequence x _kthe circular convolution sequence of (n) and channel impulse response h (n) namely because noise contribution cannot be removed, then need to obtain namely this section of time domain sequences is deducted the hangover of time domain sequences the last period, add the circulation hangover of self.

Due to two PN frame structure, in the frame head of current information frame two sections to convert through IFFT the time-domain training sequence formed by frequency domain PN sequence identical, so for second segment training sequence in the frame head of current information frame, himself hangover is also just identical with the hangover of training sequence the last period, therefore has:

${\hat{r}}_{k,\mathrm{PN}2}\left(n\right)=r_{k,\mathrm{PN}2}\left(n\right)=y_{k,\mathrm{PN}2}^{*}\left(n\right)+w_{k,\mathrm{PN}2}\left(n\right),$

For the DATA part in current information frame, then need to calculate respectively himself hangover and the last period training sequence hangover.The frame head training sequence of current information frame and next information frame is utilized to have:

r _k,PN2(n)＝y _k,PN2(n)+z _k,PN1(n)+w _k,PN2(n)，

r _k+1,PN1(n)＝y _k+1,PN1(n)+z _k,DATA(n)+w _k+1,PN1(n)，

Again because all training sequences are all identical, so

y _k,PN2(n)＝y _k+1,PN1(n)，

z _k,PN1(n)＝z _k,PN2(n)，

So

r _k+1,PN1(n)-r _k,PN2(n)＝z _k,DATA(n)-z _k,PN2(n)+w _k+1,PN1(n)-w _k,PN2(n)，

The data division of a kth information frame then can be expressed as

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=r_{k,\mathrm{DATA}}\left(n\right)+r_{k+1,\mathrm{PN}1}\left(n\right){-r}_{k,\mathrm{PN}2}\left(n\right),$

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=y_{k,\mathrm{DATA}}^{*}\left(n\right)+w_{k,\mathrm{DATA}}\left(n\right){+w}_{k+1,\mathrm{PN}1}\left(n\right)-w_{k,\mathrm{PN}2}\left(n\right),$

The training sequence of current information frame and next information frame is adopted jointly to complete for channel estimating, Received signal strength by the 2nd section of training sequence of the frame head of kth frame DATA current information frame is approximately equal to the circular convolution of training sequence and channel response, takes out the 2nd section of training sequence r of the frame head of current information frame respectively _{k, PN2}frame head the 2nd section of training sequence r of (n) and next information frame _{k+1, PN2}n (), does the time-domain signal of channel estimating corresponding to the data division that on average obtains current information frame to it namely ${\hat{r}}_{k,\mathrm{PN}}\left(n\right)=\frac{{\hat{r}}_{k,\mathrm{PN}2}\left(n\right)+{\hat{r}}_{k+1,\mathrm{PN}2}\left(n\right)}{2}=y_{k,\mathrm{PN}}^{*}\left(n\right)+\frac{w_{k,\mathrm{PN}2}\left(n\right)}{2}+\frac{w_{k+1,\mathrm{PN}2}\left(n\right)}{2}.$

Be equivalent to like this make the noise power of PN training sequence part reduce half, add the accuracy of channel estimating.In addition, because in frame structure, PN and DATA is separated in time domain, therefore alone estimated result to DATA part, to carry out equilibrium be inaccurate, the actual channel parameter of DATA part should be closer to with the interpolation of the two estimated result.So from above-mentioned two situations analysis, make of the training sequence of current information frame and next information frame and on average carry out channel estimating and make accuracy higher.

Step S22, carries out FFT conversion with the frequency domain symbol sequence obtaining time-domain signal to time-domain signal.

Particularly, to time-domain signal carry out fast fourier transform (fast fourier transform, FFT) to obtain frequency domain symbol sequence namely ${\hat{R}}_{k,\mathrm{PN}}\left(n\right)=\mathrm{FFT}\left({\hat{r}}_{k,\mathrm{PN}}\left(n\right)\right).$

Step S23, extracts the frequency domain PN sequence that frequency domain symbol sequence pair is answered.

Extract the frequency domain PN sequence C that frequency domain symbol sequence pair is answered _pN(n), in one embodiment of the invention, PN sequence C _pNn the value of () is 1 or-1.

Step S24, obtains the pre-estimation frequency domain response of channel according to frequency domain symbol sequence and frequency domain PN sequence.

