JPH08256095A - Method and device for processing signal - Google Patents

Abstract

PURPOSE: To accurately recover a source signal by correcting the Doppler deviation of a received signal in the case of communication, in which relative speed is high and changed, between a signal source and a reception station. CONSTITUTION: A signal transmitted from a high-speed mobile object 1 is received by a reception station 2, converted to an intermediate frequency by a mixer 3 and a band pass filter 4 and converted to a time sequential digital signal by an A/D converter 5. Concerning time sequential data outputted from the A/D converter 5, the time of delay caused by the delay of transmission between the signal source and the reception station is corrected by a delay time correction part 6, its phase is corrected with a central frequency as a reference by a correction value calculated by a correction value calculation part 11 and further, Fourier transformation is performed to those data by a Fourier transformation part 8. The Fourier transformed data are multiplied with the correction value, that is calculated by the correction value calculation part 11, by a frequency data phase correction part 9 and the phases of signals having respective frequency components excepting for the central frequency are corrected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital signal processing method and apparatus for wireless communication with a high-speed moving body. When the signal transmission source moves at high speed, the position relative to the signal receiving station changes, and the received signal undergoes the Doppler shift accordingly. The present invention particularly relates to a signal processing method and apparatus for correcting the above Doppler shift and restoring the original signal.

[0002]

2. Description of the Related Art In a conventional signal processing apparatus, radio communication is performed with a geostationary satellite whose relative position with respect to a ground station does not change so that Doppler shift does not occur. When Doppler shift occurs like in aircraft and ground stations, it is necessary to add the Doppler frequency correction amount according to the relative speed to the normal local oscillation frequency and convert it to an intermediate frequency signal with the influence of the Doppler effect removed. .

[0003]

By the way, since the geostationary satellite is fixed on the equator, it cannot communicate with the polar regions. For this reason, a mobile satellite is required to secure communication with the polar regions, but Doppler shift is inevitable.

Further, if the moving speed is constant, the influence of the Doppler effect can be removed by the above-mentioned conventional technique. However, in the case of a mobile satellite, the relative speed with the ground station changes, so that it is adjusted to the relative speed. There is a problem in stability and frequency accuracy when changing the Doppler correction frequency by
As the relative speed increases, the Doppler correction frequency also increases, which causes a problem in frequency accuracy.

The present invention has been made in consideration of the above-mentioned problems of the prior art, and an object of the present invention is to provide Doppler even when the relative speed is large and the relative speed changes. An object of the present invention is to provide a signal processing method and apparatus capable of correcting deviation and restoring an original signal.

[0006]

FIG. 1 is a diagram illustrating the principle of the present invention. In the figure, 1 is a high-speed moving object such as an artificial satellite, 2
Is a receiving station, 3 is a mixer for mixing the received signal with the local frequency f _L , 4 is a band pass filter for passing a signal in a specific band of the output signals of the mixer 3, and 5 is an output of the band pass filter 4. An AD converter that performs sampling and AD conversion to convert to time series data, 6 a delay time correction unit that corrects the delay time of the time series data, 7 a time series data phase correction unit that corrects the phase of the time series data, and 8 A Fourier transform unit, 9 is a frequency data phase correction unit that corrects the phase of the signal after the Fourier transform, 10 is a center frequency generation unit that generates a center frequency f ₀ that serves as a reference for correction, 11 is a delay correction value, time series data Is a correction value calculation unit for calculating the correction value and the correction value of the frequency data, and reference numeral 12 is a numerical operation processor unit for supplying data required when the correction value calculation unit 11 obtains various correction values.

In order to solve the above-mentioned problems, the invention of claim 1 of the present invention is, as shown in FIG. 1, in the environment where the relative positions of the signal source and the receiving device change with time, Signal is received, the received high-frequency signal is frequency-converted into a low-frequency signal and converted into a digital signal, and by correcting the signal extraction position from the obtained time-series digital signal, the reference position to the reception position After correcting the delay time, the phase of the corrected time-series digital signal is corrected with reference to its center frequency, and the phase-corrected time-series digital signal is Fourier-transformed. The phase is corrected to correct the Doppler shift of the received signal.

According to a second aspect of the present invention, as shown in FIG. 1, a signal from a signal source is received in an environment in which the relative positions of the signal source and the receiving station change with time, and the Doppler shift occurs. In the signal processing device for correcting the received signal received, a frequency conversion unit for converting the received high frequency signal into a low frequency signal, sampling the frequency converted signal,
An AD converter for converting to a time-series digital signal, a delay time correction unit for correcting the delay time from the reference position to the reception position by correcting the data extraction position from the time-series digital signal, and the delay time correction The time-series data phase correction unit that corrects the phase of the time-series digital signal that has been corrected with reference to the center frequency, the Fourier transform unit that performs the Fourier transform of the phase-corrected signal, and the signal after the Fourier transform for each frequency component. A frequency data phase correction unit for phase correction is provided.

According to a third aspect of the present invention, in the second aspect of the invention, the time-series data is written in the buffer memory, and the read position of the data written in the buffer memory is set as the delay time from the reference position to the reception position. (B) Sampling of the time-series digital signal by correcting the delay time and correcting the delay time, and by multiplying the time-series digital signal by the correction value and correcting the phase based on the correction center frequency. The time shift and (b) the phase of the start position of the time-series digital signal are corrected, and (c) the time-series digital signal described above (b) for each sampling data.
The time-series data phase corrector that corrects the displacement from the correction value and the phase correction by multiplying the Fourier-transformed signal by the correction value to correct the sampling time shift for each frequency component after the Fourier transform. A frequency data phase correction unit is provided.

According to a fourth aspect of the present invention, in the second or third aspect of the invention, the correction center frequency is set to a value that is an integer multiple of the sampling frequency with respect to the local frequency. According to a fifth aspect of the present invention, a reception band is divided, a correction center frequency is determined for each reception band, and the signal processing device according to the second, third or fourth aspect is provided for each band. .

The invention of claim 6 of the present invention is the invention of claims 3 and 4.
Alternatively, in the invention of claim 5, the delay time correction unit is provided with M (M = 1, 2, 3, ...) Buffer memories and the M
By sequentially writing the time-series data in each buffer memory and performing delay time correction, the delay time correction unit is M times the sampling time interval. According to a seventh aspect of the present invention, in the third, fourth, fifth or sixth aspect of the invention, the correction center frequency is 1 / M ((M = 1, 2, Three
...) is set to a value separated by an integer multiple.

The invention of claim 8 of the present invention is the method of claim 2, 3, 4, 5 or claim 6 for performing delay correction of a plurality of receiving stations and respective signals received by the plurality of receiving stations. A signal processing device and a correlation processing unit that performs a cross-correlation process on the processing results of each signal processing device are provided, and the signals received by the plurality of receiving stations are referenced based on the reference receiving station or the virtual receiving station. The delay correction is performed, and the correlation processing unit performs the cross-correlation processing of the signal subjected to the delay correction.

