CN102257752A - Receiver apparatus and symbol timing determining method - Google Patents

Abstract

A window function generating unit (21) extracts a left half section (T11) of a window function to generate a left section window function. The window function generating unit (21) then subtracts the left section window function from 1 to generate a right section window function. A processed signal generating unit (22) generates a first processed signal (S1) by multiplying a received signal by the left section window function (wk) in time series in such a manner that the starting point is set to a time point (TB) of the received signal that is past by the finite section of the left section window function from a time point (TO) that is past by a reference time interval from a given reference time point (TR). The processed signal generating unit (22) also generates a second processed signal (S2) by multiplying the received signal by the right section window function (1 - wk) in time series in such a manner that the starting point is set to a time point of the received signal that is past by the finite section of the right section window function from the reference time point (TR). An output unit (23) outputs a signal, which is obtained by adding the first processed signal (S1) to the second processed signal (S2) in time series, to a timing determining unit (3) as a received signal of a time interval (TI) from the reference time point (TR) to a time point just before the finite section of the right section window function (1 - wk).

Description

Receiving system and method for detecting symbol timing

Technical field

The present invention relates to receiving system and method for detecting symbol timing that the symbol to the multi-carrier signal (multi-carrier signal) that comprises the front pilot (preamble portion) that known symbol repeats regularly detects.

Background technology

In OFDM (the OFDM:Orthogonal Frequency Division Multiplexing) communication mode that with IEEE802.11a is representative, when the detected symbol timing, adopt the relevant treatment of the time shaft oscillogram (time axis wave pattern) of having used known symbol.Figure 12 is the frame assumption diagram of ofdm signal of the grouping of IEEE802.11a.In addition, in the following description, for convenience of explanation, ignore at the additional noise of transmission path.

As shown in figure 12, this ofdm signal comprises front pilot D1, letter head (header portion) D2 and data portion D3.Front pilot D1 comprises short front pilot D11 and the D12 of long preambles portion.Short front pilot D11 repeated storage is called the known symbol of Short Training symbol (STS:Short Training Symbol).The D12 of long preambles portion repeated storage is called the known symbol of long training symbol (LTS:Long Training Symbol).In addition, in IEEE802.11a, STS repeats 10 times, and LTS repeats 2 times.

And, in receiving system in the past, store the time shaft oscillogram of STS in advance, by obtaining this time shaft oscillogram and the correlation of received signal of input chronologically, come detected symbol regularly.

Figure 13 represents the block diagram of the detected symbol sync detection circuit regularly in the past the receiving system.As shown in figure 13, sync detection circuit comprises correlator 101, known symbol storage part 102, peak value test section 103, threshold value storage part 104, timing detection unit 105, counter 106 and OFDM demodulator circuit 107.

The processing of correlator 101 is described below.If the time shaft oscillogram of STS is s _k, received signal is s ' k, then correlation Rxx (l) is with formula (1) expression.

[formula 1]

$\mathrm{Rxx}\left(l\right)=\underset{k=0}{\overset{N-1}{\mathrm{\Σ}}}{s}_k^{\′}\·{s^{*}}_{k+l}---\left(1\right)$

Wherein, " * " represents complex conjugate (complex conjugate), k represents the sample number of the sampled value in the time domain (time domain), and N represents the hits of the FFT size (FFT size) of ofdm signal, and l represents the slip number (slide number) of the time shaft oscillogram of STS.Under the situation of IEEE802.11a, the hits of FFT size is a prerequisite with 64 fast Fourier transform (Fourier transform), thereby the time shaft oscillogram s of STS _kAs shown in Equation (2).

[formula 2]

$s_k=\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{a}_{\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{64}}^{}---\left(2\right)$

Wherein, n represents the sample number of the sampled value in the frequency domain (frequency domain), a _nThe expression Fourier coefficient.

And in IEEE802.11a, the cycle of STS is 1/4 of a FFT size, so the time shaft oscillogram s of STS _kRepresent with formula (3).

[formula 3]

$s_k=\underset{n=0}{\overset{64/4-1}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;}4\mathrm{nk}}{64}}^{}=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(3\right)$

Received signal s ' _kTherefore be subjected to the influence of the distortion (distortion) of transmission path, as the H that is expressed as of the frequency domain that establish distortion _4ne ^{J θ 4n}The time, received signal s ' _kRepresent with formula (4).

[formula 4]

$s_k^{\&prime;}=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(4\right)$

Behind formula (3) and formula (4) substitution formula (1),, obtain formula (5) based on the orthogonality of Fourier transform.

[formula 5]

$R_{\mathrm{xx}}\left(l\right)=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\left(\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4m}{e}^{j{\mathrm{\&theta;}}_{4m}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="m\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">m</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;mk}}{16}}^{}\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a_{4n}}^{*}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}(k+l)}{16}}\right)$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\left(\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4m}{e}^{j{\mathrm{\&theta;}}_{4m}}{a}_{4m}{a_{4n}}^{*}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nl}}{16}}\right)$

$=\left\{\begin{array}{c}\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{\left|a_{4n}\right|}^2\&CenterDot;\&CenterDot;\&CenterDot;[l=16l^{\&prime;}]\\ 0\&CenterDot;\&CenterDot;\&CenterDot;[l\&NotEqual;16l^{\&prime;}]\end{array}\right.---\left(5\right)$

Wherein, l ' is the positive integer more than 0.And, irrelevant during m ≠ n in formula (5) with n, become 0 about the aggregate value of k based on orthogonality.In addition, in formula (5), be similarly 0 when l ≠ 16l ' about the correlation of n.

Its result, Rxx (l) are sampled as the cycle with 16 and have correlation (correlation value).Peak value test section 103 shown in Figure 13 compares specified threshold value and correlation in order not detect noise, will be judged to be correlation peak above the correlation of threshold value.

Regularly detection unit 105 is counted the peak period of 106 pairs of peak value test section 103 detected correlation peaks of counter, detects symbol regularly according to count value.

OFDM demodulator circuit 107 is according to the 105 detected symbols timings of timing detection unit, demodulated received signal.

In addition, as the known document that is associated, known patent document 1,2.In patent documentation 1, the receiver of following purpose is disclosed, promptly under the situation of utilizing the OFDM transmit information signals, for demodulate transmitted signal correctly in receiver, and make the frequency translation of transmitter and receiver consistent accurately with the cycle of the sampling timing of transmitter and receiver with the frequency of oscillation of oscillator.

In addition, in patent documentation 2, disclose under the situation of using the frequency pilot sign (pilot symbol) that gap carrier wave (null sub-carrier) is not set, also can establish the digital communication apparatus of Frequency Synchronization at short notice reliably.

But, at received signal s ' _kComprise disturbing wave am _kSituation under, can not correctly carry out synchronous detecting.To disturbing wave am _kCarry out Fourier expansion, obtain formula (6).

[formula 6]

${\mathrm{am}}_k=\underset{n=0}{\overset{L-1}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_{\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}}{e}^j{\frac{2\mathrm{\&pi;nk}}{L}}^{}---\left(6\right)$

Wherein, AM _nThe complex amplitude (complex amplitude) of expression disturbing wave, e ^{J Θ n}The complex phase (complex phase) of expression disturbing wave, L represents the block size (block size number) of Fourier expansion.Disturbing wave am _kDo not have and be equivalent to the frequency f of self _AMFrequency indices n ₀Sampled value in the frequency domain in addition, so formula (6) can be simplified as shown in Equation (7).