Particularly, in one embodiment of the invention, due to PN sequence C _pNn the value of () is 1 or-1, therefore pre-estimation frequency domain response namely the division arithmetic carrying out channel estimating needs just can be reduced to sign bit computing:

${\hat{H}}_1\left(n\right)=\frac{{\hat{R}}_{k,\mathrm{RN}}\left(n\right)}{C_{\mathrm{PN}}\left(n\right)}={\hat{R}}_{k,\mathrm{PN}}\left(n\right)\&times;C_{\mathrm{PN}}\left(n\right).$

Step S25, carries out Fast Fourier Transform Inverse (inverse fast fourier transform, IFFT), to obtain the time-domain response of channel to pre-estimation frequency domain response.That is:

${\hat{h}}_1\left(n\right)=\mathrm{IFFT}\left({\hat{H}}_1\left(n\right)\right).$

Step S26, carries out filtering to time-domain response, to obtain the first time-domain response.

In order to further stress release treatment is to the interference of channel estimation results, by the pre-estimation frequency domain response of channel be transformed into time domain again and carry out a filtering.For the application scenarios having the main footpath of LOS, need to carry out work under low signal-to-noise ratio (0dB or lower), other multipath components so outside the main footpath of LOS are probably submerged in noise, therefore for the noise-removed filtering part in channel estimating, the effect of its filtering noise is more even more important than the effect of sifting out multipath.Therefore, in one embodiment of the invention, to time-domain response carry out filtering to comprise time-domain response middle multipath component power is set to 0 lower than the part of the power predetermined threshold in the main footpath of straight length, to obtain the first time-domain response that is,

${\hat{h}}_2\left(n\right)=\left\{\begin{array}{c}{\hat{h}}_1\left(n\right),P\left({\hat{h}}_1\left(n\right)\right)>P_{\mathrm{th}}\&CenterDot;\max \left\{P\left({\hat{h}}_1\left(n\right)\right)\right|n=1,...,N_{\mathrm{PN}}\}\\ 0,P\left({\hat{h}}_1\left(n\right)\right)<P_{\mathrm{th}}\&CenterDot;\max \left\{P\left({\hat{h}}_1\left(n\right)\right)\right|n=1,...,N_{\mathrm{PN}}\}\end{array}\right.$

The error performance of the predetermined threshold different by emulation testing, final predetermined threshold is set to 9dB, namely lower than main footpath 9dB, (this threshold value is designated as P to filtering power _th) other footpath, obtain the time-domain response of filtered channel first

Step S27, processes the first time-domain response, to obtain the second time-domain response and to carry out FFT conversion to the second time-domain response, obtains the frequency domain response of channel.

Particularly, in one embodiment of the invention, to the first time-domain response zero padding is to make the first time-domain response length equal with the length of the data division of current information frame.That is, zero padding is to n=N _dATA).

Step D, utilizes frequency response to carry out frequency domain equalization to data time-domain signal, to obtain data symbol sequence.

Utilize frequency response to the data time-domain signal obtained in step B carry out frequency domain equalization, obtain data symbol sequence

${\hat{X}}_{k,\mathrm{DATA}}\left(n\right)=\frac{\mathrm{FFT}\left({\hat{r}}_{k,\mathrm{DATA}}\left(n\right)\right)}{{\hat{H}}_2\left(n\right)}.$

In order to evaluate the performance of the TDS-OFDM channel estimation balancing method of the embodiment of the present invention, simulating, verifying is carried out to the time domain correlation under multipath model.Doppler 6 footpath model is adopted: get other footpath quantity N=5, Rice factor K=20dB, channel maximum delay τ in software emulation _max=12 μ s; For fluctuating factor P (t), if constant during channel gain, get f _a=0, if channel gain has fluctuating, get f _a=1kHz, dA=0.1.The filter implementation structure that channel model in emulation adopts as shown in Figure 3.Denoising filter effect signal in channel estimating as shown in Figure 4, is followed successively by from top to bottom: original channel estimation, filtered channel and actual reference channel.From left to right be followed successively by: time domain impulse response, frequency domain amplitude and frequency domain phase place.