According to a ninth aspect of the present invention, in the eighth aspect of the present invention, an accumulation processing section is provided, and a result of cross-correlation processing performed a plurality of times is accumulated for each frequency component. . The invention of claim 10 of the present invention is,
In the invention of claim 4, 5, 6 or 7, a receiving station for receiving a signal having a plurality of frequency components sent from a transmission source is provided, and the receiving station delays and corrects the received signal and then receives the signal. The configuration is such that the presence or absence of a specific frequency signal in the signal or the level of the specific frequency signal is determined and the received signal is decoded.

[0014]

In FIG. 1, a signal f _RF transmitted by a high-speed moving body 1 such as an artificial satellite is received by a receiving station 2, mixed with a local frequency f _L by a mixer 3, and an intermediate frequency is taken out by a band pass filter 4. The received signal converted into the intermediate frequency is sampled at a predetermined cycle and converted into a time series digital signal (hereinafter referred to as time series data) by the AD converter 5.

FIG. 2 shows an original signal A transmitted by the high-speed moving body 1.
And a received signal B subjected to Doppler shift. The black circles and double circles of signal A and signal B in the figure correspond to each other. Although the received signal B includes signals of various frequency components, the signal converted to the intermediate frequency is shown by a sine wave in FIG. As shown in the figure, the original signal A transmitted by the high-speed moving body 1 arrives at the receiving station 2 with a delay of d ₀ until the signal arrives at the receiving station 2 from the high-speed moving body 1, and the original signal A The cycle c1 is shifted to the position of the cycle c2 of the signal B as shown in the figure.

Further, the signal B undergoes Doppler shift and is out of phase (the signal C shown by the dotted line in the figure corresponds to the signal before the Doppler shift, and in the present invention, the signal is obtained by the method described later. The phase correction of B is performed to obtain the signal C shown by the dotted line). The AD converter 5 shown in FIG. 1 samples the signal B at sampling times S _t0 , S _t1 , S _t2 , ... And converts it into time-series data [1 in FIG. Although sampling is performed eight times per cycle and eight time-series data (each of these eight time-series data is hereinafter referred to as one segment) is obtained per cycle, the number of sampling points is an arbitrary value. can do〕.

Here, as shown in FIG. 2, the time at which the signal B passes through the zero point and the sampling time do not necessarily match, and the start point of the cycle c2 of the signal B on the time series data is the sampling time S _t3. Equivalent to (S _t0 to S
_If the time until _t3 is D, the delay time is d ₀ = D + ΔD). Returning to FIG. 1, the time-series data output from the AD converter 5 is sent to the delay time correction unit 6, where the delay time deviation from the reference position to the reception position is corrected.

For example, the delay time correction unit 6 is provided with a buffer memory 6a, and the time series data output from the AD converter 5 is written in the buffer memory 6a to obtain the delay correction value D obtained by the correction value calculation unit 11. According to the buffer memory 6
The delay time correction is performed by reading the time series data from the position corresponding to a. FIG. 3 is a diagram for explaining the delay time correction method. As shown in FIG. 3, the input signal is sampled and AD-converted into time-series data.
It is written in the buffer memory 6a.

When the first delay time correction value D _A is "3", the address AD of the buffer memory 6a
Data of one segment from the position of ₃ D _t3 , D _t4 , ..., D
_t10 is read. Then, the next delay time correction value D _B
Becomes "4", the address AD of the buffer memory 6a
The data D _t12 , D _{t13 of} the next segment from the position of ₁₂
..., _Dt19 is read. Similarly, data is read from the positions corresponding to the delay time correction values D _C , D _D , ..., [D _A0 , D _A1 , ..., D _A7 ], [D _B0 , D _B1 ,
, D _B7 ,], time series data is generated. In the example of the figure, the delay time correction value changes from D _A = 3 to D _B.
= 4, the data of _Dt11 is skipped and the data is lost.

[0020] Figure 2 shows how the data D _t11 as described above is missing, as shown in the figure, the data D _A0 ~
D _A7 in which data is composed of one cycle (1 segment) (black circle data in the same figure), in the next cycle, data D _t11 sampling time S _t11 is skipped,
The next one segment starts from the sampling time S _t12 (double circled data in the figure).

Returning to FIG. 1, the time-series data whose delay time is corrected by the delay time correction unit 6 is sent to the time-series data phase correction unit 7, and the phase correction is performed by the time-series data correction value obtained by the correction value calculation unit 11. Is done. Here, as described above, the signal B includes signals of various frequency components, but among the frequency components, the time-series data phase correction unit 7 is generated by the center frequency information generation unit 10. The phase at the center frequency f ₀ is corrected as follows to correct the Doppler shift. (Details of the phase correction method will be described later). By multiplying the time series data extracted from the delay time correction unit 6 by the correction value, the phase of the signal is corrected with the center frequency as a reference, and the deviation of the sampling time is corrected. By multiplying the time series data extracted from the delay time correction unit 6 by the correction value, the phase of the head position of the time series data is corrected with the center frequency as a reference. The time-series data extracted from the delay time correction unit 6 is multiplied by a correction value, and the displacement from the above correction value is corrected for each sampling data based on the center frequency.

The time-series data whose phase has been corrected by the time-series data phase correcting section 7 as described above is sent to the Fourier transforming section 8 where it is Fourier transformed and given to the frequency data phase correcting section 9. The frequency data phase correction unit 9 multiplies the frequency data correction value obtained by the correction value calculation unit 11 by the signal after the Fourier transform, and corrects the deviation of the sampling time for the signal of each frequency component other than the center frequency.

Next, the principle of the present invention will be described in detail using mathematical expressions. As shown in FIG. 4, the signal transmitted from the signal source 1 is received by the satellite station or ground station 2, 2 ', and the received RF frequency signal is mixed with the local frequency fL by the mixer 3 and converted into the intermediate frequency IF. Now, consider a case where a low-frequency signal is taken out through the bandpass filter 4. (1) Received signal Time from the signal source S until the signal reaches the reference receiving station 2 ', and Ts reaches the receiving station 2 whose position is relatively changed from the reference receiving station. The signal received by the receiving station 2 is expressed by the following equation. x _S = cos [2πf _RF {t− (T _B + T _S )}] A signal at a virtual signal receiving station (reception reference station 2 ′ in FIG. 2) whose position relative to the signal source does not change is used as a reference. Considering this, the above equation is expressed by the following equation. x _S = cos {2π
where _{f RF (t-T S)} }, T S initial value T _S0 of, (a function of the generally v time t, v is known here) the moving velocity v, and the speed of light c = 300000Km Then, Ts = (v / c) t + T _S0 x _S = cos [2πf _RF {t− (v / c) t−T _S0 }] = cos [2π [f _RF {1- (v / c)} t− f _RF · T _S0 ]] That is, the signal received by the receiving station is subjected to the Doppler shift represented by the above equation.

