WO2024105756A1 - Radar signal processing device, radar device, and radar signal processing method - Google Patents

Abstract

A radar signal processing device (5) comprises: a frequency conversion unit (53-1, 53-2... 53-M) that performs frequency conversion of a radar signal; a target extraction unit (55) that extracts a target signal component in the frequency range from the frequency-converted radar signal; an inverse frequency conversion unit (56) that performs inverse frequency conversion of the target signal component; a phase extraction unit (57) that extracts the phase of the inverse-frequency-converted target signal component; a coefficient identification unit (58) that fits the extracted phase to identify a coefficient of a term of a polynomial representing a movement parameter relating to a target movement component; and an estimation unit (59) that uses the identified coefficient of each term of the polynomial to estimate a target movement parameter.

Description

Radar signal processing device, radar device, and radar signal processing method

This disclosure relates to a radar signal processing device, a radar device, and a radar signal processing method.

Pulse Doppler radar or synthetic aperture radar (SAR) can also estimate a polynomial that gives a function of the phase by analyzing the phase of the radar signal reflected by the target and received. Using the estimated polynomial, it is possible to estimate the motion parameters of the target in ideal noise-free conditions. Here, the motion parameters are parameters related to the motion components of the target.

However, radar signals generally contain excessive noise, and analyzing the phase of a radar signal is often difficult. For example, the discrete polynomial-phase transform (DPT) described in Non-Patent Document 1 makes it possible to estimate the coefficients of each term of a polynomial that gives a function of the phase of a radar signal with an accuracy that asymptotically approaches the Cramer-Rao Bound (CRLB).

The conventional technology described in Non-Patent Document 1 cannot estimate the coefficients of each term of a polynomial with an accuracy that reaches the CRLB unless the radar signal has a relatively high S/N (signal to noise power ratio) of about 7 dB. In other words, with the conventional technology, if the radar signal contains excessive noise, the polynomial that gives the phase function cannot be estimated with high accuracy, and the target motion parameters cannot be estimated using the polynomial.

The present disclosure aims to solve the above problems and provide a radar signal processing device, a radar device, and a radar signal processing method that can estimate target motion parameters while reducing the effects of noise.

The radar signal processing device according to the present disclosure includes a frequency conversion unit that performs integration on the radar signal by frequency conversion, a target extraction unit that extracts the target's signal components in the frequency domain from the frequency-converted radar signal, a frequency inverse conversion unit that performs inverse frequency conversion on the target's signal components, a phase extraction unit that extracts the phase of the inverse-frequency-converted target's signal components, a coefficient identification unit that identifies coefficients of polynomial terms that represent motion parameters related to the target's motion components by fitting the extracted phase, and an estimation unit that estimates the target's motion parameters using the coefficients of each term of the identified polynomial.

According to the present disclosure, by frequency converting and integrating a signal, it is possible to emphasize the target signal components contained in the signal, thereby reducing the effects of noise contained in the signal in addition to the target signal components. As a result, the radar signal processing device according to the present disclosure can estimate the motion parameters of a target while reducing the effects of noise.

1 is a block diagram showing a configuration of a radar device according to a first embodiment; 1 is a block diagram showing a configuration of a radar signal processing device according to a first embodiment; 5 is a flowchart showing a process for generating processing target data in the radar signal processing device. 4 is a flowchart showing a radar signal processing method according to the first embodiment. 5A, 5B, 5C, 5D and 5E are diagrams showing characteristics of a signal processed by the radar signal processing method of FIG. 6A and 6B are blocks showing a hardware configuration for executing the functions of the radar signal processing device according to the first embodiment. FIG. 11 is a block diagram showing a configuration of a radar device according to a second embodiment. FIG. 11 is a block diagram showing a configuration of a radar signal processing device according to a second embodiment. 10 is a flowchart showing a radar signal processing method according to a second embodiment.

The radar signal of a target received by a pulse Doppler radar at time t of the slow time is expressed by the following formula (1). In the following formula (1), A is the amplitude of the received signal of the target. _r0 is the average distance between the radar and the target, _v0 is the average speed of the target, _a0 is the average acceleration of the target, and _j0 is the average jerk. The jerk is the time derivative of the acceleration. In the following explanation, the dimension in the range direction is not taken into consideration.

However, as shown in the following equation (2), the actual received signal r(t) contains a noise component n(t).

The received signal r(t) is converted into a signal R(f) shown in the following formula (3) by performing frequency conversion in the pulse direction. In the signal R(f), the signal component of the target contained in the received signal r(t) is integrated in the Doppler frequency domain. This enhances the signal component of the target, making it possible to extract the target. In the following formula (3), R(f)=F _t [ ](f) is a symbol representing frequency conversion from time t to frequency f. The signal component of the target is extracted based on the result of comparing the signal R(f) with a threshold value. By focusing on the Doppler frequency of the signal component of the target, the average velocity v ₀ can be estimated as the motion parameter of the target.

By extracting the integrated target signal components, it is possible to significantly reduce noise, which is an unwanted signal. By inserting (setting) a zero value outside the interval in which the target signal components exist, signal R(f) becomes signal R'(f). Furthermore, by performing an inverse frequency conversion on signal R'(f), it becomes signal s'(t). Compared to signal s(t), signal s'(t) retains the target signal components while the unwanted signal components are significantly reduced.

By extracting the phase sequence of the signal s'(t) while unwrapping it, a function θ(t) indicating the amount of change in phase at time t is obtained. By performing polynomial approximation on the value obtained by the function θ(t) by, for example, fitting using the least squares method, a coefficient p _k of the k-th degree term of the polynomial that gives the function θ(t) is estimated. By using the coefficient p _k according to the following formulas (4) and (5), it is possible to estimate the average acceleration a ₀ and average jerk j ₀ of the target. Since unnecessary signals are reduced in the process of extracting the signal components of the target in the Doppler frequency domain, the motion parameters of the target can be estimated with high accuracy.

