JP2008170221A - Pulse compression radar device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform pulse compression capable of flexibly handling the case even when using a plurality of repetition frequencies, without having to increase the operation scale, even with high compression and a large number of range bins. <P>SOLUTION: This device divides the receiving data for each sweep so as to have a prescribed processing range and for the processing range to have an overlapping region, respectively for each processing unit. A correlation processing is performed in the frequency range for each processing unit, and after the correlation processing, only the part where the compression pulse should be included is cut and connected. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a pulse compression radar apparatus that requires a high compression ratio and detects a long distance.

  In order to detect a long-distance target with a high range resolution by the radar apparatus, a transmission pulse having a large peak transmission power and a narrow width must be used as a transmission signal. However, if such a transmission signal cannot be obtained due to hardware limitations, a transmission pulse with a long width modulated with low peak transmission power is transmitted, and pulse compression is applied to the reflected signal from the target. By performing the processing, it is possible to obtain the same effect as when a transmission pulse having a large peak transmission power and a narrow width is used as a transmission signal. The modulation applied to the transmission signal to perform this pulse compression processing includes a linear frequency modulation method (linear FM method), a non-linear frequency modulation method (nonlinear FM method), and a code modulation method (phase modulation) using a pseudo-random code. Method).

  The pulse compression processing is generally performed in the time domain using a transversal filter configuration regardless of the transmission signal modulation method (Non-Patent Document 1).

  As shown in FIG. 4, the transversal filter includes a shift register 1 in which a plurality of delay devices 1-1 to 1-n are provided in series, a signal at each stage of the shift register 1, and a pulse compression coefficient a0. Are composed of a plurality of multipliers 2-0 to 2-n for multiplying ~ an and an adder 3 for adding the multiplication results from the multipliers 2-0 to 2-n.

  In this transversal filter, sampling data (received data) obtained by frequency-converting and A / D-converting a radar reception signal is input from the input end of the shift register 1. The signals at each stage of the shift register 1 are multiplied by the respective coefficient values a0 to an in the multipliers 2-0 to 2-n, and the multiplication results are added by the adder 3 and output. When the correlation between the sampling data to the shift register 1 and the coefficient values a0 to an is obtained, the output of the adder 3 is obtained as a pulse signal having a large amplitude and a narrow width. In FIG. 4, the signal processing is shown as one system for convenience, but since sampling data is actually supplied as complex data, there are two shift registers, and the multiplier and adder are complex multipliers, respectively. , A complex adder.

  In this transversal filter system, when a high compression ratio is obtained using a wide pulse, the number of shift register stages (ie, the number of filter taps) increases, and the operation scale of the multiplier and the adder increases.

  In addition, when the pulse compression ratio is large, correlation processing using a product in the frequency domain is generally performed (Patent Document 1).

This integration method in the frequency domain includes a Fourier transformer, a complex multiplier, a coefficient table, and an inverse Fourier transformer. Sampling data is input to a Fourier transformer, converted into frequency domain data, the frequency domain data and the coefficient value prepared in advance in the coefficient table are multiplied by a complex multiplier, and the multiplication value is inverse Fourier transformed. The data is again converted into data in the time domain by the device, and the convolution operation is executed. By this integration method in the frequency domain, a pulse signal having a large amplitude and a narrow width is obtained as an output of the inverse Fourier transformer.
Supervised by Takashi Yoshida, “Revised Radar Technology”, The Institute of Electronics, Information and Communication Engineers, March 3, 1997, first edition, second edition, p. 276-278 Japanese Patent Laying-Open No. 2005-85167

  When the conventional transversal filter method requires a very high compression ratio and wide band, the calculation scale becomes enormous and difficult to implement.

  On the other hand, in the conventional integration method in the frequency domain, in the long-distance search radar, the number of range bins for one sweep increases as the search distance increases, that is, the number of samplings increases. This increases the number of points of Fourier transform when converting to the frequency domain, increases the complexity of the sequence accompanying the increase in the operation scale of Fourier transform, memory for temporary storage of intermediate data, sine wave table, etc. There were problems such as an increase in memory capacity. In many cases, the radar apparatus has a plurality of pulse repetition frequencies PRI. However, in this case, there are problems in mounting such that a plurality of Fourier transformers having different numbers of points are required.

  Therefore, the present invention provides a pulse compression radar apparatus that performs pulse compression, which can flexibly cope with high compression and a large number of range bins without increasing the operation scale and using a plurality of repetition frequencies. The purpose is to provide.