The time t is set to the sampling frequency F_SIn total
Expressed by the counted time n, t = n / F _SAnd the FFT point
Number (Fast Fourier Transform score, Fourier in segment units)
D) In the example of FIG. 2, 8) is converted into N, m = 0, 1, 2,
, I = 0, 1, 2, ..., N−1, t = (N ·
m + i) / F_SAnd also T_s0The sampling frequency
Number F_STime K counted in_S0Is expressed as T_S0= K_s0/ F_S
Therefore, the above x_SIs expressed by the following formula. x_S= Cos [2π [f_RF{1- (v / c)} (Nm + i) / F_s -F_RF(K_S0/ F_S)]] (2) Delay time The delay for the signal receiving station to receive the same signal as the reference station.
Time d_xThen, the following formula is established. f_RF{1- (v / c)} {(Nm + i) / F_s+ (D_x/ F_S)} -F_RF(K_S0/ F_S) = F_RF{(N ・ m + i) / F_s} Therefore, d_x= {(V / c) (Nm + i) + K_S0} / {1- (v / c)} The delay time when i = 0 is d₀Then, d₀Is the following
Can be expressed as (this delay time d₀Is shown in FIG.
D₀Equivalent to). d₀= {(V / c) Nm + K_S0} / {1- (v / c)} (3) Intermediate frequency conversion The received RF frequency signal is f_RF, Local frequency f
_LAnd the signal received by the mixer 3 has the above local frequency
f_LMultiply the signal of_s
Is obtained. x_S= Cos [2π [f_RF{1- (v / c)} t-f_RF・ T_S0]] ・ Cos 2πf_Lt = 1/2 [cos [2π [f_RF{1- (v / c)} t-f_Lt-f_RF・ T_S0]] + Cos [2π [f_RF{1- (v / c)} t + f_Lt-f_RF・ T_S0]]] In the band pass filter 4, the component of the second term in the above equation is filtered.
Filter and leave only the first term and double the amplitude.
Then, xs is as follows. x_S= Cos [2π [f_RF{1- (v / c)} t-f_Lt-f_RF・ T_S0]] Here, t = (N · m + i) / F_S, T_S0= K_s0/ F_S
Then x_SCan be transformed as follows. x_S= Cos [2π [[f_RF{1- (v / c)}-f_L] {(N ・ m + i) / F_S } -F_RF(K_s0/ F_S)]] (4) Delay correction Delay time d₀Is divided into an integer part D and a decimal part ΔD,
Delay time d shown in (2)₀Can be expressed as
It The above D and ΔD are the same as D and ΔD in FIG.
Correspond. d₀= {(V / c) Nm + K_S0} / {1- (v / c)} ≡D + ΔD where x shown in (3) above_sIf only D is corrected,
Corrected signal x_s'Is represented by the following formula. x_s'= Cos [2π [[f_RF{1- (v / c)}-f_L] [(N ・ m + i + D) / F_S] -F_RF(K_s0/ F_S)]] By modifying the above equation, the following equation is obtained. x_s'= Cos [2π [(f_RF-F_L) (N ・ m + i) / F_S -F_RF(V / c) i / F_S-F_RF{1- (v / c)} · ΔD / F_S -F_L・ D / F_S]] Here, as shown in FIG. 5, the reference center frequency is f
₀And the center frequency f ₀And the reception frequency f_RFThe difference of f_0k,
Local frequency f_LAnd the reception frequency f_RFDifference with f_kTosu
Then f_RF= F_L+ F_k, F_RF= F₀+ F_0kTo represent
Can be.

Here, the condition that the difference between the center frequency f ₀ and the local frequency f _L is an integral multiple of the sampling frequency F _s , that is, f _L = f ₀ −F _s · B (B = 1, 2, ...) Given the conditions in _{which), cos (2πf L · D} / F S) = cos _{[2π (f 0 -F S · B} ) D / F S ] = cos _{[2π (f 0 · D / F} S -F _S · B · D / F _S ] = cos (2πf ₀ · D / F _S ).

Further, considering the condition of D = D _m · M (M = 1, 2, ...), f _L = f ₀ − (F _s / M) · B
(B = 1,2, ...) and _{when, cos (2πf L · D /} F S) = cos _{[2π {f 0 - (F s} / M) · B} D / F S) = cos [2 [pi {f ₀ · D / F _S − (F _s / M) · B} D _m · M / F _S ] = cos [2π (f ₀ · D / F _S −B · D _m )] = cos (2πf ₀ · D / F _S ) This is because the delay correction value D is an integer multiple of the integer value M. Then
The difference between the local frequency f _L and the center frequency f ₀ is F _S / M
It indicates that (F _S is a sampling frequency) may be an integral multiple.

[0027] Now, said in _{x s' f RF = f L} + f k, f
Substituting _RF = f ₀ + f _0k and transforming the corrected signal x _s ′ gives: x _s ′ = cos [2π [(f _RF −f _L ) (N · m + i) / F _S −f _RF (v / c) i / F _S −f _RF {1- (v / c)} · ΔD / F _S −f ₀ · D / F _S ]] = cos [2π [f _K (N · m + i) / F _S −f ₀ (v / c) i / F _S −f _0k (v / c) i / F _{S -f 0 · ΔD / F S} + f 0 (v / c) · ΔD / F S -f 0k · ΔD / F S + f 0k (v / c) · ΔD / F S -f 0 · D / F S] ] = cos _{[2π [f K (N · m} + i) / F S: Fourier transform target wave _{-f 0 (v / c) i} / F S: phase correction _{(ΔP · i) -f 0k (} v / c) i / F _S: error _{-f 0 · d 0 / F S} : phase correction _{(P 0) + f 0 (} v / c) · ΔD / F S: phase correction _{(ΔP · ΔW) -f 0k ·} ΔD / F S: Delay residual correction (ΔW) + f _0k (v / c) · ΔD / F _S ]]: error where f _k = k (F _s / N), f ₀ = k ₀ (F _S /
N), f _0k = (k−k ₀ ) (F _s / N), x _s ′ = cos [2π [k {m + (i / N)}: Fourier transform target wave −k ₀ (v / c) ) I / N: Phase correction (ΔP · i) − (k−k ₀ ) (v / c) i / N: Error −k ₀ · d ₀ / N: Phase correction (P ₀ ) + k ₀ (v / c) ) .ΔD / N: Phase correction (ΔP · ΔW)-(k−k ₀ ) · ΔD / N: Delay residual correction (ΔW) + (k−k ₀ ) (v / c) · ΔD / N]] : Error Further, (k−k ₀ ) (v / c) i / N and (k−k ₀ ).
If the term of (v / c) · ΔD / N is ignored as an error, the above equation becomes as follows. x _s ′ = cos [2π [k {m + (i / N)}: Fourier transform target wave −k ₀ (v / c) i / N: Phase correction (ΔP · i) −k ₀ · d ₀ / N: Phase correction (P ₀ ) + k ₀ (v / c) · ΔD / N: Phase correction (ΔP · ΔW)-(k−k ₀ ) · ΔD / N]]: Delay residual correction (ΔW) In the above formula, The first item is a signal to be subjected to Fourier transform, and the second and subsequent items are signals to be subjected to phase correction.