Next, the estimation of velocity in the azimuth direction by the SAR will be described.
The Doppler rate is given by the following equation (6): In the following equation (6), _vp is the moving speed of the platform, θ is the squint angle, λ is the wavelength, and _R0 is the closest distance between the radar and the target. The Doppler rate assumes that the target is stationary.

By introducing the target's velocity in the azimuth direction, _vt, into the above equation (6), the Doppler rate is expressed by the following equation (7): By using the target's Doppler rate shown in the following equation (7), it is possible to estimate the target's velocity component in the azimuth direction.

The signal component of the target in the Doppler frequency domain obtained from the received signal having a Doppler rate is expressed by the following formula (8) by performing range compression on the received signal and further correcting range cell migration. In the following formula (8), S(r, _fd ) is the range-Doppler frequency component of the target signal. r is the range, and _fd is the Doppler frequency. A(r, _fd ) is the amplitude fluctuation obtained by the range envelope obtained by range compression and the antenna pattern in the azimuth direction. _Kt is the Doppler rate including the velocity component in the azimuth direction of the target. _f0 is the center frequency. _αfd is a linear phase term generated according to the imaging position in the azimuth direction in the SAR image and the velocity component in the range direction, and n(r, _fd ) is a term representing the noise component.

By multiplying the received signal by a reference function for azimuth compression, the signal components of the target change as shown in the following equation (9). In the following equation (9), K _o is the Doppler rate of the reference function. If the target does not have a velocity component in the azimuth direction, K _t =K ₀ , and the second-order component of the phase term of the Doppler frequency component disappears. As a result, the target is ideally imaged on the SAR image. The SAR image is generated by performing an azimuth inverse frequency transform on the signal components obtained in this way.

The signal components of the target are extracted from the SAR image, and the extracted signal components are subjected to azimuth FFT (frequency transformation) to obtain the Doppler frequency components of the target. The phase of the Doppler frequency components of the target is unwrapped, and polynomial approximation is performed on the function θ(t) obtained by unwrapping. The polynomial obtained by polynomial approximation is found by the least squares method, and by using the above formula (9), the coefficient _p2 of the second-order term of the polynomial shown in the following formula (10) is obtained.

There is a certain relationship between the coefficient _p2 of the quadratic term of the polynomial that gives the function θ(t) and the target velocity _vt in the azimuth direction. By using the coefficient _p2 shown in the above formula (10), it is possible to estimate the target velocity _vest in the azimuth direction shown in the following formula (11). In the following formula (11), _Kt is expressed by the following formula (12).

Embodiment 1.
Fig. 1 is a block diagram showing a configuration of a radar device 1 according to a first embodiment. In Fig. 1, the radar device 1 is a device that is fixedly installed on the ground or installed on a platform such as an aircraft, for example, and estimates motion parameters of a target, and includes an antenna unit 2, a transmission/reception circuit 3, a storage device 4, and a radar signal processing device 5. The antenna unit 2 includes a transmission element antenna that radiates radio waves of a transmission signal, and a reception element antenna that receives the incoming radio waves. For example, the antenna unit 2 includes one or more transmission element antennas and one or more reception element antennas. In addition, one element antenna may function as both the transmission element antenna and the reception element antenna.

The transmission/reception circuit 3 transmits radio waves of a radar signal from the antenna unit 2, receives the radio waves arriving at the antenna unit 2, and generates a received signal of the radar signal. The received signal generated by the transmission/reception circuit 3 is stored as raw data in the memory unit 41 provided in the memory unit 4. The memory unit 4 is connected between the transmission/reception circuit 3 and the radar signal processing device 5 by wired communication or wireless communication. The radar signal processing device 5 estimates the motion parameters of the target using the raw data stored in the memory unit 41. Note that the transmission/reception circuit 3 may output the raw data directly to the radar signal processing device 5 without going through the memory unit 4.

As shown in FIG. 1, the transmission/reception circuit 3 includes a switching unit 31, an amplifier unit 32a, an amplifier unit 32b, a multiplier unit 33a, a multiplier unit 33b, an oscillator unit 34, a filter unit 35, an A/D converter unit 36, and a signal generator unit 37. The switching unit 31 is a circuit that switches between transmission and reception. For example, during transmission, the switching unit 31 connects the transmitting element antenna in the antenna unit 2 to the amplifier unit 32a, and disconnects the receiving element antenna in the antenna unit 2 from the amplifier unit 32b. Furthermore, during reception, the switching unit 31 connects the receiving element antenna to the amplifier unit 32b, and disconnects the transmitting element antenna from the amplifier unit 32a.

The amplifier 32a amplifies the transmission signal. The transmission signal amplified by the amplifier 32a is output to the antenna unit 2 via the switching unit 31. The amplifier 32b acquires the radio signal received by the antenna unit 2 via the switching unit 31, amplifies the acquired signal, and outputs it to the multiplier 33b. The multiplier 33a multiplies the signal generated by the signal generator 37 by the carrier wave generated by the oscillator 34.

When the signal generated by the signal generating unit 37 is multiplied by a carrier wave, it becomes a transmission signal whose frequency has been up-converted. The multiplier unit 33b multiplies the signal amplified by the amplifier unit 32b by the carrier wave generated by the oscillator unit 34. When the signal received by the antenna unit 2 is multiplied by a carrier wave, it becomes a signal whose frequency has been down-converted. The oscillator unit 34 generates a carrier wave and outputs it to the multiplier unit 33a and the multiplier unit 33b.

The filter unit 35 suppresses signal components outside the used frequency band from the down-converted signal. The A/D conversion unit 36 generates a received signal by analog/digital conversion of the signal that has passed through the filter unit 35. The received signal converted into a digital signal by the A/D conversion unit 36 is stored in the storage unit 41 as raw data.
In addition, the A/D conversion unit 36 may directly output the raw data of the received signal to the radar signal processing device 5 .

