Radar device, radar control method, and radar control program

Abstract

A radar device (100) comprises: a pulse compression unit (26) that uses a first reference signal having the same modulation band as a repeatedly transmitted chirp signal to pulse-compress each of a plurality of reception signals obtained as a result of each of the chirp signals being reflected by a target, and uses a plurality of types of second reference signals including signals each having a modulation band and a central frequency different from those of the chirp signal to individually pulse-compress each of the plurality of reception signals; a target detecting unit (28) that detects the distance and approximate speed of the target from a plurality of first pulse-compressed signals obtained by pulse compression by means of the first reference signal; and a target speed calculating unit (30) that, on the basis of the distance and the approximate speed of the target detected by the target detecting unit (28), calculates, for each of the plurality of types of second reference signals obtained by individually pulse-compressing each of the plurality of reception signals by means of the plurality of types of second reference signals, a target speed, which is the speed of the target, from a second pulse-compressed signal composed of a plurality of compressed signals related to each of the plurality of reception signals.

Description

Radar device, radar control method, and radar control program

[0001] The present disclosure relates to a radar device, a radar control method, and a radar control program.

[0002] A chirp radar is known that detects targets with high sensitivity and high range resolution by repeatedly transmitting a chirp signal whose frequency changes over time, pulse-compressing the received signal reflected by the target, and detecting peaks from the Fourier transform of multiple pulse-compressed signals.The target's velocity can also be calculated because the Doppler frequency due to the target's movement can be determined from the position of the detected peak.However, if the target's velocity exceeds the velocity range corresponding to the observable Doppler frequency determined by the center frequency and repetition period of the transmitted signal, the target's velocity will be aliased back within the observable velocity range, which poses a problem: velocity ambiguity.

[0003] As a technique for resolving velocity ambiguity, for example, Non-Patent Document 1 discloses a technique for resolving velocity ambiguity in FMCW (frequency modulated continuous wave) radar, which repeatedly and continuously transmits chirp signals. The technique in Non-Patent Document 1 transmits and receives one type of frequency modulated continuous wave, detects a target from the results of 2D-FFT (fast Fourier transform), and detects the peak fast-time frequency corresponding to the distance and the ambiguous velocity. Since the detected velocity is the actual velocity of the target that is folded back an indefinite number of times within the observable velocity range, multiple velocity candidates before the folding back are calculated. Next, for each speed candidate, a 2D-DFT (discrete Fourier transform) is calculated so that the peak fast-time frequency of the partial signal of the transmitted frequency-modulated continuous wave and the component corresponding to the speed candidate are integrated, and the speed candidate with the maximum amplitude in the 2D-DFT calculation result is set as the target speed.

[0004] Zhang, Cheng, et al. “Velocity ambiguity resolution for wideband automatic millimeter wave radar: a carrier frequency multiplexing ”Journal of Electromagnetic Waves and Applications 34.3 (2020): 375-389.

[0005] However, in the technology described in Non-Patent Document 1, only one type of part of the transmitted signal is used when calculating a speed without ambiguity from multiple speed candidates, which has the problem of increasing the probability of selecting an incorrect speed candidate when the S / N ratio of the received signal is low, i.e., increasing the probability of failing to resolve the ambiguity.

[0006] An object of the present disclosure is to provide a radar device, a radar control method, and a radar control program that can suppress the occurrence of velocity ambiguity.

[0007] The radar device of the present disclosure includes a pulse compression unit that pulse-compresses each of a plurality of received signals, which are the result of receiving a plurality of reflected signals obtained by reflecting each of repeatedly transmitted chirp signals off a target, using a first reference signal having the same modulation band as the chirp signal, and pulse-compresses each of the plurality of received signals individually using a plurality of types of second reference signals including signals having modulation bands and center frequencies different from those of the chirp signal; a target detection unit that detects the distance and approximate velocity of the target from a plurality of first-pulse-compressed signals obtained by pulse-compressing each of the plurality of received signals using the first reference signal; and a target velocity calculation unit that calculates the target velocity, which is the velocity of the target, based on the distance and approximate velocity of the target detected by the target detection unit from second-pulse-compressed signals that are obtained by individually pulse-compressing each of the plurality of received signals using the plurality of types of second reference signals and that are composed of a plurality of pulse-compressed signals corresponding to each of the plurality of received signals for each type of second reference signal.

[0008] A radar control method according to the present disclosure is a computer-executable radar control method, comprising the steps of: pulse-compressing each of a plurality of received signals, which are the result of receiving a plurality of reflected signals obtained by reflecting each of repeatedly transmitted chirp signals off a target, using a first reference signal having the same modulation band as the chirp signal; and individually pulse-compressing each of the plurality of received signals using a plurality of types of second reference signals, including signals having modulation bands and center frequencies different from those of the chirp signal; detecting a distance and an approximate velocity of the target from a plurality of first-pulse-compressed signals obtained by pulse-compressing using the first reference signal; and calculating a target velocity, which is the velocity of the target, based on the detected distance and approximate velocity of the target from second-pulse-compressed signals composed of a plurality of pulse-compressed signals for each of the plurality of received signals for each type of second reference signal, which are obtained by individually pulse-compressing each of the plurality of received signals using the plurality of types of second reference signals.

[0009] The radar control program of the present disclosure causes a computer to execute the following steps: pulse-compressing each of a plurality of received signals, which are the result of receiving each of a plurality of reflected signals obtained by reflecting each of repeatedly transmitted chirp signals off a target, using a first reference signal having the same modulation band as the chirp signal, and individually pulse-compressing each of the plurality of received signals using a plurality of types of second reference signals including signals each having a modulation band and a center frequency different from those of the chirp signal; detecting the distance and approximate velocity of the target from the plurality of first-pulse-compressed signals obtained by pulse-compressing using the first reference signal; and calculating the target velocity, which is the velocity of the target, based on the detected distance and approximate velocity of the target from second-pulse-compressed signals composed of a plurality of pulse-compressed signals corresponding to each of the plurality of received signals for each type of second reference signal, which are obtained by individually pulse-compressing each of the plurality of received signals using the plurality of types of second reference signals.

[0010] According to the present disclosure, it is possible to provide a radar device, a radar control method, and a radar control program that can suppress the occurrence of velocity ambiguity by performing signal processing using multiple reference signals that have a different modulation band from the transmitted chirp signal and each include signals with different center frequencies.

[0011] FIG. 1 is a configuration diagram of a radar device according to a first embodiment. FIG. 2 is a process image diagram in the radar device according to the first embodiment. FIG. 3 is a flowchart showing an example of a process for resolving velocity ambiguity using coherent addition of DFT in the radar device according to the first embodiment. FIG. 4 is a flowchart showing an example of a process for resolving velocity ambiguity using absolute value addition of DFT in the radar device according to the first embodiment. FIG. 5 is a flowchart showing an example of a process for resolving velocity ambiguity using coherent addition of DFT in the radar device according to the second embodiment. FIG. 6 is a graph showing a performance ratio with respect to the modulation bandwidth of a reference signal β in the radar device according to the third embodiment. FIG. 7 is a flowchart showing an example of a process for resolving velocity ambiguity using two reference signals β in the radar device according to the third embodiment.

[0012] Radar devices according to embodiments 1, 2, 3, 4, and 5 will be described below with reference to the drawings. The following embodiments are merely examples, and the embodiments can be combined as appropriate and each embodiment can be modified as appropriate.