The second term is a phase correction amount determined by each sampling time i, which corresponds to ΔP in FIG. 2 and is corrected for each sampling data. The third term is a phase correction amount determined by the time delay d _0, which corresponds to P ₀ in FIG. 2 described above and is a phase correction amount at the head position of each time series data. The fourth term is a phase correction amount determined by the sampling time shift ΔD.

Further, the fifth term is the frequency after the Fourier transform.
The phase correction amount corrected by the numerical data phase correction unit.
It (5) Phase correction ΔP = k₀(V / c) / N, P₀= K₀・ D₀/ N
Then, the above equation is transformed as follows. x_s'= Cos [2π [k {m + (i / N)} − ΔP · i−P₀+ ΔP · ΔD − (k−k₀) .ΔD / N]] where e^jx= Cos x + jsin x, e^-jx= Cos x-j
From sin x, cos x = 1/2 · (e^jx+ E^-jx)
Therefore, if the amplitude is doubled, the above x_s'Is the real part
It can be transformed into the following equation consisting of an imaginary part. x_s'= Exp [j2π [k {m + (i / N)} − ΔP · i−P₀ + ΔP ・ ΔD- (kk₀). [Delta] D / N]] + exp [-j2 [[k {m + (i / N)}-[Delta] P.i-P₀ + ΔP ・ ΔD- (kk₀) .ΔD / N]] ΔP, P in the first term of the above equation₀To delete, exp
[J2π (ΔP · i + P ₀-ΔP · ΔD)]
Therefore, exp [j2π (ΔP · i + P₀−
ΔP · ΔD)], the center frequency k₀Phase correction with
Then, it becomes as follows. x_s″ = Exp [j2π [k {m + (i / N)} − (k−k₀). [Delta] D / N]] + exp [-j2 [[k {m + (i / N)}-2. [Delta] P.i-2.P₀ + 2 · ΔP · ΔD- (kk₀) .ΔD / N]] In the above equation, the second term is the residual difference, which will be described later.
Are discarded after the Fourier transform. Also, (k in the first term
-K₀) -ΔD / N is the delay residual, which will be described later.
It is corrected after performing the Fourier transform.

As described above, in the time-series data phase correction unit 7 shown in FIG. 1, the correction amount exp [j2π (ΔP · i + P ₀ −ΔP) obtained by the correction value calculation unit 11 is calculated for x _s ′.
・ ΔD)] is applied to correct the phase of the time series data. (6) Fourier transform The Fourier transform of the above x _s ″ is given by the following equation: X _{S (f)} = Σx _s ″ exp {−j2πf (i / N)} · Δ (i / N) where Σ is From i = m to i = [m + {(N-1) /
N}].

Here, δ (f) = ∫exp (j2πf
t) and dt, if omitted X S of _{(f) (f) is} X _{S (f) is} denoted as X _S, X _S is as follows. X _S = exp [−j2π {(k−k ₀ ) · ΔD / N}] · δ [f−k] + exp [−j2π {−2 · ΔP · i−2 · P ₀ + 2 · ΔP · ΔD- ( for _{k-k 0) · ΔD /} N] ] · [delta] [f + k] (7) the delay residual correction Furie converted signal X _S, wherein (similar to 5), the X _s exp [j2? {(k- k ₀ ) ・ ΔD /
N}] is applied to perform delay residual correction.

X _s ″ = X _S · exp [j2π {(k−k ₀ ) · ΔD / N}] = δ (f−k) + exp [−j2π {−2 · ΔP · i−2 · P ₀ +2 · ΔP · ΔD-2 · ( k-k 0) · ΔD / N] ] · [delta] [f + k], where the second term is the error fraction, from the Fourier transform unit 8 shown in FIG. 1, [delta] Only the output on the (f−k) side is taken out, and the δ [f + k] side is discarded.

The frequency data phase correction section 9 shown in FIG. 1 outputs the above exp [j2π {(k−k ₀ ) · ΔD / N}] to the output on the δ (f−k) side of the Fourier transform unit 8. Then, the delay residual is corrected to obtain the phase-corrected δ (f−k).

[0034]

FIG. 6 is a diagram showing the overall configuration of the first embodiment of the present invention. In FIG. 6, the high-speed moving body 1, the receiving station 2, the mixer 3, and the bandpass filter 4 shown in FIG. Although omitted, it has the same configuration as that of FIG. 1 except for the illustrated portion. In the figure, 5 is an AD converter for sampling the signal converted into the intermediate frequency and converting it into time series data, 6
Is a delay time correction unit, and the delay time correction unit 6 includes a buffer memory 6a and a write address counter 6 that determines a write address WADRS for writing data to the buffer memory 6a.
b, and the data written in the buffer memory 6a is read address RADRS output by the correction value calculation unit 11.
Read from.

Reference numeral 7 is a time series data phase correction unit,
The phase correction unit 7 includes multipliers 7a and 7b, and a delay time correction unit
The correction value calculation unit 1 is added to the time series data x read from 6
1 output time-series data correction value Pr + jPi (e above)
xp [j2π (ΔP · i + P ₀-ΔP ・ ΔD)]
To set the phase of the time series data to the center frequency.
Correct accordingly.

Reference numeral 8 is a Fourier transform unit, 9 is a frequency data phase correction unit, and the frequency data phase correction unit 9 is a multiplier 9
a, 9b, 9c, 9d and adders 9e, 9f, and the correction value Wr + jWi of the frequency data output by the correction value calculation unit 11 is added to the signal Fr + jFi after the Fourier transform (the above ex
p [j2π {(k−k ₀ ) · ΔD / N}])
And the deviation of the sampling time for the signal of each frequency component other than the center frequency is corrected.

FIG. 7 is a block diagram of the correction value calculation unit and the compensator shown in FIG.
Numerical calculation process for calculating parameters for positive value calculation
It is a figure which shows the structure of a spacer part. In the figure, 12 is a number
The numerical value arithmetic processor unit 12
Is a display / keyboard 121 and a LAN control unit.
122, hard disk 123, and floppy disk
And the parameter d required for calculating the correction value ₀, M,
..., EXT storing N. Register 125 and the above parameters
Data N and F required to calculate the meter_S, F₀,
T_S0, V₀, V₁Instruction data memory for storing ,, etc.
126, etc., which are MPU 127 and system bus
Connected through.

Then, the MPU 127 uses the data and the like stored in the command data memory 126 based on the instructions / data and the like given from the display / keyboard 121 and the like to obtain the parameters d ₀ , M, and
..., N is calculated, and EXT. It is stored in the register 125.
Reference numeral 11 is a correction value calculation unit, and the correction value calculation unit 11 includes a counter 111 and a shifter 112 that generates an integer part D and a decimal part ΔD of the delay correction value.