Baseband transmission system structure as shown in Figure 5.Subcarrier spacing is 5kHz, and sub-carrier number is 2048, and single frequency domain PN training sequence length is 128, and subframe lengths is 2304, and frame structure as shown in Figure 2.Over-sampling rate is 4, intertexture use 48 × 64 rectangular interweaving, chnnel coding adopts (3840 of 1/6 code check, 640) low-density checksum (Low-densityParity-check, LDPC) code, modulation system is the Quadrature Phase Shift Keying (Quadrature Phase ShiftKeying, QPSK) of gray mappings.

Transmission system in theory busy channel bandwidth is BW=2048 × 5kHz=10.24MHz, and the channel maximum delay that two PN frame structure can be tackled can not exceed the length of individualized training sequence, therefore calculates according to system parameters the channel maximum delay that can tolerate to be: ${\mathrm{\&tau;}}_{\max }=\frac{128}{2048\&times;5\mathrm{kHz}}=12.5\mathrm{\&mu;s}.$

In addition, the channel estimation process of time-domain training sequence, can be similar to the discrete sampling regarded as channel parameter, can estimate that the maximum channel Parameters variation speed that this information frame structure can be tackled is about a superframe is made up of a superframe-synchronized head and several information frames, it is initial thick synchronous when superframe-synchronized head is used for Received signal strength, just thin synchronous and other process are carried out by the frame head of each information frame self afterwards, therefore when calculated data transmission rate, superframe-synchronized head can temporarily be ignored in the signal length of whole transmission, and calculated data transmission rate is about: $v_{\mathrm{data}}\&ap;\frac{2048}{2304}\&times;2\mathrm{bit}/\mathrm{symbol}\&times;10.24\mathrm{MHz}\&times;\frac{640}{3840}\&ap;3.034\mathrm{MHz}.$

Superframe-synchronized head adopts the known array identical with information frame length, two sections of identical known arrays are put in superframe-synchronized head, and two sections of sequence time delays, 1/2 frame length, carry out the lower time domain auto-correlation of computation complexity at two buffer memorys of receiving terminal and available time delay 1/2 frame length and carry out the thick synchronous of signal.Frequency domain PN sequence (128 point) is obtained time-domain training sequence by IFFT conversion, uses two identical time-domain training sequences as the frame head of Frame; Again the data symbol stream after coded modulation is grouped into 2,048 one groups, transforms to time domain through IFFT, and frame head composition time domain data frame.In order to promote the accuracy of synchronizing sub-frame and channel estimating, be the twice of DATA part by the power ascension of two PN frame head, namely the temporal amplitude of two PN frame head is multiplied by coefficient

The emulation testing of one embodiment of the invention respectively for awgn channel, time constant multipath channel, fast time variant multipath channel, BER-SNR curve is as shown in Figure 6.The TDS-OFDM channel estimation balancing method of the embodiment of the present invention, when channel gain is constant, error performance distance AWGN thresholding is about 0.6dB, can reach 10 when SNR=-1.3dB ^-5the following error rate; When channel impulse response amplitude has the fluctuating of 10% and fluctuating frequency is 1kHz, error performance distance AWGN thresholding is about 1dB, can reach 10 when SNR=-0.8dB ^-5the following error rate.Transmitted data rates reaches more than 3Mbps, can meet the transmission demand of video image.

According to the TDS-OFDM channel estimation balancing method of the embodiment of the present invention, utilize the frame head of current information frame and the frame head of next information frame that are close to the data division in information frame, and the simplification division feature of the time-domain training sequence effectively utilizing frequency domain PN sequence to form, achieve the independence of channel and estimate with balanced exactly.The TDS-OFDM channel estimation balancing method of proposition of the present invention can be tackled up to the High-speed mobile Channel condition under 1kHz Doppler effect, channel estimating can adapt to fast time variant feature, and realize simple, solve quick estimation and the equalization problem of ground TV broadcast transmission channel under high-speed mobile.

The TDS-OFDM channel estimation balancing system 100 of second aspect present invention embodiment, as shown in Figure 7, comprises data acquisition module 10, data processing module 20, channel estimation module 30 and balance module 40.

Data acquisition module 10, for receiving and storing information frame, extracts the frame head of the frame head of current information frame, data division and next information frame.Data processing module 20 for processing the data division of current information frame, with obtain the frame head eliminating current information frame hangover interference data time-domain signal.Channel estimation module 30 is estimated for the channel corresponding to the data division of current information frame, to obtain the frequency domain response of channel.Balance module 40 carries out frequency domain equalization to data time-domain signal, to obtain data symbol sequence for utilizing frequency response.

