JP2010286337A - Correlation processing device, correlation processing method, pulse compression device and target detector - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a pulse compression device which does not need rearrangement, even when multiple additions of result obtained by a correlation processing in a frequency domain is performed. <P>SOLUTION: A Fourier transformation section 51 calculates a received echo signal y[n] by discretization of a received echo signal y[t] and generates a received echo signal ym[k], in a frequency domain by the Fourier transformation by each section. A correlation processing section 52 performs pulse compression by a correlation processing of each section of a replica signal Re[k] stored in a memory 521 and outputs a pulse compressed signal Zm[k]. At this time, q replica signal Re[k] is obtained, for example, by subjecting the waveform of a transmitted pulse signal x[n] to time reversal, delaying the transmitted pulse signal x[n] by the pulse length and performing complex conjugate processing. Each of the calculated sections of the calculated pulse compressed signal Zm[k] is converted into a pulse compressed signal zm[n] in a time domain, and multiple addition is simply performed thereon along the time series. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a correlation processor that performs autocorrelation processing on a correlated signal with a replica signal based on the correlated signal, and pulse compression that uses the correlation processor to pulse-compress the received signal with the replica signal of the transmission signal. The present invention relates to a device and a target detection device.

  Conventionally, various target detection apparatuses that transmit a pulse signal to a detection target area and detect a target from a reflected signal have been put into practical use. Among such target detection apparatuses, as shown in Patent Document 1, there is an apparatus that improves the distance resolution by performing pulse compression processing on a received echo signal using an FM chirp transmission pulse signal. One such pulse compression process is a matched filter executed in the frequency domain. This matched filter sets a filter coefficient in the frequency domain based on a replica signal composed of a complex conjugate signal of a transmission pulse signal, and multiplies the received echo signal converted to the frequency domain by the filter coefficient to perform pulse compression processing. Execute. That is, pulse compression processing is executed by performing correlation processing between the replica signal and the received echo signal in the frequency domain.

  Here, when performing the target detection described above, a reception period having a length corresponding to the detection distance is provided, and pulse compression processing is performed on the entire received echo signal within the reception period. Depending on the length, if the pulse compression process is performed with the entire reception period as one section, there may be a problem such as an increase in calculation load. For this reason, the reception period is divided into a plurality of partial sections, pulse compression processing is performed for each partial section, and the pulse compression results obtained for each partial section are overlapped to perform pulse compression for the entire reception period. A method has been devised.

Japanese Patent Publication No. 6-90277

  However, the conventional pulse compression processing by overlap addition has the following problems.

FIG. 5 shows a transmission pulse signal x [n] included in a received echo signal when performing pulse compression processing in the conventional frequency domain, a time-inverted signal x [−n] of the transmission pulse signal, and a complex conjugate signal of the transmission pulse signal. It is a figure which shows the relationship in the time domain of replica signal x ^* [n] which consists of, and pulse compression signal z [t].

  Further, FIG. 6 shows a state on the time axis of the pulse compression signal zm [n] of each partial section when the conventional pulse compression processing by overlap addition is performed, and rearrangement processing for each pulse compression signal generated at this time. It is a figure for demonstrating.

First, the pulse compression processing in the frequency domain includes X [k] that is the frequency domain signal of the transmission pulse signal x [n] included in the received echo signal and the transmission pulse signal X [k] in this frequency domain that is a replica signal. ] Is a correlation process with a replica signal X ^* [k] which is a complex conjugate signal. Accordingly, in the frequency domain, the pulse compression signal Z [k] is given by the following equation.

Z [k] = X [k] · X ^* [k] − (1)
N means the number of DFT data points.

Here, if the transmission signal x [n] in the time domain of the transmission pulse signal X [k] is a waveform that rises at a timing of “0” on the time axis as shown in FIG. The signal on the time axis of the replica signal X ^* [k] is a signal x obtained by taking the time conjugate of the transmission signal x [n] in the time domain to x [−n] and taking the complex conjugate of the x [−n]. ^* [-N].

  However, when time reversal occurs, no waveform appears in the negative region on the time axis as shown by the broken line in FIG. 5B, and in reality, as shown by the solid line in FIG. 5B. As a result, a waveform having the last timing of the DFT period as the falling timing appears.

For this reason, the replica signal x ^* [− n] on the time axis appears as a waveform whose falling timing is the last timing of the DFT period, as shown in FIG.

