Signal parameter estimation device and signal parameter estimation method

Abstract

This signal parameter estimation device is configured to comprise: a candidate value setting unit (1) that sets a plurality of candidate values for a chirp rate representing the ratio of chirp frequency width to chirp time width of a reception signal, which is an observation signal of a chirp signal that is the target for parameter estimation; and a candidate value selection unit (2) that multiplies the reception signal by opposite-phase components of each of the candidate values set by the candidate value setting unit (1), converts the reception signals after the opposite-phase component multiplication to signals of a frequency region, and selects, from among the plurality of candidate values, the candidate value for which an electrical power peak value of the signal of the frequency region is relatively larger. Moreover, the signal parameter estimation device comprises: a frequency width estimation unit (3) that multiplies the reception signal, which is the observation signal, by the opposite-phase component of the candidate value selected by the candidate value selection unit (2), converts the reception signal after the opposite phase component multiplication to a signal of the frequency region, and, on the basis of frequency intervals included in the post-conversion signal of the frequency region which correspond to a plurality of electrical power peak values, estimates the chirp frequency width; and a time width estimation unit (4) that estimates the chirp time width from the chirp frequency width estimated by the frequency width estimation unit (3) and the candidate value selected by the candidate value selection unit (2).

Description

Signal parameter estimation device and signal parameter estimation method

[0001] The present disclosure relates to a signal parameter estimation device and a signal parameter estimation method.

[0002] Conventionally, there has been a signal parameter estimation device (hereinafter referred to as a "conventional signal parameter estimation device") that converts a chirp signal into a frequency domain signal and estimates the signal parameters of the chirp signal based on the frequency domain signal. Patent Document 1 discloses a technique for estimating the parameters of a chirp signal using a multi-component chirp parameter estimation algorithm.

[0003] Special table 2021-515212 publication

[0004] Conventional signal parameter estimation devices have a problem in that, when the power of a chirp signal is low, equal to or less than the noise power, it is difficult to distinguish between the peak power of the chirp signal and the noise power, and therefore it may be impossible to estimate the signal parameters of the chirp signal based on the frequency domain signal. Even if it is assumed that the multi-component chirp parameter estimation algorithm disclosed in Patent Document 1 can be applied to conventional signal parameter estimation devices, it is difficult to estimate the signal parameters of the chirp signal when the power of the chirp signal is low, equal to or less than the noise power.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a signal parameter estimation device that can estimate the signal parameters of a chirp signal even if the power of the chirp signal is low, equal to or less than the noise power.

[0006] A signal parameter estimation device according to the present disclosure includes a candidate value setting unit that sets multiple candidate values ​​for a chirp rate, which indicates the ratio of a chirp frequency width to a chirp time width of a received signal, which is an observation signal of a chirp signal to be estimated, a candidate value selection unit that multiplies the received signal by an anti-phase component of each candidate value set by the candidate value setting unit, converts the received signal after multiplication with the anti-phase component into a frequency-domain signal, and selects from the multiple candidate values ​​a candidate value that relatively increases the power peak value of the frequency-domain signal. The signal parameter estimation device also includes a frequency width estimation unit that multiplies the received signal, which is an observation signal, by the anti-phase component of the candidate value selected by the candidate value selection unit, converts the received signal after multiplication with the anti-phase component into a frequency-domain signal, and estimates a chirp frequency width based on the intervals of frequencies corresponding to the multiple power peak values ​​included in the converted frequency-domain signal, and a time width estimation unit that estimates a chirp time width from the chirp frequency width estimated by the frequency width estimation unit and the candidate value selected by the candidate value selection unit.

[0007] According to the present disclosure, it is possible to estimate the signal parameters of a chirp signal even if the power of the chirp signal is low, equal to or less than the noise power.

[0008] FIG. 1 is a configuration diagram showing a signal parameter estimation device according to a first embodiment; FIG. 2 is a hardware configuration diagram showing hardware of the signal parameter estimation device according to the first embodiment; FIG. 3 is a hardware configuration diagram of a computer when the signal parameter estimation device is realized by software, firmware, or the like; and FIG. 4 is a flowchart showing a signal parameter estimation method, which is a processing procedure of the signal parameter estimation device. ｋ ) is an explanatory diagram showing an example of a chirp signal x(t ｋ ) and replica signal ref(t ｋ ) and | mod(t ). FIG. 10 is an explanatory diagram showing an example of a frequency domain signal including two power peak values. ａ , t ｗ ) | < t ｗ / 2, the replica signal ref(t ｋ ) and chirp signal x(t ｋ ) is an explanatory diagram showing an example of | mod (t ａ , t ｗ ) | ≧ t ｗ / 2, the replica signal ref(t ｋ ) and chirp signal x(t ｋ ) is an explanatory diagram showing an example of a first power peak value Peak １ and the second power peak value Peak ２ Fig. 10 is an explanatory diagram showing an example of the signal parameter estimation device according to the third embodiment. Fig. 11 is a hardware configuration diagram showing hardware of the signal parameter estimation device according to the third embodiment.

[0009] In order to explain the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0010] Embodiment 1 Fig. 1 is a configuration diagram showing a signal parameter estimation device according to embodiment 1. Fig. 2 is a hardware configuration diagram showing hardware of the signal parameter estimation device according to embodiment 1. The signal parameter estimation device shown in Fig. 1 includes a candidate value setting unit 1, a candidate value selection unit 2, a frequency width estimation unit 3, and a time width estimation unit 4.

[0011] The candidate value setting unit 1 is realized by, for example, the candidate value setting circuit 11 shown in Fig. 2. The candidate value setting unit 1 sets multiple candidate values ​​for the chirp rate, which indicates the ratio of the chirp frequency width to the chirp time width of the received signal, which is the observed signal of the chirp signal whose specifications are to be estimated. The candidate value setting unit 1 outputs the multiple candidate values ​​to the candidate value selection unit 2.

[0012] The candidate value selection unit 2 is realized, for example, by the candidate value selection circuit 12 shown in FIG. 2 . The candidate value selection unit 2 acquires a received signal, which is an observation signal of a chirp signal whose parameters are to be estimated, and acquires multiple candidate values ​​from the candidate value setting unit 1. The candidate value selection unit 2 multiplies the received signal by the negative-phase component of each candidate value and converts the received signal after multiplication with the negative-phase component into a frequency domain signal. The candidate value selection unit 2 selects, from the multiple candidate values, a candidate value that produces a relatively large power peak value for the frequency domain signal. The candidate value selection unit 2 outputs the selected candidate value to each of the frequency width estimation unit 3 and the time width estimation unit 4.