[formula 7]

${\mathrm{am}}_k={\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{n_0}}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j---\left(7\right)$

Its result is at this disturbing wave am _kUnder the situation about existing, received signal s ' _kNeed to change as shown in Equation (8).

[formula 8]

$s_k^{\&prime;}+{\mathrm{am}}_k=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}+{\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{n_0}}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j---\left(8\right)$

Study this received signal s ' _k+ am _kTime shaft oscillogram s with STS _kBetween dependency relation.According to the orthogonality of Fourier transform, comprise disturbing wave am _kReceived signal s ' _k+ am _kWith time shaft oscillogram s _kBetween correlation Rxx (l) with formula (9) expression.

[formula 9]

$\mathrm{Rxx}\left(l\right)=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}(s_k^{\&prime;}+{\mathrm{am}}_k){s^{*}}_{k+l}$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\{\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}$

$+{\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{n_0}}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a^{*}}_{4-m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}\}$

$=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{\left|a_{4n}\right|}^2{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nl}}{16}}$

$+{\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{{\mathrm{<figure-callout\; id="0"\; label="n"\; filenames="HPA00001390193200041.png,HPA00001390193200071.png"\; state="\{\{state\}\}">n</figure-callout>}}_0}}\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\left\{{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}\right\}---\left(9\right)$

In formula (9), first shown in the row second from the bottom with do not have a disturbing wave am shown in the formula (5) _kSituation under the correlation unanimity, but second shown in the row last becomes error, makes symbol timing detection performance degradation.Herein, in the error shown in second, disturbing wave am _kAmplitude composition AMn ₀Be main component, if disturbing wave am _kModulated, then change according to its modulating frequency.And the error shown in second also changes according to the cycle of STS.

As mentioned above, at disturbing wave am _kUnder the situation about existing, occur becoming big situation, make the detection difficult treatment of correlation peak, have the problem of symbol timing detection performance degradation based on the correlation of the error shown in second.In addition, in the method for patent documentation 1,2, record does not suppress the content of disturbing wave.

Patent documentation 1: Japan Patent open communique spy open flat 8-223132 number

Patent documentation 2: Japan Patent open communique spy open 2000-22660 number

Summary of the invention

Even the object of the present invention is to provide a kind of symbol receiving system and method for detecting symbol timing regularly that under the situation that disturbing wave exists, also can detect accurately.

Receiving system involved in the present invention detects the symbol timing of the multi-carrier signal of the front pilot that comprises the known symbol repetition, comprise: the window function handling part, utilization has the window function of the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol, carries out window function to received signal and handles; And timing test section, obtain the correlation peak between the time shaft oscillogram that described window function handling part carries out received signal that window function handles and described known symbol, based on this correlation peak, detect the symbol timing of described multi-carrier signal, wherein, described window function handling part comprises: window function generating unit, the left-half interval or the right half part interval of taking out described window function, window function based on taking out generates left interval window function and right interval window function; The processing signals generating unit, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of the finite interval of recalling described right interval window function from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And efferent, with described first processing signals and described second processing signals chronologically added signal as the finite interval that dates back described right interval window function from described benchmark constantly before till received signal, export described timing test section to.

In addition, method for detecting symbol timing involved in the present invention detects the symbol timing of the multi-carrier signal of the front pilot that comprises the known symbol repetition, may further comprise the steps: the window function that window function treatment step, utilization have the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol carries out the window function processing to received signal; And timing detects step, obtain by described window function treatment step and be carried out correlation peak between the time shaft oscillogram of the received signal of processing and described known symbol, and regularly based on the symbol of the described multi-carrier signal of this correlation peak detection, wherein, described window function treatment step may further comprise the steps: window function generates step, take out the left-half interval or the right half part interval of described window function, generate left interval window function and right interval window function based on the window function that takes out; Processing signals generates step, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of the finite interval of recalling described right interval window function from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And the output step, with described first processing signals and described second processing signals added signal chronologically, as the received signal till before the finite interval that dates back described right interval window function from described benchmark constantly and export.

Description of drawings

Fig. 1 represents the block diagram of the receiving system of embodiment of the present invention 1.

Fig. 2 is the oscillogram of an example of expression window function.

Fig. 3 is the block diagram of the detailed structure of expression window function handling part shown in Figure 1.

Fig. 4 is the key diagram that the window function of window function handling part shown in Figure 3 is handled.

Fig. 5 (a) is the concept map during with the number that does not carry out the received signal addition FFT size that window function handles, and (b) is the concept map during with the number that carried out the received signal addition FFT size that window function handles.

Fig. 6 represents the block diagram of the receiving system of embodiment of the present invention 2.

Fig. 7 is the figure of result of calculation of the correlation of expression correlator shown in Figure 6.

Fig. 8 represents the block diagram of the receiving system of embodiment of the present invention 3.

Fig. 9 represents to be used for the power spectrum (power spectra) of subcarrier of the generation of STS.

Figure 10 represented to superpose correlated results of received signal of the disturbing wave that is carried out AM modulation that with 3.4MHz is carrier wave.

Figure 11 represented to superpose correlated results of signal vector of the disturbing wave that is carried out AM modulation that with 20.2MHz is carrier wave.

Figure 12 is the frame assumption diagram of multi-carrier signal of the grouping (packet) of IEEE802.11a.

Figure 13 represents the block diagram of the detected symbol sync detection circuit regularly in the past the receiving system.

Embodiment

(execution mode 1)

Below, the receiving system of embodiment of the present invention 1 is described.Fig. 1 represents the block diagram of the receiving system of embodiment of the present invention 1.Receiving system shown in Figure 1 for example is the receiving system of IEEE802.11a standard, is to receive the receiving system that a plurality of subcarriers to the arrowband (narrow band) that has orthogonality relation on frequency axis carry out digital modulation and carry out the multi-carrier signal of multiplexing (multiplexing).

This multi-carrier signal is OFDM (Orthogonal Frequency Division Multiplexing, the OFDM) signal of IEEE802.11a standard, as shown in figure 12, comprises front pilot D1, letter head D2 and data portion D3.Front pilot D1 for example is that PLCP (Physical Layer Convergence Protocol, Physical layer convergence protocol) is leading, and storage is used to establish the synchronous symbol of reception.Letter head D2 storage letter header.Data portion D3 storage sends the data symbol of object.

Front pilot D1 comprises short front pilot D11 and the D12 of long preambles portion.Short front pilot D11 storage is as the Short Training symbol (STS) of known symbol.The D12 of long preambles portion stores long training symbol (LTS).STS and LTS are respectively the known symbols of receiver side.STS is mainly used in detected symbol sign synchronization regularly and the coarse adjustment of AFC (Automatic Frequency Control, automatic frequency control).LTS is mainly used in fine setting and the channel of AFC and infers.

In the IEEE802.11a standard, 10 STS of short front pilot D11 storage are 0.8 μ S during the STS.In addition, the D12 of long preambles portion stores 2 LTS, is 3.2 μ S during the LTS.And, in the present embodiment, use 10 STS detected symbol regularly.

The letter header for example comprises the transmission speed and the data length of the data symbol of storing among the data portion D3.