Then, the counter 111 counts up at each sampling time and generates an address value corresponding to the sampling time and an integer value i indicating what number data is in one segment. Further, the shifter 111 is the EXT. The delay time d ₀ given from the register 125 and the EXT. From the integer value M given from the register 125, the integer part D and the decimal part ΔD of the delay correction value that is a value that is an integer multiple of the integer value M is generated.

In this embodiment, the delay correction value D is set to an integral multiple of the integer value M, and the difference between the local frequency f _L and the center frequency f ₀ is F _S / M (F _S Is a sampling frequency) is an integer multiple. Also, 113
a, 113b, 113c, 113d are adders, 114
a, 114b, 114c are multipliers, 115 is a divider, 1
16a and 116b are table search units, and the table search units 116a and 116b use the input value x as an address and output exp (-j2πx) [= cos (2πx) -jsin.
(2πx)], and when the input x is given, the exp (-
j2πx) is output.

It should be noted that the data stored in the read-only memory should be set to a value that corrects a conversion error (so-called DC offset bias) of the AD converter or the like in advance to cancel the steady error. You can Next, the operation of the embodiment shown in FIGS. 6 and 7 will be described. In FIG. 6, the received signal converted to an intermediate frequency by a mixer and a band pass filter (not shown) is AD
Sampling and AD conversion are performed by the converter 5, and the result is given to the delay time correction unit 6.

The write address counter 6b of the delay time correction unit 6 writes the write address WA to the buffer memory 6a.
The DRS is output, and the time series data output by the AD converter 5 is written to the write address WADRS of the buffer memory 6a at each sampling time. On the other hand, the counter 111 of the correction value calculation unit 11 counts up at each sampling time, and the count value is added by the adder 113a.
Is added to the integer part D of the delay correction value that is output to the shifter 112.

As a result, the adder 113a adds the address value corresponding to the sampling time (corresponding to the address value of the write address counter 6b) and the delay correction value D, that is, the delay-corrected read from the buffer memory 6a. Generate address RADRS. The read address RADRS is given to the buffer memory 6a of the delay time correction unit 6 shown in FIG. 6, and the time series data is read from the position corresponding to the address RADRS of the buffer memory 6a. As a result, the delay time correction shown in FIG. 3 is performed.

On the other hand, the multiplier 1 of the correction value calculation unit 11 of FIG.
14a is an integer value i output from the counter 111 and a numerical operation
The EXT. Given from register 125
ΔP (ΔP = k as described above₀(V / c) / N
And a) are output and ΔP · i is output. Also, the multiplier 1
14b multiplies ΔD output from the shifter 112 and ΔP described above.
And outputs ΔP · ΔD, and the adder 113c outputs the EXT. Les
P given by Dista 125 ₀(As mentioned above, P₀
= K₀・ D₀/ N).

Then, in the adder 113b, -ΔP is calculated from the output of the multiplier 114a and the output of the adder 113c.
Obtain i-P ₀ + ΔP · ΔD (= x). The above x is given to the table search unit 116a, converted into exp (-j2πx) corresponding to x as described above, and the phase correction value Pr + jPi (exp [-j2π) of the time series data is converted.
(Corresponding to (−ΔP · i−P ₀ + ΔP · ΔD)]).

Correction value Pr + jP of the above-mentioned time series data
i is given to the time series data phase correction unit 7 shown in FIG.
The time-series data phase correction unit 7 multiplies the correction value of the time-series data and the time-series data x read from the buffer memory 6a by the multipliers 7a and 7b to obtain (x · Pr + j
x.Pi) to obtain phase-corrected time series data with the center frequency as a reference.

The output of the time-series data phase correction unit 7 is given to the Fourier transform unit 8. The Fourier transform unit 8 Fourier-transforms the time-series data in segment units and outputs Fourier-transformed data Fr + jFi. On the other hand, the adder 113d of the correction value calculation unit 11 uses the EXT. Register 125
I given from k ₀ and the counter 111 supplied from
(K−k ₀ ) based on
k ₀ ) is divided by N to output (k−k ₀ ) / N.

The multiplier 114c multiplies (k−k ₀ ) / N which is the output of the divider 116 and ΔD given from the shifter 112 to obtain ΔD · (k−k ₀ ) / N (= x). Ask. The above-mentioned x is given to the table search unit 116b, and exp (-j2πx) corresponding to x is given here as described above.
To the phase correction value Wr + jWi of the time series data
(Corresponding to exp [−j2π (ΔD · (k−k ₀ ) / N)]) is obtained.

Correction value Wr + jW of the above frequency data
i is given to the frequency data phase correction unit 9 shown in FIG.
The frequency data phase correction unit 9 multiplies the correction values Wr + jWi of the frequency data by the multipliers 9a to 9d and the Fourier-transformed data Fr + jFi, and adds them by the adders 9e and 9f (Fr.Wr-Fi. Wi) + j (Fi ・ Wr +
Fr.multidot.Wi) is obtained, and the deviation of the sampling time for the signal of each frequency component other than the center frequency is corrected.

The output of the frequency data phase correction section 9 is further given to a computer or the like (not shown) for necessary analysis. As described above, in this embodiment, after the delay time of the reception signal converted into the intermediate frequency and converted into the time-series digital data is corrected, the phase of the corrected time-series digital signal is adjusted to the center frequency. Is used as a reference to perform Fourier transform, and the signal after Fourier transform is also subjected to phase correction for each frequency component, so the Doppler shift that occurs in high-speed mobile communication in which the relative speed changes is corrected digitally. Therefore, it is possible to perform stable communication with a high-speed moving body whose relative speed changes.

Further, since digital signal processing is performed instead of analog signal processing, adjustment and the like like analog signal processing are not required, and stable operation and reproducibility can be expected. Further, the delay correction value D is set to an integer multiple of the integer value M, and the difference between the local frequency f _L and the center frequency f ₀ is set to an integer multiple of F _S / M (F _S is a sampling frequency), which simplifies the calculation process. It can be carried out.

In the above embodiment, the example in which the digital processing is performed by hardware has been shown, but if the processing speed can be satisfied, the digital processing can be partially performed by using a general-purpose computer. In addition, in the correction calculation, the integer part may be unnecessary and only the decimal part may be needed. In this case, the method of calculating by the floating point arithmetic mechanism that uses the integer value and leaves only the decimal part is used. can do.

[0053] Further, when calculating the correction value of the time series data, - realizing _{(-P 0 + ΔP · ΔD)} -ΔP · i by (P ₀ + ΔP · ΔD) to continue to accumulate the -DerutaP It is possible to reduce the number of multipliers in this case. FIG. 8 is a diagram showing a second embodiment of the present invention. This embodiment is an embodiment in which a reception band is divided, a correction center frequency is set for each band, and phase correction is performed for each band. Is shown.

In FIG. 8, 2 is a receiving station, 3 is a mixer for mixing the local frequency f _L with the receiving frequency, 4 and 4'are band pass filters for passing intermediate frequencies, and 5 and 5'are converted to intermediate frequencies. It is an AD converter for sampling and AD converting the received signal. Further, 10 'is a center frequency generation unit, and the center frequency generation unit 10' adds the local frequency f _L to the product of the sampling frequency F _S and predetermined integer values B ₁ and B _{2 to} generate the first and second frequencies. Center frequency f ₀₁ , f
Generate ₀₂ .