  That is, when the pulse compression processing by the discretized signal in the frequency domain given by the above equation (1) is represented by the processing on the time axis, z [n] is the discretized pulse on the time axis. As a compressed signal, the following equation is obtained.

  By performing such a convolution operation, the pulse compression signal z [t] in the time domain is the end point of the waveform N before the distance 0 ends at the last timing N-1 of the DFT period, as shown in FIG. The waveform is shifted to a partial waveform starting from the timing of distance 0 and a partial waveform starting from the last timing N-1 of the DFT period. Furthermore, the context of each partial waveform in the time domain is reversed.

  Therefore, the reception echo signal y [n] having the same waveform component as the transmission pulse signal x [n] is divided in the time domain for each partial reception echo signal ym [n] (m is a positive integer). When the pulse compression processing is performed, each pulse compression signal zm [n] corresponding to each reception echo signal ym [n] has its partial relationship in the time domain reversed as shown in FIG. And it becomes a separated waveform. For example, the pulse compression signal z1 [n] is assumed to be a partial pulse compression signal z11 [n], z12 [n], z13 [n] in time series. The partial pulse compression signal z11 [n] corresponds to a period from when the replica signal Re [n] starts to overlap in the period of the partial reception echo signal y1 [n] until it completely overlaps. The partial pulse compression signal z12 [n] corresponds to a period in which the entire replica signal Re [n] completely overlaps the reception echo signal y1 [n] period. The partial pulse compression signal z13 [n] corresponds to a period from when the replica signal Re [n] starts to deviate from the period of the received echo signal y1 [n] until it completely deviates.

  In this case, the pulse compression signals z12 [n] and z13 [n] are continuously output in the time domain, and thereafter, the pulse compression signals z11 [n] are output separated in time.

  Therefore, when performing the above-described overlap addition processing, as shown in FIG. 6, partial pulse compression signals zm1 [n], zm2 [n], and zm3 [n] of each pulse compression signal zm [n] It must be sorted into the normal context in the area. For example, in the case of the pulse compression signal z1 [n], the pulse compression signal z11 [n] on the rear side in the time axis is continuously placed before (the past side) the pulse compression signals z12 [n] and z13 [n]. It must be replaced so as to obtain a typical waveform. For this reason, the calculation load for the overlap addition process increases.

  In view of such problems, an object of the present invention is to provide a correlation processor that does not need to be rearranged even when the results of correlation processing in the frequency domain are overlap-added, and pulse compression using the correlation processor. It is in realizing an apparatus and a target detection apparatus.

  The present invention relates to a correlation processor that correlates a pulsed correlated signal with a replica signal. The correlation processor includes a conversion unit, a correlation processing unit, and an inverse conversion unit. The conversion unit converts the correlated signal from the time domain to the frequency domain. The correlation processing unit inverts the pulse-like signal that is the source of the correlated signal in the time domain, delays it by the pulse length, and converts the same waveform as the complex conjugate processed waveform into the frequency domain Correlation processing is performed by multiplying the frequency domain correlated processing signal by the replica signal. The inverse transform unit inversely transforms the signal after the correlation processing from the frequency domain to the time domain.

  By setting the replica signal in this way, the pulse compression waveform is delayed by the pulse length, and the signal after the pulse compression is not a waveform like the conventional technique shown in FIG. It is output in a continuous waveform along the time axis as shown in FIG.

  In addition, the conversion unit of the correlation processor of the present invention divides the correlated signal into a plurality of continuous sections in the time domain and performs conversion to the frequency domain. The correlation processing unit performs correlation processing for each correlated process signal divided into sections. The inverse transform unit performs inverse transform on each of the individually correlated signals. Further, the correlation processor includes an overlapping addition processing unit that adds the signals subjected to the inverse transformation so as to be continuous in the time domain.

  In this configuration, since the signal after correlation processing in each section has a continuous waveform along the time axis, when performing overlapping addition in which these are added in the time domain, the signals after correlation processing divided into sections are displayed. There is no need to rearrange them along the time axis.

  In the present invention, the pulse-like correlated signal is a signal having a symmetrical waveform in the time domain. In the correlation processor, the replica signal is complex-conjugated to the pulse-like signal that is the source of the correlated signal. It is obtained by processing.

  In this configuration, when the pulse-like correlated signal is a signal having a symmetrical waveform in the time domain, the waveform does not change even if time reversal is performed, so that it is not necessary to perform time reversal processing. Thereby, since there is no need for delay processing, a replica signal can be generated only by complex conjugate processing.