Data portion D3 for example comprises the OFDM symbol that specifies number that holds data symbol D32 and holds the OFDM symbol that specifies number of pilot signal (PS) D33.Each OFDM symbol comprises protection interval (GI:Guard Interval) D31 that is arranged on front end (leading end).GID31 is the redundant signals (redundant signal) that is provided with for fear of intersymbol interference.And, the signal of (during being PSD33 rear end certain under the situation of PSD33) during GID31 is duplicated (manifolding) data symbol D32 rear end certain.

For example, in the IEEE802.11a standard, during the data symbol D32 3.2 μ S, during the GID31 0.8 μ S.

Turn back to Fig. 1, receiving system comprises acceptance division 1, window function handling part 2, timing test section 3, known symbol storage part 4 and OFDM demodulator circuit 5.

Acceptance division 1 receives received signal via the transmission path of appointment, after the simulation process of carrying out appointment, by carrying out the analog digital conversion, generates the received signal (receiving vector) of numeral.In the present embodiment, acceptance division 1 is that 16 mode is sampled to received signal with the hits of the one-period of STS for example.In addition, the received signal during certain is stored in the figure buffer slightly the past.

As transmission path, for example can adopt in wired or wireless any.As wired, can adopt various order wires or be used for the power line etc. of power line transmission communication.

Window function handling part 2 uses the window function of appointment to carry out window function to received signal and handles.

Herein, window function w _kHave the short finite interval T1 of positive integer base period doubly that likens to the cycle of the STS of known symbol, and have symmetrical time shaft waveform, near the central authorities of finite interval, its value is big more along with more.In addition, window function w _kMaximum be below 1.

In the present embodiment, reference period (should be base period) is 4 times of cycle of STS, and hits is 16 * 4=64.This reference period (should be base period) and FFT (Fast Fourier Transform, fast Fourier transform) size, the OFDM symbol period of promptly removing GI equates.

In the present embodiment, as window function w _kFor example adopt Gaussian window (Gaussian window), but be not limited thereto, also can adopt rectangular window (rectangular window), Gaussian window, Hanning window (Hann window), Hamming window (Hamming window), Blackman window (Blackman window), kaiser window (Kaiser window), bartlett window (Bartlett window), window index various window functions such as (exponential window).

Fig. 2 (a) and (b) are the window function w that adopt in the expression present embodiment _kThe oscillogram of an example.Fig. 2 (a),

(b) in, the longitudinal axis is represented window function w _kValue, the transverse axis express time.The window function w of Fig. 2 (a) and (b) _kThe hits of finite interval T1 be 14 of even numbers.

Shown in Fig. 2 (a) and (b), finite interval T1 represent from the sampled point PL of left end play till the sampled point PR of right-hand member during.In addition, the interval T11 of left-half is the interval of being stipulated by the sampled point of left-half among the finite interval T1, and the interval T12 of right half part is the interval of being stipulated by the sampled point of right half part among the finite interval T1.

In Fig. 2 (a), there is sampled point on the longitudinal axis, the sampled point number in the left side the sampled point on the longitudinal axis is 7, and the sampled point number on the right side the sampled point on the longitudinal axis is 6, and the number left and right sides of sampled point is asymmetric.Therefore, under the situation of Fig. 2 (a), comprise from sampled point PL to the longitudinal axis sampled point that the left side is nearest 7 sampled points during be the interval T11 of left-half, comprise 7 sampled points from the sampled point on the longitudinal axis to sampled point PR during be the interval T12 of right half part.In addition, under the situation of Fig. 2 (a), the value of the sampled point on the longitudinal axis is 1.

On the other hand, in Fig. 2 (b), do not have sampled point on the longitudinal axis, the sampled point number in longitudinal axis left side and the sampled point number on right side are 7, equate about the number of sampled point.Therefore, under the situation of Fig. 2 (b), comprise from sampled point PL to the longitudinal axis sampled point that the left side is nearest 7 sampled points during be the interval T11 of left-half, comprise 7 sampled points from the nearest sampled point in longitudinal axis right side to sampled point PR during be the interval T12 of right half part.On the other hand, under the situation of Fig. 2 (b), owing to do not have sampled point on the longitudinal axis, so the equal less than 1 of value of the nearest sampled point in the value of the nearest sampled point in longitudinal axis left side and longitudinal axis right side.

In Fig. 2 (a) and (b), for convenience of explanation, only show 14 sampled points, in the following description, suppose to have p+1 sampled point among the interval T11 of left-half, there be p+1 sampled point among the interval T12 of right half part, there be 2p+2 sampled point altogether.

Fig. 3 is the block diagram of the detailed structure of expression window function handling part 2.Fig. 4 (a) and (b) are key diagrams that the window function of window function handling part 2 is handled.As shown in Figure 3, window function handling part 2 comprises window function generating unit 21, processing signals generating unit 22 and efferent 23.

Window function generating unit 21 is taken out window function w _kThe interval T11 of left-half generate left interval window function w _k, and based on this left side interval window function w _k, generate right interval window function 1-w _kHerein, shown in Fig. 2 (a) and (b), window function generating unit 21 is taken out p+1 the sample point of the interval T11 of left-half, generates left interval window function w _kAnd, establish left interval window function w _kThe value of each sample point be w _k, based on 1-w _k, generate right interval window function 1-w _k

Therefore, in the case, left interval window function w _k, right interval window function 1-w _kFinite interval be the interval T11 of left-half.

In addition, shown in Fig. 4 (a), processing signals generating unit 22 with from the benchmark of appointment constantly the TR moment TB of received signal of finite interval that recalls the moment TO of base period and recall left interval window function again as starting point, with left interval window function w _kMultiply each other with received signal chronologically, generate the first processing signals S1.

In addition, shown in Fig. 4 (a), processing signals generating unit 22 with from benchmark constantly TR recall moment of received signal of finite interval of right interval window function as starting point, with right interval window function 1-w _kMultiply each other with received signal chronologically, generate the second processing signals S2.

Shown in Fig. 4 (b), efferent 23 is with the first processing signals S1 and second processing signals S2 added signal chronologically, as from benchmark constantly TR to dating back to right interval window function 1-w _kFinite interval before till during the received signal of T1, and be output in regularly test section 3.

In addition, window function handling part 2 is for example carried out above-mentioned processing by each sampling period.

Turn back to Fig. 1, regularly test section 3 is obtained by window function handling part 2 and is carried out correlation peak between the time shaft oscillogram (time axis wave pattern) of received signal that window function handles and STS, based on this correlation peak, detect the symbol timing of multi-carrier signal.

Particularly, regularly test section 3 comprises correlator 31, peak value test section 32, threshold value storage part 33, timing detection unit 34 and counter 35.

Correlator 31 is obtained the correlation based on the dependency relation between the time shaft oscillogram of the STS of storage in received signal that is carried out the window function processing and the known symbol storage part 4, and is output in peak value test section 32.

Peak value test section 32 detects from the correlation peak of the correlation of correlator 31 outputs.Herein, peak value test section 32 the correlation from correlator 31 output be in the assign thresholds a of storage the threshold value storage part 33 and assign thresholds b (scope between the b＜a) with interior situation under, judge that this correlation is a correlation peak.In view of the above, can prevent from noise is come out as correlation peak detection.