[0013] 1 is a block diagram showing an example of the configuration of a radar device 100 according to embodiment 1. As shown in Fig. 1, the radar device 100 includes a DSS (Direct Digital Synthesizer) 1, a control device 2, a mixer 3, a high-frequency signal generating circuit 4, an amplifier 5, a duplexer 6, an antenna 7, an amplifier 8, a mixer 9, an amplifier 10, an A / D converter 11, and a signal processing circuit 12. The DSS 1, the control device 2, the high-frequency signal generating circuit 4, the A / D converter 11, and the signal processing circuit 12 are each connected to a common reference signal generating circuit and operate in synchronization with each other, and the mixers 3 and 9 each have a function of filtering out undesired signals after signal mixing. Each of the DSS 1, control device 2, mixer 3, high-frequency signal generating circuit 4, amplifier 5, transmit / receive switch 6, antenna 7, amplifier 8, mixer 9, amplifier 10, A / D converter 11, and signal processing circuit 12 functions as either a signal generating unit 20, a signal transmitting unit 22, a signal receiving unit 24, a pulse compressing unit 26, a target detecting unit 28, or a target speed calculating unit 30, depending on its role in signal processing.

[0014] The signal generating unit 20 repeatedly generates chirp signals and is composed of a DDS 1, a control device 2, a mixer 3, and a high-frequency signal generating circuit 4. The DDS 1 generates a chirp signal based on the chirp start frequency, modulation bandwidth, period, and initial phase of the period of the chirp signal output from the control device 2. The chirp signal output from the DDS 1 is input to the mixer 3. The mixer 3 sequentially mixes the chirp signal with the high-frequency signal generated by the high-frequency signal generating circuit 4, generating a chirp signal with a converted center frequency.

[0015] The signal transmitting unit 22 is a unit that repeatedly transmits the chirp signal generated by the signal generating unit 20, and is composed of an amplifier 5, a duplexer 6, and an antenna 7. The chirp signal is amplified by the amplifier 5 and transmitted from the antenna 7 via the duplexer 6.

[0016] The signal receiving unit 24 receives multiple reflected signals resulting from the chirp signals repeatedly transmitted by the signal transmitting unit and reflected from targets, and is composed of an antenna 7, a duplexer 6, an amplifier 8, a mixer 9, an amplifier 10, and an A / D converter 11. The chirp signals transmitted from the antenna 7 are reflected from targets and received by the antenna 7. The received signals pass through the duplexer 6 and are amplified by the amplifier 8. The amplified received signals are input to the mixer 9. The mixer 9 mixes the amplified received signals with the high-frequency signal generated by the high-frequency signal generating circuit 4 and filters out unwanted signals, thereby generating a received signal with a converted center frequency. The center-frequency-converted received signal is amplified by the amplifier 10 and converted into a digital signal by the A / D converter 11.

[0017] The pulse compressor 26 is a component that performs pulse compression on the received signals that have been received by the signal receiver 24 and converted into digital signals, and is comprised of the signal processing circuit 12. The multiple received signals are each pulse-compressed by the signal processing circuit 12 using a first reference signal (hereinafter referred to as "reference signal α") having the same modulation band as the transmitted chirp signal, and are also individually pulse-compressed using multiple reference signals β-m included in a second reference signal (hereinafter referred to as "reference signal β") that has a different modulation band from the transmitted chirp signal and includes a signal with a different center frequency.

[0018] The target detection unit 28 is a component that detects targets from a plurality of pulse-compressed signals (referred to in the claims as "first-pulse-compressed signals" or "plural first-pulse-compressed signals") obtained by pulse-compressing the received signals using the reference signal α by the pulse compression unit 26, and is configured with the signal processing circuit 12. The signal processing circuit 12 performs Doppler FFT on the plurality of pulse-compressed signals to create a range-Doppler map, and further detects peaks in the range-Doppler map to calculate the target distance and the velocity at which velocity ambiguity may occur. The velocity calculated by the target detection unit 28 is an approximate velocity (approximate velocity) whose accuracy is not guaranteed. In the first embodiment, as will be described later, the target velocity calculation unit 30 calculates the target velocity, which is the velocity of the target, based on the distance and approximate velocity of the target detected by the target detection unit 28 from pulse-compressed signals (referred to in the claims as "second pulse-compressed signals") that are composed of multiple compressed signals related to each of the multiple received signals for each of the multiple types of reference signals β, which are obtained by individually pulse-compressing each of the multiple received signals using multiple types of reference signals β.

[0019] The target velocity calculation unit 30, which is a component that resolves velocity ambiguity in the velocity detected by the target detection unit, is composed of a signal processing circuit 12. The signal processing circuit 12 determines target velocity candidates by adding or subtracting an integer multiple of twice the Nyquist velocity (the maximum velocity that can be observed without velocity ambiguity due to aliasing) within a predetermined range to or from the estimated velocity. For each type of pulse-compressed signal, which is individually pulse-compressed using a reference signal β, the signal processing circuit 12 calculates a DFT (Discrete Fourier Transform) for a portion corresponding to the distance of the target detected by the target detection unit, using the central pulse-compressed signal as a reference, while correcting for a Doppler frequency difference that is proportional to the center frequency of the reference signal β used for pulse compression and proportional to the target velocity candidate. After correcting the phase of each type of DFT, the signal processing circuit 12 coherently adds the results of each type of DFT. The signal processing circuit 12 then outputs the target velocity candidate that maximizes the coherent addition result of the DFT as the target velocity.

[0020] The signal processing circuit 12, which functions as the pulse compression unit 26, the target detection unit 28, and the target velocity calculation unit 30, is a type of computer, and each of the CPU (Central Processing Unit), main memory, input / output interface 230, and memory unit is connected to one another via a system bus.

[0021] The CPU is an integrated circuit (IC) that performs arithmetic processing. Instead of a CPU, a computing element such as a digital signal processor (DSP) or a graphics processing unit (GPU) may be used. By executing a radar control program, the CPU functions as the pulse compressor 26, target detector 28, and target velocity calculator 30. The radar control program is provided, for example, on a recording medium on which it is recorded.

[0022] Next, a method for detecting a target in a normal chirp radar will be described. For simplicity, it is assumed that the target moves at a constant speed in a straight line relative to the radar device 100, as shown in the following equation (1). In the following equation (1), x ０ represents the initial position of the target, and v represents the velocity of the target.

[0023]

[0024] The chirp signal repeatedly transmitted by the chirp radar is expressed by the following equation (2): In the following equation (2), B is the modulation bandwidth, T ｐ is the pulse width, f ｃ represents the carrier frequency.

[0025]

[0026] The received signal of the wave reflected by the target can be approximated by the following equation (3): In the following equation (3), Δf is the Doppler frequency, expressed as Δf = -2v / λ, and is proportional to the target velocity v. λ is the wavelength of the carrier wave, λ = c / f ｃ and c is the speed of light.

[0027]

[0028] ζ(t) related to equation (3) is the reception time of the signal transmitted at time t, which is expressed by the following equation (4).

[0029]

[0030] In chirp radar, pulse compression, which is a cross-correlation between the received signal and the transmitted signal, is performed to improve the distance resolution of the target reflected wave. Pulse compression may be performed on both the transmitted and received signals in the baseband signal state. Detection sensitivity is then improved by performing a Doppler FFT, i.e., coherent integration, between the pulse-compressed chirp signals repeatedly transmitted. The target velocity can be calculated from the peak position detected on the range-Doppler map (hereinafter referred to as "RD map" in Figures 2, 3, 4, 5, 6, and 8), which is the result of the Doppler FFT. However, if the target velocity is greater than the carrier frequency f, which is the center frequency of the transmitted signal, ｃ and the repetition period T ＰＲＩ If the observable velocity range determined by v is exceeded, the target velocity will be folded back into the observable velocity range, i.e., velocity ambiguity will occur. The observable velocity range is defined by v Ｎｙｑ = λ / 4T ＰＲＩ The Nyquist velocity v Ｎｙｑ Therefore, ±v Ｎｙｑ The range is.