[0013] The frequency width estimator 3 is realized by, for example, the frequency width estimation circuit 13 shown in FIG. 2 . The frequency width estimator 3 acquires a received signal, which is an observation signal, and acquires a selected candidate value from the candidate value selector 2. The frequency width estimator 3 multiplies the received signal by the anti-phase component of the candidate value selected by the candidate value selector 2, and converts the received signal after multiplication with the anti-phase component into a frequency domain signal. The frequency width estimator 3 calculates multiple power peak values ​​included in the converted frequency domain signal. The frequency width estimator 3 estimates a chirp frequency width based on the intervals between frequencies corresponding to the multiple power peak values. The frequency width estimator 3 outputs information indicating the chirp frequency width to the time width estimator 4.

[0014] The time width estimator 4 is realized by, for example, the time width estimator circuit 14 shown in Fig. 2. The time width estimator 4 acquires the selected candidate value from the candidate value selector 2 and acquires information indicating the chirp frequency width from the frequency width estimator 3. The time width estimator 4 estimates the chirp time width from the chirp frequency width and the candidate value selected by the candidate value selector 2.

[0015] 1, it is assumed that each of the components of the signal parameter estimation device, that is, a candidate value setting unit 1, a candidate value selection unit 2, a frequency width estimation unit 3, and a time width estimation unit 4, is realized by dedicated hardware as shown in Fig. 2. That is, it is assumed that the signal parameter estimation device is realized by a candidate value setting circuit 11, a candidate value selection circuit 12, a frequency width estimation circuit 13, and a time width estimation circuit 14. Each of the candidate value setting circuit 11, the candidate value selection circuit 12, the frequency width estimation circuit 13, and the time width estimation circuit 14 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 thereof.

[0016] The components of the signal parameter estimation device are not limited to those realized by dedicated hardware, and the signal parameter estimation device may be realized by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the memory of a computer. The computer refers to hardware that executes a program, and includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

[0017] 3 is a hardware configuration diagram of a computer when the signal parameter estimation device is realized by software, firmware, etc. When the signal parameter estimation device is realized by software, firmware, etc., a program for causing a computer to execute the respective processing procedures of the candidate value setting unit 1, the candidate value selecting unit 2, the frequency width estimation unit 3, and the time width estimation unit 4 is stored in a memory 21. A processor 22 of the computer then executes the program stored in the memory 21.

[0018] 2 shows an example in which each of the components of the signal parameter estimation device is realized by dedicated hardware, while Fig. 3 shows an example in which the signal parameter estimation device is realized by software, firmware, etc. However, this is merely an example, and some of the components in the signal parameter estimation device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0019] Next, the operation of the signal parameter estimation device shown in Fig. 1 will be described. The signal parameter estimation device shown in Fig. 1 estimates, for example, an unknown chirp signal x(t ｋ 4 is a flowchart showing a signal parameter estimation method, which is a processing procedure of the signal parameter estimation device.

[0020] The candidate value setting unit 1 calculates a chirp signal x(t ｋ ) is the observed signal of the received signal x ｏｂｓ (t) chirp time width t ｗ Chirp frequency width f ｗ A plurality of candidate values ​​of the chirp rate α, which indicates the ratio of １ , ..., α Ｍ M is an integer equal to or greater than 2. １ , ..., α Ｍ are different values. The candidate value setting unit 1 may use any method for setting the plurality of candidate values. For example, a method may be used in which a plurality of chirp rates within a range of chirp rates is randomly set as candidate values. ｋ ) is observed at time t ｋ is the observed signal, and the received signal x ｏｂｓ (t) is the observation time t ｋ The candidate value setting unit 1 sets a plurality of candidate values ​​α １ , ..., α Ｍ is output to the candidate value selection unit 2.

[0021] FIG. 5 shows the unknown chirp signal x(tｋ 5 is an explanatory diagram showing an example of a time domain signal (t) in a time domain (t). In FIG. 5, the horizontal axis represents time [sec] and the vertical axis represents frequency [Hz]. ｗ is the chirp signal x(t ｋ ) chirp time width, f ｗ is the chirp signal x(t ｋ ) chirp frequency width, t ａ is the chirp signal x(t ｋ ) start time, f ｃ is the chirp signal x(t ｋ ) center frequency, T ｏｂｓ is the chirp signal x(t ｋ ) observation time length, T ＦＭＣＷ is the chirp signal x(t ｋ ) is the length of time that exists.

[0022] The candidate value selector 2 selects the received signal x ｏｂｓ (t), and a plurality of candidate values ​​α １ , ..., α Ｍ The candidate value selection unit 2 obtains the candidate value α as shown on the right side of the following equation (1). ｍ (m=1, ..., M) ｏｂｓ (t), and the received signal after multiplication with the antiphase-phase component is converted into a frequency domain signal by, for example, FFT (Fast Fourier Transform) (step ST2 in FIG. 4).

[0023] The candidate value selection unit 2 selects the power peak value peak(α) of the converted frequency domain signal as shown in the following equation (1): ｍ ) is calculated (step ST3 in FIG. 4). In equation (3), ｋ The received signal x ｏｂｓ Since the entire (t) is subjected to FFT, the subscript k of t is omitted. ｍ By repeating the calculation process of M times, M peak power values ​​peak(α １ ) ~ peak (α Ｍ ) is calculated.

[0024]

[0025] The candidate value selection unit 2 selects M candidate values ​​α １ , ..., α Ｍ Among these, the candidate value α that makes the power peak value of the frequency domain signal relatively large is selected. ｏｐｔ (Step ST4 in FIG. 4). Specifically, the candidate value selection unit 2 selects M peak power values ​​peak(α １ ) ~ peak (α Ｍ ) the candidate value α corresponding to the maximum power peak value ｍ α ｏｐｔ Here, the candidate value selection unit 2 searches for a candidate value α ｏｐｔ The candidate value α corresponding to the maximum power peak value is ｍ However, this is only an example, and the candidate value selection unit 2 may select a candidate value α that makes the power peak value relatively large within a practically acceptable range. ｏｐｔ The candidate value α corresponding to the second largest power peak value is ｍ The candidate value selection unit 2 may search for the candidate value α ｏｐｔ to the frequency width estimation unit 3 and the time width estimation unit 4, respectively.