Particularly, in one embodiment of the invention, as shown in Figure 2, K information frame comprises the frame head of current information frame (as the PN1 before DATA in Fig. 2, PN2), the frame head (PN1, PN2 as after DATA in 2 figure) of data division and next information frame.Like this, two sections of PN training sequences of the frame head of the current information frame of K information frame and the data division of K information frame constitute the information frame of two PN frame structure.

In one embodiment of the invention, data processing module 20 comprises: pretreatment unit 22 and go interference units 24.

Pretreatment unit 22 carries out alignment operation for the 2nd section of Received signal strength of training sequence position of the frame head by current information frame and the Received signal strength of the 1st section of training sequence position of the frame head of next information frame and the data division of current information frame.Interference units 24 is gone to carry out linear operation for the 2nd section of Received signal strength of training sequence position of the frame head to current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame and the data division of current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame.

(1) pretreatment unit 22 is for the Received signal strength r of the 2nd section of training sequence position of the frame head by current information frame _{k, PN2}the Received signal strength r of the 1st section of training sequence position of the frame head of (n) and next information frame _{k+1, PN1}the data division r of (n) and current information frame _{k, DATA}n () carries out alignment operation, by r _{k, PN2}(n), r _{k+1, PN1}(n) and r _{k, DATA}n () carries out alignment operation, make the figure place of three identical.

(2) go interference units 24 for the Received signal strength r of the 2nd section of training sequence position of the frame head to current information frame _{k, PN2}the Received signal strength r of the 1st section of training sequence position of the frame head of (n), next information frame _{k+1, PN1}the data division r of (n) and current information frame _{k, DATA}n () carries out linear operation, to obtain the data time-domain signal of the hangover interference of the frame head eliminating current information frame data time-domain signal computing formula be expressed as follows:

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=r_{k,\mathrm{DATA}}\left(n\right)+r_{k+1,\mathrm{PN}1}\left(n\right){-r}_{k,\mathrm{PN}2}\left(n\right).$

In one embodiment of the invention, channel estimation module 30 comprises: time domain interpolation unit 31, first FFT unit 32, PN retrieval unit 33, sign bit arithmetic element 34, an IFFT unit 35, filter unit 36 and the second FFT unit 37.Particularly,

(1) time domain interpolation unit 31 carries out linear interpolation arithmetic for the Received signal strength of the 2nd section of Received signal strength of training sequence position of the frame head to current information frame and the 2nd section of training sequence position of the frame head of next information frame, with the time-domain signal of channel estimating corresponding to the data division obtaining current information frame.

From the information frame that time domain receives, kth section receives information frame r _kn () can be regarded as and to be transmitted x by the kth section of correspondence _k(n) response y after channel _k(n), kth-1 hangover z of segment signal _k-1(n) and noise w _kn () is formed.That is,

r _k(n)＝y _k(n)+z _k-1(n)+w _k(n),n＝1,...,N _k

Wherein, y _k(n)=x _k(n) * h (n), n=1 ..., N _k, z _k(n-N _k)=x _k(n) * h (n), n > N _k.

In order to accurate reconstruction data symbol sequence, it is desirable to the time-domain signal received to be built into its own transmission sequence x _kthe circular convolution sequence of (n) and channel impulse response h (n) namely because noise contribution cannot be removed, then need to obtain namely this section of time domain sequences is deducted the hangover of time domain sequences the last period, add the circulation hangover of self.

Due to two PN frame structure, in the frame head of current information frame two sections to convert through IFFT the time-domain training sequence formed by frequency domain PN sequence identical, so for second segment training sequence in the frame head of current information frame, himself hangover is also just identical with the hangover of training sequence the last period, therefore has:

${\hat{r}}_{k,\mathrm{PN}2}\left(n\right)=r_{k,\mathrm{PN}2}\left(n\right)=y_{k,\mathrm{PN}2}^{*}\left(n\right)+w_{k,\mathrm{PN}2}\left(n\right),$