Reference numerals 101 and 102 denote first and second processing sections, and the first and second processing sections 101 and 102 respectively have arithmetic processor sections 12 and 12 'and correction value calculation sections 11 and 1.
1 ', a delay time correction unit 6, 6', a time series data phase correction unit 7, 7 ', a Fourier transform unit 8, 8', and a frequency data phase correction unit 9, 9'are provided, and the time series data phase correction unit The parts 7 and 7'include the first and second center frequencies f ₀₁ and f.
_{Using 02} as a reference, the phase of each time-series data in the divided reception band is corrected.

Other configurations and operations are shown in FIGS.
The processing units 101 and 102 are the same as those shown in FIG.
After delay correction by 6 ', the phase of the time-series data is corrected by the time-series data phase correction section 7, 7'with reference to the first and second center frequencies f ₀₁ , f ₀₂ . Then, after the phase-corrected time series data is Fourier-transformed by the Fourier transform units 8 and 8 ′,
The frequency data phase correction unit 9, 9'performs phase correction for each frequency.

In the present embodiment, as described above, the reception band is divided, and the center frequency serving as a reference for correction is determined in each band, and the phase correction is performed.
The error due to the displacement from the center frequency can be reduced, and highly accurate correction can be performed. In the above embodiment, the example in which the reception band is divided into two has been shown, but the number of divisions of the reception band can be any number of 2 or more, and more accurate correction can be made according to the number of divisions. It becomes possible to do.

FIG. 9 is a diagram showing a third embodiment of the present invention. In this embodiment, two buffer memories are provided in the delay time correction section 6 and the two buffer memories are alternately used. Thus, an embodiment is shown in which a low-speed memory can be used as the buffer memory. In the figure, 2 is a receiving station, 3 is a mixer for mixing the local frequency f _L with the receiving frequency, 4 is a band pass filter for passing an intermediate frequency, and 5 is an AD for sampling & AD converting the received signal converted to the intermediate frequency. It is a converter. Further, 10 is a center frequency generation unit, and the center frequency generation unit 10 adds the local frequency f _L to the product of the sampling frequency F _S and a predetermined integer value B / M to generate the center frequency f ₀ .

Reference numeral 12 is a numerical operation processor unit, 11 is a correction value calculation unit, and the correction value calculation unit 11 is provided with a counter 111, a shifter 112 and an adder 113a, and by the shifter 112 as in the embodiment shown in FIG. Do / (log ₂ M) is calculated to obtain D and ΔD, and the delay correction value D is added to the address value output from the counter 11 to generate a read address from the buffer memory. Although the other configuration of the correction value calculation unit 11 is omitted, it has the same configuration as that shown in FIG.

Reference numeral 14 is a clock frequency dividing circuit, which counts the system clock by the counter 114a, and the decoder 1
14b decodes the first and second clocks CK
1 and CK2 are output. 15 is a demultiplexer, and the demultiplexer 15 has the clocks CK1 and CK.
Flip-flop FF triggered by 2 respectively
1, FF11, FF2, and FF22, the time series data output from the AD converter 5 is distributed to the buffer memories 6a and 6a 'provided in the delay time correction unit 6.

Reference numeral 6 denotes a delay time correction unit, and the delay time correction unit 6 of this embodiment is the first and second buffer memories 6
a, 6a 'and a multiplexer 6d that operates according to the clock CK1, write time series data alternately in the buffer memories 6a, 6a' and read the written time series data alternately from the correction position, as will be described later. Output via the multiplexer 6d for delay time correction.

Reference numeral 7 is a time-series data phase correction unit, and the time-series data phase correction unit 7 corrects the phase of the time-series data as in the first embodiment shown in FIG. Although the configuration of the time-series data phase correction unit 7 and the subsequent components is not shown in FIG. 9, this embodiment also has the same configuration as that of the first embodiment. FIG. 10 is a time chart showing the operation of this embodiment, and the operation of this embodiment will be described with reference to FIG.

The clock divider circuit 14 divides the system clock to generate the clocks CK1 and CK2 shown in FIG. On the other hand, from the AD converter 5, the time series data D _t0 , D _t1 , D _t2 , D _t3 , D _t4 , ...
Are supplied to the demultiplexer 15. The flip-flop FF1 of the demultiplexer 15 fetches the time series data D _t0 , D _t2 in synchronization with the rising edge of the clock CK1, and the flip-flop FF2 synchronizes with the rising edge of the clock CK2 in time series data D _t1 , D _t . _{Take in t3} .

[0064] time-series data taken into the flip-flop FF1, FF2 is taken into the flip-flop FF 11, FF22 at the rising edge of the clock CK1, the flip-flop FF 11, when the FF22 as shown in FIG. 10-series data D _t0, D _t2 , ..., Time series data D _t1 , D
_t3 , ... are set. Flip-flops FF11, F
The time series data set in F22 is sent to the delay time correction unit 6 and written in the buffer memories 6a and 6b according to the address output by the write control unit 6c.

Then, the time-series data written in the buffer memories 6a and 6b are read out from the address position outputted by the correction value calculation section 11, and are synchronized with the clock CK1 and alternately from the multiplexer 6d as shown in FIG. (Note that FIG. 10 shows a case where the time-series data is read without delay time correction). In the present embodiment, two buffer memories are provided in the delay time correction unit 6 as described above, and the two buffer memories are used alternately, so a low-speed memory can be used as the buffer memory. Becomes

Further, as a buffer memory, a memory in which one word is n bits is used, and these are used as a plurality of buffer memories to perform delay time correction. In the above embodiment, an example in which two buffer memories are provided is shown, but the number of buffer memories is not limited to two, and M buffer memories can be provided and used sequentially. This makes it possible to use a lower speed memory as the buffer memory.

FIG. 11 is a diagram showing a fourth embodiment of the present invention. In this embodiment, the receiving stations are provided at two locations, the received data of the respective receiving stations are processed, and then the output signals of those receiving stations are processed. An example is shown in which the movement distances due to crustal movements at two points can be precisely measured by correlating and accumulating. In FIG. 11, 1 is a signal source of a high-speed moving object, 2 and 2'are receiving stations, 103 and 104 are signal processing devices, and the signal processing devices are those shown in the first to third embodiments. Can be used.

Reference numeral 16 is a correlation processing unit,
Is a multiplier 16a, 16b, 16c, 16d and an adder 16
e, 16f, and the respective signal processing devices 103, 1
04 output phase-corrected signals Ar + jAi and Br
After multiplying each term of + jBi by addition, (Ar + jA
i) * (Br-jBi) = (Ar * Br-Ai * Bi)
+ J (Ar.Bi + Ai.Br) is obtained, and the real part Cr and the imaginary part Ci showing the cross-correlation result of the signals received at the above two locations are output.