[0026]

[0027] The frequency width estimation unit 3 estimates the received signal x ｏｂｓ (t), and the candidate value selection unit 2 selects the candidate value α ｏｐｔ The frequency width estimation unit 3 obtains the candidate value α as shown in the following equation (3). ｏｐｔ The reverse phase component of the received signal x ｏｂｓ (t) and converts the received signal after multiplication with the antiphase component into a frequency domain signal z(f).

[0028]

[0029] As shown in Fig. 6, the converted frequency domain signal z(f) contains multiple power peak values. Fig. 6 is an explanatory diagram showing multiple power peak values ​​contained in the converted frequency domain signal z(f). In Fig. 6, the horizontal axis represents frequency [Hz] and the vertical axis represents power [dB]. In the example of Fig. 6, three power peak values ​​are contained, and the frequencies corresponding to the three power peak values ​​are f Ｌ , f Ｍ , f Ｈ The frequency f Ｌ and frequency f Ｍ The distance between ｗ and the frequency f Ｍ and frequency f Ｈ The distance between ｗ The frequency width estimation unit 3 estimates the chirp frequency width f ｗ The frequencies f corresponding to the three power peak values ​​are Ｌ , f Ｍ , f Ｈ The frequency interval f ｗ (Step ST5 in FIG. 4). The frequency width estimation unit 3 estimates the chirp frequency width f ｗ The time width estimation unit 4 outputs information indicating the above.

[0030] The time width estimation unit 4 receives the candidate value α from the candidate value selection unit 2. ｏｐｔ and obtains the chirp frequency width f ｗ The time width estimation unit 4 obtains information indicating the chirp frequency width f ｗ and candidate value α ｏｐｔ From this, the chirp time width t ｗ (Step ST6 in FIG. 4). The time width estimation unit 4 estimates the chirp frequency width f ｗ and chirp time width t ｗ The chirp frequency width f is displayed on a display device (not shown), for example. ｗ and chirp time width t ｗ Each of the chirp signals x(t ｋ ) signal specifications.

[0031]

[0032] In the first embodiment described above, the signal parameter estimation device is configured to include a candidate value setting unit 1 that sets multiple candidate values ​​for the chirp rate, which indicates the ratio of the chirp frequency width to the chirp time width of a received signal, which is an observation signal of a chirp signal to be estimated, and a candidate value selection unit 2 that multiplies the received signal by the negative-phase-sequence component of each candidate value set by the candidate value setting unit 1, converts the received signal after multiplication with the negative-phase-sequence component into a frequency-domain signal, and selects from the multiple candidate values ​​a candidate value that relatively increases the power peak value of the frequency-domain signal. The signal parameter estimation device also includes a frequency width estimation unit 3 that multiplies the received signal, which is an observation signal, by the negative-phase-sequence component of the candidate value selected by the candidate value selection unit 2, converts the received signal after multiplication with the negative-phase-sequence component into a frequency-domain signal, and estimates the chirp frequency width based on the interval between frequencies corresponding to the multiple power peak values ​​included in the converted frequency-domain signal, and a time width estimation unit 4 that estimates the chirp time width from the chirp frequency width estimated by the frequency width estimation unit 3 and the candidate value selected by the candidate value selection unit 2. Therefore, the signal parameter estimation device can estimate the signal parameters of a chirp signal even if the power of the chirp signal is low, equal to or lower than the noise power. In the first embodiment, the signal parameter estimation device estimates the chirp parameters, multiplies the signal by its inverse phase, and then converts the signal into a frequency domain signal, which makes it easier for the signal power to build up. Multiplying the inverse phase is equivalent to canceling the chirp component of the received signal. As a result, the signal-to-noise gain (SNR) is improved, making it possible to detect the signal parameters even for low-SNR signals.

[0033] Second Embodiment In a second embodiment, a signal parameter estimation device including a start time estimation unit 9 that estimates the start time of a chirp signal and a center frequency identification unit 10 that estimates the center frequency of a chirp signal will be described.

[0034] FIG. 7 is a configuration diagram showing a signal parameter estimation device according to embodiment 2. In FIG. 7, the same reference numerals as in FIG. 1 indicate the same or corresponding parts, and detailed description thereof will be omitted. FIG. 8 is a hardware configuration diagram showing hardware of the signal parameter estimation device according to embodiment 2. In FIG. 8, the same reference numerals as in FIG. 2 indicate the same or corresponding parts, and detailed description thereof will be omitted. The signal parameter estimation device shown in FIG. 7 includes a candidate value setting unit 1, a candidate value selecting unit 2, a frequency width estimating unit 3, a time width estimating unit 4, a pulse number estimating unit 5, a replica signal creating unit 6, a signal converting unit 7, a time offset calculating unit 8, a start time estimating unit 9, and a center frequency identifying unit 10.

[0035] The pulse number estimation unit 5 is realized by, for example, a pulse number estimation circuit 15 shown in Fig. 8. The pulse number estimation unit 5 estimates the number of pulses, which is the number of chirp signals, from the number of power peak values ​​included in the frequency domain signal after conversion by the frequency width estimation unit 3. The pulse number estimation unit 5 outputs the estimated number of pulses to the replica signal creation unit 6.

[0036] The replica signal generator 6 is realized by, for example, the replica signal generator circuit 16 shown in Fig. 8. The replica signal generator 6 acquires information indicating the chirp frequency width from the frequency width estimator 3, acquires information indicating the chirp time width from the time width estimator 4, and acquires the number of pulses from the pulse number estimator 5. The replica signal generator 6 generates a replica signal of the chirp signal using the chirp frequency width, the chirp time width, and the number of pulses. The replica signal generator 6 outputs the replica signal to each of the signal converter 7 and the start time estimator 9.

[0037] The signal converter 7 is realized by, for example, the signal converter circuit 17 shown in Fig. 8 . The signal converter 7 acquires a received signal, which is an observation signal of the chirp signal whose parameters are to be estimated, and acquires a replica signal from the replica signal generator 6. The signal converter 7 multiplies the replica signal by the received signal and converts the signal resulting from the multiplication of the replica signal by the received signal into a frequency domain signal. The signal converter 7 outputs the converted frequency domain signal to each of the time offset calculator 8 and the center frequency identifyr 10.

[0038] The time offset calculation unit 8 is realized by, for example, the time offset calculation circuit 18 shown in FIG. 8 . The time offset calculation unit 8 acquires the converted frequency domain signal from the signal conversion unit 7. The time offset calculation unit 8 calculates two power peak values ​​contained in the frequency domain signal converted by the signal conversion unit 7. The time offset calculation unit 8 calculates a first time offset between the replica signal and the chirp signal and a second time offset between the replica signal and the chirp signal based on the two power peak values. The time offset calculation unit 8 outputs information indicating the first time offset and information indicating the second time offset to the start time estimation unit 9.