For the DATA part in current information frame, then need to calculate respectively himself hangover and the last period training sequence hangover.The frame head training sequence of current information frame and next information frame is utilized to have:

r _k,PN2(n)＝y _k,PN2(n)+z _k,PN1(n)+w _k,PN2(n)，

r _k+1,PN1(n)＝y _k+1,PN1(n)+z _k,DATA(n)+w _k+1,PN1(n)，

Again because all training sequences are all identical, so

y _k,PN2(n)＝y _k+1,PN1(n)，

z _k,PN1(n)＝z _k,PN2(n)，

So

r _k+1,PN1(n)-r _k,PN2(n)＝z _k,DATA(n)-z _k,PN2(n)+w _k+1,PN1(n)-w _k,PN2(n)，

The data division of a kth information frame then can be expressed as

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=r_{k,\mathrm{DATA}}\left(n\right)+r_{k+1,\mathrm{PN}1}\left(n\right){-r}_{k,\mathrm{PN}2}\left(n\right),$

${\hat{r}}_{k,\mathrm{DATA}}\left(n\right)=y_{k,\mathrm{DATA}}^{*}\left(n\right)+w_{k,\mathrm{DATA}}\left(n\right){+w}_{k+1,\mathrm{PN}1}\left(n\right)-w_{k,\mathrm{PN}2}\left(n\right),$

The training sequence of current information frame and next information frame is adopted jointly to complete for channel estimating, Received signal strength by the 2nd section of training sequence of the frame head of kth frame DATA current information frame is approximately equal to the circular convolution of training sequence and channel response, takes out the 2nd section of training sequence r of the frame head of current information frame respectively _{k, PN2}frame head the 2nd section of training sequence r of (n) and next information frame _{k+1, PN2}n (), does the time-domain signal of channel estimating corresponding to the data division that on average obtains current information frame to it namely ${\hat{r}}_{k,\mathrm{PN}}\left(n\right)=\frac{{\hat{r}}_{k,\mathrm{PN}2}\left(n\right)+{\hat{r}}_{k+1,\mathrm{PN}2}\left(n\right)}{2}=y_{k,\mathrm{PN}}^{*}\left(n\right)+\frac{w_{k,\mathrm{PN}2}\left(n\right)}{2}+\frac{w_{k+1,\mathrm{PN}2}\left(n\right)}{2}.$

Be equivalent to like this make the noise power of PN training sequence part reduce half, add the accuracy of channel estimating.In addition, because in frame structure, PN and DATA is separated in time domain, therefore alone estimated result to DATA part, to carry out equilibrium be inaccurate, the actual channel parameter of DATA part should be closer to with the interpolation of the two estimated result.So from above-mentioned two situations analysis, make of the training sequence of current information frame and next information frame and on average carry out channel estimating and make accuracy higher.

(2) first FFT unit 32 are for carrying out FFT conversion with the frequency domain symbol sequence obtaining time-domain signal to time-domain signal.

Particularly, to time-domain signal carry out FFT conversion to obtain frequency domain symbol sequence namely ${\hat{R}}_{k,\mathrm{PN}}\left(n\right)=\mathrm{FFT}\left({\hat{r}}_{k,\mathrm{PN}}\left(n\right)\right).$

(3) PN retrieval unit 33 frequency domain PN sequence of answering for extracting frequency domain symbol sequence pair.

Extract the frequency domain PN sequence C that frequency domain symbol sequence pair is answered _pN(n), in one embodiment of the invention, PN sequence C _pNn the value of () is 1 or-1.

(4) sign bit arithmetic element 34 is for obtaining the pre-estimation frequency domain response of described channel according to frequency domain symbol sequence and frequency domain PN sequence.

Particularly, in one embodiment of the invention, due to PN sequence C _pNn the value of () is 1 or-1, therefore pre-estimation frequency domain response namely the division arithmetic carrying out channel estimating needs just can be reduced to sign bit computing:

${\hat{H}}_1\left(n\right)=\frac{{\hat{R}}_{k,\mathrm{RN}}\left(n\right)}{C_{\mathrm{PN}}\left(n\right)}={\hat{R}}_{k,\mathrm{PN}}\left(n\right)\&times;C_{\mathrm{PN}}\left(n\right).$

(5) the one IFFT unit 35 for carrying out IFFT conversion to pre-estimation frequency domain response, to obtain the time-domain response of channel.Namely

${\hat{h}}_1\left(n\right)=\mathrm{IFFT}\left({\hat{H}}_1\left(n\right)\right).$

(6) filter unit 36 is for carrying out filtering to time-domain response, to obtain the first time-domain response.