Here, if the relative positions of the receiving stations 2 and 2'change, the correlation result will deviate depending on the moving distance. Therefore, as described above, by obtaining the correlation result of the signals received by the receiving stations installed at different points,
It is possible to obtain the relative moving distance of the receiving station due to crustal movements or the like. Reference numeral 17 denotes an accumulation processing unit. The accumulation processing unit 17 includes adders 17a and 17b, a buffer memory 17d including a number of memories corresponding to the number of Fourier transform points, a buffer memory write control unit 17c, and a buffer memory. Read-out control unit 17e.

Then, the correlation value output from the correlation processing section 16 is given to the adders 17a and 17b, and the adders 17a and 17b
In 17b, it is added with the correlation value read from the buffer memory 17d and stored in the buffer memory 17d. At that time, the correlation value at the time of the previous Fourier transform is read from the buffer memory 17d and added to the current correlation value, and (Dr + jDi) ← (Dr-1 + jDi-1 + Cr + j
Ci) is obtained, and the process of storing the added value in the buffer memory 17d is repeated. That is, the buffer memory 17d stores the accumulated value of the correlation value for each FFT unit.

Therefore, the random noise included in the received signal can be removed, and the signal can be restored even when the signal becomes weak relative to the noise. Here, if the standard deviation of the noise component that is normally distributed is σ, the value of the input data is log ₂ (3σ)
When expressed as the number of bits of, the number of bits required when accumulated W times is statistically log ₂ (3σ√W).

Therefore, from the number of bits of the input data,
The number of bits of the accumulation result register may be increased by log ₂ (√W) or more, or may be the number of bits required to express a value obtained by accumulating valid signals. As described above, in this embodiment, the signals received by the two receiving stations are delay-corrected by the method shown in the above embodiment, and the correlation results thereof are obtained. It is possible to find the distance traveled by the crustal movements of.

Further, by accumulating the correlation results, it is possible to reduce the influence of noise, and it is possible to accurately restore the signal even if the signal becomes weak with respect to the noise in the ultra long distance communication. . In the above embodiment, the example of removing the noise component by accumulating the correlation results is shown. However, the same is true even if the accumulation processing unit shown in the present embodiment is added to the first to third embodiments. The effect of noise can be eliminated.

FIG. 12 is a diagram showing a fifth embodiment of the present invention. In this embodiment, a transmission source for transmitting a signal having two frequency components is installed in a high-speed moving body or the like which is a signal source. An example is shown, and this embodiment is particularly suitable for application to ultra-long-distance communication with a high-speed flying object such as space communication. FIG. 12 shows a transmission source installed in an artificial satellite or the like. 20, 21 are oscillators that output carrier signals f1, f2, 22 is a transmission data generator, and 23, 23 'are DF.
The DF converters 23 and 23 ', which are converters, are composed of gates, and carrier signals f output from the oscillators 20 and 21 are generated.
1, f2 are modulated with transmission data.

Reference numeral 24 is a signal synthesizing unit
Is to combine the output signals of the DF converters 23 and 23 ′, and the combined signal is transmitted from the transmitter 25. FIG. 13 is a time chart showing the operation of this embodiment, and this embodiment will be described with reference to FIG. The transmission data output from the transmission data generation unit 22 are transmission data br + 1 and br + 0 consisting of 1 or 0 as shown in FIG. 13, and these transmission data are given to the DF conversion units 23 and 23 '.

The DF converters 23 and 23 'transmit data br +
Only when 1 and br + 0 are 1, the carrier signals output from the oscillators 20 and 21 are passed. As a result, transmission data f1 and f2 as shown in FIG. 13 are generated. Signal synthesizer 2
Reference numeral 4 synthesizes the transmission data f1 and f2 and outputs the transmission wave shown in FIG. The output of the signal synthesizer 24 is the output of the transmitter 2.
5, and is transmitted from the transmission unit 25.

The signal transmitted from the transmitter 25 is received by, for example, a receiving station installed on the ground or the like, and
It is processed by the signal processing device shown in the fourth embodiment.
Here, as shown in the figure, the transmission data br + 1, br + 0
Both correspond to transmission data 0, transmission data br + 1 = 0, br + 0 = 1 correspond to transmission data 1, and transmission data br + 1 = 1, br + 0 = The case of 0 corresponds to the transmission data 2, and the transmission data br + 1 =
The case of 1, br + 0 = 1 is made to correspond to the transmission data 3.

FIG. 14 is a diagram showing the result of decoding the received signal at the receiving station. As shown in the figure, the transmission data br +
When 1 and br + 0 are 0 (time period from t0 to t2 in FIG. 14),
The reception station does not receive the signals of the frequencies f1 and f2, and the decoding result of the received data is 0. Br + 1 of the transmitted data
When becomes 1, the signal of frequency f1 is received by the receiving station (time period from t3 to t5 in FIG. 14), and the receiving station decodes the received data as 1.

When br + 0 becomes 1 in the transmission data, the signal of frequency f2 is received by the receiving station (t in FIG. 14).
In the period from 6 to t8), the receiving station decodes the received data as 2. Further, when both br + 1 and br + 0 of the transmission data become 1, the receiving station receives the signal in which the frequency f1 and the frequency f2 are mixed (the period from t9 to t11 in FIG. 14), and the receiving station Decode the received data as 3.

As described above, in this embodiment, the signal having a plurality of frequency components is transmitted, and the receiving station decodes the signal depending on the presence or absence of the specific frequency component.
Even when receiving a weak electric wave buried in noise, such as ultra-long distance communication such as space communication, communication can be performed at a high communication speed without being affected by noise. In particular, by using the signal processing device shown in the first to fourth embodiments, it is possible to accurately process the reception wave in which two frequency signals are superposed as shown in FIG.

[0081]

As described above, the following effects can be obtained in the present invention. (1) Convert the received signal to an intermediate frequency and convert it to time-series digital data, correct the delay time of the received signal, and then correct the phase of the corrected time-series digital signal based on its center frequency. Since the Fourier transform is performed and the phase of the signal after the Fourier transform is corrected for each frequency component, the Doppler shift that occurs in high-speed mobile communication that changes the relative speed can be corrected digitally. Communication with a high-speed moving body whose speed changes can be stably performed.