[0031] 2 is a process image diagram of the radar device 100 according to the first embodiment. In the first embodiment, as in a normal chirp radar, a chirp signal is repeatedly transmitted and received K times, as shown in (1) of FIG. 2, and each received signal r ｋFor (t) (k = 1, ..., K), pulse compression (2) in Fig. 2 and Doppler FFT (3) in Fig. 2 are performed using a reference signal α with the same modulation band as the transmitted chirp signal, and the target distance x^ (hereinafter, in the formula, "^" is placed above "x") and approximate velocity v^ (hereinafter, in the formula, "^" is placed above "v"), which may have velocity ambiguity, are calculated as shown in Fig. 2 (4). In addition, velocity ambiguity is resolved by signal processing shown in (5) to (9) in Fig. 2. The target distance x^ and approximate velocity v^ can be detected from the peaks obtained by integrating multiple pulse-compressed signals, the phase of which has changed due to the movement of the target, at Doppler frequencies corresponding to the target distance and velocity using Doppler FFT, as shown in Fig. 2 (4).

[0032] To resolve velocity ambiguity, multiple reference signals including signals with different modulation bands and center frequencies from the transmitted chirp signal are used, as shown in (5) of Fig. 2, and pulse compression is utilized, as shown in (6) of Fig. 2. In the first embodiment, reference signals β-m (m = 1, ..., M) obtained by dividing the transmitted chirp signal into M parts are used as reference signals β. Reference signal β-m, defined as the signal from the pulse start time to the pulse end time, is expressed by the following equation (5), and the pulse start time and pulse end time are expressed by the following equations (6).

[0033]

[0034]

[0035] The center frequency f of the reference signal β-m ｍ is expressed by the following equation (7).

[0036]

[0037] In the first embodiment, the received signal r ｋ The signal after pulse compression by the reference signal β-m for (t) is expressed as x ｋ，ｍ (t) (k=1, . . . , K, m=1, . . . , M).

[0038] Since pulse compression is a cross-correlation process, pulse compression using a reference signal β-m for a received signal obtained by receiving a reflected wave of a transmitted chirp signal produces results equivalent to pulse compression using a reference signal β-m for a received signal obtained by transmitting a reference signal β-m and receiving a reflected wave from a target. In other words, the Doppler frequency of the received signal obtained by receiving a reflected wave of a transmitted chirp signal is Δf ｍ =-2vf ｍ In the first embodiment, the center frequency f of the reference signal β-m is ｍ (central frequency f ｍ Doppler frequency Δf ｍ The velocity ambiguity is resolved using the difference between

[0039] Specifically, the approximate velocity v^ where velocity ambiguity may occur and the Nyquist velocity v Ｎｙｑ The speed is calculated by adding or subtracting an integer multiple of twice the speed of the target. The speed within the assumed target speed range is designated as the target speed candidate v ｎ (n = 1, ..., N). The approximate velocity v^ is usually calculated as a discrete value, but it is desirable to estimate and use an accurate velocity by Lagrange interpolation using the Doppler FFT values ​​of v^ and three points including both sides of v^. Other estimation methods such as spline interpolation may also be used. Then, the pulse-compressed signal x by each reference signal β-m is calculated. ｋ，ｍ The pulse-compressed signal x at the target detection time t^ (hereinafter, in the formulas, "^" is placed above "t") corresponding to the target distance x^ of (t) ｋ，ｍ For each target speed candidate v ｎ As shown in (8) of FIG. 2, the DFT is calculated to cancel the difference in Doppler frequency due to the Doppler frequency difference. As shown in (9) of FIG. 2, the target velocity candidate v that maximizes the coherent addition result is calculated. ｎ is calculated as the target velocity v~ (hereinafter, in the formula, "~" is placed above "v"). The method for determining the target candidate v~ is expressed as the following formula (8). f in formula (8) ｍ is the center frequency f of the reference signal β-m ｍ Therefore, as a result, the target velocity candidate v that maximizes the coherent addition result for all reference signals β-m isｎ is calculated as the target speed v~.

[0040]

[0041] For coherent addition, it is not necessary to use pulse-compressed signals using all reference signals β-m. For example, pulse-compressed signals using reference signals β-m whose center frequencies are close to the reference signal α have little effect on the calculation of equation (8), and therefore may not be used when prioritizing reduction in the amount of calculation. Furthermore, signals obtained by unequally dividing the transmission chirp signal may be used as reference signals β-m, or reference signals β-m may overlap each other.

[0042] In equation (8), the value of k where the exponent of the exponential function becomes 0, i.e., the base point of the DFT, is set to the median value of K, which is the number of chirp signals transmitted K times, i.e., the largest integer not exceeding K / 2 plus 1, and the signal after pulse compression by the reference signal β of the received signal related to the transmission signal (chirp signal) whose transmission order is the center is used as the reference. The reason for this is that the phases of the signals after pulse compression by different reference signals β-m are aligned at their peak positions but shift in proportion to the difference in the center frequencies of the reference signals β-m as they move away from the peak positions. Therefore, the difference in Doppler frequency is corrected using the signal after pulse compression whose peak position is near the target distance x^ as the base point. Furthermore, although the target detection time t^ is usually calculated as a discrete value, if the exact peak position t~ (hereinafter, in the equation, "~" above "t") of the range-Doppler map is different from t^, the phases of the signals after pulse compression by different reference signals β-m will also be shifted. Therefore, it is desirable to estimate the accurate peak position t of the pulse-compressed signal in advance by Lagrange interpolation or the like using t^ and three points including both sides thereof, calculate the phase shift between the pulse-compressed signals of reference signals β-m and β-m' as shown in the following equation (9), and correct the phase shift when calculating equation (8). Other methods such as spline interpolation may also be used to estimate the peak position of the pulse-compressed signal.

[0043]

[0044] To summarize the above, an example of the process of resolving velocity ambiguity using coherent addition of DFT in the radar device 100 of the first embodiment is shown in Fig. 3. In step S100, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the received signal received by the signal receiver 24 using the reference signal α. As will be described later, in the first embodiment, the procedure of step S106 is performed in parallel with the procedure of step S100.

[0045] In step S101, the signal processing circuit 12, which is the target detection unit 28, performs Doppler FFT on the pulse-compressed signal.

[0046] In step S102, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target detection time t^ and the approximate velocity v^ at which velocity ambiguity may occur from the range-Doppler map.

[0047] In step S103, the signal processing circuit 12, which is the target velocity calculation unit 30, corrects the phase shift due to the difference between the target detection time t^ and the exact peak position of the range-Doppler map.

[0048] In step S104, the accuracy of the estimated velocity v^, which may contain velocity ambiguity, is improved. Specifically, although the estimated velocity v^ is usually calculated as a discrete value, the accuracy of the estimated velocity v^ is improved by estimating an accurate velocity by Lagrangian interpolation or the like using Doppler FFT values ​​of v^ and three points including both sides of v^.

[0049] In step S105, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the approximate velocity v^ in which velocity ambiguity may occur and the Nyquist velocity v^. Ｎｙｑ The speed is calculated by adding or subtracting an integer multiple of twice the speed of the target. The speed within the assumed target speed range is designated as the target speed candidate v ｎ Calculation is performed as (n=1, . . . , N).

[0050] In step S106, in parallel with the procedure in step S100, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the signal received by the signal receiver 24 using the reference signal β.

[0051] In step S107, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the velocity candidate v calculated in step S105. ｎ The target velocity v is calculated using the pulse-compressed signal x by the reference signal β obtained in step S106, and the process ends. Specifically, as shown in the above equation (8), the pulse-compressed signal x by the reference signal β-m at the target detection time t is calculated using the ｋ，ｍ For each target speed candidate v ｎ The DFT is calculated so as to cancel the difference in Doppler frequency due to the difference in velocity between the target velocity candidate v and the target velocity candidate v, which maximizes the coherent addition result of each DFT. ｎ is calculated as the target speed v~.

[0052] In equation (8), the results of DFT are coherently added, but absolute value addition may also be used. When absolute value addition is used, phase shift correction is not required. In such a case, the target velocity v is calculated using the following equation (10).