In order to further stress release treatment is to the interference of channel estimation results, by the pre-estimation frequency domain response of channel be transformed into time domain again and carry out a filtering.For the application scenarios having the main footpath of LOS, need to carry out work under low signal-to-noise ratio (0dB or lower), other multipath components so outside the main footpath of LOS are probably submerged in noise, therefore for the noise-removed filtering part in channel estimating, the effect of its filtering noise is more even more important than the effect of sifting out multipath.Therefore, in one embodiment of the invention, to time-domain response carry out filtering to comprise time-domain response middle multipath component power is set to 0 lower than the part of the power predetermined threshold in the main footpath of straight length, to obtain the first time-domain response that is,

${\hat{h}}_2\left(n\right)=\left\{\begin{array}{c}{\hat{h}}_1\left(n\right),P\left({\hat{h}}_1\left(n\right)\right)>P_{\mathrm{th}}\&CenterDot;\max \left\{P\left({\hat{h}}_1\left(n\right)\right)\right|n=1,...,N_{\mathrm{PN}}\}\\ 0,P\left({\hat{h}}_1\left(n\right)\right)<P_{\mathrm{th}}\&CenterDot;\max \left\{P\left({\hat{h}}_1\left(n\right)\right)\right|n=1,...,N_{\mathrm{PN}}\}\end{array}\right.$

The error performance of the predetermined threshold different by emulation testing, final predetermined threshold is set to 9dB, namely lower than main footpath 9dB, (this threshold value is designated as P to filtering power _th) other footpath, obtain the time-domain response of filtered channel first in one embodiment of the invention, by filter as shown in Figure 3, filtering is carried out to time-domain response.

(7) second FFT unit 37 for processing the first time-domain response, to obtain the second time-domain response and to carry out FFT conversion to the second time-domain response, to obtain the frequency domain response of channel.

Particularly, in one embodiment of the invention, to the first time-domain response zero padding is to make the first time-domain response length equal with the length of the data division of current information frame.That is, zero padding is to n=N _dATA).

Balance module 40 carries out frequency domain equalization to data time-domain signal, to obtain data symbol sequence for utilizing frequency response.

Utilize frequency response to the data time-domain signal obtained in data processing module 20 carry out frequency domain equalization, obtain data symbol sequence

${\hat{X}}_{k,\mathrm{DATA}}\left(n\right)=\frac{\mathrm{FFT}\left({\hat{r}}_{k,\mathrm{DATA}}\left(n\right)\right)}{{\hat{H}}_2\left(n\right)}.$

In order to evaluate the performance of the TDS-OFDM channel estimation balancing system of the embodiment of the present invention, simulating, verifying is carried out to the time domain correlation under multipath model.Doppler 6 footpath model is adopted: get other footpath quantity N=5, Rice factor K=20dB, channel maximum delay τ in software emulation _max=12 μ s; For fluctuating factor P (t), if constant during channel gain, get f _a=0, if channel gain has fluctuating, get f _a=1kHz, dA=0.1.The filter implementation structure that channel model in emulation adopts as shown in Figure 3, send signal x (t) and be first multiplied by doppler phase twiddle factor exp (jwt), the main footpath LOS part in direct-view footpath is multiplied by channel gain a, the delayer that other path portion is D by 5 time delays generates, other path portion is multiplied by channel gain c, then be multiplied by channel fluctuations factor P (t), finally add white Gaussian noise.Filter effect signal in channel estimation module 30 as shown in Figure 4, is followed successively by from top to bottom: original channel estimation, filtered channel and actual reference channel.From left to right be followed successively by: time domain impulse response, frequency domain amplitude and frequency domain phase place.

Baseband transmission system structure as shown in Figure 5.Subcarrier spacing is 5kHz, and sub-carrier number is 2048, and single frequency domain PN training sequence length is 128, and subframe lengths is 2304, and frame structure as shown in Figure 2.Over-sampling rate is 4, intertexture use 48 × 64 rectangular interweaving, chnnel coding adopts (3840 of 1/6 code check, 640) low-density checksum (Low-densityParity-check, LDPC) code, modulation system is the Quadrature Phase Shift Keying (Quadrature Phase ShiftKeying, QPSK) of gray mappings.