Herein, threshold value a for example adopts the value after the correlation that the received signal that sends and be not applied disturbing wave from immediate transmitter obtains adds to a certain degree surplus.In view of the above, can make the correlation that obtains according to the received signal that is coupled with the energy of disturbing wave invalid.In addition, threshold value b for example can adopt from farthest transmitter send and have can the burbling noise signal the correlation that obtains of peaked received signal.In view of the above, can get rid of The noise, and make synchronous sensitivity maximization.

Regularly detection unit 34 is counted the hits of the peak intervals of 35 pairs of peak value test section 32 detected correlation peaks of counter.And regularly detection unit 34 is under the situation of the peak intervals of 16 samplings of the one-period that is equivalent to STS that detects predetermined number of times repeatedly, and the detection that detects this initial correlation peak that detects repeatedly is used as symbol regularly constantly.

Known symbol storage part 4 is stored the time shaft oscillogram as the STS of known symbol in advance.This time shaft oscillogram is the numerical data of for example being represented by 16 sampled values.In addition, known symbol is stored the time shaft oscillogram of STS in advance, but is not limited thereto, and also can store the value of symbol of frequency domain in advance.In the case, can between known symbol storage part 4 and correlator 31, the known symbol generating unit be set, this known symbol generating unit is carried out IFFT (Inverse Fast Fourier Transform to the value of symbol of the frequency domain of storage in the known symbol storage part 4, invert fast fourier transformation) comes axle oscillogram computing time, and be output in correlator 31.

OFDM demodulator circuit 5 carries out from the OFDM demodulation process of the received signal of acceptance division 1 output according to timing detection unit 34 detected synchronization timings.Herein, OFDM demodulator circuit 5 carries out the removing of front pilot, FFT, phase place correction to received signal, is carried out processing such as the demodulation of complex symbol (complex symbol) of primary modulation and error correction with BPSK, QPSK etc.In view of the above, data symbol D32 is restored.

Then, the processing details to receiving system shown in Figure 1 describes.At first, instructions window function handling part 2 uses the situation that the common window function of the window function in 64 sampling periods different with above-mentioned explanation is handled.

In window function is handled, generally speaking, calculate in advance and preserve and as the window function w of the measure-alike hits of the FFT of the multi-carrier signal of object _k, go forward side by side and be about to this window function w _kHandle with the inner product (inner product) that received signal multiplies each other chronologically according to each sampling.When establishing the received signal that is carried out this inner product processing is s ' _k, and for convenience of explanation, when the transfer function of supposing transmission path is l, s ' _kRepresent with formula (10).

[formula 10]

$s_k^{\&prime;}=w_k\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(10\right)$

In the case, the time shaft oscillogram s of STS _kAnd carried out correlation Rxx (l) between the received signal that window function handles with formula (11) expression according to formula (10).

[formula 11]

$\mathrm{Rxx}\left(l\right)=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}{s}_k^{\&prime;}{s^{*}}_{k+l}=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}{w}_k\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a_{4m}}^{*}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}{w}_k\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4n}{a^{*}}_{4m}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;ml}}{16}}---\left(11\right)$

Herein, be 64 ofdm signal because the imagination received signal is the hits of FFT size, so window function w _kHave 64 sampled points.In addition, the number of sampled point is 64, is even number, thereby window function w _kBe serving as that boundary is with respect to axisymmetric function of time between the 31st sampling and the 32nd sampling.Therefore, w _k=w _63-kSet up.

Fig. 5 (a) is the concept map during with the number that does not carry out the received signal addition FFT size that window function handles.Fig. 5 (b) is the concept map during with the number that has carried out the received signal addition FFT size that window function handles.In addition, in Fig. 5 (a) and (b), the each point in the drawing is represented the sampled value of received signal, and the longitudinal axis is represented the imaginary axis, and transverse axis is represented real axis.

Shown in Fig. 5 (a), based on the orthogonality of ofdm signal, not carry out after received signal that window function handles is added the number of FFT size, aggregate value (summation) is 0.In addition, according to formula (10), the denominator of the index of exponential function is 16.In addition, the hits of the FFT size of received signal is 64 of k=0～63.Therefore, behind the number with received signal addition FFT size, shown in Fig. 5 (a), received signal goes up 4 weeks of rotation at complex plane (complex plane).

On the other hand, shown in the stain of Fig. 5 (b), the received signal of having carried out the window function processing samples the 31st middle sample amplitudes from the 0th and becomes gradually greatly, changes with helical form.And, shown in the white point of Fig. 5 (b), being sampled as boundary with the 32nd of centre and transferring to and reducing, amplitude diminishes gradually, changes with helical form.

That is, carry out after window function handles, the imaginary number composition of received signal has symmetry shown in Fig. 5 (b), cancel out each other behind the number of addition FFT size, and aggregate value is almost 0, keeps orthogonality.The opposing party, the real number composition does not have symmetry shown in Fig. 5 (b), so behind the number of addition FFT size, aggregate value is not 0 also, and orthogonality is destroyed.

Thus, after received signal was carried out the window function processing, orthogonality was destroyed, and produced between the subcarrier and disturbed, and contained more error in the OFDM symbol after the demodulation, can't correctly carry out demodulation.

To this, in the present embodiment, carry out the window function shown in above-mentioned Fig. 4 (a) and (b) and handle.Below, specifically describe the window function shown in Fig. 4 (a) and (b) and handle.In the following description, suppose that at Fig. 4 (a) and (b) the hits of base period is 64, p=31.Thus, at the window function w of Fig. 2 (a) and (b) _kIn, the hits of finite interval T1 is 2p+2=64.In addition, benchmark constantly TR be the 64th sampling of i base period.In addition, window function w _kHits be even number 64, thereby, be the 63rd sampling, then window function w with sampled point PR if be the 0th sampling with sampled point PL _kBecome between the 31st sampling and the 32nd sampling to serve as the roughly function of symmetry of boundary.

At first, shown in Fig. 2 (a) and (b), window function generating unit 21 is taken out window function w _kSampled point in the sampled point of the interval T11 of left-half generate left interval window function w _k, and establish left interval window function w _kEach sampled point be w _k, based on 1-w _k, generate right interval window function 1-w _k

[formula 12]

$s_k^{\&prime;}=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}(k=-p~\mathrm{0,64}-p~64)---\left(12\right)$

Then, as shown in Equation (12), processing signals generating unit 22 is taken out the individual sampled point of p+1 (=32) of 0 sampling of the individual sampling to the of the-p (=-31) of received signal.

Then, as shown in Equation (13), processing signals generating unit 22 is with left interval window function w _kThe 0th sampling to the p the sampling p+1 sampled point be set at left interval window function w _kK=-p～0, make the left interval window function w of setting _kAct on p+1 the sampled point of the-p sampling of received signal to the 0th sampling.

[formula 13]

$\underset{k=-p\&RightArrow;0}{s_k^{\&prime;}}=w_k\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(13\right)$

Then, as the formula (12), processing signals generating unit 22 is taken out the individual sampled point of p+1 (=32) of 64 samplings of the individual sampling to the of 64-p (=33) of received signal.

Then, as shown in Equation (14), processing signals generating unit 22 is with right interval window function 1-w _kThe 0th sampling to the p the sampling p+1 sampled point be set at right interval window function 1-w _kK=64-p～64, make the right interval window function 1-w of setting _kAct on p+1 the sampled point of 64-p sampling of received signal to the 64th sampling.