[0053]

[0054] 4 shows an example of a process for resolving velocity ambiguity using absolute value addition of DFT. In step S200, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the received signal received by the signal receiver 24 using the reference signal α. As will be described later, in the first embodiment, the procedure of step S205 is performed in parallel with the procedure of step S200.

[0055] In step S201, the signal processing circuit 12, which is the target detection unit 28, performs Doppler FFT on the pulse-compressed signal.

[0056] In step S202, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target detection time t^ and the approximate velocity v^ at which velocity ambiguity may occur from the range-Doppler map.

[0057] In step S203, the accuracy of the estimated velocity v^, which may contain velocity ambiguity, is improved. Specifically, although the estimated velocity v^ is usually calculated as a discrete value, the accuracy of the estimated velocity v^ is improved by estimating an accurate velocity by Lagrangian interpolation or the like using Doppler FFT values ​​of v^ and three points including both sides of v^.

[0058] In step S204, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the approximate velocity v^ in which velocity ambiguity may occur and the Nyquist velocity v^. Ｎｙｑ The speed is calculated by adding or subtracting an integer multiple of twice the speed of the target. The speed within the assumed target speed range is designated as the target speed candidate v ｎ Calculation is performed as (n=1, . . . , N).

[0059] In step S205, in parallel with the procedure in step S200, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the signal received by the signal receiver 24 using the reference signal β.

[0060] In step S206, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the velocity candidate v calculated in step S204. ｎ The target velocity v is calculated using the pulse-compressed signal x by the reference signal β obtained in step S205, and the process ends. Specifically, as shown in the above equation (10), the pulse-compressed signal x by the reference signal β-m at the target detection time t ｋ，ｍ For each target speed candidate v ｎ The DFT is calculated so as to cancel the difference in Doppler frequency due to the difference in the ... ｎ is calculated as the target speed v~.

[0061] In the case of coherent addition, the addition effect is large, so a high probability of ambiguity resolution is obtained when the model error is small, and in the case of absolute value addition, robustness against model error is high.

[0062] As described above, the radar device 100, radar control method, and radar control program according to the first embodiment include the signal generation unit 20 that generates a chirp signal, the signal transmission unit 22 that repeatedly transmits the generated chirp signal, the signal reception unit 24 that receives a plurality of reflected signals resulting from the chirp signals repeatedly transmitted by the signal transmission unit 22 being reflected by a target, the pulse compression unit 26 that pulse-compresses each of the plurality of received signals received by the signal reception unit 24 using a reference signal α and individually pulse-compresses each of the plurality of received signals using a reference signal β, the target detection unit 28 that detects a target from the plurality of pulse-compressed signals that have been pulse-compressed by the pulse compression unit 26 using the reference signal α, and the target velocity calculation unit 30 that uses the distance and velocity of the target detected by the target detection unit 28 to calculate an approximate velocity v^ of the target from the plurality of types of pulse-compressed signals that have been individually pulse-compressed by the pulse compression unit 26 using the reference signal β. The target velocity calculation unit 30 calculates an approximate velocity v^ of the target detected by the target detection unit 28 and a Nyquist velocity v^ within a predetermined range. Ｎｙｑ The speed obtained by adding or subtracting an integer multiple of twice the target speed candidate v ｎ Then, for the portion of the plurality of types of pulse-compressed signals obtained by individually pulse-compressing the received signal using the reference signal β, which corresponds to the distance x^ of the target detected by the target detection unit 28, the DFT is calculated based on the central pulse-compressed signal while correcting the difference in Doppler frequency, which is proportional to the approximate velocity v^ of the target for each type of reference signal β and proportional to the center frequency of the reference signal β used for pulse compression, and after correcting the phase of each type of DFT, the DFTs of each type are coherently added to obtain the target velocity candidate v that maximizes the coherent addition result of the DFT. ｎ is output as the target velocity v~, so that the velocity of the target that exceeds the observable velocity range determined by the center frequency and repetition period of the transmission signal can be calculated with high probability.

[0063] Next, a second embodiment will be described. The radar device of the second embodiment differs from the radar device 100 of the first embodiment in that the signal processing circuit 12 functions as a target velocity correction unit 32 (described later), and the signal processing circuit 12, which is the target velocity calculation unit 30, resolves velocity ambiguity of the velocity detected by the target detection unit using an iterative method. However, the signal generation unit 20, signal transmission unit 22, signal reception unit 24, pulse compression unit 26, and target detection unit 28 are the same as those of the first embodiment, and the radar device 100 of the first embodiment also includes the DSS 1, control unit 2, mixer 3, high-frequency signal generation circuit 4, amplifier 5, transmission / reception switch 6, antenna 7, amplifier 8, mixer 9, amplifier 10, A / D converter 11, and signal processing circuit 12. Therefore, the same components as those of the first embodiment are denoted by the same reference numerals as those of the first embodiment, and detailed description thereof will be omitted.

[0064] The target velocity calculation unit 30 is a component that resolves velocity ambiguity in the velocity detected by the target detection unit 28, and is composed of the signal processing circuit 12. The signal processing circuit 12 iteratively calculates and outputs the target velocity that maximizes the result of coherent addition after correcting the phase of the DFT calculated based on the central pulse-compressed signal, while correcting the difference in Doppler frequency that is proportional to the center frequency of the reference signal β and proportional to the velocity of the target, for the portion of the multiple types of multiple pulse-compressed signals that are individually pulse-compressed using the reference signal β that corresponds to the distance to the detected target.

[0065] The target speed correction unit 32 is a part that corrects the target speed for which the speed ambiguity has been resolved by the target speed calculation unit 30, and is composed of the signal processing circuit 12. The signal processing circuit 12 corrects the target speed for which the speed ambiguity has been resolved to the speed before it was folded back into the observable speed range.

[0066] As described above, the target velocity calculation unit 30 uses an iterative method to resolve the velocity ambiguity of the velocity detected by the target detection unit 28. Specifically, in the iterative method, the pulse-compressed signal x ｋ，ｍ (t) target detection time t^ pulse compressed signal x ｋ，ｍFor (t^), calculate ζ- (hereinafter, "-" will be placed above "ζ" in the formula) that maximizes X(ζ) shown in formula (11), which is the result of coherent addition of DFT calculated so as to cancel out phase changes due to Doppler frequency. If an increase in the amount of calculation is not a problem, ω^ (hereinafter, "^" will be placed above "ω" in the formula), which is a constant based on v^ in formula (11), may be added to formula (11) as an optimization variable as angular frequency ω. Also, as in the first embodiment, it is desirable to calculate the phase shift between the pulse-compressed signals of reference signals β-m and β-m' as shown in formula (9) above, and correct the phase shift when calculating formula (11).

[0067]

[0068] In equation (11), ω^ is an angular frequency for canceling the phase change due to the estimated velocity v^, which may cause the detected velocity ambiguity, and g ｍ is the center frequency offset of the reference signal β-m (f ｍ -f ｃ ) to the modulation bandwidth B of the transmitted chirp signal. As in the first embodiment, it is desirable to use an accurate velocity estimated by Lagrangian interpolation or the like for the approximate velocity v^. The target velocity v- (hereinafter, in the equations, a "-" is placed above "v") after resolving the ambiguity is calculated using the following equation (12).

[0069]

[0070] The iterative method used in the second embodiment can utilize various methods for solving continuous optimization problems, including gradient methods such as the steepest descent method, Newton's method, and interior point method. As an example, in the case of the Newton's method, the target equation is set to the following formula (13), and the maximum value of X(ζ) is calculated.

[0071]

[0072] The initial value of the Newton method is, for example, ζ shown in the following equation (14): ０ can be used. ｍ ^ (hereinafter, in the formula, "^" is placed above "ω") is the pulse-compressed signal x ｋ，ｍ is the peak angular frequency of the Doppler FFT of (t).