[0039] The start time estimation unit 9 is realized by, for example, the start time estimation circuit 19 shown in Fig. 8. The start time estimation unit 9 includes a replica signal correction unit 9a, a first peak value calculation unit 9b, a second peak value calculation unit 9c, and a start time estimation processing unit 9d. The start time estimation unit 9 acquires information indicating the first time shift and information indicating the second time shift from the time shift calculation unit 8. The start time estimation unit 9 estimates the start time of the chirp signal using the first time shift and the second time shift.

[0040] The replica signal corrector 9a acquires the replica signal from the replica signal generator 6 and acquires information indicating the first time shift and information indicating the second time shift from the time shift calculator 8. The replica signal corrector 9a corrects the replica signal using the first time shift and outputs the replica signal corrected using the first time shift as a first replica signal to the first peak value calculator 9b. The replica signal corrector 9a corrects the replica signal using the second time shift and outputs the replica signal corrected using the second time shift as a second replica signal to the second peak value calculator 9c.

[0041] The first peak value calculation unit 9b acquires a received signal, which is an observation signal of the chirp signal whose parameters are to be estimated, and acquires a first replica signal from the replica signal correction unit 9a. The first peak value calculation unit 9b multiplies the first replica signal by the received signal. The first peak value calculation unit 9b converts the signal obtained by multiplying the first replica signal by the received signal into a frequency domain signal. The first peak value calculation unit 9b calculates a first power peak value included in the frequency domain signal. The first peak value calculation unit 9b outputs the first power peak value to the start time estimation processing unit 9d.

[0042] The second peak value calculation unit 9c acquires the received signal and acquires the second replica signal from the replica signal correction unit 9a. The second peak value calculation unit 9c multiplies the second replica signal by the received signal. The second peak value calculation unit 9c converts the signal obtained by multiplying the second replica signal by the received signal into a frequency domain signal. The second peak value calculation unit 9c calculates a second power peak value included in the frequency domain signal. The second peak value calculation unit 9c outputs the second power peak value to the start time estimation processing unit 9d.

[0043] The start time estimation processing unit 9d acquires information indicating the first time offset and information indicating the second time offset from the time offset calculation unit 8. The start time estimation processing unit 9d acquires the first power peak value from the first peak value calculation unit 9b and the second power peak value from the second peak value calculation unit 9c. If the first power peak value is equal to or greater than the second power peak value, the start time estimation processing unit 9d estimates the start time of the chirp signal based on the first time offset. If the first power peak value is less than the second power peak value, the start time estimation processing unit 9d estimates the start time of the chirp signal based on the second time offset.

[0044] The center frequency identifying unit 10 is realized by, for example, the center frequency identifying circuit 20 shown in FIG. 8 . The center frequency identifying unit 10 acquires the converted frequency domain signal from the signal converting unit 7. The center frequency identifying unit 10 acquires the first power peak value from the first peak value calculating unit 9 b and the second power peak value from the second peak value calculating unit 9 c. If the first power peak value is equal to or greater than the second power peak value, the center frequency identifying unit 10 identifies the frequency corresponding to the first power peak value in the converted frequency domain signal as the center frequency of the chirp signal. If the first power peak value is less than the second power peak value, the center frequency identifying unit 10 identifies the frequency corresponding to the second power peak value in the converted frequency domain signal as the center frequency of the chirp signal.

[0045] 7, it is assumed that each of the components of the signal parameter estimation device, namely, the candidate value setting unit 1, the candidate value selection unit 2, the frequency width estimation unit 3, the time width estimation unit 4, the pulse number estimation unit 5, the replica signal creation unit 6, the signal conversion unit 7, the time offset calculation unit 8, the start time estimation unit 9, and the center frequency identification unit 10, is realized by dedicated hardware as shown in Fig. 8. In other words, it is assumed that the signal parameter estimation device is realized by the candidate value setting circuit 11, the candidate value selection circuit 12, the frequency width estimation circuit 13, the time width estimation circuit 14, the pulse number estimation circuit 15, the replica signal creation circuit 16, the signal conversion circuit 17, the time offset calculation circuit 18, the start time estimation circuit 19, and the center frequency identification circuit 20. Each of the candidate value setting circuit 11, the candidate value selection circuit 12, the frequency width estimation circuit 13, the time width estimation circuit 14, the pulse number estimation circuit 15, the replica signal creation circuit 16, the signal conversion circuit 17, the time offset calculation circuit 18, the start time estimation circuit 19, and the center frequency identification circuit 20 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0046] The components of the signal parameter estimation device are not limited to those realized by dedicated hardware, and the signal parameter estimation device may be realized by software, firmware, or a combination of software and firmware. When the signal parameter estimation device is realized by software, firmware, or the like, a program for causing a computer to execute the respective processing procedures of the candidate value setting unit 1, the candidate value selecting unit 2, the frequency width estimating unit 3, the time width estimating unit 4, the pulse number estimating unit 5, the replica signal creating unit 6, the signal converting unit 7, the time offset calculating unit 8, the start time estimating unit 9, and the center frequency identifying unit 10 is stored in a memory 21 shown in FIG. 3. Then, a processor 22 shown in FIG. 3 executes the program stored in the memory 21.

[0047] 8 shows an example in which each of the components of the signal parameter estimation device is realized by dedicated hardware, while Fig. 3 shows an example in which the signal parameter estimation device is realized by software, firmware, etc. However, this is merely an example, and some of the components in the signal parameter estimation device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0048] Next, the operation of the signal parameter estimation device shown in Fig. 7 will be described. The components other than the pulse number estimation unit 5, replica signal creation unit 6, signal conversion unit 7, time offset calculation unit 8, start time estimation unit 9, and center frequency identification unit 10 are the same as those of the signal parameter estimation device shown in Fig. 1. Therefore, only the operations of the pulse number estimation unit 5, replica signal creation unit 6, signal conversion unit 7, time offset calculation unit 8, start time estimation unit 9, and center frequency identification unit 10 will be described here.

[0049] The pulse number estimation unit 5 acquires the converted frequency domain signal z(f) from the frequency width estimation unit 3. The pulse number estimation unit 5 counts the number N of power peak values ​​contained in the converted frequency domain signal z(f). That is, the pulse number estimation unit 5 counts the number N of power peak values ​​by determining that signal components in the converted frequency domain signal z(f) whose power is equal to or greater than a threshold Th are power peak values. The threshold Th may be stored in an internal memory of the pulse number estimation unit 5, for example, or may be provided from outside the signal parameter estimation device shown in FIG. 7. The pulse number estimation unit 5 outputs the number N of power peak values ​​to the replica signal creation unit 6 as the number of pulses of the chirp signal.