[formula 14]

$\underset{k=64-p\&RightArrow;64}{s_k^{\&prime;}}=(1-w_k)\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(14\right)$

Then, as shown in Equation (15), efferent 23 is with the s ' of formula (13) _kThe s ' of (k=-p～0) and formula (14) _k(k=64-p～64) addition is exported as 64-p sampling of k=to the sampled point of the 64th sampling of received signal.

[formula 15]

$\underset{k=-p\&RightArrow;0}{s_k^{\&prime;}}+\underset{k=64-p\&RightArrow;64}{s_k^{\&prime;}}$

$=w_k\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}+(1-w_k)\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}---\left(15\right)$

$=\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}$

At the received signal s ' that obtains based on formula (15) _kIn, w _kdo not have so received signal s ' of the number of FFT size _kShown in Fig. 5 (a) with respect to the initial point point symmetry be configured.Its result handles received signal s ' by this window function _kCan keep orthogonality.

,, considered the situation of k=-p～64 herein as the scope of k, but k=-p～16q (q is the integer value more than 1, (wherein, the w of p+1≤16q/2) _kSize be 2p+2) set up too.

As mentioned above, adopt above-mentioned window function to handle, can keep received signal s ' _kOrthogonality, can correctly take out the OFDM symbol.

Then, consider s ' to received signal _kDisturbing wave am _kBe suitable for the situation of the window function processing of present embodiment.In the following description, for convenience of explanation, only consider complementary sampled point k=-1,63.At first, the hits of establishing the FFT size is 64, to disturbing wave am _kCarry out Fourier expansion, disturbing wave such as formula (16) expression.

[formula 16]

${\mathrm{am}}_k={\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{n_0}}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j=\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nk}}{64}}+{\mathrm{\&epsiv;}}_k---\left(16\right)$

Wherein, ε _kThe disturbing wave and actual disturbing wave am of Fourier expansion carried out in expression _kBetween error.Make window function w _kAct on k=-1, the disturbing wave am of 63 sampled point _kAfter, obtain formula (17).

[formula 17]

$w_{-1}{\mathrm{am}}_{-1}=w_{-1}\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}(-1)}{64}}+w_{-1}{\mathrm{\&epsiv;}}_{-1}$

$=(1-w_{-1})\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}\left(63\right)}{64}}+(1-w_{-1}){\mathrm{\&epsiv;}}_{63}---\left(17\right)$

Above-mentioned w _-1Am _-1And w ₆₃Am ₆₃Be added as shown in Equation (18).

[formula 18]

$w_{-1}{\mathrm{am}}_{-1}+w_{63}{\mathrm{am}}_{63}$

$=w_{-1}(\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}(-1)}{64}}+{\mathrm{\&epsiv;}}_{-1})$

$+(1-w_{-1})(\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}\left(63\right)}{64}}+{\mathrm{\&epsiv;}}_{63})---\left(18\right)$

As shown in Equation (18), ε _-1-ε _-63Item coefficient w arranged _-1Herein, w _-1Satisfy 0＜w _-1≤ 1.Thus, handle, can make ε based on above-mentioned window function _-1-ε _-63Influence be reduced to w _-1Doubly, can suppress disturbing wave am _k

In addition, as the k=-2 of other complementary sampled point, 62, k=-3,61, k=-4,60 ... also similarly represent, can utilize window function w with formula (18) _kSuppress disturbing wave am _k

As mentioned above, according to the receiving system of execution mode 1, received signal s ' can not destroyed _kThe situation of orthogonality under suppress disturbing wave, detected symbol is regularly accurately.

In addition, in the above description, window function generating unit 21 is taken out the interval T11 of left-half, generates left interval window function w _k, but the invention is not restricted to this.That is, window function generating unit 21 also can be taken out window function w _kThe interval T12 of right half part generate right interval window function w _k, and based on this right side interval window function w _k, generate left interval window function 1-w _k

And shown in Fig. 2 (a) and (b), window function generating unit 21 is taken out p+1 the sampled point of the interval T12 of right half part, generates right interval window function w _kAnd the value of establishing each sampled point of right interval window function is w _k, based on 1-w _k, generate left interval window function 1-w _kGet final product.In the case, right interval window function w _k, left interval window function 1-w _kFinite interval be the interval T12 of right half part.

In addition, in the above description, generate second processing signals after the first processing signals S1 and be illustrated at first to generate, but be not limited thereto, generate first processing signals after also can at first generating the second processing signals S2.These modes are also applicable in the following execution mode.

(execution mode 2)

The receiving system of embodiment of the present invention 2 is a feature so that the different respectively a plurality of window function handling parts 2 of base period to be set.In addition, in the present embodiment, the part identical with execution mode 1 omitted its explanation.

Fig. 6 represents the block diagram of the receiving system of embodiment of the present invention 2.As shown in Figure 6, receiving system comprises three window function handling parts 201 to 203.

The hits of the base period of window function handling part 201 and window function w _kThe hits of finite interval T1 be 3 times of STS promptly 64.The hits of the base period of window function handling part 202 and window function w _kThe hits of finite interval T1 be 2 times of STS promptly 32.The hits of the base period of window function handling part 203 and window function w _kThe hits of finite interval T1 be 1 times of STS promptly 16.In addition, the processing details of window function handling part 201 to 203 is identical with the window function handling part 2 of execution mode 1, therefore omits explanation.

Regularly test section 3 was obtained from the cycle of the correlation peak of each received signal of window function handling part 201 to 203 outputs, approached the correlation peak of received signal in the cycle of STS most based on the cycle of correlation peak, and the symbol that detects multi-carrier signal regularly.

Particularly, regularly test section 3 comprises and window function handling part 201 to 203 corresponding three correlators 311 to 313, peak value test section 321 to 323 and peak counter 351 to 353.In addition, regularly test section 3 comprises threshold value storage part 33 and timing detection unit 34.

Correlator 311 to 313 calculates respectively from the correlation between the time shaft oscillogram of the received signal of window function handling part 201 to 203 outputs and STS, and is output in peak value test section 321 to 323.

Peak value test section 321 to 323 detects respectively from the correlation peak of the correlation of correlator 311 to 313 outputs.In addition, the detection method of the correlation peak of peak value test section 321 to 323 is identical with the peak value test section 32 of execution mode 1, therefore omits detailed explanation.

Peak counter 351 to 353 is counted the hits of the peak intervals of peak value test section 321 to 323 detected correlation peaks respectively.

Regularly detection unit 34 any in peak counter 351 to 353 detects under the situation of peak intervals of 16 samplings of the one-period that is equivalent to STS, with this peak counter as gazing at peak counter, and will be from being set at valid interval with this detection for example interval of 32 samplings constantly of gazing at the initial correlation peak of the corresponding correlator of peak counter (should be the peak value test section).

And, regularly detection unit 34 is in this valid interval, when gazing at peak counter can count the peak intervals of 16 samplings repeatedly the time, will constantly regularly detect with this detection of gazing at the initial correlation peak of the corresponding correlator of peak counter as symbol.In addition, be not 16 though gaze at the count value of the peak intervals of peak counter, regularly detection unit 34 also can, when for example being positioned in 16 ± 1 the scope, peak intervals is judged to be effectively.