[0073]

[0074] The target speed correction unit 32 corrects the target speed v- calculated by the target speed calculation unit 30 to the speed v~ before it is folded back into the approximate speed v^, which may contain speed ambiguity detected by the target detection unit 28, using the following equation (15):

[0075]

[0076] To summarize the above, an example of a process for resolving velocity ambiguity using coherent addition of DFT in the radar device according to the second embodiment is shown in Fig. 5. In step S300, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression using the reference signal α on the signal received by the signal receiver 24. As will be described later, in the second embodiment, the procedure of step S305 is performed in parallel with the procedure of step S300.

[0077] In step S301, the signal processing circuit 12, which is the target detection unit 28, performs Doppler FFT on the pulse-compressed signal.

[0078] In step S302, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target detection time t^ and the approximate velocity v^ at which velocity ambiguity may occur from the range-Doppler map.

[0079] In step S303, the signal processing circuit 12, which is the target velocity calculation unit 30, corrects the phase shift due to the difference between the target detection time t^ and the exact peak position of the range-Doppler map.

[0080] In step S304, the accuracy of the estimated velocity v^, which may contain velocity ambiguity, is improved. Specifically, although the estimated velocity v^ is usually calculated as a discrete value, the accuracy of the estimated velocity v^ is improved by estimating an accurate velocity by Lagrangian interpolation or the like using Doppler FFT values ​​of v^ and three points including both sides of v^.

[0081] In step S305, in parallel with the procedure in step S300, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the signal received by the signal receiver 24 using the reference signal β.

[0082] In step S306, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the initial value ζ of the Newton method in accordance with the above-mentioned equation (14). ０ Calculate.

[0083] In step S307, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the pulse-compressed signal x by each reference signal β-m at the target detection time t^. ｋ，ｍ For (t^), ζ- that maximizes X(ζ) shown in equation (11), which is the coherent addition result of DFT calculated so as to cancel out the phase change due to the Doppler frequency, is calculated using Newton's method to find ζ- that satisfies the above equation (13).

[0084] In step S308, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target velocity v- after resolving the ambiguity using the above-mentioned equation (12).

[0085] In step S309, the signal processing circuit 12, which is the target speed correction unit 32, corrects the target speed v- calculated by the target speed calculation unit 30 to the speed v~ before it is folded back to the approximate speed v^ detected by the target detection unit 28, which may be causing speed ambiguity, using the above-mentioned equation (15), and then ends the processing.

[0086] In equation (11), the results of DFT are coherently added, but absolute value addition may also be performed. In absolute value addition, the following equation (16) is used instead of equation (11).

[0087]

[0088] 6 shows an example of a process for resolving velocity ambiguity using absolute value addition of DFT. In step S400, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the received signal received by the signal receiver 24 using the reference signal α. As will be described later, in the second embodiment, the procedure of step S405 is performed in parallel with the procedure of step S400.

[0089] In step S401, the signal processing circuit 12, which is the target detection unit 28, performs Doppler FFT on the pulse-compressed signal.

[0090] In step S402, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target detection time t^ and the approximate velocity v^ at which velocity ambiguity may occur from the range-Doppler map.

[0091] In step S403, the accuracy of the estimated velocity v^, which may contain velocity ambiguity, is improved. Specifically, although the estimated velocity v^ is usually calculated as a discrete value, the accuracy of the estimated velocity v^ is improved by estimating an accurate velocity by Lagrangian interpolation or the like using Doppler FFT values ​​of v^ and three points including both sides of v^.

[0092] In step S404, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the initial value ζ of the Newton method in accordance with the above-mentioned equation (14). ０ Calculate.

[0093] In step S405, in parallel with the procedure in step S200, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the signal received by the signal receiver 24 using the reference signal β.

[0094] In step S406, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the pulse-compressed signal x by each reference signal β-m at the target detection time t^. ｋ，ｍ For (t^), the ζ- that maximizes X(ζ) shown in equation (16), which is the result of adding the absolute values ​​of the DFT calculated so as to cancel out the phase change due to the Doppler frequency, is calculated using Newton's method to find the ζ- that satisfies the above equation (13).

[0095] In step S407, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target velocity v- after resolving the ambiguity using the above-mentioned equation (12).

[0096] In step S408, the signal processing circuit 12, which is the target speed correction unit 32, corrects the target speed v- calculated by the target speed calculation unit 30 to the speed v~ before it is folded back to the approximate speed v^ detected by the target detection unit 28, which may be causing speed ambiguity, using the above-mentioned equation (15), and then ends the processing.

[0097] As described above, according to the radar device, radar control method, and radar control program of the second embodiment, the target velocity calculation unit 30 calculates and outputs a target velocity that maximizes the result of coherent addition after correcting the phase of the DFT calculated based on the central pulse-compressed signal, while correcting the Doppler frequency difference, which is proportional to the target velocity and proportional to the center frequency of the reference signal β used for pulse compression, for each type of pulse-compressed signal, for portions corresponding to the detected target distance x^. The target velocity correction unit 32 corrects the target velocity v-, for which velocity ambiguity has been resolved, to the velocity v~ before aliasing of the detected velocity that has been aliased within the observable velocity range. Therefore, it is possible to calculate the velocity of a target that exceeds the observable velocity range determined by the center frequency and repetition period of the transmission signal with a high probability. In particular, even when the assumed target velocity range is wide and the number of target velocity candidates increases, the radar device of the second embodiment can calculate the target velocity with a small amount of calculation.

[0098] Third Embodiment Next, a third embodiment will be described. The radar device of the third embodiment differs from the radar device of the second embodiment in that the pulse compressor 26 performs pulse compression using two signals in the modulation bands at both ends of the modulation band of the reference signal α as the reference signal β, and the target velocity calculator 30 iteratively calculates a correction frequency at which the DFT is maximized for each of the signals after pulse compression using the two reference signals β, and calculates the target velocity from the difference between the two correction frequencies. However, the signal generator 20, signal transmitter 22, signal receiver 24, target detector 28, and target velocity corrector 32 are the same as those of the second embodiment, and the radar device of the second embodiment includes the DSS 1, control device 2, mixer 3, high-frequency signal generator circuit 4, amplifier 5, duplexer 6, antenna 7, amplifier 8, mixer 9, amplifier 10, A / D converter 11, and signal processing circuit 12. Therefore, the same components as those of the second embodiment are denoted by the same reference symbols as those of the second embodiment, and detailed description thereof will be omitted.

[0099] As described above, the pulse compressor 26 performs pulse compression using two signals having modulation bandwidths at both ends of the modulation bandwidth of the reference signal α as the reference signal β. As will be described later, the modulation bandwidths of the two reference signals β are both in the range of 1 / 5 to 1 / 2 of the modulation bandwidth of the reference signal α, with the modulation bandwidth thought to be optimal being 1 / 3.

[0100] The target velocity calculation unit 30 is a unit that resolves velocity ambiguity in the velocity detected by the target detection unit 28, and is composed of the signal processing circuit 12. The signal processing circuit 12 iteratively calculates the correction frequency that maximizes the DFT for each of the portions of two types of multiple pulse-compressed signals that are individually pulse-compressed using two reference signals β, corresponding to the distance to the detected target, and calculates the target velocity from the difference between the two correction frequencies.

[0101] Hereinafter, a method will be described in which the target velocity calculation unit 30 calculates, by an iterative method, the correction frequency at which the DFT is maximized for each of the signals after pulse compression using the two reference signals β, and calculates the target velocity from the difference between the two correction frequencies.

[0102] The signal x after pulse compression using two reference signals β-m (m = 1, 2) ｋ，ｍ(t) target detection time t^ pulse compressed signal x ｋ，ｍ For each of (t^), the DFT is calculated so as to cancel the phase change due to the Doppler frequency, and the X shown in the following equation (17) is ｍ ω that maximizes (ω) ｍ - (hereinafter, in the formula, "-" is placed above "ω") is calculated.