The pulse compression radar apparatus according to claim 1 transmits a modulated pulsed transmission signal to the outside, receives a reflection signal reflected outside as a reception signal, and transmits the reception signal and the transmission by the pulse compression apparatus. In a pulse compression radar device that performs pulse compression by performing correlation processing based on a signal,
The pulse compression device comprises:
Overlap processing unit data creating means for dividing the received data based on the received signal into processing units having a predetermined processing range and the processing ranges having mutually overlapping regions, and sequentially outputting as processing unit received data;
Fourier transform means for performing Fourier transform for each processing unit reception data sequentially output from the overlap processing unit data creating means, multiplication means for multiplying the output of the Fourier transform means and a pulse compression coefficient based on the transmission signal, An inverse Fourier transform means for performing an inverse Fourier transform on the output of the multiplication means, and a correlation means for performing pulse compression processing for each processing unit received data;
From the pulse compression processing output for each processing unit received data of the correlating means, the part that should contain the compressed pulse is deleted and other unnecessary parts are deleted, and the parts that should contain the sequentially obtained compressed pulse are connected, and the radar And a compressed data cut-out / connection means for generating data for one sweep.

  The pulse compression radar apparatus according to claim 2 is the pulse compression radar apparatus according to claim 1, wherein the processing range is at least twice the width of the transmission pulse, and the overlap region is a width of the transmission pulse. It is the above.

  According to the pulse compression radar apparatus of the present invention, the received data for each sweep is divided into processing units so that the processing ranges have a predetermined processing range and the processing ranges have an overlap region with each other, and a frequency is set for each processing unit. By performing correlation processing in a region and cutting out and connecting only portions that should contain compression pulses after the correlation processing, it is possible to seamlessly compress all data for one sweep without increasing the operation scale . In addition, since the processing unit of the correlation processing is reduced, the same compression processing is repeated without changing the number of points of the Fourier transform for a plurality of pulse repetition periods PRI as well as the reduction of the operation scale such as Fourier transform. Can adapt.

  Hereinafter, embodiments for implementing the pulse compression radar apparatus of the present invention will be described. In the present invention, in the correlation processing using the product of the Fourier transform output of the received data in the frequency domain and the pulse compression coefficient based on the transmission signal, the received data for one sweep is reduced in order to reduce the number of correlation data performed at a time. Divide each processing unit for processing.

  In the division processing, it is assumed that the reflected pulse portion in the received data divided for each processing unit straddles the boundary of the processing unit. In this case, the correlation processing is performed using data including the missing reflected pulse portion. Therefore, it is necessary to exclude the result of the correlation processing of that part. In order to make up for the lack of correlation processing results due to this exclusion, data of processing units in which received data overlaps is generated, thereby eliminating unnecessary data areas after correlation processing and seamlessly connecting correlation processing data.

  FIG. 1 is a block diagram showing a configuration of a pulse compression device in a pulse compression radar device of the present invention.

  A pulse-shaped transmission signal (transmission pulse) subjected to modulation is transmitted from a transmission device of a pulse compression radar device (not shown) toward an external target (target) every predetermined pulse repetition period PRI. The transmission pulse has a pulse width: τ, and linear frequency modulation is applied from the lower limit frequency f1 (= fc−Δf / 2) to the upper limit frequency f2 (= fc + Δf / 2) with respect to the center frequency fc. It has been. That is, the transmission pulse is an up-chirp signal having a pulse width: τ, a center frequency: fc, and a frequency modulation width: Δf.

  The transmission signal may be a down chirp signal whose frequency decreases linearly from the upper limit frequency f2 to the lower limit frequency f1, or may be a non-linear frequency modulation signal whose frequency changes nonlinearly. The code modulation signal may be a code modulation signal that is discretely code-modulated by a code sequence that takes a discrete value as modulation within the transmission pulse.

  The pulse compression radar apparatus receives a reception signal including a reflection pulse (echo pulse) in which the transmission signal is reflected by an external target. The received signal is frequency-converted by a frequency converting means, phase-detected by a detecting means, and converted into a digital signal by an analog-digital converting means to obtain received data rx (t). The received data includes a real part (I signal) and an imaginary part (Q signal) by phase detection by the detection means. However, in order to simplify the description, the real part (I signal) is simply used as received data. explain.

  This received data rx (t) is input to the pulse compression device of FIG. The overlap processing unit creation means 11 receives the received data for one sweep (radar pulse repetition period PRI), and the received data rx (t) has a predetermined processing range, and the processing ranges are overlapped with each other. Each processing unit B is divided and output as processing unit reception data to the FFT unit 12 which is a Fourier transform unit. The processing range of the processing unit B may be, for example, twice or more the transmission pulse width τ, and the upper limit may be arbitrary, but may be, for example, the pulse repetition period PRI or less. Further, it is preferable that the overlap region is not less than 1 times the width τ of the transmission pulse. Note that the upper limit of the overlap area is within the length of the processing range.

  The FFT means 12 performs Fourier transform on each of the processing unit received data that is sequentially input to convert it into frequency domain received data.