The signal generating unit 37 generates a signal to be emitted as a radio wave of a transmission signal into the space in which the target exists. For example, the signal generating unit 37 generates a signal based on a pulse signal. Note that while FIG. 1 shows a case in which the transmission/reception circuit 3 includes the signal generating unit 37, the signal generating unit 37 may be included in a control device provided separately from the transmission/reception circuit 3.

Fig. 2 is a block diagram showing the configuration of the radar signal processing device 5 according to the first embodiment. As shown in Fig. 2, the radar signal processing device 5 includes a pulse compression unit 51, compensation units 52-1, 52-2, ..., 52-M, frequency conversion units 53-1, 53-2, ..., 53-M, averaging units 54-1, 54-2, ..., 54-M, a target extraction unit 55, a frequency inverse conversion unit 56, a phase extraction unit 57, a coefficient identification unit 58, and an estimation unit 59.

The pulse compression unit 51 inputs raw data from the memory unit 41 or the transmission/reception circuit 3, and performs pulse compression on the received signal, which is the raw data. For example, if the signal generation unit 37 generates a chirp pulse signal, the pulse compression unit 51 performs pulse compression on the received chirp pulse signal.

Compensation units 52-1, 52-2, ..., 52-M compensate for target motion components in the received signal. M is an integer equal to or greater than 1. For example, compensation units 52-1, 52-2, ..., 52-M compensate for changes in the Doppler frequency domain caused by the generation of target motion components, thereby improving the power of the spectrum after the Doppler frequency.

The frequency conversion units 53-1, 53-2, ..., 53-M each perform frequency conversion on the input received signal. For example, the received signal is converted into a signal in the Doppler frequency domain by the frequency conversion units 53-1, 53-2, ..., 53-M. The averaging units 54-1, 54-2, ..., 52-M perform averaging of the frequency-converted received signal. For example, the averaging units 54-1, 54-2, ..., 52-M perform moving averaging in the Doppler frequency direction on the received signal converted into a signal in the Doppler frequency domain.

The target extraction unit 55 extracts the target's signal components in the frequency domain from the frequency-converted received signal. For example, the target extraction unit 55 extracts the target's signal components in the Doppler frequency domain using a signal sequence obtained by averaging the received signal. Then, the target extraction unit 55 extracts, from among the signals obtained by averaging, the signal whose signal value is maximized due to the target's signal component.

The frequency inverse conversion unit 56 performs an inverse frequency conversion on the target signal components extracted by the target extraction unit 55. For example, the frequency inverse conversion unit 56 performs an inverse frequency conversion on the target signal components to convert them back into a time domain signal. The phase extraction unit 57 extracts the phase of the target signal components inverse frequency converted by the frequency inverse conversion unit 56. For example, the phase extraction unit 57 unwraps the phase of the target signal components to obtain a phase function.

The coefficient identification unit 58 identifies the coefficients of the polynomial terms representing the motion parameters of the target by fitting the phase extracted by the phase extraction unit 57. For example, the coefficient identification unit 58 finds a polynomial that gives a function of the phase by fitting the extracted phase using the least squares method, and identifies the coefficients of each term of the found polynomial. The estimation unit 59 estimates the motion parameters of the target using the coefficients of each term of the polynomial identified by the coefficient identification unit 58.

The radar signal processing device 5 reduces the influence of noise by frequency converting the received signal and integrating the signal components of the target. Therefore, the radar signal processing device 5 only needs to include the frequency conversion units 53-1, 53-2, ..., 53-M, the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient specification unit 58, and the estimation unit 59 among the components shown in Fig. 2, and does not need to include any other components.
In addition, the pulse compressor 51, the compensation units 52-1, 52-2, . . . , 52-M, and the averaging units 54-1, 54-2, . . . , 54-M may be provided in an external device capable of exchanging data with the radar signal processing device 5.

The process of generating raw data (received signals) which is data to be processed by the radar signal processing device 5 will be described in detail. Fig. 3 is a flowchart showing the process of generating data to be processed by the radar signal processing device 5. First, the signal generating unit 37 generates a signal based on a pulse signal (step ST1). For example, the signal generating unit 37 generates a chirp pulse signal as a signal to be transmitted, and outputs the signal to the multiplier 33a.
It should be noted that the signal generated by the signal generating unit 37 is not limited to a chirp pulse signal, but may be a simple pulse signal or another type of pulse signal.

The multiplier 33a upconverts the frequency of the signal generated by the signal generator 37 by multiplying the signal by the carrier wave generated by the oscillator 34 (step ST2). The upconverted transmission signal is output to the amplifier 32a. The amplifier 32a amplifies the power of the input transmission signal (step ST3). The transmission signal amplified by the amplifier 32a is output to the switch 31. During transmission, the switch 31 connects the transmitting element antenna in the antenna unit 2 to the amplifier 32a. As a result, the transmission signal is output from the amplifier 32a to the antenna unit 2 via the switch 31. The antenna unit 2 radiates the transmission signal into space as radio waves (step ST4).

The radio waves of the transmission signal radiated into space are scattered or reflected by targets in space, and the scattered or reflected radio waves arrive at antenna unit 2. Antenna unit 2 receives the arriving radio waves (step ST5). During reception, switching unit 31 connects the receiving element antenna in antenna unit 2 to amplifier unit 32b. Therefore, the signal received as a radio wave by antenna unit 2 is output to amplifier unit 32b via switching unit 31. Amplifier unit 32b amplifies the power of the input signal (step ST6).

The multiplier 33b multiplies the signal amplified by the amplifier 32b by the carrier wave generated by the oscillator 34, thereby down-converting the frequency of the signal (step ST7).
The filter unit 35 suppresses signal components outside the used frequency band from the signal down-converted by the multiplication unit 33b (step ST8). The signal that has passed through the filter unit 35 is output to the A/D conversion unit 36. The A/D conversion unit 36 performs A/D conversion on the signal that has passed through the filter unit 35 to generate a digital reception signal (step ST9). The reception signal generated by the A/D conversion unit 36 is stored in the storage unit 41 as raw data.