Transmission system in theory busy channel bandwidth is BW=2048 × 5kHz=10.24MHz, and the channel maximum delay that two PN frame structure can be tackled can not exceed the length of individualized training sequence, therefore calculates according to system parameters the channel maximum delay that can tolerate to be: ${\mathrm{\&tau;}}_{\max }=\frac{128}{2048\&times;5\mathrm{kHz}}=12.5\mathrm{\&mu;s}.$

In addition, the channel estimating of time-domain training sequence, can be similar to the discrete sampling regarded as channel parameter, can estimate that the maximum channel Parameters variation speed that this information frame structure can be tackled is about superframe is made up of a superframe-synchronized head and several information frames, initial thick synchronous when superframe-synchronized head is used for Received signal strength, afterwards just by the frame head of each information frame self carry out thin synchronous and other process.Therefore, when calculated data transmission rate, superframe-synchronized head can temporarily be ignored in the signal length of whole transmission, and calculated data transmission rate is about: $v_{\mathrm{data}}\&ap;\frac{2048}{2304}\&times;2\mathrm{bit}/\mathrm{symbol}\&times;10.24\mathrm{MHz}\&times;\frac{640}{3840}\&ap;3.034\mathrm{MHz}.$

Superframe-synchronized head adopts the known array identical with information frame length, two sections of identical known arrays are put in superframe-synchronized head, and two sections of sequence time delays, 1/2 frame length, carry out the lower time domain auto-correlation of computation complexity at two buffer memorys of receiving terminal and available time delay 1/2 frame length and carry out the thick synchronous of signal.Frequency domain PN sequence (128 point) is obtained time-domain training sequence by IFFT conversion, uses two identical time-domain training sequences as the frame head of Frame; Again the data symbol stream after coded modulation is grouped into 2,048 one groups, transforms to time domain through IFFT, and frame head composition time domain data frame.In order to promote the accuracy of synchronizing sub-frame and channel estimating, be the twice of DATA part by the power ascension of two PN frame head, namely the temporal amplitude of two PN frame head is multiplied by coefficient

The emulation testing of one embodiment of the invention respectively for awgn channel, time constant multipath channel, fast time variant multipath channel, BER-SNR curve is as shown in Figure 6.The TDS-OFDM channel estimation balancing method of the embodiment of the present invention, when channel gain is constant, error performance distance AWGN thresholding is about 0.6dB, can reach 10 when SNR=-1.3dB ^-5the following error rate; When channel impulse response amplitude has the fluctuating of 10% and fluctuating frequency is 1kHz, error performance distance AWGN thresholding is about 1dB, can reach 10 when SNR=-0.8dB ^-5the following error rate.Transmitted data rates reaches more than 3Mbps, can meet the transmission demand of video image.

According to the TDS-OFDM channel estimation balancing system of the embodiment of the present invention, utilize the frame head of current information frame and the frame head of next information frame that are close to the data division in information frame, and the simplification division feature of the time-domain training sequence effectively utilizing frequency domain PN sequence to form, achieve the independence of channel and estimate with balanced exactly.The TDS-OFDM channel estimation balancing method of proposition of the present invention can be tackled up to the High-speed mobile Channel condition under 1kHz Doppler effect, channel estimating can adapt to fast time variant feature, and realize simple, solve quick estimation and the equalization problem of ground TV broadcast transmission channel under high-speed mobile.

In the description of this specification, specific features, structure, material or feature that the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to describe in conjunction with this embodiment or example are contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not must for be identical embodiment or example.And the specific features of description, structure, material or feature can combine in one or more embodiment in office or example in an appropriate manner.In addition, when not conflicting, the feature of the different embodiment described in this specification or example and different embodiment or example can carry out combining and combining by those skilled in the art.

Although illustrate and describe embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, and those of ordinary skill in the art can change above-described embodiment within the scope of the invention, revises, replace and modification.

Claims (12)

1. a TDS-OFDM channel estimation balancing method, is characterized in that, comprises the following steps:

A, receives and stores information frame, extracts the frame head of the frame head of current information frame, data division and next information frame;

B, processes the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame;

C, the channel corresponding to the data division of described current information frame is estimated, to obtain the frequency domain response of described channel; And

D, utilizes described frequency response to carry out frequency domain equalization, to obtain data symbol sequence to described data time-domain signal.

2. the method for claim 1, is characterized in that, step B comprises:

The Received signal strength of the 1st section of training sequence position of the 2nd of the frame head of described current information frame the section of Received signal strength of training sequence position and the frame head of next information frame described and the data division of described current information frame are carried out alignment operation;

Linear operation is carried out to the Received signal strength of the 2nd section of training sequence position of the frame head of described current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame described and the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame.