[0050] The replica signal generator 6 receives the chirp frequency width f ｗ and obtains information indicating the chirp time width t ｗ and obtains the number of pulses N from the pulse number estimation unit 5. The replica signal generation unit 6 obtains the information indicating the chirp frequency width f ｗ and chirp time width t ｗ and the number of pulses N, the chirp signal x(t ｋ ) replica signal ref(t ｋ ) to create a chirp signal x(t ｋ ) replica signal ref(t ｋ ) is a chirp signal x(t ｋ ) corresponds to the mathematical model of

[0051] In formula (5), t ｋ is the time, and A is the chirp signal x(t ｋ ) and, for example, the amplitude value of 10 SNR [dB] / 20.

[0052] Replica signal ref(t ｋ ) is the chirp signal x(t ｋ ), the replica signal ref(t ｋ ) chirp frequency width is f ｗ and the replica signal ref(t ｋ ) chirp time width is t ｗIn addition, the replica signal ref(t ｋ The number of pulses in the chirp signal x(t ｋ ) start time t ａ Since is unknown, the replica signal ref(t ｋ ) start time t ａ is set to "0". The chirp signal x(t ｋ ) start time t ａ is non-zero, the chirp signal x(t ｋ ) and replica signal ref(t ｋ ) are the time-series data t ａ FIG. 9 shows the chirp signal x(t ｋ ) and replica signal ref(t ｋ 9 is an explanatory diagram showing an example of the replica signal ref(t ｋ ) to the signal conversion unit 7 and the start time estimation unit 9, respectively.

[0053] The signal converter 7 converts the received signal x ｏｂｓ (t), and the replica signal generator 6 generates a replica signal ref(t ｋ The signal converter 7 obtains the replica signal ref(t ｋ ) and the received signal x ｏｂｓ The signal converter 7 multiplies the replica signal ref(t ｋ ) and the received signal x ｏｂｓ (t) after multiplication with the signal ref(t ｋ ) x ｏｂｓ (t) is subjected to FFT, for example, to obtain the multiplied signal ref(t ｋ ) x ｏｂｓ The signal converter 7 converts (t) into a frequency domain signal, and outputs the converted frequency domain signal to the time offset calculator 8 and the center frequency identifier 10.

[0054] The time lag calculation unit 8 obtains the converted frequency domain signal from the signal conversion unit 7. ｋ ) and replica signal ref(t ｋ) are the time-dependent ａ 10, the converted frequency domain signal has two power peak values ​​P Ｈｉｇｈｅｒ , P Ｌｏｗｅｒ FIG. 10 is an explanatory diagram showing an example of a frequency domain signal containing two power peak values. In FIG. 10, the horizontal axis represents frequency [Hz] and the vertical axis represents power [dB]. The power peak value P Ｈｉｇｈｅｒ is the frequency f shown in FIG. Ｈｉｇｈｅｒ is the power corresponding to the peak power value P Ｌｏｗｅｒ is the frequency f shown in FIG. Ｌｏｗｅｒ is the power corresponding to

[0055] The time lag calculation unit 8 calculates the two power peak values ​​P Ｈｉｇｈｅｒ , P Ｌｏｗｅｒ The time lag calculation unit 8 calculates the two peak power values ​​P Ｈｉｇｈｅｒ , P Ｌｏｗｅｒ Based on this, the replica signal ref(t ｋ ) and chirp signal x(t ｋ ) and the first time difference t ａ１ and replica signal ref(t ｋ ) and chirp signal x(t ｋ ) and the second time difference t ａ２ and calculate.

[0056] Hereinafter, the first time lag t ａ１ and the second time difference t ａ２ The calculation process of the two peak power values ​​P Ｈｉｇｈｅｒ , P Ｌｏｗｅｒ Each of the replica signals ref(t ｋ ) and chirp signal (t ｋ ) and the time difference t ａ and chirp time width t ｗ FIG. 11 shows the relationship between |mod(t ａ , t ｗ ) | < t ｗ / 2, the replica signal ref(t ｋ ) and chirp signal (t ｋ) is an explanatory diagram showing an example of |mod(t ａ , t ｗ ) | < t ｗ / 2, the peak power value P Ｌｏｗｅｒ is expressed as Nt ａ is proportional to the peak power value P Ｈｉｇｈｅｒ is expressed as N(t ｗ -t ａ ) in equations (7) and (8). ａ Gat ａ１ If so, then, from the relationship between equations (7) and (8), the first time shift t ａ１ is expressed as the following equation (9).

[0057]

[0058] FIG. 12 shows the relationship between |mod(t ａ , t ｗ ) | ≧ t ｗ / 2, the replica signal ref(t ｋ ) and chirp signal (t ｋ ) is an explanatory diagram showing an example of |mod(t ａ , t ｗ ) | ≧ t ｗ / 2, the peak power value P Ｌｏｗｅｒ is expressed as N(t ｗ -t ａ ) and the peak power value P Ｈｉｇｈｅｒ is expressed as Nt ａ In equations (10) and (11), t ａ Gat ａ２ Then, from the relationship between equations (10) and (11), the second time difference t ａ２ is expressed as the following equation (12).

[0059]

[0060] The time lag calculation unit 8 calculates the first time lag t ａ１ and calculates the second time difference t ａ２ The time lag calculation unit 8 calculates the first time lag t ａ１and the second time difference t ａ２ and information indicating the start time to the start time estimation unit 9.

[0061] The start time estimation unit 9 receives the first time lag t ａ１ and the second time difference t ａ２ The start time estimation unit 9 obtains information indicating the first time difference t ａ１ and the second time difference t ａ２ and the start time t of the chirp signal ａ Estimate.

[0062] Hereinafter, the start time t of the chirp signal estimated by the start time estimation unit 9 ａ The replica signal correction unit 9a receives the replica signal ref(t ｋ ) is acquired, and the first time lag t ａ１ and the second time difference t ａ２ The replica signal corrector 9a obtains information indicating the first time shift t ａ１ Using the replica signal ref(t ｋ ) is corrected. ａ１ The replica signal ref(t ｋ ) １ is the time when the signal starts. ａ１ The replica signal corrector 9a corrects the replica signal when the second time shift t ａ２ Using the replica signal ref(t ｋ ) is corrected. ａ２ The replica signal ref(t ｋ ) ２ is the time when the signal starts. ａ２ The replica signal corrector 9a corrects the replica signal when the first time shift t ａ１ The replica signal ref(t ｋ ) １ The replica signal corrector 9a outputs the second time shift t ａ２ The replica signal ref(t ｋ) ２ is output as a second replica signal to the second peak value calculation unit 9c.