Next, the radar signal processing method according to the first embodiment will be described.
Fig. 4 is a flow chart showing the radar signal processing method according to the first embodiment, showing the process of estimating the motion parameters of a target. Fig. 5A, Fig. 5B, Fig. 5C, Fig. 5D and Fig. 5E are diagrams showing the characteristics of a signal processed by the radar signal processing method of Fig. 4.
When the signal generated by the signal generating unit 37 is a chirp pulse, the pulse compressing unit 51 performs pulse compression on the raw data acquired from the storage unit 41 (step ST1A).
The raw data corresponding to each signal received by the antenna unit 2 is designated as received signal s ₀ (n, h), where n is the range cell number and h is the pulse number.

For example, the pulse compressor 51 performs a frequency conversion in the range direction on the received signal _s0 (n,h), multiplies the signal component obtained by this frequency conversion by a reference function for range compression, and then performs an inverse frequency conversion in the range direction. As a result, the received signal _s0 (n,h) becomes a pulse-compressed signal _s1 (r,h), where r is a range bin.

The compensation units 52-1, 52-2, ..., 52-M compensate for the acceleration component of the pulse-compressed signal s ₁ (r, h) (step ST2A). For example, the compensation units 52-1, 52-2, ..., 52-M prepare a ₁ , ..., a _m , ..., a _M as candidates for the primary estimation result of the target acceleration, and compensate for the acceleration component according to the following formula (13). In the following formula (13), s _2,m (r, h) is a signal obtained by compensating for the acceleration component of the pulse-compressed signal s ₁ (r, h). N _h is the number of pulses, T is the pulse repetition period, and λ is the wavelength of the signal.
The pulse-compressed signal s ₁ (r, h) is compensated for the acceleration component to become a signal a having a waveform shown in Fig. 5A. This signal a contains a target signal component and a noise component.

When a candidate for the primary estimation result of the target acceleration has a value close to the average acceleration a ₀ , the signal s _2,m (r, h) reduces the spread of the target velocity component in the Doppler frequency domain due to the generation of the acceleration component. Therefore, the compensation units 52-1, 52-2, ..., 52-M compensate for the motion components such as acceleration of the pulse-compressed signal s ₁ (r, h), thereby improving the power of the signal spectrum after frequency conversion by the frequency conversion units 53-1, 53-2, ..., 53-M.
Moreover, it is also possible to set a _m in s _2,m (r, h) that shows a high peak power after frequency conversion as a candidate for an estimation result of an acceleration component having a value close to the average acceleration a ₀ .
Although the motion component to be compensated for is acceleration, other motion components may be used for compensation, or a plurality of motion components may be compensated for.

The frequency conversion units 53-1, 53-2, ..., 53-M perform frequency conversion in the pulse direction on the signal s _2,m (r, h) after compensation for the acceleration component, to generate a signal S _2,m (r, f) (step ST3A), where f is the Doppler frequency.
The signal S _2,m (r, f) has a spectrum b shown in Fig. 5B. As shown in Fig. 5B, the frequency band corresponding to the target signal component is a peak band.

Next, the averaging units 54-1, 54-2, ..., 54-M perform moving averages in the Doppler frequency direction on the power components of the signal S _2,m (r, f) to generate a signal _pm (r, f) (step ST4A). This reduces the influence of noise power components in the spectrum when extracting the target signal components from the Doppler frequency spectrum of the received signal.

For example, the averaging units 54-1, 54-2, . . . , 54-M perform moving averaging for each pixel _nw in the Doppler frequency direction on the power components of the signal S _2,m (r, f).
The value of _nw may be any value, but may be set according to the following formula (14) and formula (15). In the following formula (14) and formula (15), _vreso is the _velocity resolution. Also, _nw is obtained by dividing the maximum value of the velocity spread due to acceleration remaining after acceleration compensation by the velocity resolution _vreso , and rounding the obtained value to an integer value. _vreso is a measure of the width of the Doppler frequency spectrum. Here, it is assumed that _am is set at equal intervals. In _pm (r,f) obtained by performing moving average, noise is reduced due to the effect of moving average.

In addition, the averaging units 54-1, 54-2, ..., 54-M may use the signal sequence obtained by moving averaging as weights, and multiply S _2,m (r,f) by these weights to calculate _pm (r,f). Although performing moving averaging can suppress noise, if there is no target speed spread, the reduction in signal power becomes significant. Therefore, by multiplying the signal S _2,m (r,f) by the weights obtained by performing moving averaging, noise can be reduced because the weights are obtained by moving averaging. Furthermore, since _pm (r,f) is calculated by multiplying the signal S _2,m (r,f) by the weights, it is also possible to mitigate the reduction in signal power.

The target extraction unit 55 acquires the maximum value _pm (r,f) from among the multiple pm(r,f) calculated by each of the averaging units 54-1, 54-2, ..., 54-M, and acquires _pmmax (r,f) having the maximum power from the acquired _pm (r,f). Here, the target extraction unit 55 acquires _{the mmax} corresponding to the Doppler frequency bin _fmax , range bin _rmax , and value of m of the signal pmmax(r,f) having the maximum power. The primary estimation result of the acceleration component corresponding to mmax is denoted as a _mmax .

Furthermore, the target extraction unit 55 extracts the target signal component from the signal S2 _,mmax ( _rmax ,f) according to the following formula (16) (step ST5A). In the following formula (16), _fw is the Doppler frequency width corresponding to the Doppler frequency bin number _nw . Furthermore, when the number of frequency conversion points (number of FFT points) is N, the noise power is reduced to _nw /N. The length of the signal component cut out by the target extraction unit 55 is _nw . However, the cut-out length itself may be any length.