[0103]

[0104] The target velocity v− after resolving the ambiguity is ω ｍ - is calculated using the following equation (18). ω is the angular frequency, and when divided by 2π, it becomes the frequency, so the target speed v- shown in equation (18) is calculated from the difference between the two correction frequencies. ｍ (m=1, 2) are the center frequencies of the reference signals β-m (m=1, 2), respectively.

[0105]

[0106] As in the second embodiment, the iterative method used in the third embodiment can utilize various methods for solving continuous optimization problems, including gradient methods such as the steepest descent method, the Newton method, and the interior point method. As an example, in the case of the Newton method, the target equation is set to the following equation (19), and the maximum value of X(ω) is calculated. The initial value is, for example, ω ｍ ^ can be used.

[0107]

[0108] Next, the performance when the modulation bandwidth of reference signal β is x times the modulation bandwidth of reference signal α will be explained using the performance ratio (dB) expressed as the product of the ratio of the frequency difference, which is an observable quantity, and the S / N ratio, with x = 1 / 2 as the reference.

[0109] The modulation bandwidth of reference signal α is set to 1. When the modulation bandwidth of reference signal β is set to x times the modulation bandwidth of reference signal α, the center frequency difference between the two signals of reference signal β is 1−x, as shown in the following equation (20).

[0110]

[0111] When the modulation bandwidth of reference signal β is x times the modulation bandwidth of reference signal α, the frequency difference changes by the value (dB) shown in the following equation (21) compared to when x=½.

[0112]

[0113] Furthermore, when the modulation bandwidth is increased by a factor of x, the S / N ratio of the signal after pulse compression changes by a value (dB) shown in the following equation (22).

[0114]

[0115] Combining the two changes indicated by equations (21) and (22), the performance ratio is expressed by the following equation (23).

[0116]

[0117] A graph of the performance ratio versus modulation bandwidth of reference signal β is shown in Figure 7. Compared to when x = 1 / 2, performance improves when the modulation bandwidth of reference signal β is between 1 / 5 (0.2) and 1 / 2 (0.5), and the peak position is particularly pronounced when the modulation bandwidth is 1 / 3 (0.33), where the slope is 0. Therefore, it is desirable to set the modulation bandwidth of reference signal β in the range of 1 / 5 to 1 / 2 of the modulation bandwidth of reference signal α, preferably 1 / 3.

[0118] To summarize the above, an example of the process for resolving velocity ambiguity in the radar device of the third embodiment is shown in Fig. 8. In step S500, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the received signal received by the signal receiver 24 using the reference signal α. As will be described later, in the third embodiment, the procedure of step S503 is performed in parallel with the procedure of step S500.

[0119] In step S501, the signal processing circuit 12, which is the target detection unit 28, performs Doppler FFT on the pulse-compressed signal.

[0120] In step S502, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target detection time t^ and the approximate velocity v^ at which velocity ambiguity may occur from the range-Doppler map.

[0121] In step S503, in parallel with the procedure in step S500, the signal processing circuit 12, which is the pulse compressor 26, performs pulse compression on the signal received by the signal receiver 24 using the reference signal β.

[0122] In step S504, the signal processing circuit 12, which is the target speed calculation unit 30, calculates the initial value ω ｍ Calculate the initial value ω ｍ As described in the second embodiment, ^ is the pulse-compressed signal x ｋ，ｍ is the peak angular frequency of the Doppler FFT of (t).

[0123] In step S505, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the pulse-compressed signal x by each reference signal β-m at the target detection time t^. ｋ，ｍ X, which is a DFT calculated to cancel the phase change due to the Doppler frequency for (t^) ｍ ω that maximizes (ω) ｍ -, and ω that satisfies the above equation (17) ｍ - is calculated using Newton's method.

[0124] In step S506, the signal processing circuit 12, which is the target velocity calculation unit 30, calculates the target velocity v- after resolving the ambiguity using the above-mentioned equation (18).

[0125] In step S507, the signal processing circuit 12, which is the target speed correction unit 32, corrects the target speed v- calculated by the target speed calculation unit 30 to the speed v~ before it is folded back to the approximate speed v^ detected by the target detection unit 28, which may be causing speed ambiguity, using the above-mentioned equation (15), and then ends the processing.

[0126] As described above, according to the radar device, radar control method, and radar control program of the third embodiment, the target velocity calculation unit 30 iteratively calculates the correction frequency at which the DFT is maximized for each of the portions of the two types of multiple pulse-compressed signals that have been individually pulse-compressed using the two reference signals β, which correspond to the distance x^ of the detected target, and calculates the target velocity from the difference between the two correction frequencies. This makes it possible to calculate with a high probability the velocity of a target that exceeds the observable velocity range determined by the center frequency and repetition period of the transmission signal. In particular, even when the expected target velocity range is wide and the number of target velocity candidates increases, the radar device 100 of the third embodiment can calculate the target velocity with a small amount of calculation, and further, since there are only two reference signals β, the amount of calculation required for pulse compression and the iterative method can be reduced.

[0127] Next, a fourth embodiment will be described. The radar device according to the fourth embodiment differs from the radar device according to the second embodiment in that the radar device according to the fourth embodiment further includes a target velocity group calculation unit 34. However, the signal generation unit 20, the signal transmission unit 22, the signal reception unit 24, the pulse compression unit 26, the target detection unit 28, the target velocity calculation unit 30, and the target velocity correction unit 32 are the same as those in the second embodiment, and the radar device according to the second embodiment also includes the DSS 1, the control unit 2, the mixer 3, the high-frequency signal generation circuit 4, the amplifier 5, the transmission / reception switch 6, the antenna 7, the amplifier 8, the mixer 9, the amplifier 10, the A / D converter 11, and the signal processing circuit 12. Therefore, the same components as those in the second embodiment are denoted by the same reference numerals as those in the second embodiment, and detailed description thereof will be omitted.

[0128] The target speed group calculation unit 34 is a part that calculates a target speed group that includes the target speed corrected by the target speed correction unit 32 (corrected target speed) and a speed obtained by adding or subtracting an integer multiple of twice the Nyquist speed within a predetermined range to or from the target speed, and calculates a plurality of target speeds together with probabilities that represent their likelihoods, and is composed of the signal processing circuit 12. The signal processing circuit 12 calculates the lower limit of the standard deviation of the target speed error from the S / N ratio of the pulse-compressed signal, and assumes that the distribution of the target speeds follows a normal distribution, calculates a probability that represents the likelihood of each target speed in the target speed group, and outputs the probability together with the target speed group.

[0129] Hereinafter, the target speed group calculation unit 34 sets the target speeds corrected by the target speed correction unit 32 and speeds obtained by adding or subtracting an integer multiple of twice the Nyquist speed within a predetermined range to or from the target speeds as a target speed group, and calculates a plurality of target speeds together with probabilities that represent their likelihoods. A method for doing this will be described.

[0130] First, when deriving the lower limit of the standard deviation of the error in the target speed calculated by the target speed calculation unit 30, the signal model for target speed estimation in the second embodiment is defined by the following equations (24) and (25).

[0131]

[0132]

[0133] x in formula (24) and formula (25) ｋ，ｍ ^ (hereinafter, in the formula, "^" is placed above "x") represents the pulse-compressed signal x by each reference signal β-m. ｋ，ｍ (t) target detection time t^ pulse compressed signal x ｋ，ｍ is a signal model representing (t^). ｋ，ｍ represents the pulse-compressed signal generated from the received signal, and x ｋ，ｍ ^ complex noise n ｋ，ｍ In the above model, x ｋ，ｍ The amplitude of ^ is fixed to 1, and the S / N ratio of the signal after pulse compression is reflected in σ according to equation (25). Also, φ in equation (24) is the initial phase.

[0134] Complex noise n(n ｋ，ｍ ) is expressed by the matrix shown in the following equation (26).

[0135]

[0136] The logarithmic likelihood function L of the above signal model is expressed by the following equation (27).

[0137]

[0138] In formula (27), n ＊ represents the conjugate transpose matrix of n, and vec(n) is an operator that rearranges the elements of a matrix into a column vector.