Also, the conversion unit of the correlation processor of the present invention discretizes the correlated signal to convert data having a pulse length or longer in ²ⁿ units.

In this configuration, FFT (Fast Fourier Transform) processing can be performed by performing frequency conversion on data longer than the pulse length in units of ²ⁿ . As a result, faster conversion processing is possible.

  The present invention also relates to a pulse compression device, and the pulse compression device includes the correlation processor described above. The correlation processing unit of this correlation processor uses an FM chirp transmission pulse signal as a base signal, and uses a signal obtained by performing time inversion processing, delay processing, and complex conjugate processing on the transmission pulse signal as a replica signal. Perform correlation processing.

  In this configuration, a pulse compression device using an FM chirp transmission pulse signal is shown as a specific device using the above-described correlation processor. Then, by performing the pulse compression process using the correlation process described above, it is possible to reduce the calculation load during the pulse compression process without performing the conventional rearrangement process.

  The present invention also relates to a target detection device, which includes the above-described pulse compression device and displays the signal on a display screen using a signal after pulse compression obtained by correlation processing. An image generation unit for generating an image is provided.

  In this configuration, as a specific device using the above-described pulse compression device, a target detection device that displays a target echo image from a reception echo signal obtained by transmitting the transmission pulse signal in a detection region is shown. Yes. And the calculation load can be reduced also as a target detection apparatus by using the above-mentioned pulse compression process.

  According to the present invention, even if the result of correlation processing of a correlated signal such as a received signal in the frequency domain is subjected to overlap addition processing, the data rearrangement processing within each correlation processing result need not be performed. The calculation load can be reduced. Thereby, the calculation load of the overlap addition type pulse compression processing can also be reduced.

2 is a block diagram showing a main circuit configuration of a radar apparatus 11 and a block diagram showing a main circuit configuration of a pulse compression unit 15. FIG. The concept of generating the signal x ^* [Np-1-n] in the discrete time domain that becomes the replica signal Re [k] in the frequency domain of the present application and the correlation with the signal x ^* [Np-1-n] It is a figure which shows the waveform of the pulse compression signal in the time domain at the time of processing. It is a figure which shows the state on the time-axis of the pulse compression signal of each partial area at the time of performing the correlation process (pulse compression) and overlap addition of this embodiment. It is the figure which showed the relationship between the waveform of the transmission pulse signal at the time of using the transmission pulse signal x [n] of a symmetrical waveform in a time domain, and the complex conjugate signal x ^* [n] which is a replica signal. A replica signal composed of a transmission pulse signal x [n] included in a received echo signal when performing pulse compression processing in the conventional frequency domain, a time-inverted signal x [−n] of the transmission pulse signal, and a complex conjugate signal of the transmission pulse signal It is a figure which shows the relationship in the time domain of x ^* [n] and the pulse compression signal z [t]. It is a figure for demonstrating the state on the time-axis of the pulse compression signal of each partial area at the time of performing the pulse compression process by the conventional overlap addition, and the rearrangement process for every pulse compression signal produced at this time.

  A correlation processor, a pulse compression device, and a target detection device according to an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, the radar apparatus 11 using a transmission pulse signal having a constant amplitude that is an FM chirp method will be described as an example of a target detection apparatus.

  FIG. 1A is a block diagram showing a main circuit configuration of the radar apparatus 11, and FIG. 1B is a block diagram showing a main circuit configuration of the pulse compression unit 15.

  The radar apparatus 11 includes a transmission unit 12, a DPX (duplexer) 13, an antenna 14, a pulse compression unit 15, and an image generation unit 16. The pulse compression unit 15 corresponds to a “pulse compression device” of the present invention and also corresponds to a “correlation processor”.

  The transmission unit 12 includes an oscillating element such as a semiconductor or a magnetron, and transmits a transmission pulse signal x [t] having a pulsed waveform of the FM chirp method at a predetermined transmission timing corresponding to the transmission period. The transmission pulse signal x [t] output from the transmission unit 12 is output to the DPX 13.