In the following formula (16), as shown in Fig. 5C, the target extraction unit 55 sets the value of 0 to components other than the target signal component to produce a spectrum c. That is, although the target is extracted in D(f), 0 is inserted in sections other than the extracted section, reducing unnecessary signals such as noise.
This makes it possible to estimate high-order movement parameters such as speed, acceleration, and jerk with high accuracy even if the received signal has a low S/N ratio.

The frequency inverse conversion unit 56 performs an inverse frequency conversion on D(f) calculated by the target extraction unit 55 to generate a signal d(t) (step ST6A). As described above, the noise components are reduced, so that the signal d(t) becomes the signal d with the waveform shown in FIG. 5D. The signal d contains the target signal components, but the noise components are reduced.

The phase extraction unit 57 unwraps the phase of the signal d(t) to generate a phase function θ(t) (step ST7A). For example, the phase extraction unit 57 calculates the distance change characteristic Δr(t) expressed by the following equation (17) for the function θ(t). By analyzing this phase Δr(t), information on the change in the relative distance between the radar and the target can be obtained. The phase Δr(t) becomes a signal e with a waveform shown in FIG. 5E.

Next, the coefficient specifying unit 58 performs fitting on Δr(t) by the least squares method (step ST8A), thereby obtaining a polynomial that gives the function θ(t) and specifying the coefficient p _k of the kth degree term of the polynomial.

The estimation unit 59 uses coefficients _p1 , _p2 , and _p3 among the coefficients _pk of the kth degree term of the polynomial to estimate the estimated velocity v _est , estimated acceleration a _est , and estimated jerk j _est of the target according to the following equations (18), (19), and (20) (step ST9A).

It should be noted that the coefficient specifying unit 58 may use other fitting methods for fitting Δr(t) instead of the least squares method, as long as the same effect as above can be obtained.
The estimation unit 59 outputs the estimated speed v _est , the estimated acceleration a _est and the estimated jerk j _est to complete the movement.

In the above explanation, the processing has been limited to one range bin r _max , but the estimation accuracy may be improved by estimating the motion parameters in multiple range bins and calculating the average or median of the obtained values.

In addition, although the series of pulse trains is processed all at once, it is also possible to process them separately and then combine the results.

Furthermore, although the estimated motion parameters are velocity, acceleration, and jerk, higher-order motion components may also be estimated.

Next, a hardware configuration for realizing the functions of the radar signal processing device 5 will be described.
The functions of the pulse compression unit 51, the compensation units 52-1, 52-2, ..., 52-M, the frequency conversion units 53-1, 53-2, ..., 53-M, the averaging units 54-1, 54-2, ..., 54-M, the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient specification unit 58, and the estimation unit 59 included in the radar signal processing device 5 are realized by processing circuits. That is, the radar signal processing device 5 includes a processing circuit for executing the processes from step ST1A to step ST9A shown in Fig. 4. The processing circuit may be dedicated hardware, or may be a CPU (Central Processing Unit) that executes a program stored in a memory.

FIG. 6A is a block diagram showing a hardware configuration that realizes the functions of the radar signal processing device 5. FIG. 6B is a block diagram showing a hardware configuration that executes software that realizes the functions of the radar signal processing device 5. In FIGS. 6A and 6B, the input interface 100 is an interface that relays data that the radar signal processing device 5 acquires from the memory unit 41. The output interface 101 is an interface that relays the estimation results of the target motion parameters that are output from the estimation unit 59 to an external device.

When the processing circuit is the dedicated hardware processing circuit 102 shown in FIG. 6A, the processing circuit 102 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination of these. The functions of the pulse compression unit 51, the compensation units 52-1, 52-2, ..., 52-M, the frequency conversion units 53-1, 53-2, ..., 53-M, the averaging units 54-1, 54-2, ..., 54-M, the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient identification unit 58, and the estimation unit 59 provided in the radar signal processing device 5 may be realized by separate processing circuits, or these functions may be realized together by a single processing circuit.

When the processing circuit is the processor 103 shown in FIG. 6B, the functions of the pulse compression unit 51, compensation units 52-1, 52-2, ..., 52-M, frequency conversion units 53-1, 53-2, ..., 53-M, averaging units 54-1, 54-2, ..., 54-M, target extraction unit 55, frequency inverse conversion unit 56, phase extraction unit 57, coefficient identification unit 58, and estimation unit 59 provided in the radar signal processing device 5 are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 104.

The processor 103 reads out and executes the programs stored in the memory 104 to realize the functions of the pulse compression unit 51, compensation units 52-1, 52-2, ..., 52-M, frequency conversion units 53-1, 53-2, ..., 53-M, averaging units 54-1, 54-2, ..., 54-M, target extraction unit 55, frequency inverse conversion unit 56, phase extraction unit 57, coefficient identification unit 58 and estimation unit 59 provided in the radar signal processing device 5. For example, the radar signal processing device 5 includes a memory 104 for storing a program that, when executed by the processor 103, results in the processing of steps ST1A to ST9A shown in FIG. 4 being performed. These programs cause the computer to execute the procedures or methods of processing performed by the pulse compression unit 51, the compensation unit 52-1, 52-2, ..., 52-M, the frequency conversion unit 53-1, 53-2, ..., 53-M, the averaging unit 54-1, 54-2, ..., 54-M, the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient specification unit 58, and the estimation unit 59. The memory 104 may be a computer-readable storage medium that stores programs for causing the computer to function as the pulse compression unit 51, the compensation unit 52-1, 52-2, ..., 52-M, the frequency conversion unit 53-1, 53-2, ..., 53-M, the averaging unit 54-1, 54-2, ..., 54-M, the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient specification unit 58, and the estimation unit 59.

Memory 104 may be, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically-EPROM) (registered trademark), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, etc.