[0063] The first peak value calculation unit 9b calculates the received signal x ｏｂｓ (t), and the replica signal correction unit 9a outputs the first replica signal ref(t ｋ ) １ The first peak value calculation unit 9b obtains the first replica signal ref(t ｋ ) １ and the received signal x ｏｂｓ The first peak value calculation unit 9b multiplies the first replica signal ref(t ｋ ) １ and the received signal x ｏｂｓ (t) after multiplication with the signal ref(t ｋ ) １ x ｏｂｓ (t) is subjected to FFT, for example, to obtain the multiplied signal ref(t ｋ ) １ x ｏｂｓ The first peak value calculation unit 9b converts the first power peak value Peak (t) contained in the frequency domain signal. １ The first peak value calculation unit 9b calculates the first power peak value Peak １ to the start time estimation processing unit 9d.

[0064] The first replica signal ref(t ｋ ) １ and the chirp signal x(t ｋ ) the first power peak value Peak １ On the other hand, the number of the first replica signal ref(t ｋ ) １ and the chirp signal x(t ｋ ) the first power peak value Peak １ The number of the first power peak values ​​Peak contained in the frequency domain signal is two. １The number of peaks can be determined by checking the number of signal components whose power is equal to or greater than the threshold value Th′, as shown in Fig. 13. The threshold value Th′ may be stored in an internal memory of the first peak value calculation unit 9b, or may be provided from outside the signal parameter estimation device shown in Fig. 7.

[0065] The second peak value calculation unit 9c calculates the received signal x ｏｂｓ (t), and the replica signal correction unit 9a outputs a second replica signal ref(t ｋ ) ２ The second peak value calculation unit 9c obtains the second replica signal ref(t ｋ ) ２ and the received signal x ｏｂｓ The second peak value calculation unit 9c multiplies the second replica signal ref(t ｋ ) ２ and the received signal x ｏｂｓ (t) after multiplication with the signal ref(t ｋ ) ２ x ｏｂｓ (t) is subjected to FFT, for example, to obtain the multiplied signal ref(t ｋ ) ２ x ｏｂｓ The second peak value calculation unit 9c converts the second power peak value Peak (t) contained in the frequency domain signal. ２ The second peak value calculation unit 9c calculates the second power peak value Peak ２ to the start time estimation processing unit 9d.

[0066] The second replica signal ref(t ｋ ) ２ and the chirp signal x(t ｋ ) the second power peak value Peak ２ On the other hand, the number of the second replica signal ref(t ｋ ) ２ and the chirp signal x(t ｋ ) the second power peak value Peak ２ The number of the second power peak values ​​Peak contained in the frequency domain signal is two.２ The number of peaks is determined by checking the number of signal components whose power is equal to or greater than the threshold value Th'. The threshold value Th' may be stored in an internal memory of the second peak value calculation unit 9c, or may be provided from outside the signal parameter estimation device shown in FIG.

[0067] The start time estimation processing unit 9d receives the first time lag t ａ１ and the second time difference t ａ２ The start time estimation processing unit 9d acquires information indicating the first power peak value Peak from the first peak value calculation unit 9b. １ and the second peak value calculation unit 9c calculates the second power peak value Peak ２ The start time estimation processing unit 9d acquires the first power peak value Peak １ is the second peak power value Peak ２ If so, the start time of the chirp signal t ａ is the first time difference t ａ１ The start time estimation processing unit 9d estimates that the first power peak value Peak １ is the second peak power value Peak ２ If it is less than t, the start time of the chirp signal is ａ is the second time difference t ａ２ The start time t of the chirp signal estimated by the start time estimation processing unit 9d is estimated to be equivalent to ａ is displayed on a display device (not shown), for example.

[0068] FIG. 13 shows the first power peak value Peak １ and the second power peak value Peak ２ 13 is an explanatory diagram showing an example of the first power peak value Peak. In FIG. 13, the horizontal axis represents frequency [Hz] and the vertical axis represents power [dB]. In the example of FIG. 13, the first power peak value Peak １ is the second peak power value Peak ２ In this case, the first power peak value Peak １ The number of the second peak power value Peak is one. ２ The number of the first peak power value is two. １is the second peak power value Peak ２ If it is less than the first peak power value Peak １ The number of peak power values ​​is two, and the second peak power value is ２ The number of is one.

[0069] The center frequency identifying unit 10 acquires the converted frequency domain signal from the signal converting unit 7. The center frequency identifying unit 10 receives the first power peak value Peak from the first peak value calculating unit 9b. １ and the second peak value calculation unit 9c calculates the second power peak value Peak ２ The center frequency identifying unit 10 obtains the first power peak value Peak １ is the second peak power value Peak ２ If this is the case, the chirp signal x(t ｋ ) center frequency f ｃ In the transformed frequency domain signal, the first power peak value Peak １ The center frequency identifying unit 10 identifies the frequency corresponding to the first power peak value Peak １ is the second peak power value Peak ２ If less than, the chirp signal x(t ｋ ) center frequency f ｃ In the transformed frequency domain signal, the second power peak value Peak ２ The frequency corresponding to the chirp signal x(t ｋ ) center frequency f ｃ is displayed on a display device (not shown), for example.

[0070] In the second embodiment described above, the signal parameter estimation device is configured to include: a replica signal generator 6 that generates a replica signal of the chirp signal using the chirp frequency width estimated by the frequency width estimator 3, the chirp time width estimated by the time width estimator 4, and the number of pulses estimated by the pulse number estimator 5; a signal converter 7 that multiplies the replica signal generated by the replica signal generator 6 by the received signal and converts the resulting signal into a frequency-domain signal; a time offset calculator 8 that calculates a first time offset between the replica signal and the chirp signal and a second time offset between the replica signal and the chirp signal based on two power peak values ​​contained in the frequency-domain signal converted by the signal converter 7; and a start time estimator 9 that estimates the start time of the chirp signal using the first time offset and the second time offset. Therefore, even if the power of the chirp signal is low, equal to or less than the noise power, the signal parameter estimation device can estimate not only the signal parameters of the chirp signal but also the start time of the chirp signal. The start time of the chirp signal is also a signal parameter of the chirp signal. In the second embodiment, as in the first embodiment, the SNR is improved, and it becomes possible to detect signal parameters even for low SNR signals.