[0139] By differentiating the logarithmic likelihood function L shown in equation (27), the Fisher information I expressed by the following equation (28) is obtained.

[0140]

[0141] The lower bound of the standard deviation of the error in the target velocity is the (2,2) element of the inverse matrix of the Fisher information I (I －１ ) ２２ is expressed by the following equation (29) using

[0142]

[0143] Under conditions of an S / N ratio that allows the radar to detect targets while suppressing false detections, the distribution of detected target velocities can be considered to follow a normal distribution. Therefore, when the target velocities are defined by the following equation (30), the probability representing the likelihood of each target velocity can be calculated by the following equation (31).

[0144]

[0145]

[0146] As described above, according to the radar device, radar control method, and radar control program of the fourth embodiment, the target velocity group calculation unit 34 sets the target velocity group to the target velocity corrected by the target velocity correction unit 32 and a velocity obtained by adding or subtracting an integer multiple of twice the Nyquist velocity within a predetermined range to or from the target velocity, and calculates multiple target velocities together with probabilities representing their likelihoods. This makes it possible to quantitatively determine the reliability of the calculated target velocities. Furthermore, it is possible to improve the reliability of the start and maintenance of tracking processing, which is usually performed in a later stage of the radar device.

[0147] Even if the target speed group calculation unit 34 is added to the first embodiment, it can be configured to calculate a plurality of target speeds together with probabilities that indicate the likelihood of each speed, as in the fourth embodiment.

[0148] Fifth Embodiment Next, a fifth embodiment will be described. The radar device according to the fifth embodiment differs from the radar device according to the third embodiment in that the radar device according to the fifth embodiment further includes a target velocity group calculation unit 34. However, the signal generation unit 20, the signal transmission unit 22, the signal reception unit 24, the pulse compression unit 26, the target detection unit 28, the target velocity calculation unit 30, and the target velocity correction unit 32 are the same as those in the third embodiment, and the radar device according to the third embodiment also includes the DSS 1, the control unit 2, the mixer 3, the high-frequency signal generation circuit 4, the amplifier 5, the transmission / reception switch 6, the antenna 7, the amplifier 8, the mixer 9, the amplifier 10, the A / D converter 11, and the signal processing circuit 12. Therefore, the same components as those in the third embodiment are denoted by the same reference numerals as those in the third embodiment, and detailed description thereof will be omitted.

[0149] The target speed group calculation unit is a part that calculates a target speed group consisting of the target speed corrected by the target speed correction unit and a speed obtained by adding or subtracting an integer multiple of twice the Nyquist speed within a predetermined range to or from that speed, and calculates a plurality of target speeds together with probabilities that represent their likelihoods, and is composed of a signal processing circuit 12. The signal processing circuit 12 calculates the lower limit of the standard deviation of the target speed error from the S / N ratio of the pulse-compressed signal, and assumes that the distribution of the target speeds follows a normal distribution, calculates a probability that represents the likelihood of each target speed in the target speed group, and outputs it together with the target speed group.

[0150] Hereinafter, the target speed group calculation unit 34, as in the fourth embodiment, sets the target speed corrected by the target speed correction unit 32 and a speed obtained by adding or subtracting an integer multiple of twice the Nyquist speed within a predetermined range to or from the target speed as a target speed group, and calculates a plurality of target speeds together with probabilities that represent their likelihoods. A method for doing this will be described.

[0151] First, when deriving the lower limit of the standard deviation of the error in the target speed calculated by the target speed calculation unit 30, the signal model for target speed estimation in the third embodiment is defined by the following equations (32) and (33).

[0152]

[0153]

[0154] x in formula (32) and formula (33)ｋ ^ (hereinafter, in the formula, "^" is placed above "x") represents the pulse-compressed signal x by each reference signal β-m. ｋ，ｍ (t) target detection time t^ pulse compressed signal x ｋ，ｍ is a signal model representing (t̂). ｋ represents the pulse-compressed signal generated from the received signal, and x ｋ ^ complex noise n ｋ In the above model, x ｋ The amplitude of ^ is fixed to 1, and the S / N ratio of the pulse-compressed signal is reflected in σ according to equation (33). Also, φ in equation (32) is the initial phase.

[0155] Complex noise n(n ｋ ) is calculated using the vector shown in the following equation (34).

[0156]

[0157] The logarithmic likelihood function L of the above signal model is expressed by the following equation (35): ＊ is the conjugate transpose of n.

[0158]

[0159] By differentiating the logarithmic likelihood function L shown in equation (35), the Fisher information I expressed by the following equation (36) is obtained.

[0160]

[0161] The lower bound of the standard deviation of the error in the target velocity is the (1,1) element of the inverse matrix of the Fisher information I (I －１ ) １１ In equation (37), f ｍ (m=1, 2) are the center frequencies of the reference signals β-m (m=1, 2), respectively, and T ＰＲＩ indicates the repetition period of the transmission signal.

[0162]

[0163] Under conditions of an S / N ratio that allows the radar to detect targets while suppressing false detections, the distribution of detected target velocities can be considered to follow a normal distribution. Therefore, when the target velocities are defined by the following equation (38) as in the fourth embodiment, the probability representing the likelihood of each target velocity can be calculated by the following equation (39) as in the fourth embodiment.

[0164]

[0165]

[0166] As described above, according to the radar device, radar control method, and radar control program of the fifth embodiment, the target velocity group calculation unit 34 sets the target velocity group to the target velocity corrected by the target velocity correction unit 32 and a velocity obtained by adding or subtracting an integer multiple of twice the Nyquist velocity within a predetermined range to or from the target velocity, and calculates a plurality of target velocities together with probabilities representing their likelihoods. This makes it possible to quantitatively determine the reliability of the calculated target velocities. Furthermore, it is possible to improve the reliability of the start and maintenance of tracking processing, which is usually performed in a later stage of the radar device.

[0167] 1 DSS (Digital Direct Synthesis Oscillator), 2 Control device, 3, 9 Mixer, 4 High frequency signal generating circuit, 5, 8, 10 Amplifier, 6 Transmission / reception switch, 7 Antenna, 11 A / D converter, 12 Signal processing circuit, 20 Signal generating unit, 22 Signal transmitting unit, 24 Signal receiving unit, 26 Pulse compressing unit, 28 Target detecting unit, 30 Target velocity calculating unit, 32 Target velocity correcting unit, 34 Target velocity group calculating unit, 100 Radar device.

Claims

1. A pulse compression unit that pulse-compresses each of a plurality of received signals, which are the result of receiving each of a plurality of reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at the target, using a first reference signal with the same modulation bandwidth as the chirp signal, and individually pulse-compresses each of the plurality of received signals using a plurality of types of second reference signals, each of which has a different modulation bandwidth and center frequency from the chirp signal. A target detection unit that detects the distance to the target and the approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A target speed calculation unit calculates the target speed, which is the speed of the target, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals related to each of the plurality of received signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal, based on the distance to the target detected by the target detection unit and the approximate speed. A signal generation unit that generates the chirp signal, A signal transmission unit that repeatedly transmits the aforementioned chirp signal, A signal receiving unit that receives multiple reflected signals, each of which is reflected at the target, from the chirp signal repeatedly transmitted by the signal transmitting unit. Equipped with, The target speed calculation unit calculates a discrete Fourier transform for each of the second pulse-compressed signals, corresponding to the distance of the target, correcting for the difference in Doppler frequencies that are proportional to the center frequency of the second reference signal used for pulse compression and proportional to the target speed, for each type of second reference signal, and calculates the target speed at which the amplitude is maximum in the calculation result of the discrete Fourier transform. The target speed calculation unit uses the estimated speed and speeds obtained by adding or subtracting an integer multiple of twice the Nyquist speed within a predetermined range to the estimated speed as target speed candidates. For the portion of the second pulse-compressed signal corresponding to the target distance, it calculates a discrete Fourier transform based on the center frequency of the second reference signal and corrects for the difference in the Doppler frequency proportional to the target speed candidate, for each type of second reference signal. The target speed candidate with the maximum amplitude in the calculation result of the discrete Fourier transform is calculated as the target speed. A radar device characterized by the following features.