Fig. 7 (a) to (d) is the figure of result of calculation of the correlation of expression correlator, (a) expression is not to carrying out the received signal that window function is handled, correlation when getting the cross-correlation mutually with the time shaft oscillogram of STS (b) is represented the correlation that correlator shown in Figure 6 311 to 313 calculates respectively to (d).In addition, in (d), the longitudinal axis is represented correlation at Fig. 7 (a), transverse axis hits express time.In addition, in (d), threshold value is represented above-mentioned threshold value b at Fig. 7 (a).

In Fig. 7 (a), to play the peak intervals that then detects till the correlation peak be not 16 samplings from detecting correlation peak at first, therefore can't detect symbol regularly.

In Fig. 7 (b), detecting correlation peak to the hits that detects till the next correlation peak at first from peak value test section 321 is 16, therefore sets from the detection of the initial correlation peak valid interval of 32 samplings constantly.And, in this valid interval, twice repeatedly of the peak intervals of 16 samplings.Therefore, regularly detection unit 34 regularly detects the moment that correlator 311 (should be peak value test section 321) detects correlation peak at first as symbol.That is symbol timing detection success.

In Fig. 7 (c), the same with Fig. 7 (a), can't detect symbol regularly.In Fig. 7 (d), the same with Fig. 7 (b), the success of symbol timing detection.

In above-mentioned formula (18), if ε _-63+ w _-1(ε _-1-ε _-63)=0, then the effect of window function processing reaches maximum, can suppress disturbing wave am _kSet up ε in order to make this formula _-1And ε _-63Need satisfy formula (19).

[formula 19]

${\mathrm{\&epsiv;}}_{63}\&GreaterEqual;0,{\mathrm{\&epsiv;}}_{-1}-{\mathrm{\&epsiv;}}_{63}\&le;0\&DoubleLeftRightArrow;{\mathrm{\&epsiv;}}_{63}\&GreaterEqual;0,{\mathrm{\&epsiv;}}_{-1}\&le;0$

${\mathrm{\&epsiv;}}_{63}\&le;0,{\mathrm{\&epsiv;}}_{-1}-{\mathrm{\&epsiv;}}_{63}\&GreaterEqual;0\&DoubleLeftRightArrow;{\mathrm{\&epsiv;}}_{63}\&le;0,{\mathrm{\&epsiv;}}_{-1}\&GreaterEqual;0---\left(19\right)$

But, ε _kDepend on disturbing wave am _kPhase place and frequency, so the relation of formula (19) is not necessarily set up.To this, in execution mode 2, different three the window function handling parts 201 to 203 of base period are set respectively.In view of the above, any the peak value test section in peak value test section 321 to 323 can detect correlation peak accurately, and detected symbol regularly can improve the effect that window function is handled more reliably.

(execution mode 3)

Compare with the receiving system of execution mode 2, the receiving system of execution mode 3 is a feature so that two known symbol storage parts to be set.Fig. 8 represents the block diagram of the receiving system of execution mode 3.

Known symbol storage part 41,42 is stored respectively by combination and is used for a plurality of subcarriers of generation of STS and the different multiple time shaft oscillogram of frequency band that generates in advance.In IEEE802.11a, use the time shaft oscillogram of 12 subcarriers generation STS in 64 subcarriers.

Fig. 9 (a) to (c) expression is used for the power spectrum of subcarrier of the time shaft waveform map generalization of STS.The subcarrier of Fig. 9 (a) expression time shaft waveform map generalization that is used for STS in the past, the subcarrier of the time shaft waveform map generalization that is used for STS of Fig. 9 (b) expression known symbol storage part 41 storages, the subcarrier of the time shaft waveform map generalization that is used for STS of Fig. 9 (c) expression known symbol storage part 42 storages.

Shown in Fig. 9 (a), the time shaft oscillogram of STS in the past uses 12 subcarriers to generate.Therefore, in 12 subcarriers, be loaded with under the situation of disturbing wave in the frequency band of any subcarrier, can't detect symbol accurately regularly.

To this, shown in Fig. 9 (b), known symbol storage part 41 is stored the time shaft oscillogram of 3 and the STS that 3 subcarriers that amount to 6 two ends frequency bands generate from the right from the left side of 12 subcarriers using Fig. 9 (a) in advance.In addition, shown in Fig. 9 (c), known symbol storage part 42 is stored the time shaft oscillogram of the STS that the 4th to the 9th subcarrier that amounts to 6 intermediate frequency band from the left side of 12 subcarriers using Fig. 9 (a) generate in advance.

Turn back to Fig. 8, correlator 311 to 313 is obtained the correlation between the time shaft oscillogram of the STS that stores respectively from the received signal of window function handling part 201 to 203 output and known symbol storage part 41.

Correlator 314 to 316 is obtained the correlation between the time shaft oscillogram of the STS that stores respectively from the received signal of window function handling part 201 to 203 output and known symbol storage part 42.

Peak value test section 321 to 323 detects respectively from the correlation peak of the correlation of correlator 311 to 313 outputs.In addition, peak value test section 324 to 326 detects respectively from the correlation peak of the correlation of correlator 314 to 316 outputs.In addition, the detection method of the correlation peak of peak value test section 321 to 326 is identical with the peak value test section 32 of execution mode 1, therefore omits detailed explanation.

Peak counter 351 to 356 is counted the hits of the peak intervals of peak value test section 321 to 323 detected correlation peaks respectively.

Regularly detection unit 34 is respectively according to the peak intervals of peak counter 351 to 356 countings, uses the method detected symbol timing identical with execution mode 2.That is, regularly detection unit 34 approaches the time shaft oscillogram in cycle of STS and the correlation peak between the received signal most based on the peak period according to the correlation peak of peak counter 351 to 356 countings, and the symbol that detects multi-carrier signal regularly.

Each output threshold value a and threshold value b in 33 pairs of peak value test sections 321 to 326 of threshold value storage part.

Then, use numerical expression that the processing of the correlator 311 to 316 of present embodiment is described.At first, suppose window function processing, disturbing wave am by window function handling part 201 to 203 _kIn error ε _kBe decreased to negligible degree, then can be similar to as shown in Equation (20).

[formula 20]

${\mathrm{am}}_k={\mathrm{AM}}_{n_0}{e}^{j{\mathrm{\&Theta;}}_{n_0}}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\&ap;\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nk}}{64}}---\left(20\right)$

Suppose disturbing wave am herein, _kThe low territory side that is present in the above-mentioned two ends frequency band of n=0～15, the disturbing wave am of n=16～63 _kCan ignore.In the case, disturbing wave am _kCan be similar to by enough formula (21).

[formula 21]

${\mathrm{am}}_k\&ap;\underset{n=0}{\overset{63}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;n}\left(63\right)}{64}}\&ap;\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nk}}{64}}---\left(21\right)$

In addition, the time shaft oscillogram of the STS of known symbol storage part 41 storages is represented with formula (22).Wherein, the sampling interval of m is 4 times of sampling interval of n.

[formula 22]

$\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}---\left(22\right)$

On the other hand, the time shaft oscillogram of the STS of known symbol storage part 42 storages is represented with formula (23).

[formula 23]

$\underset{m=0}{\overset{3}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}+\underset{m=12}{\overset{15}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}---\left(23\right)$

Be not loaded with disturbing wave am _kFormula (22) STS the time shaft oscillogram with comprise disturbing wave am _kReceived signal s ' _k+ am _kBetween correlation Rxx (l) with formula (24) expression.In addition,, be not loaded with disturbing wave, so the s ' shown in first row of formula (24) at n=16～63 places because the supposition disturbing wave is stated from the low territory side of the two ends frequency band of n=0～15 _k+ am _kIn n=16～63 o'clock, am _k=0.Therefore, in formula (24), save the calculating of n=16～63.