[0071] In the second embodiment, the signal parameter estimation device is configured to include a center frequency identification unit 10 that identifies the frequency corresponding to the first power peak value as the center frequency of the chirp signal if the first power peak value is equal to or greater than the second power peak value, and identifies the frequency corresponding to the second power peak value as the center frequency of the chirp signal if the first power peak value is less than the second power peak value. Therefore, even if the power of the chirp signal is low, equal to or less than the noise power, the signal parameter estimation device can estimate the signal parameters of the chirp signal and can also identify the center frequency of the chirp signal. The center frequency of the chirp signal is also a signal parameter of the chirp signal.

[0072] Third Embodiment In a third embodiment, a signal parameter estimation device will be described in which a start time estimation unit 9' further includes a start time search unit 9e.

[0073] FIG. 14 is a configuration diagram showing a signal parameter estimation device according to embodiment 3. In FIG. 14, the same reference numerals as those in FIGS. 1 and 7 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. FIG. 15 is a hardware configuration diagram showing hardware of a signal parameter estimation device according to embodiment 3. In FIG. 15, the same reference numerals as those in FIGS. 2 and 8 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. The signal parameter estimation device shown in FIG. 14 includes a candidate value setting unit 1, a candidate value selecting unit 2, a frequency width estimating unit 3, a time width estimating unit 4, a pulse number estimating unit 5, a replica signal creating unit 6, a signal converting unit 7, a time offset calculating unit 8, a start time estimating unit 9′, and a center frequency identifying unit 10.

[0074] The start time estimation unit 9′ is realized by, for example, a start time estimation circuit 19′ shown in Fig. 15. The start time estimation unit 9′ includes a replica signal correction unit 9a, a first peak value calculation unit 9b, a second peak value calculation unit 9c, a start time estimation processing unit 9d, and a start time search unit 9e.

[0075] The start time search unit 9e calculates the start time t of the chirp signal estimated by the start time estimation processing unit 9d. ａ The start time search unit 9e uses values ​​including the initial value to calculate the replica signal ref(t ｋ The value including the initial value includes, for example, the initial value and values ​​around the initial value. The start time search unit 9e generates the replica signal ref(t ｋ ) and the received signal x ｏｂｓ (t) and calculate the correlation with the replica signal ref(t ｋ ) and the received signal x ｏｂｓ (t), for example, the correlation with the replica signal ref(t ｋ ) and the received signal x ｏｂｓ The start time search unit 9e multiplies the replica signal ref(t ｋ ) and the received signal x ｏｂｓ Based on the correlation with (t), the start time t of the chirp signal is ａ The start time search unit 9e searches for the start time t estimated by the start time estimation processing unit 9d. ａThis is to obtain a more accurate start time than

[0076] 14, it is assumed that each of the components of the signal parameter estimation device, namely, the candidate value setting unit 1, the candidate value selection unit 2, the frequency width estimation unit 3, the time width estimation unit 4, the pulse number estimation unit 5, the replica signal creation unit 6, the signal conversion unit 7, the time offset calculation unit 8, the start time estimation unit 9′, and the center frequency identification unit 10, is realized by dedicated hardware as shown in Fig. 15. In other words, it is assumed that the signal parameter estimation device is realized by the candidate value setting circuit 11, the candidate value selection circuit 12, the frequency width estimation circuit 13, the time width estimation circuit 14, the pulse number estimation circuit 15, the replica signal creation circuit 16, the signal conversion circuit 17, the time offset calculation circuit 18, the start time estimation circuit 19′, and the center frequency identification circuit 20. Each of the candidate value setting circuit 11, the candidate value selection circuit 12, the frequency width estimation circuit 13, the time width estimation circuit 14, the pulse number estimation circuit 15, the replica signal creation circuit 16, the signal conversion circuit 17, the time offset calculation circuit 18, the start time estimation circuit 19′, and the center frequency identification circuit 20 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0077] The components of the signal parameter estimation device are not limited to those realized by dedicated hardware, and the signal parameter estimation device may be realized by software, firmware, or a combination of software and firmware. When the signal parameter estimation device is realized by software, firmware, or the like, a program for causing a computer to execute the respective processing procedures of the candidate value setting unit 1, the candidate value selecting unit 2, the frequency width estimating unit 3, the time width estimating unit 4, the pulse number estimating unit 5, the replica signal creating unit 6, the signal converting unit 7, the time offset calculating unit 8, the start time estimating unit 9′, and the center frequency identifying unit 10 is stored in a memory 21 shown in FIG. 3. Then, a processor 22 shown in FIG. 3 executes the program stored in the memory 21.

[0078] 15 shows an example in which each of the components of the signal parameter estimation device is realized by dedicated hardware, while Fig. 3 shows an example in which the signal parameter estimation device is realized by software, firmware, etc. However, this is merely an example, and some of the components in the signal parameter estimation device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0079] Next, the operation of the signal parameter estimation device shown in Fig. 14 will be described. Since the components other than the start time search unit 9e are the same as those of the signal parameter estimation device shown in Fig. 7, only the operation of the start time search unit 9e will be described here.

[0080] The start time search unit 9e receives the start time t of the chirp signal from the start time estimation processing unit 9d. ａ The start time search unit 9e obtains the start time t ａ The start time search unit 9e uses each of the values ​​including the initial value to calculate the replica signal ref(t ｋ The value including the initial value includes, for example, the initial value and values ​​around the initial value. ｋ The process of generating the replica signal ref(t) by the replica signal generating unit 6 is the same as that of generating the replica signal ref(t) except that values ​​including the initial value are used. ｋ 14, the start time search unit 9e receives the chirp frequency width f from the frequency width estimation unit 3 in the same way as the replica signal creation unit 6. ｗ and obtains information indicating the chirp time width t ｗ and obtains the pulse number N from the pulse number estimation unit 5.

[0081] The start time search unit 9e generates each replica signal ref(t ｋ ) and the received signal x ｏｂｓ (t) and calculate the correlation with the replica signal ref(t ｋ ) and the received signal x ｏｂｓ (t), for example, the correlation with the replica signal ref(t ｋ) and the received signal x ｏｂｓ (t) and the result of the multiplication is subjected to FFT. The start time search unit 9e determines the start time t ａ Specifically, the start time search unit 9e finds the value (initial value or a peripheral value) associated with the highest correlation among the calculated correlations at the start time t ａ The start time t of the chirp signal searched by the start time search unit 9e is specified as the search result. ａ is displayed on a display device (not shown), for example.