  In the coefficient table 13, pulse compression coefficients obtained as a result of Fourier transform of the complex conjugate of the transmission data tx (t) based on the transmission signal are calculated and stored in advance.

  The multiplication unit 14 multiplies the reception data converted into the frequency domain of the processing unit reception data from the FFT unit 12 and the pulse compression coefficient from the coefficient table 13 to obtain a multiplication result (product value).

  The IFFT means 15 which is an inverse Fourier transform means performs inverse Fourier transform on the multiplication result from the multiplication means 14 and returns it to the time domain data again, and obtains a pulse compression processing output subjected to pulse compression processing for each processing unit received data.

  The FFT means 12, the coefficient table 13, the multiplication means 14, and the IFFT means 15 constitute a correlation means 20 that performs pulse compression processing for each processing unit received data. The pulse compression processing output from the correlation means 20 includes a portion that should contain a compressed pulse and other unnecessary portions.

  The compressed data cut-out / concatenating means 16 includes compressed pulses that are sequentially obtained by leaving the part where the compressed pulse should be included from the pulse compression processing output for each processing unit received data of the correlating means and deleting other unnecessary parts. The power parts are concatenated to generate concatenated data ry (t) for one sweep of the radar.

  The processing unit B and its overlap region will be described with reference to FIG. In the example of FIG. 2, the processing unit B is set to a little over twice the transmission pulse width τ, and the overlap region is set to a little over one time the transmission pulse width τ.

  When the reflected pulse (echo pulse; its width is the same as the transmission pulse width, “τ”) falls within the range of the processing unit B as shown in FIG. 2A, the reflected pulse is correctly compressed. Pc appears at the head position of the reflected pulse as shown in FIG. The position of the compression pulse Pc is in the first half of the processing unit B. Note that delays due to the processing time of the correlator 20 and the like naturally occur, but here the delays due to the processing time and the like are ignored.

  Instead of appearing at the head position of the reflected pulse, the compressed pulse Pc is changed to any position within the reflected pulse such as the intermediate position or the end position of the reflected pulse by changing the phase data of the pulse compression coefficient. It can also be located. However, in the radar, it is preferable that the compression pulse Pc appears at the head position of the reflected pulse when ranging the target. Therefore, the position example shown in FIG. 2 is recommended.

  As shown in FIG. 2B, when the reflected pulse is almost in the middle of the processing unit B, the compression pulse Pc appears at the head position of the reflected pulse as shown in FIG. In the part.

  As shown in FIG. 2C, when the reflected pulse enters all the latter half of the processing unit B, the compressed pulse Pc appears at the head position of the reflected pulse as shown in FIG. It becomes the second half part. The (c) compression pulse Pc is included in the latter half of the processing unit B, that is, other unnecessary portions, and will be removed later.

  When the reflected pulse is partially applied to the first half of the processing unit B as shown in FIG. 2D, an abnormal compression pulse Pc ′ is generated in the processing unit B where no reflected pulse exists as shown in FIG. Located in the second half. This is because in the case of correlation using Fourier transform, a circular correlation is performed in which cyclic convolution processing is performed. Further, the compressed pulse Pc ′ has a smaller amplitude and a wider width than the one shown in FIG. 5A because only a part of the reflected pulse is used for the correlation processing.

  Further, when the reflected pulse is partially applied to the latter half of the processing unit B as shown in FIG. 2E, the abnormal compression pulse Pc ′ is positioned at the head of the reflected pulse as shown in FIG. However, since only a part of the reflected pulse is used for the correlation processing, the compression pulse Pc ′ has a smaller amplitude and the same as that of FIG. The width becomes wider.

  In this way, the compressed pulse Pc ′ obtained by compressing the reflected pulse partially applied to the processing unit B is not sufficiently compressed and has a small amplitude, and the compressed pulse Pc ′ appears at an incorrect position. There is also.

  In the present invention, the compression pulse Pc ′ with insufficient compression and small amplitude and the incorrect compression pulse Pc ′ appearing at an incorrect position are removed, and only the compression pulse Pc with the correct amplitude and position is employed. .

  For this purpose, the overlap processing unit creation means 11 receives the received data for one sweep (radar pulse repetition period PRI), and converts the received data rx (t) into a processing range that is, for example, twice or more the transmission pulse width τ. And their processing ranges are divided for each processing unit B having an overlap region that is at least one times the transmission pulse width τ.

  Further, in the compressed data extraction / concatenation means 16, the portion that should contain the compressed pulse Pc correctly compressed from the pulse compression processing output for each processing unit received data of the correlation means 20, that is, the first half of the processing unit B is left. , Other unnecessary portions, that is, the latter half of the processing unit B is deleted. In this way, the data ry (t) in which one sweep of the radar is seamlessly connected is generated by concatenating the portions where the compression pulse Pc obtained sequentially, that is, the first half of the processing unit B, is connected.