[formula 24]

$\mathrm{Rxx}\left(l\right)=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}(s_k^{\&prime;}+{\mathrm{am}}_k){s^{*}}_{k+l}$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\{\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}\underset{m=0}{\overset{15}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}$

$+\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}A{M}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;}\mathrm{<figure-callout\; id="64"\; label="n\; k"\; filenames="HPA00001390193200041.png"\; state="\{\{state\}\}">n</figure-callout>}k}{64}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}\}$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\{\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4\mathrm{<figure-callout\; id="2"\; label="n\; e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">n</figure-callout>}}{e}^j{\frac{2\mathrm{\&pi;nk}}{16}}^{}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}$

$+\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nk}}{64}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}\}$

$=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}\{\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{a}_{4n}{a^{*}}_{4m}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;ml}}{16}}$

$+\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{e}^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nk}}{64}}{a^{*}}_{4m}{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;m}(k+l)}{16}}\}$

$=\underset{n=4}{\overset{11}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{4n}}{\left|a_{4n}\right|}^2{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nl}}{16}}+\underset{n=0}{\overset{15}{\mathrm{\&Sigma;}}}\underset{m=4}{\overset{11}{\mathrm{\&Sigma;}}}{\mathrm{AM}}_n{e}^{j{\mathrm{\&Theta;}}_n}{a^{*}}_{2m}{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">e</figure-callout>}}^j{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;}2\mathrm{<figure-callout\; id="64"\; label="ml"\; filenames="HPA00001390193200041.png"\; state="\{\{state\}\}">ml</figure-callout>}}{64}}$

$=\underset{n=4}{\overset{11}{\mathrm{\&Sigma;}}}{H}_{4n}{e}^{j{\mathrm{\&theta;}}_{2n}}{\left|a_{4n}\right|}^2{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HPA00001390193200041.png,HPA00001390193200111.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;nl}}{16}}---\left(24\right)$

Herein, in second of the row second from the bottom of formula (24), n=0～15, there is not the condition that satisfies n=4m in m=4～11, are 0 therefore.

Therefore, at disturbing wave am _kBe stated under the situation of low territory side of two ends frequency band of n=0～15, because the time shaft oscillogram of the STS of known symbol storage part 42 storages is the time shaft oscillogram of intermediate frequency band, so, then can make disturbing wave am if use this time shaft oscillogram to carry out relevant treatment _kBe 0.Thus, correlator 314 to 316 shown in Figure 8 is compared with correlator 311 to 313, can obtain correlation accurately.

On the other hand, at disturbing wave am _kBe stated under the situation of intermediate frequency band of n=15～48, because the time shaft oscillogram of the STS of known symbol storage part 41 storages is the time shaft oscillogram of two ends frequency band, so the correlator shown in Figure 8 311 to 313 that uses this time shaft oscillogram to carry out relevant treatment is compared with correlator 314 to 316, can obtain correlation accurately.

Figure 10 represented to superpose correlated results of received signal of the disturbing wave that is carried out AM modulation that with 3.4MHz is carrier wave.In addition, Figure 11 correlated results of signal vector of the disturbing wave that is carried out AM modulation that with 20.2MHz is carrier wave of having represented to superpose.

In Figure 10 and Figure 11, first tabulation shows that use is based on the correlated results under the situation of the time shaft oscillogram of the STS of the subcarrier generation of Fig. 9 (a), secondary series is represented use based on the correlated results under the situation of the time shaft oscillogram of the STS of subcarrier (subcarrier of the intermediate frequency band) generation of Fig. 9 (c), and the 3rd tabulation shows that use is based on the correlated results under the situation of the time shaft oscillogram of the STS of subcarrier (subcarrier of the two ends frequency band) generation of Fig. 9 (b).

In addition, in Figure 10 and Figure 11, first line display does not carry out the correlated results under the situation that window function handles, and the 2nd to 4 row represents that respectively the hits of base period is the correlated results under 64,32,16 the situation.

That is, in Figure 10 and Figure 11, the correlated results of the 2nd to 4 row of secondary series is represented the correlated results of correlator shown in Figure 8 314 to 316 respectively, and the correlated results of tertial the 2nd to 4 row is represented the correlated results of correlator shown in Figure 8 311 to 313 respectively.

In Figure 10, the interference wave frequency is 3.4MHz, because the frequency band of disturbing wave belongs to the low territory side of the two ends frequency band of n=0～15, therefore as can be known, in the 2nd to 4 row of the secondary series that surrounds with quadrilateral frame, obtain good correlated results, the success of symbol timing detection.

In Figure 11, the interference wave frequency is 20.2MHz, and the frequency band of disturbing wave belongs to the intermediate frequency band of n=16～48, therefore as can be known, in tertial the 2nd to 4 row that surrounds with quadrangle, obtains good correlated results, the success of symbol timing detection.

As mentioned above, receiving system according to present embodiment, carried out the received signal that window function is handled at three different respectively window functions of the hits that utilizes base period, obtain the correlation of time shaft oscillogram of the STS of the time shaft oscillogram of STS of intermediate frequency band and two ends frequency band, in these correlations, use can precision best detection peak correlation detected symbol at interval regularly therefore can precision more detect symbol regularly in the highland.

In addition, in execution mode 3, the time shaft oscillogram of STS has adopted two kinds, but also can adopt more than three kinds.In the case, the different subcarrier rise time axle oscillogram of combination frequency just can respectively.

In addition, in execution mode 2,3, the window function handling part has adopted three kinds, but is not limited thereto, and also can adopt more than four kinds.In the case, the different value of the integral multiple of the one-period of the base period of each window function handling part employing STS just can.

The technical characterictic of above-mentioned receiving system can be summarized as follows.

(1) receiving system involved in the present invention detect the multi-carrier signal that comprises the front pilot that known symbol repeats symbol regularly, comprise: the window function handling part, utilization has the window function of the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol, carries out window function to received signal and handles; And timing test section, obtain the correlation peak between the time shaft oscillogram that described window function handling part carries out received signal that window function handles and described known symbol, based on this correlation peak, detect the symbol timing of described multi-carrier signal, wherein, described window function handling part comprises: the window function generating unit, take out the left-half interval or the right half part interval of described window function, based on the window function that takes out, generate as the left interval window function of the window function in described left-half interval with as the right interval window function of the window function in described right half part interval; The processing signals generating unit, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of recalling described right half part interval from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And efferent, with described first processing signals and described second processing signals chronologically added signal as the finite interval that dates back described right interval window function from described benchmark constantly before till received signal, export described timing test section to.

In addition, method for detecting symbol timing involved in the present invention detects the symbol timing of the multi-carrier signal of the front pilot that comprises the known symbol repetition, may further comprise the steps: the window function that window function treatment step, utilization have the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol carries out the window function processing to received signal; And timing detects step, obtain by described window function treatment step and be carried out correlation peak between the time shaft oscillogram of the received signal of processing and described known symbol, and regularly based on the symbol of the described multi-carrier signal of this correlation peak detection, wherein, described window function treatment step may further comprise the steps: window function generates step, take out the left-half interval or the right half part interval of described window function, generate left interval window function and right interval window function based on the window function that takes out; Processing signals generates step, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of the finite interval of recalling described right interval window function from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And the output step, with described first processing signals and described second processing signals added signal chronologically, as the received signal till before the finite interval that dates back described right interval window function from described benchmark constantly and export.