2. The target speed calculation unit calculates a target speed candidate that maximizes the sum of the calculation results of the discrete Fourier transform for each type of second reference signal. The radar device according to feature 1.

3. The target speed calculation unit calculates the discrete Fourier transform of the second pulse-compressed signal based on the second pulse-compressed signal of the received signal related to the chirp signal which is in the middle of the transmission order, corrects the phase of the discrete Fourier transform for each type of second reference signal, and then coherently adds the calculated discrete Fourier transforms for each type of second reference signal. The radar device according to feature 2.

4. The target speed calculation unit adds the absolute values ​​of the calculation results of the discrete Fourier transform for each type of second reference signal. The radar device according to feature 2.

5. A pulse compression unit that pulse-compresses each of a plurality of received signals, which are the result of receiving each of a plurality of reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at a target, with a first reference signal having the same modulation bandwidth as the chirp signal, and individually pulse-compresses each of the plurality of received signals with a plurality of types of second reference signals, each of which has a different modulation bandwidth and center frequency from the chirp signal, A target detection unit that detects the distance to the target and the approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A target speed calculation unit calculates the target speed, which is the speed of the target, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals related to each of the plurality of received signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal, based on the distance to the target detected by the target detection unit and the approximate speed. Equipped with, The target speed calculation unit calculates a discrete Fourier transform by correcting the difference in Doppler frequencies proportional to the center frequency of the second reference signal for each type of second reference signal, for the portion of the second pulse-compressed signal corresponding to the distance of the target detected by the target detection unit, and iteratively calculates the correction frequency that maximizes the calculation result of the discrete Fourier transform, and calculates the target speed from the correction frequency. The system includes a target speed correction unit that corrects the target speed calculated by the target speed calculation unit to the speed before it is converted back to the approximate speed detected by the target detection unit. A radar device characterized by the following features.

6. A target speed correction unit that corrects the target speed calculated by the target speed calculation unit to the speed before it is converted back to the approximate speed detected by the target detection unit, The target speed group calculation unit calculates a group of target speeds, which are the corrected target speed corrected by the target speed correction unit and the speed obtained by adding or subtracting an integer multiple of twice the Nyquist speed to the corrected target speed, along with the probability representing the likelihood of each target speed. Equipped with The radar device according to feature 5.

7. A pulse compression unit that pulse-compresses each of a plurality of received signals, which are the result of receiving each of a plurality of reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at a target, with a first reference signal having the same modulation bandwidth as the chirp signal, and individually pulse-compresses each of the plurality of received signals with a plurality of types of second reference signals, each of which has a different modulation bandwidth and center frequency from the chirp signal, A target detection unit that detects the distance to the target and the approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A target speed calculation unit calculates the target speed, which is the speed of the target, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals related to each of the plurality of received signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal, based on the distance to the target detected by the target detection unit and the approximate speed. Equipped with, The target speed calculation unit calculates a discrete Fourier transform (DTF) on the portion of the second pulse-compressed signal corresponding to the distance of the target detected by the target detection unit, correcting for the difference in Doppler frequencies proportional to the center frequency of the second reference signal for each type of second reference signal. The unit then iteratively calculates the correction frequency that maximizes the sum of the results of the discrete Fourier transforms for each type of second reference signal, and calculates the target speed from the correction frequency. A radar device characterized by the following features.

8. The target speed calculation unit calculates the discrete Fourier transform of the second pulse-compressed signal based on the second pulse-compressed signal of the received signal related to the chirp signal which is in the middle of the transmission order, corrects the phase of the discrete Fourier transform for each type of second reference signal, and then coherently adds the calculated discrete Fourier transforms for each type of second reference signal. The radar device according to feature 7.

9. The target speed calculation unit adds the absolute values ​​of the calculation results of the discrete Fourier transform for each type of second reference signal. The radar device according to feature 7.

10. A pulse compression unit that pulse-compresses each of a plurality of received signals, which are the result of receiving each of a plurality of reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at a target, with a first reference signal having the same modulation bandwidth as the chirp signal, and individually pulse-compresses each of the plurality of received signals with a plurality of types of second reference signals, each of which has a different modulation bandwidth and center frequency from the chirp signal, A target detection unit that detects the distance to the target and the approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A target speed calculation unit calculates the target speed, which is the speed of the target, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals related to each of the plurality of received signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal, based on the distance to the target detected by the target detection unit and the approximate speed. Equipped with, The pulse compression unit performs pulse compression using two signals, one at each end of the modulation band of the first reference signal, as the second reference signal. The target speed calculation unit iteratively calculates the correction frequency that maximizes the discrete Fourier transform for each of the second pulse-compressed signals using the two second reference signals, and calculates the target speed from the difference between the two calculated correction frequencies. A radar device characterized by the following features.

11. The target speed calculation unit determines that the modulation bandwidth of each of the two second reference signals is 1 / 5 to 1 / 2, preferably 1 / 3, of the modulation bandwidth of the first reference signal. The radar device according to feature 10.

12. The system includes a target speed correction unit that corrects the target speed calculated by the target speed calculation unit to the speed before it is converted back to the approximate speed detected by the target detection unit. A radar device according to any one of claims 7 to 11, characterized by the following:

13. A target speed group calculation unit that calculates a group of target speeds, which are the corrected target speed corrected by the target speed correction unit and the speed obtained by adding or subtracting an integer multiple of twice the Nyquist speed to the corrected target speed, along with a probability representing the likelihood of each target speed. The radar device according to feature 12.

14. A radar control method performed by a computer, The steps include: pulse-compressing each of the multiple received signals, which are the result of receiving each of the multiple reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at the target, with a first reference signal having the same modulation bandwidth as the chirp signal; and individually pulse-compressing each of the multiple received signals with a second reference signal of a plurality of types, each of which has a different modulation bandwidth and center frequency from the chirp signal; The steps include detecting the target distance and approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A step of calculating the target speed, which is the speed of the target, based on the detected distance to the target and the estimated speed, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal. It has, In the pulse compression step, pulse compression is performed using two signals, one at each end of the modulation band of the first reference signal, as the second reference signal. In the step of calculating the target speed, the correction frequency at which the discrete Fourier transform is maximized for each of the second pulse-compressed signals using the two second reference signals is calculated using an iterative method, and the target speed is calculated from the difference between the two calculated correction frequencies. A radar control method characterized by the following features.

15. The steps include: pulse-compressing each of the multiple received signals, which are the result of receiving each of the multiple reflected signals obtained when each of the repeatedly transmitted chirp signals is reflected at the target, with a first reference signal having the same modulation bandwidth as the chirp signal; and individually pulse-compressing each of the multiple received signals with a second reference signal of a plurality of types, each of which has a different modulation bandwidth and center frequency from the chirp signal; The steps include detecting the target distance and approximate speed from a plurality of first pulse-compressed signals obtained by pulse compression using the first reference signal, A step of calculating the target speed, which is the speed of the target, based on the detected distance to the target and the estimated speed, from a second pulse-compressed signal, which is composed of a plurality of pulse-compressed signals for each of the plurality of received signals, obtained by individually pulse-compressing each of the plurality of received signals with the plurality of types of second reference signals for each type of second reference signal. A radar control program that causes a computer to execute, In the pulse compression step, pulse compression is performed using two signals, one at each end of the modulation band of the first reference signal, as the second reference signal. In the step of calculating the target speed, the correction frequency at which the discrete Fourier transform is maximized for each of the second pulse-compressed signals using the two second reference signals is calculated using an iterative method, and the target speed is calculated from the difference between the two calculated correction frequencies. A radar control program characterized by the following features.