  FIG. 3 is a diagram for explaining operations of creating compressed data, extracting compressed data therefrom, and connecting extracted data in the pulse compression processing apparatus.

  The transmission data tx (t) in FIG. 3 (a) is based on a transmission signal (time width τ) transmitted from the transmission device for each pulse repetition period PRI. Here, the center frequency fc is zero and the upper limit is set. In the baseband region, the frequency f2 is + Δf / 2 and the lower limit frequency f1 is −Δf / 2. This transmission data tx (t) is transmitted at time t0.

  The reception data rx (t) in FIG. 5B is based on the reception signal and is shown in the baseband region as with the transmission data tx (t). This is an example in which reflection data having a head position at time td is obtained in the reception data rx (t). Time points t1, t2, t3, t4, and t5 are written so that the reception data rx (t) is divided into time intervals slightly longer than the transmission pulse width τ.

  The signals I, II, III, and IV before correlation in FIG. 5C are over the received data rx (t) at time points t0 to t2, time points t1 to t3, time points t2 to t4, and time points t3 to t5. Each of them is cut out by the lap processing unit creation means 11. Each of the correlation processing units I to IV has a length that is at least twice the transmission pulse width τ and an overlap region that is at least one time the transmission pulse width τ.

  These pre-correlation signals I, II, III, and IV are sequentially correlated by the correlator 20 (FIG. (D)), and the post-correlation signals i, ii, iii, and iv of FIG. obtain.

  Since the reflected pulse in the received data rx (t) exists across the time point t3, the compressed pulse Pc is not generated in the correlated signal i. In the post-correlation signal ii, incorrect compressed data Pc ′ is generated in the latter half (between time points t2 and t3). The post-correlation signal iii is the first half (between time points t2 and t3), and compressed data Pc that is correctly compressed is obtained at time point td, which is the head position of the reflected pulse. In the post-correlation signal iv, incorrect compressed data Pc ′ is generated in the latter half (between time points t4 and t5).

  From these compressed data Pc and Pc ′, the compressed data cutting and concatenating means 16 removes the incorrect compressed data Pc ′ and extracts only the correctly compressed compressed data Pc so that the post-correlation signals i, ii, Only the first half parts i · f, ii · f, iii · f, and iv · f of iii and iv are cut out and connected to obtain a connected signal ry (t) as shown in FIG.

  The processing range of the processing unit B has been described as being, for example, two times or more of the transmission pulse width τ and one time or less of the pulse repetition period PRI, but the processing range is one time or more of the transmission pulse width τ and the pulse repetition period PRI. If it is 1 times or less, the present invention can be realized. Even in this case, the overlap region may be one or more times the transmission pulse width τ.

The block diagram which shows the pulse compression apparatus of the pulse compression radar apparatus which concerns on this invention The figure explaining the relationship between the reflected pulse and the compression pulse in the processing unit The figure explaining creation of compression data in the pulse compression processing, and the cutting and connecting operation The figure which shows the structure of the conventional transversal filter

Explanation of symbols

11 Overlap processing unit creation means 12 FFT means 13 Coefficient table 14 Multiplication means 15 IFFT means 16 Compressed data extraction / concatenation means 20 Correlation means 1 Shift registers 1-1 to 1-n Delay devices 2-0 to 2-n Multiplication 3 Adder

Claims (2)

Transmits a modulated pulsed transmission signal to the outside, receives the reflected signal reflected from the outside as a reception signal, and performs pulse compression by performing correlation processing based on the reception signal and the transmission signal by a pulse compression device. In the pulse compression radar device to perform,
The pulse compression device comprises:
Overlap processing unit data creating means for dividing the received data based on the received signal into processing units having a predetermined processing range and the processing ranges having mutually overlapping regions, and sequentially outputting as processing unit received data;
Fourier transform means for performing Fourier transform for each processing unit reception data sequentially output from the overlap processing unit data creating means, multiplication means for multiplying the output of the Fourier transform means and a pulse compression coefficient based on the transmission signal, An inverse Fourier transform means for performing an inverse Fourier transform on the output of the multiplication means, and a correlation means for performing pulse compression processing for each processing unit received data;
From the pulse compression processing output for each processing unit received data of the correlating means, the part that should contain the compressed pulse is deleted and other unnecessary parts are deleted, and the parts that should contain the sequentially obtained compressed pulse are connected, and the radar A pulse compression radar apparatus comprising compressed data cutout / connection means for generating data for one sweep.   The pulse compression radar apparatus according to claim 1, wherein the processing range is at least twice the width of the transmission pulse, and the overlap region is at least the width of the transmission pulse.

Images