Further, since digital signal processing is performed instead of analog signal processing, adjustment and the like such as analog signal processing are not required, and stable operation and reproducibility can be expected. In particular, by recording the AD-converted data on a digital tape recorder or the like, the signal processing can be repeated. Therefore, even if there is an uncertain factor in the speed information or the like, it can be repeated as many times as possible while changing the parameter, and highly accurate correction processing can be performed. (2) The error due to displacement from the center frequency is reduced by dividing the reception band into multiple parts, determining the center frequency that is the reference for correction for each band, and performing the correction process for each band. Therefore, it is possible to perform highly accurate correction. (3) The delay time correction unit is provided with M buffer memories, and time-series data is sequentially written into the M buffer memories to perform delay time correction, whereby the delay time correction unit is M times the sampling time interval. And a low-speed memory can be used as the buffer memory. Further, it becomes possible to use an n-bit memory as a buffer memory. (4) By providing multiple receiving stations and correcting the signals received by the multiple receiving stations and performing cross-correlation processing, changes in the relative position between multiple receiving stations due to crustal movements, etc. can be accurately performed. Can be measured. (5) An accumulation processing unit is provided, and the cross-correlation processing results of multiple times are
Random noise components included in the received signal can be removed by accumulating for each frequency component, and signals that are weak relative to the noise component can be restored in ultra long distance communication. You can (6) Transmit a signal with multiple frequency components from the transmission source,
By detecting the presence or absence of a specific frequency signal in the signal received by the receiving station or the level of the specific frequency signal and decoding the signal, it is possible to communicate with a high-speed flying object like space communication and super long distance communication. Thus, when receiving a weak radio wave, the signal can be restored at a high communication speed.

[Brief description of drawings]

FIG. 1 is a principle diagram of the present invention.

FIG. 2 is a diagram showing a signal transmitted and a signal received by a high-speed moving body.

FIG. 3 is a diagram illustrating a method of delay time correction in a delay time correction unit.

FIG. 4 is a diagram illustrating a relationship between a signal source, a receiving station, and a reference receiving station.

FIG. 5 is a diagram showing a relationship between a center frequency f ₀ and a local frequency f _L.

FIG. 6 is a diagram showing an overall configuration of a first exemplary embodiment of the present invention.

FIG. 7 is a diagram showing configurations of a correction value calculation unit and a numerical operation processor unit.

FIG. 8 is a diagram showing a second embodiment of the present invention.

FIG. 9 is a diagram showing a third embodiment of the present invention.

FIG. 10 is a time chart showing the operation of the third embodiment.

FIG. 11 is a diagram showing a fourth embodiment of the present invention.

FIG. 12 is a diagram showing a fifth embodiment of the present invention.

FIG. 13 is a time chart showing the operation of the fifth embodiment.

FIG. 14 is a diagram showing a decoding result of a received signal in the fifth embodiment.

[Explanation of symbols]

 1 High-speed moving object 2, 2'Reception station 3 Mixer 4, 4 'Band pass filter 5, 5' AD converter 6, 6 'Delay time correction unit 6a, 6a' Buffer memory 7, 7 'Time series data phase correction unit 8.8 'Fourier transform section 9, 9'Frequency data phase correction section 10, 10' Center frequency generation section 11, 11 'Correction value calculation section 12, 12' Numerical operation processor section 14 Clock division circuit 15 Demultiplexer 16 Correlation Processor 17 Accumulation processor 17d Buffer memory 20,21 Oscillator 22 Transmission data generator 23,23 'DF converter 24 Signal combiner 101,102 Processor 103,104 Signal processor 111 Counter 112 Shifter 121 Display / keyboard 122 LAN controller 123 Hard disk 124 Floppy disk, 125 EX T. Register 126 Instruction data memory 127 MPU

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Noboru Morita 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited (72) Inventor Toshiharu Kawanishi, 1015, Kamedotachu, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited ( 72) Inventor Tetsuro Nagashima 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited

Claims (10)

[Claims] 1. A digital signal, which receives a signal from the signal source and frequency-converts the received high-frequency signal into a low-frequency signal under an environment in which the relative positions of the signal source and the receiving device change with time. After correcting the delay time from the reference position to the reception position by correcting the signal extraction position from the obtained time-series digital signal, the phase of the corrected time-series digital signal is adjusted to its center frequency. The signal processing method is characterized by correcting the phase-corrected time-series digital signal by Fourier transform and correcting the Fourier-transformed signal for each frequency component to correct the Doppler shift of the received signal. . 2. A signal processing apparatus for receiving a signal from a signal source in an environment in which the relative positions of the signal source and the receiving station change with time, and correcting the received signal subjected to Doppler shift. , A frequency converter for converting the received high frequency signal into a low frequency signal, an AD converter for sampling the frequency converted signal and converting it into a time series digital signal, and correcting the data extraction position from the time series digital signal By doing so, a delay time correction unit that corrects the delay time from the reference position to the reception position, and a time-series data phase correction unit that corrects the phase of the delay-time corrected time-series digital signal with reference to the correction center frequency. A Fourier transform unit for Fourier transforming the phase-corrected signal, and a frequency data phase corrector unit for phase-correcting the Fourier-transformed signal for each frequency component A signal processing device comprising: 3. A delay time correction unit that writes time-series data in a buffer memory, corrects a read position of data written in the buffer memory according to a delay time from a reference position to a reception position, and corrects the delay time. By multiplying the time-series digital signal by the correction value and correcting the phase based on the center frequency, (a) time-series digital signal sampling time shift and (b) time-series digital signal start position Correct the phase of
(C) For each sampling data, the time-series data phase correction unit that corrects the displacement of the time-series digital signal from the correction value according to (a) above, and the phase is corrected by multiplying the Fourier-transformed signal by the correction value. Accordingly, the signal processing device according to claim 2, further comprising a frequency data phase correction unit that corrects the deviation of the sampling time for each frequency component after the Fourier transform. 4. The signal processing device according to claim 2, wherein the correction center frequency is set to a value that is an integer multiple of the sampling frequency with respect to the local frequency. 5. A signal processing device, characterized in that a reception band is divided, a correction center frequency is determined for each reception band, and the signal processing device according to claim 2, 3 or 4 is provided for each band. . 6. The delay time correction unit has M (M = 1, 2,
3, ...) buffer memories are provided, and time-series data is sequentially written into the M buffer memories to perform delay time correction, whereby the delay time correction unit is M times the sampling time interval. The signal processing device according to claim 3, 4, or 5. 7. The corrected center frequency is 1 / M ((M = 1,2,3,3 of the sampling frequency with respect to the local frequency.
.) Is set to a value that is an integral multiple of the distance apart from each other, the signal processing apparatus according to claim 3, 4, 5, or 6. 8. A signal processing apparatus according to claim 2, 3, 4, 5 or 6, wherein a plurality of receiving stations and delay correction of respective signals received by the plurality of receiving stations are performed, and each signal processing. A correlation processing unit that performs cross-correlation processing of the processing result of the device is provided, and the delay correction of the signals received by the plurality of receiving stations is performed based on the reference receiving station or the virtual receiving station, and the correlation processing unit is provided. 2. A signal processing device, characterized by performing cross-correlation processing of a signal for which delay correction has been performed in (1). 9. The signal processing apparatus according to claim 8, wherein an accumulation processing unit is provided, and a result of cross-correlation processing performed a plurality of times is accumulated for each frequency component. 10. A reception station for receiving a signal having a plurality of frequency components sent from a transmission source, wherein the reception station delay-corrects the received signal, and then the presence or absence of a specific frequency signal in the received signal. Or the signal processing apparatus according to claim 2, wherein the received signal is decoded by determining the level of the specific frequency signal.