3. the method for claim 1, is characterized in that, step C comprises:

Linear interpolation arithmetic is carried out to the Received signal strength of the Received signal strength of the 2nd section of training sequence position of the frame head of described current information frame and the 2nd section of training sequence position of the frame head of next information frame described, with the time-domain signal of channel estimating corresponding to the data division obtaining described current information frame;

FFT conversion is carried out with the frequency domain symbol sequence obtaining described time-domain signal to described time-domain signal;

Extract the frequency domain PN sequence that described frequency domain symbol sequence pair is answered;

The pre-estimation frequency domain response of described channel is obtained according to described frequency domain symbol sequence and described frequency domain PN sequence;

IFFT conversion is carried out to described pre-estimation frequency domain response, to obtain the time-domain response of described channel;

Filtering is carried out to described time-domain response, to obtain the first time-domain response;

Described first time-domain response is processed, to obtain the second time-domain response and to carry out FFT conversion to described second time-domain response, to obtain the frequency domain response of described channel.

4. method as claimed in claim 3, it is characterized in that, the value of described frequency domain PN sequence is 1 or-1.

5. method as claimed in claim 3, is characterized in that, carries out filtering comprise multipath component power in described time-domain response is set to 0, to obtain described first time-domain response lower than the part of the power predetermined threshold in the main footpath of straight length described time-domain response.

6. method as claimed in claim 3, is characterized in that, carry out process comprise described first time-domain response:

To described first time-domain response zero padding to make the length of described first time-domain response equal with the length of the data division of described current information frame.

7. a TDS-OFDM channel estimation balancing system, is characterized in that, comprising:

Data acquisition module, for receiving and storing information frame, extracts the frame head of the frame head of current information frame, data division and next information frame;

Data processing module, for processing the data division of described current information frame, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame;

Channel estimation module, estimates for the channel corresponding to the data division of described current information frame, to obtain the frequency domain response of described channel; And

Balance module, carries out frequency domain equalization, to obtain data symbol sequence for utilizing described frequency response to described data time-domain signal.

8. system as claimed in claim 7, it is characterized in that, described data processing module comprises:

Pretreatment unit, carries out alignment operation for the Received signal strength of the 2nd section of training sequence position of the frame head by described current information frame and the Received signal strength of the 1st section of training sequence position of the frame head of next information frame described and the data division of described current information frame; And

Go interference units, linear operation is carried out, to obtain the data time-domain signal of the hangover interference of the frame head eliminating described current information frame for the Received signal strength of the 2nd section of training sequence position of the frame head to described current information frame, the Received signal strength of the 1st section of training sequence position of the frame head of next information frame described and the data division of described current information frame.

9. system as claimed in claim 7, it is characterized in that, described channel estimation module comprises:

Time domain interpolation unit, Received signal strength for the Received signal strength of the 2nd section of training sequence position of the frame head to described current information frame and the 2nd section of training sequence position of the frame head of next information frame described carries out linear interpolation arithmetic, with the time-domain signal of channel estimating corresponding to the data division obtaining described current information frame;

First FFT unit, for carrying out FFT conversion with the frequency domain symbol sequence obtaining described time-domain signal to described time-domain signal;

PN retrieval unit, for extracting the frequency domain PN sequence that described frequency domain symbol sequence pair is answered;

Sign bit arithmetic element, for obtaining the pre-estimation frequency domain response of described channel according to described frequency domain symbol sequence and described frequency domain PN sequence;

One IFFT unit, for carrying out IFFT conversion to pre-estimation frequency domain response, to obtain the time-domain response of described channel;

Filter unit, for carrying out filtering to described time-domain response, to obtain the first time-domain response; And

Second FFT unit, for processing described first time-domain response, to obtain the second time-domain response and to carry out FFT conversion to described second time-domain response, to obtain the frequency domain response of described channel.

10. system as claimed in claim 9, it is characterized in that, the value of described frequency domain PN sequence is 1 or-1.

11. systems as claimed in claim 9, is characterized in that, described filter unit is also for being set to 0, to obtain the first time-domain response by multipath component power in described time-domain response lower than the part of the power predetermined threshold in the main footpath of straight length.

12. systems as claimed in claim 9, is characterized in that, described second FFT unit also for:

To described first time-domain response zero padding to make the length of described first time-domain response equal with the length of the data division of described current information frame.