According to these structures, take out the left-half of window function or the interval of right half part, the window function based on taking out generates left interval window function and right interval window function.And, will recall the moment of base period constantly from the benchmark of appointment, and moment of received signal of finite interval of recalling left interval window function again as starting point, left interval window function is multiplied each other with received signal chronologically, generate first processing signals.In addition, will from moment of received signal of finite interval that benchmark is recalled right interval window function constantly as starting point, right interval window function is multiplied each other with received signal chronologically, generate second processing signals.

And,, export as the received signal till before the finite interval that dates back right interval window function from benchmark constantly with first processing signals and described second processing signals added signal chronologically.

In view of the above, can be when keeping the orthogonality of received signal, the disturbing wave that comprises in the inhibition received signal, therefore detected symbol timing accurately.

(2) comparatively it is desirable to, described window function generating unit is taken out the window function in described left-half interval, and the window function that takes out is generated as described left interval window function, generates described right interval window function by deducting described left interval window function from 1.

According to this structure, take out the window function in left-half interval, the window function that takes out is generated as left interval window function, and, generate right interval window function based on 1-left side interval window function.Therefore, can generate left interval window function and the right interval window function that finite interval equates by simple the processing.

(3) comparatively it is desirable to, described window function generating unit is taken out the window function in described right half part interval, and the window function that takes out is generated as described right interval window function, generates described left interval window function by deducting described right interval window function from 1.

According to this structure, take out the window function in right half part interval, the window function that takes out is generated as right interval window function, and, generate left interval window function based on the right interval window function of 1-.Therefore, can generate left interval window function and the right interval window function that finite interval equates by simple the processing.

(4) comparatively it is desirable to, described window function handling part exists a plurality of, the described base period of each window function handling part is different, described timing test section is obtained from the peak period of the correlation peak of each received signal of each window function handling part output, approach most the correlation peak of received signal in the cycle of described known symbol based on described peak period, the symbol that detects described multi-carrier signal regularly.

According to this structure, the different respectively window function handling part of a plurality of base periods is set.Therefore, can utilize regularly from the received signal detected symbol of the window function handling part output that can suppress to disturb intensity of wave, can precision more the highland detected symbol is regularly.

(5) comparatively it is desirable to, also comprise the known symbol storage part, storage is used for a plurality of subcarriers of generation of described known symbol and the multiple time shaft oscillogram that generates in advance, frequency band is different by combination, wherein, described timing test section is obtained from the received signal of described window function handling part output and the peak period of the correlation peak between each time shaft oscillogram, approach most the time shaft oscillogram in cycle of described known symbol and the correlation peak between the described received signal based on described peak period, the symbol that detects described multi-carrier signal regularly.

According to this structure, a plurality of time shaft oscillograms of the known symbol that service band is different are respectively obtained correlation peak.Therefore, can use the time shaft oscillogram of the frequency band that is not loaded with disturbing wave to obtain correlation peak, utilize the correlation peak that uses this time shaft oscillogram to obtain to come detected symbol regularly, thus can more highland detected symbol timing of precision.

(6) comparatively it is desirable to, described timing test section comes out with interior correlation the scope of appointment as correlation peak detection, and utilizes this correlation peak to obtain described peak period.

According to this structure, will come out as correlation peak detection with interior correlation in the scope of appointment, and use this correlation peak to obtain peak period, therefore detected symbol timing accurately.

Claims (7)

1. a receiving system detects the symbol timing of the multi-carrier signal of the front pilot that comprises the known symbol repetition, it is characterized in that comprising:

Window function handling part, utilization have the window function of the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol, carry out window function to received signal and handle; And

Test section is regularly obtained the correlation peak between the time shaft oscillogram that described window function handling part carries out received signal that window function handles and described known symbol, based on this correlation peak, detects the symbol timing of described multi-carrier signal, wherein,

Described window function handling part comprises:

The window function generating unit, the left-half interval or the right half part interval of taking out described window function, the window function based on taking out generates left interval window function and right interval window function;

The processing signals generating unit, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of the finite interval of recalling described right interval window function from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And

Efferent, with described first processing signals and described second processing signals chronologically added signal as the finite interval that dates back described right interval window function from described benchmark constantly before till received signal, export described timing test section to.

2. receiving system according to claim 1 is characterized in that:

Described window function generating unit is taken out the window function in described left-half interval, and the window function that takes out is generated as described left interval window function, generates described right interval window function by deducting described left interval window function from 1.

3. receiving system according to claim 1 is characterized in that:

Described window function generating unit is taken out the window function in described right half part interval, and the window function that takes out is generated as described right interval window function, generates described left interval window function by deducting described right interval window function from 1.

4. according to each described receiving system in the claim 1 to 3, it is characterized in that:

Described window function handling part exists a plurality of,

The described base period of each window function handling part is different,

Described timing test section is obtained from the peak period of the correlation peak of each received signal of each window function handling part output, approach most the correlation peak of received signal in the cycle of described known symbol based on described peak period, the symbol that detects described multi-carrier signal regularly.

5. according to each described receiving system in the claim 1 to 4, it is characterized in that also comprising:

Known symbol storage part, storage are used for a plurality of subcarriers of generation of described known symbol and the multiple time shaft oscillogram that generates in advance, frequency band is different by combination, wherein,

Described timing test section is obtained from the received signal of described window function handling part output and the peak period of the correlation peak between each time shaft oscillogram, approach most the time shaft oscillogram in cycle of described known symbol and the correlation peak between the described received signal based on described peak period, the symbol that detects described multi-carrier signal regularly.

6. according to each described receiving system in the claim 1 to 5, it is characterized in that:

Described timing test section comes out with interior correlation the scope of appointment as correlation peak detection, and utilizes this correlation peak to obtain described peak period.

7. a method for detecting symbol timing detects the symbol timing of the multi-carrier signal of the front pilot that comprises the known symbol repetition, it is characterized in that may further comprise the steps:

The window function that window function treatment step, utilization have the finite interval shorter than the base period of the integral multiple in the cycle of described known symbol carries out the window function processing to received signal; And

Regularly detect step, obtain the correlation peak between the time shaft oscillogram of the received signal that is carried out processing by described window function treatment step and described known symbol, and based on the symbol timing of the described multi-carrier signal of this correlation peak detection, wherein,

Described window function treatment step may further comprise the steps:

Window function generates step, and the left-half interval or the right half part interval of taking out described window function generate left interval window function and right interval window function based on the window function that takes out;

Processing signals generates step, with moment of received signal of finite interval of recalling the moment of described base period constantly from the benchmark of appointment and recalling described left interval window function again as starting point, described left interval window function multiplied each other with described received signal chronologically generate first processing signals, and with moment of the described received signal of the finite interval of recalling described right interval window function from described benchmark constantly as starting point, described right interval window function multiplied each other with described received signal chronologically generates second processing signals; And

The output step with described first processing signals and described second processing signals added signal chronologically, is exported as the received signal till before the finite interval that dates back described right interval window function from described benchmark constantly.

Images