  The DPX 13 transmits the transmission pulse signal x [t] from the transmission unit 12 to the antenna 14. The antenna 14 radiates the supplied transmission pulse signal x [t] to the outside while continuing to rotate at a constant speed at a predetermined rotation speed, and a period in which the transmission pulse signal x [t] is not supplied (reception period) Receives a reflection signal from an external target and outputs a reception echo signal y [t] to the DPX 13. The DPX 13 outputs the received echo signal y [t] from the antenna 14 to the pulse compression unit 15. This received echo signal y [t] corresponds to the “correlated signal” of the present invention.

  The pulse compression unit 15 is a so-called matched filter, and will be described in detail later. As will be described in detail later, the received echo signal y [n] that has been discretized during the reception period is divided into a plurality of sections and subjected to pulse compression in the frequency domain. After the processing is performed, the pulse compression signal z [n] is calculated by performing overlap addition.

  The image generation unit 16 generates an image to be displayed on the display screen using the pulse compression signal z [n] from the pulse compression unit 15.

  Next, the pulse compression unit 15 will be described in more detail. As shown in FIG. 1B, the pulse compression unit 15 includes a Fourier transform unit 51 corresponding to the “transformer” of the present invention, a correlation processing unit 52, and an inverse Fourier transform corresponding to the “inverse transform unit” of the present invention. A unit 53 and an overlap addition unit 54 are provided.

  The Fourier transform unit 51 acquires and stores the received echo signal y [n] by discretizing the received echo signal y [t] on the time axis input from the antenna 14 via the DPX 13 at a predetermined sampling timing. To do. Here, n is a sampling number. The Fourier transform unit 51 divides the received echo signal y [n] into sections for each predetermined number of sampling data.

  The Fourier transform unit 51 sequentially performs a Fourier transform on the reception echo signal ym [n] in each partial section to generate a reception echo signal Ym [k] in the frequency domain. Here, m is a partial section number, and k is a discrete frequency. The Fourier transform unit 51 provides the reception echo signal Ym [k] in the frequency domain to the correlation processing unit 52 for each partial section.

At this time, the Fourier transform unit 51 sets, on the delay side of the received echo signal ym [n] in each partial section, a dummy sampling data group consisting of “0” necessary for correlation processing described later as the pulse length of the transmission pulse signal. After adding the corresponding number, section division is set so that the number of sampling data in each partial section is ²ⁿ , and Fourier transform is executed. Thereby, the Fourier transform unit 51 can perform a fast Fourier transform (FFT) process on the received echo signal ym [n] of each partial section. As a result, the received echo signal ym [n] in each partial section can be converted to the received echo signal Ym [k] in the frequency domain at high speed.

The correlation processing unit 52 includes a memory 521 and a multiplier 522.
The correlation processing unit 52 multiplies the received echo signal Ym [k] by the multiplier 522 by the replica signal Re [k] for each discrete frequency stored in the memory 521, and performs correlation processing. Zm [k] = N · Re [k] · Ym [k] is output. N means the number of DFT data points.

  The signal Zm [k] after this correlation processing becomes a pulse compression signal in the frequency domain. In the following description, it will be referred to as a “pulse compression signal”. In addition, here, correlation processing is functionally shown, but these processing stores each discrete frequency component of the replica signal Re [k] as a filter coefficient in the memory 521 that is a filter bank, and the filter coefficient The multiplier 522 can be realized as a matched filter that filters the received echo signal Ym [k].

The replica signal Re [k] stored in the memory 521 is specifically set by the following method. FIG. 2 shows the concept of generating the signal x ^* [Np-1-n] in the discrete time domain that becomes the replica signal Re [k] in the frequency domain of the present application, and the signal x ^* [Np-1-n ] Is a figure which shows the waveform of the pulse compression signal in the time domain at the time of performing a correlation process.

(1) A signal x [−n] indicated by a broken line in FIG. 2B obtained by time-reversing the waveform of the discretized transmission pulse signal x [n] shown in FIG. Here, the signal x [−n] is actually a waveform having the last timing of the DFT period as the falling timing (the solid line waveform in FIG. 2B) as described above. Can be regarded as a dashed waveform.
(2) A signal x [Np-1-n] shown in FIG. 2C is calculated by delaying the time-inverted signal x [-n] by the pulse length Np of the transmission pulse signal x [n]. .
(3) Replicate the complex conjugate signal x ^* [Np-1-n] shown in FIG. 2D, which is obtained by performing complex conjugate processing on the signal x [Np-1-n] subjected to time inversion and delay processing. Calculated as signal Re [n].
(4) The complex conjugate signal x ^* [Np−1−n], which is the replica signal Re [n], is converted into a frequency domain function Re [k] = FFT [x ^* [Np−1−n]] and discrete Set to replica signal at frequency.