[0082] In the third embodiment described above, the signal parameter estimation device shown in Fig. 14 is configured so that the start time estimation unit 9' uses the start time of the chirp signal estimated by the start time estimation processing unit 9d as an initial value, creates a replica signal of the chirp signal using values ​​including the initial value, and searches for the start time of the chirp signal based on the correlation between the replica signal and the received signal. Therefore, the signal parameter estimation device shown in Fig. 14 can determine the start time of the chirp signal more accurately than the signal parameter estimation device shown in Fig. 7.

[0083] In addition, the present disclosure allows for free combination of the respective embodiments, modification of any of the components of the respective embodiments, or omission of any of the components of the respective embodiments.

[0084] The signal parameter estimation device according to the present disclosure can be used to estimate the signal parameters of a chirp signal even if the power of the chirp signal is low, equal to or less than the noise power.

[0085] REFERENCE SIGNS LIST 1 candidate value setting unit, 2 candidate value selection unit, 3 frequency width estimation unit, 4 time width estimation unit, 5 pulse number estimation unit, 6 replica signal creation unit, 7 signal conversion unit, 8 time offset calculation unit, 9, 9' start time estimation unit, 9a replica signal correction unit, 9b first peak value calculation unit, 9c second peak value calculation unit, 9d start time estimation processing unit, 9e start time search unit, 10 center frequency identification unit, 11 candidate value setting circuit, 12 candidate value selection circuit, 13 frequency width estimation circuit, 14 time width estimation circuit, 15 pulse number estimation circuit, 16 replica signal creation circuit, 17 signal conversion circuit, 18 time offset calculation circuit, 19, 19' start time estimation circuit, 20 center frequency identification circuit, 21 memory, 22 processor.

Claims

1. A signal parameter estimation device comprising: a candidate value setting unit that sets multiple candidate values ​​for a chirp rate, which indicates the ratio of a chirp frequency width to a chirp time width of a received signal, which is an observation signal of a chirp signal to be estimated; a candidate value selection unit that multiplies the received signal by an anti-phase component of each of the candidate values ​​set by the candidate value setting unit, converts the received signal after multiplication with the anti-phase component into a frequency domain signal, and selects from the multiple candidate values ​​a candidate value that produces a relatively large power peak value of the frequency domain signal; a frequency width estimation unit that multiplies the received signal, which is the observation signal, by the anti-phase component of the candidate value selected by the candidate value selection unit, converts the received signal after multiplication with the anti-phase component into a frequency domain signal, and estimates the chirp frequency width based on the intervals of frequencies corresponding to the multiple power peak values ​​included in the converted frequency domain signal; and a time width estimation unit that estimates the chirp time width from the chirp frequency width estimated by the frequency width estimation unit and the candidate value selected by the candidate value selection unit.

2. The signal parameter estimation device according to claim 1, further comprising a pulse number estimation unit that estimates the number of pulses, which is the number of the chirp signal, from the number of power peak values ​​contained in the converted frequency domain signal.

3. The signal parameter estimation device according to claim 2, further comprising: a replica signal generation unit that generates a replica signal of the chirp signal using the chirp frequency width estimated by the frequency width estimation unit, the chirp time width estimated by the time width estimation unit, and the number of pulses estimated by the pulse number estimation unit; a signal conversion unit that multiplies the replica signal generated by the replica signal generation unit by the received signal that is the observed signal, and converts the multiplied signal of the replica signal and the received signal into a frequency domain signal; a time offset calculation unit that calculates a first time offset between the replica signal and the chirp signal and a second time offset between the replica signal and the chirp signal based on two power peak values ​​contained in the frequency domain signal converted by the signal conversion unit; and a start time estimation unit that estimates the start time of the chirp signal using the first time offset and the second time offset.

4. The start time estimation unit includes: a replica signal correction unit that corrects the replica signal using the first time offset and outputs the replica signal corrected using the first time offset as a first replica signal; and corrects the replica signal using the second time offset and outputs the replica signal corrected using the second time offset as a second replica signal; a first peak value calculation unit that multiplies the first replica signal by a received signal that is the observation signal, converts the signal after multiplication of the first replica signal by the received signal into a frequency domain signal, and calculates a first power peak value included in the frequency domain signal; and a second peak value calculation unit that multiplies the second replica signal by the received signal that is the observation signal, converts the signal after multiplication of the second replica signal by the received signal into a frequency domain signal, and calculates a second power peak value included in the frequency domain signal. a start time estimation processing unit that estimates the start time of the chirp signal based on the first time offset if the first power peak value is equal to or greater than the second power peak value, and that estimates the start time of the chirp signal based on the second time offset if the first power peak value is less than the second power peak value.

5. A signal parameter estimation device according to claim 4, further comprising a center frequency specifying unit that specifies, if the first power peak value is equal to or greater than the second power peak value, a frequency corresponding to the first power peak value as the center frequency of the chirp signal, and that specifies, if the first power peak value is less than the second power peak value, a frequency corresponding to the second power peak value as the center frequency of the chirp signal.

6. The signal parameter estimation device according to claim 4, further comprising a start time search unit that sets the start time of the chirp signal estimated by the start time estimation processing unit as an initial value, creates a replica signal of the chirp signal using a value including the initial value, and searches for the start time of the chirp signal based on the correlation between the replica signal and the received signal.

7. A signal parameter estimation method in which a candidate value setting unit sets multiple candidate values ​​for a chirp rate indicating the ratio of a chirp frequency width to a chirp time width of a received signal, which is an observation signal of a chirp signal to be estimated; a candidate value selection unit multiplies the received signal by an anti-phase component of each candidate value set by the candidate value setting unit, converts the received signal after multiplication with the anti-phase component into a frequency domain signal, and selects from the multiple candidate values ​​a candidate value that relatively increases the power peak value of the frequency domain signal; a frequency width estimation unit multiplies the received signal, which is the observation signal, by the anti-phase component of the candidate value selected by the candidate value selection unit, converts the received signal after multiplication with the anti-phase component into a frequency domain signal, and estimates the chirp frequency width based on the interval between frequencies corresponding to multiple power peak values ​​included in the converted frequency domain signal; and a time width estimation unit estimates the chirp time width from the chirp frequency width estimated by the frequency width estimation unit and the candidate value selected by the candidate value selection unit.

Images