Some of the functions of the pulse compression unit 51, compensation unit 52-1, 52-2, ..., 52-M, frequency conversion unit 53-1, 53-2, ..., 53-M, averaging unit 54-1, 54-2, ..., 54-M, target extraction unit 55, frequency inverse conversion unit 56, phase extraction unit 57, coefficient identification unit 58 and estimation unit 59 provided in the radar signal processing device 5 may be realized by dedicated hardware, and other functions may be realized by software or firmware. For example, the functions of the pulse compression unit 51, the compensation units 52-1, 52-2, ..., 52-M, the frequency conversion units 53-1, 53-2, ..., 53-M, and the averaging units 54-1, 54-2, ..., 54-M may be realized by the processing circuit 102, which is dedicated hardware, and the functions of the target extraction unit 55, the frequency inverse conversion unit 56, the phase extraction unit 57, the coefficient identification unit 58, and the estimation unit 59 may be realized by the processor 103 reading and executing a program stored in the memory 104. In this way, the processing circuit can realize the above functions by hardware, software, firmware, or a combination of these.

As described above, the radar signal processing device 5 according to the first embodiment includes frequency conversion units 53-1, 53-2, ..., 53-M that frequency convert the radar signal, a target extraction unit 55 that extracts a target signal component in the frequency domain from the frequency-converted radar signal, a frequency inverse conversion unit 56 that inversely converts the target signal component, a phase extraction unit 57 that extracts the phase of the inversely frequency-converted target signal component, a coefficient specification unit 58 that specifies a coefficient of a polynomial term that represents a motion parameter related to the target motion component by fitting the extracted phase, and an estimation unit 59 that estimates the target motion parameter using the coefficient of each term of the specified polynomial. By frequency converting and integrating the signal, the target signal component included in the signal can be emphasized, so that it is possible to reduce the influence of noise included in the signal in addition to the target signal component. As a result, the radar signal processing device 5 can estimate the target motion parameter while reducing the influence of noise.
Furthermore, by applying the radar signal processing device 5 to a pulse Doppler radar, it is possible to estimate high-order movement parameters such as the acceleration and jerk of a target with high accuracy while reducing the effects of noise contained in the received signal.
In conventional techniques, it was only possible to estimate motion parameters at intervals equal to the compensation amount for compensating for the signal with the motion components of the target. However, the radar signal processing device 5 frequency-converts the received signal to reduce the effects of noise, and then analyzes the phase of the target's signal components. This makes it possible to estimate the motion parameters of the target with high accuracy by interpolating the conventional compensation intervals.

In the radar signal processing device 5 according to the first embodiment, the phase extraction unit 57 unwraps the phase of the target signal component. This makes it possible to calculate a function that indicates the phase of the target signal component.

In the radar signal processing device 5 according to the first embodiment, the target extraction unit 55 sets zero values to components other than the target signal components. In this way, noise reduction is performed by inserting zeros in the frequency domain, and averaging is not used, so that no signal information is lost and it is possible to estimate the motion parameters with high accuracy. This is equivalent to arranging filters that reduce noise on a Doppler frequency filter in parallel on the signal, selecting the most reliable signal from among multiple noise-reduced signals, and analyzing the phase. Noise is reduced by extracting the target signal components and inserting zero values into other bands, so it is possible to achieve reliable estimation of motion parameters without increasing the amount of calculation.

In the radar signal processing device 5 according to the first embodiment, the coefficient determination unit 58 fits the extracted phase using the least squares method. This makes it possible to calculate a polynomial that gives a function of the phase.

The radar signal processing device 5 according to the first embodiment includes compensation units 52-1, 52-2, ..., 52-M that compensate for the target motion components in the radar signal. By compensating the received signal with the target motion components, it is possible to improve the power of the spectrum of the received signal after frequency conversion.

The radar signal processing device 5 according to the first embodiment includes averaging units 54-1, 54-2, ..., 54-M that perform averaging of the frequency-converted radar signal. The target extraction unit 55 uses the signal sequence obtained by averaging the radar signal to extract the target signal components in the frequency domain. This reduces the influence of noise power components in the spectrum, making it easier to identify the target signal components from the Doppler frequency spectrum of the received signal.

In the radar signal processing device 5 according to the first embodiment, the target extraction unit 55 extracts the signal components of the target using a signal obtained by multiplying the radar signal by the signal sequence. In this way, by multiplying the radar signal by the signal sequence obtained by the averaging process as a weight, it is possible to mitigate the reduction in signal power.

The radar device 1 according to the first embodiment includes an antenna unit 2, a transmission/reception circuit 3 that transmits radio waves of a radar signal from the antenna unit 2 and receives the radio waves arriving at the antenna unit 2 to generate a received signal of the radar signal, and a radar signal processing device 5 that inputs the received signal and estimates motion parameters related to the motion components of the target extracted from the received signal. This allows the radar device 1 to estimate the motion parameters of the target while reducing the effects of noise.

The radar signal processing method according to the first embodiment includes a step in which frequency conversion units 53-1, 53-2, ..., 53-M frequency convert the radar signal, a step in which a target extraction unit 55 extracts a target signal component in the frequency domain from the frequency-converted radar signal, a step in which a frequency inverse conversion unit 56 inversely frequency converts the target signal component, a step in which a phase extraction unit 57 extracts the phase of the inverse frequency-converted target signal component, a step in which a coefficient identification unit 58 identifies coefficients of polynomial terms representing motion parameters related to the target motion components by fitting the extracted phase, and a step in which an estimation unit 59 estimates the target motion parameters using the identified coefficients of each term of the polynomial. By having the radar signal processing device 5 execute this method, it is possible to estimate the target motion parameters while reducing the effects of noise.

Embodiment 2.
Fig. 7 is a block diagram showing the configuration of a radar device 1A according to embodiment 2. In Fig. 7, the same components as those in Fig. 1 are given the same reference numerals, and duplicated explanations will be omitted. In Fig. 7, the radar device 1A is a device that is provided on a moving platform such as an aircraft or an artificial satellite, estimates motion parameters of a target, and includes an antenna unit 2, a transmission/reception circuit 3A, a storage device 4A, and a radar signal processing device 5A.
In addition, the radar device 1A may be configured by providing the antenna unit 2, the transmission/reception circuit 3A, and the memory device 4A on a platform, providing a radar signal processing device 5A as a ground station, and connecting the platform and the ground station via a communication line.