  When correlation processing is performed using such a replica signal Re [k], the pulse compression signal z [t] in the time domain is delayed by the pulse length Np of the transmission pulse signal x [n]. It is not a discontinuous waveform as shown in FIG. 5 of the above-mentioned prior art but a continuous waveform as shown in FIG.

  Accordingly, the received echo signal ym [n] is sequentially correlated in the frequency domain, and inverse Fourier transformed to the time domain to obtain the time domain pulse compressed signal zm [n], thereby obtaining the pulse compressed signal zm. [N] has a continuous waveform in the time domain, as shown in FIG.

  FIG. 3 is a diagram illustrating a state on the time axis of the pulse compression signal of each partial section when the correlation processing (pulse compression) and overlap addition of the present embodiment are performed.

  For example, when the pulse compression signal z1 [n] for the reception echo signal y1 [n] in the partial section is the partial pulse compression signal z11 [n], z12 [n], z13 [n] as described above. In the time domain, it is possible to obtain a pulse compression signal z1 [n] having a waveform in which these partial pulse compression signals z11 [n], z12 [n], and z13 [n] are continuously output in time series. it can.

  The inverse Fourier transform unit 53 obtains a pulse compression signal zm [n] in the time domain by performing inverse Fourier transform on the pulse compression signal Zm [k] of each partial section output from the correlation processing unit 52.

  The overlap addition unit 54 acquires the pulse compression signal zm [n] of each partial section, and performs overlap addition processing so that each pulse compression signal zm [n] continues in time series in the time domain. More specifically, as shown in FIG. 3B, the overlap adder 54 includes a partial pulse compression signal z13 [n] on the rear end side of the pulse compression signal z1 [n] and the pulse compression signal z1. The partial pulse compression signal z21 [n] on the front end side of the pulse compression signal z2 [n] continuous to [n] is added so as to overlap in the time domain. Further, the overlap addition unit 54 includes a partial pulse compression signal z23 [n] on the rear end side of the pulse compression signal z2 [n] and a pulse compression signal z3 [n] continuous to the pulse compression signal z2 [n]. The partial pulse compression signal z31 [n] on the front end side is added so as to overlap in the time domain. And the duplication addition part 54 performs such duplication addition processing over a reception period.

  By performing such processing, it is not necessary to rearrange the data within the pulse compression signal of each partial section as shown in the prior art. As a result, the pulse compression processing using the correlation processing in the frequency domain and the overlap addition method can be executed at high speed with a lighter calculation load than in the past. By reducing the calculation load on the pulse compression processing and increasing the speed in this way, it is possible to reduce the calculation load and increase the speed of the entire target detection process.

  In the above description, the transmission pulse signal of the FM chirp method is simply described. However, when a transmission pulse signal having a symmetrical waveform in the time domain is used, a replica signal that is simply a complex conjugate of the transmission pulse signal is used. Also good.

FIG. 4 is a diagram showing a relationship between waveforms of a transmission pulse signal and a complex conjugate signal x ^* [n] that is a replica signal when a transmission pulse signal x [n] having a symmetrical waveform in the time domain is used. .

  When the waveform of the transmission pulse signal x [n] is symmetric in the time domain, the time-reversed waveform is naturally the same. Therefore, the time-reversed complex conjugate signal used as the replica signal is simply the same as the complex conjugate signal. For this reason, no time reversal processing is required, and no delay processing is required as long as time reversal processing is not required.

  Thus, when the waveform of the transmission pulse signal x [n] is symmetric in the time domain, the replica signal calculated only from the complex conjugate process is the same as the replica signal that has been subjected to the time inversion process, the delay process, and the complex conjugate process. It becomes the same signal. As a result, a replica signal for making the pulse compression signal continuous can be set by simple processing.

  In the above description, the example in which the replica signal is set by performing time inversion processing, delay processing, and complex conjugate processing on the transmission pulse signal has been shown, but as a result, the transmission pulse signal is subjected to time inversion processing, delay processing, As long as a waveform equivalent to the complex conjugate processed waveform can be obtained, the replica signal may be set by another method.

  Moreover, although the case where the number of correlation processing units 52 is one is shown in the above description, a plurality of correlation processing units 52 are provided, and for each correlation processing unit 52, the received echo signal Ym [ k] correlation processing may be performed.