7, the transmission/reception circuit 3A includes a switching unit 31, amplifiers 32a and 32b, multipliers 33a and 33b, an oscillator 34, a filter unit 35, an A/D converter 36, a signal generator 37, and an image generator 38. The image generator 38 generates an SAR image using the received signal, which is a digital signal converted by the A/D converter 36. A method for generating an SAR image is described in, for example, Reference 1.
(Reference 1)
I. G. Cumming and F. H. Wong, Digital Processing of Synthetic Aperture Radar Data. Norwood, MA: Artech House, 2005.

The storage device 4A includes a storage unit 41A. The storage unit 41A stores SAR images. The radar signal processing device 5A extracts target signal components from the SAR images stored in the storage unit 41A. The radar signal processing device 5A may also extract target signal components from SAR images acquired directly from the transmission/reception circuit 3A.

Fig. 8 is a block diagram showing the configuration of a radar signal processing device 5A according to embodiment 2. As shown in Fig. 8, the radar signal processing device 5A includes a target extraction unit 55A, a frequency conversion unit 60, a phase extraction unit 57A, a coefficient specification unit 58A, and an estimation unit 59A.
The functions of the target extraction unit 55A, frequency conversion unit 60, phase extraction unit 57A, coefficient specification unit 58A, and estimation unit 59A included in the radar signal processing device 5A are realized by a processing circuit. That is, the radar signal processing device 5A includes a processing circuit for executing the processes from step ST1B to step ST6B shown in Fig. 9 described later. The processing circuit may be dedicated hardware, or may be a CPU that executes a program stored in a memory.

The target extraction unit 55A extracts the target signal components from the SAR image. The frequency conversion unit 60 performs frequency conversion on the target signal components extracted by the target extraction unit 55A. The phase extraction unit 57A extracts the phase of the target signal components frequency converted by the frequency conversion unit 60.

The coefficient identification unit 58A identifies the coefficients of the polynomial terms that represent the motion parameters related to the motion components of the target by fitting the phases extracted by the phase extraction unit 57A. The estimation unit 59A estimates the motion parameters of the target using the coefficients of the polynomial terms identified by the coefficient identification unit 58A.

FIG. 9 is a flowchart showing a radar signal processing method according to the second embodiment.
The target extraction unit 55A detects moving targets from the SAR image using a method such as CFAR (constant false alarm rate) or MTI (moving target indicator) (step ST1B). Here, the range bin is r, the azimuth bin is a, and the pixel of the SAR image is i(r, a).
Among those corresponding to the target extracted from the SAR image, the range bin with the maximum power is designated as r _max and the azimuth bin as a _max .

The target extraction unit 55A extracts a signal sequence d(a) corresponding to the signal component of the target, which is expressed by the following formula (21), from the SAR image (step ST2B), where c is a constant that determines the sample length for extracting the azimuth bin.

d(a) is the target signal component with reduced noise and clutter due to the insertion of zero values. If c is large, unwanted components such as noise or clutter will be mixed in, reducing reliability, but if it is small, the number of sample points will be small, reducing reliability, so it must be set appropriately. For the set c, the length for extracting azimuth bins is determined to be 2c+1.

The frequency conversion unit 60 performs azimuth frequency conversion on the signal sequence d(a) to generate S(f) (step ST3B). The second-order component of the phase term of the signal sequence S(f) is given the characteristics expressed by the above formula (9) due to the target's speed in the azimuth direction.

The phase extraction unit 57A extracts and unwraps the phase term of the signal sequence S(f) to calculate the phase sequence θ(f) (step ST4B).
The coefficient specifying unit 58A performs the least squares method to perform polynomial approximation on the phase sequence θ(f) to derive the quadratic coefficient _p2 (step ST5B). Note that the end of the sequence is the end of the antenna pattern in the azimuth direction, and may be excluded because it has a poor signal-to-noise power ratio.
Furthermore, as long as the same effect as above can be obtained, a fitting method other than the least squares method may be used.
The estimation unit 59A uses the above equations (11) and (12) from the quadratic coefficient _p2 to calculate v _est , which is the estimated velocity value in the azimuth direction of the target (step ST6B).

Although the processing has been described with reference to only one range bin r _max , the estimation accuracy may be improved by estimating the motion parameters in a plurality of range bins and finding the average or median of the obtained values.

As described above, the radar signal processing device 5A according to the second embodiment includes a target extraction unit 55A that extracts the signal components of the target from the SAR image, a frequency conversion unit 60 that frequency converts the extracted signal components of the target, a phase extraction unit 57A that extracts the phase of the frequency-converted signal components of the target, a coefficient identification unit 58A that identifies the coefficients of the polynomial terms that represent the motion parameters related to the motion components of the target by fitting the extracted phase, and an estimation unit 59A that estimates the motion parameters of the target using the coefficients of each term of the identified polynomial. This allows the radar signal processing device 5 to extract the target from the SAR image while reducing the influence of noise, and to estimate the motion parameters of the target. Furthermore, even if there is only one SAR image, it is possible to estimate the velocity in the azimuth direction of the detected target while reducing the influence of noise.

In the radar signal processing device 5A according to the second embodiment, the phase extraction unit 57A unwraps the phase of the target signal component. This makes it possible to calculate a function that indicates the phase of the target signal component.

In the radar signal processing device 5A according to the second embodiment, the coefficient determination unit 58A fits the extracted phase using the least squares method. This makes it possible to calculate a polynomial that gives a function of the phase.

The radar device 1A according to the second embodiment includes an antenna unit 2, a transmission/reception circuit 3A that transmits radar signal radio waves from the antenna unit 2 and receives the radio waves arriving at the antenna unit to generate a SAR image, and a radar signal processing device 5A that inputs the SAR image and estimates motion parameters related to the motion components of the target extracted from the SAR image. This allows the radar device 1A to estimate the motion parameters of the target from the SAR image while reducing the effects of noise.