  Further, in the above description, an example in which the overlap addition process is performed is shown. However, the effect of not having to rearrange the pulse compression data for each partial section is applicable even when the overlap addition process is not performed. Can do. Therefore, the above-described effect can be obtained even with a pulse compression device that does not perform overlap addition processing.

  In the above description, the pulse compression device that performs pulse compression of the reception signal with the replica signal based on the transmission pulse signal has been described as an example. However, if the device performs autocorrelation processing on the correlated signal in the frequency domain, Configuration can be applied.

  In the above description, the radar device is shown as an example of the target detection device. However, the above configuration is applied to any device such as a sonar device that uses the result of autocorrelation processing of the correlated signal in the frequency domain. be able to.

  In the above description, the pulse compression unit 15 has been described as individual functional units such as the Fourier transform unit 51, the correlation processing unit 52, the inverse Fourier transform unit 53, and the overlap addition unit 54. Need not be formed individually. That is, the pulse compression unit 15 may be configured by one IC, and the pulse compression process using the correlation processing method as described above may be stored as a program and executed by the IC.

11-radar apparatus, 12-transmission unit, 13-DPX, 14-antenna, 15-pulse compression unit, 16-target detection unit, 51-Fourier transform unit, 52-correlation processor, 53-inverse Fourier transform unit, 54-overlapping adder, 521-memory, 522-multiplier

Claims (9)

A correlation processor for correlating a pulsed correlated signal with a replica signal,
A conversion unit for converting the correlated signal from the time domain to the frequency domain;
A replica signal obtained by inverting the pulse-shaped signal that is the source of the correlated signal in the time domain, delaying by the pulse length, and converting the same waveform as the complex conjugate processed waveform into the frequency domain, A correlation processing unit that performs correlation processing by multiplying the correlated signal in the frequency domain;
An inverse transform unit that inversely transforms the signal after the correlation processing from the frequency domain to the time domain;
A correlation processor. The transform unit divides the correlated signal into a plurality of continuous sections in the time domain, performs transform to the frequency domain,
The correlation processing unit performs the correlation processing for each correlated processing signal divided into sections,
The inverse transform unit performs the inverse transform on each of the individually correlated signals,
The correlation processor according to claim 1, further comprising an overlapping addition processing unit that adds the signals subjected to the inverse transformation so as to be continuous in the time domain. The pulsed correlated signal is a signal having a symmetrical waveform in the time domain,
The correlation processor according to claim 1, wherein the replica signal is obtained by performing complex conjugate processing on a pulsed signal that is a source of the correlated signal. The correlation processor according to any one of claims 1 to 3, wherein the conversion unit discretizes the correlated signal to convert data having a pulse length or longer in ²ⁿ units. While comprising the correlation processor according to any one of claims 1 to 4,
The correlation processing unit uses an FM chirp transmission pulse signal as the original signal, and a signal obtained by performing time inversion processing, delay processing, and complex conjugate processing on the transmission pulse signal as the replica signal, A pulse compression device that performs correlation processing. While comprising the pulse compression device according to claim 5,
A target detection apparatus comprising an image generation unit that generates an image to be displayed on a display screen using a signal after pulse compression obtained by the correlation processing. A correlation processing method for correlating a pulsed correlated signal with a replica signal,
Transforming the correlated signal from the time domain to the frequency domain;
A replica signal obtained by inverting the pulse-shaped signal that is the source of the correlated signal in the time domain, delaying by the pulse length, and converting the same waveform as the complex conjugate processed waveform into the frequency domain, Performing correlation processing by multiplying the correlated signal in the frequency domain;
A step of inversely transforming the signal after the correlation processing from the frequency domain to the time domain;
A correlation processing method. The step of converting to the frequency domain divides the correlated signal into a plurality of continuous sections in the time domain, and performs conversion to the frequency domain.
The step of performing the correlation processing performs the correlation processing for each correlated processing signal divided into sections,
The inverse transforming step performs the inverse transform on each individually correlated signal,
Furthermore, the correlation processing method of Claim 7 which added the overlap addition process step which adds the signal by which the said inverse transformation was performed so that it may continue in a time domain. The pulsed correlated signal is a signal having a symmetrical waveform in the time domain,
The correlation processing method according to claim 7, wherein the step of performing the correlation processing obtains the replica signal by performing complex conjugate processing on a pulse-shaped signal that is a source of the correlated processing signal.

Images