The radar signal processing method according to the second embodiment includes a step in which a target extraction unit 55A extracts a signal component of the target from a SAR image, a step in which a frequency conversion unit 60 frequency converts the extracted signal component of the target, a step in which a phase extraction unit 57A extracts the phase of the frequency-converted signal component of the target, a step in which a coefficient identification unit 58A identifies coefficients of polynomial terms expressing motion parameters related to the motion components of the target by fitting the extracted phase, and a step in which an estimation unit 59A estimates the motion parameters of the target using the coefficients of each term of the identified polynomial. This makes it possible to estimate the motion parameters of the target from the SAR image while reducing the effects of noise.

It is possible to combine the embodiments, modify any of the components in each embodiment, or omit any of the components in each embodiment.

The radar signal processing device according to the present disclosure can be used, for example, in various radar devices.

1, 1A radar device, 2 antenna section, 3, 3A transmission/reception circuit, 4, 4A storage device, 5, 5A radar signal processing device, 31 switching section, 32a, 32b amplifier section, 33a, 33b multiplication section, 34 oscillation section, 35 filter section, 36 A/D conversion section, 37 signal generation section, 38 image generation section, 41, 41A storage section, 51 pulse compression section, 52-1, 52-2, ..., 52-M Compensation unit, 53-1, 53-1, ..., 53-M, 60 Frequency conversion unit, 54-1, 54-2, ..., 54-M Averaging unit, 55, 55A Target extraction unit, 56 Frequency inverse conversion unit, 57, 57A Phase extraction unit, 58, 58A Coefficient identification unit, 59, 59A Estimation unit, 100 Input interface, 101 Output interface, 102 Processing circuit, 103 Processor, 104 Memory.

Claims (14)

1. a frequency conversion unit that converts the frequency of a radar signal;
   a target extraction unit that extracts a signal component of a target in a frequency domain from the frequency-converted radar signal;
   a frequency inverse transform unit that inversely frequency-transforms the target signal component;
   a phase extraction unit for extracting a phase of the inverse frequency converted signal component of the target;
   a coefficient specifying unit that specifies coefficients of polynomial terms that represent motion parameters related to the motion components of the target by fitting the extracted phase;
   an estimation unit that estimates a motion parameter of the target by using a coefficient of each term of the identified polynomial.
2. The radar signal processing device according to claim 1 , wherein the phase extraction unit unwraps a phase of the signal component of the target.
3. The radar signal processing device according to claim 1 or 2, wherein the target extraction unit sets a value of 0 to components other than the signal components of the target.
4. The radar signal processing device according to claim 1 , wherein the coefficient specifying unit performs fitting of the extracted phase by a least squares method.
5. 5. The radar signal processing device according to claim 1, further comprising a compensating unit that compensates for a motion component of the target in the radar signal.
6. an averaging unit that performs averaging of the frequency-converted radar signal;
   4. The radar signal processing device according to claim 3, wherein the target extraction unit extracts a signal component of the target in a frequency domain by using a signal sequence obtained by averaging the radar signal.
7. 7. The radar signal processing device according to claim 6, wherein the target extraction unit extracts a signal component of the target by using a signal obtained by multiplying the radar signal by the signal sequence.
8. a target extraction unit that extracts a signal component of a target from a SAR image;
   a frequency conversion unit that converts the frequency of the extracted signal component of the target;
   a phase extraction unit for extracting a phase of the frequency-converted signal component of the target;
   a coefficient specifying unit that specifies coefficients of polynomial terms that represent motion parameters related to the motion components of the target by fitting the extracted phase;
   an estimation unit that estimates a motion parameter of the target by using a coefficient of each term of the identified polynomial.
9. The radar signal processing device according to claim 8 , wherein the phase extraction unit unwraps the phase of the signal component of the target.
10. The radar signal processing device according to claim 8 or 9, wherein the coefficient specifying unit performs fitting of the extracted phase by a least squares method.
11. An antenna portion;
    a transmission/reception circuit that causes the antenna unit to transmit radio waves of the radar signal and receives the radio waves arriving at the antenna unit to generate a received signal of the radar signal;
    8. A radar device comprising: a radar signal processing device according to claim 1 , which receives the received signal and estimates a motion parameter related to a motion component of the target extracted from the received signal.
12. An antenna portion;
    a transmission/reception circuit that causes the antenna unit to transmit radio waves of a radar signal and receives the radio waves arriving at the antenna unit to generate the SAR image;
    11. A radar device comprising: a radar signal processing device according to claim 8, which receives the SAR image and estimates motion parameters relating to motion components of the target extracted from the SAR image.
13. A radar signal processing method for a radar signal processing device, comprising:
    A frequency conversion unit converts the frequency of a radar signal;
    A target extraction unit extracts a signal component of a target in a frequency domain from the frequency-converted radar signal;
    A step in which a frequency inverse transform unit inversely frequency transforms the target signal component;
    A phase extraction unit extracts a phase of the inverse frequency converted target signal component;
    A step in which a coefficient specifying unit specifies coefficients of polynomial terms representing motion parameters related to the motion components of the target by fitting the extracted phase;
    an estimation unit estimating a motion parameter of the target by using a coefficient of each term of the identified polynomial.
14. A radar signal processing method for a radar signal processing device, comprising:
    A step in which a target extraction unit extracts a signal component of a target from a SAR image;
    A step of a frequency conversion unit frequency converting the extracted signal component of the target;
    A phase extraction unit extracts a phase of the frequency-converted target signal component;
    A step in which a coefficient specifying unit specifies coefficients of polynomial terms representing motion parameters related to the motion components of the target by fitting the extracted phase;
    an estimation unit estimating a motion parameter of the target by using a coefficient of each term of the identified polynomial.

Images