Signal identification device, signal suppression device, signal identification and suppression device, parameter identification method, parameter identification program, and recording medium

Abstract

This signal identification device (1) is provided with: a sweep cycle estimation unit (12a) that estimates a sweep cycle in which a sweep of a linear frequency modulation (LFM) wave in a digital reception signal obtained by converting an incoming wave comprising an LFM wave into a digital signal is repeated; a chirp rate estimation unit (12b) that estimates a chirp rate, which is the frequency change rate of the LFM wave in the digital reception signal; and an initial frequency estimation unit (12c) that estimates the initial frequency of the LFM wave in the digital reception signal.

Description

Signal identification device, signal suppression device, signal identification / suppression device, parameter identification method, parameter identification program, and recording medium

[0001] The present disclosure relates to a signal identification device, a signal suppression device, a signal identification and suppression device, a parameter identification method, a parameter identification program, and a recording medium.

[0002] In the field of wireless communications, Linear Frequency Modulation (LFM) signals, also known as chirp signals, are widely used, and the accuracy and speed of parameter estimation of chirp signals are extremely important. Non-Patent Document 1 proposes a method for estimating parameters with high accuracy and high speed.

[0003] The chirp signal parameter estimation proposed in Non-Patent Document 1 is a method in which the frequency of the chirp signal is first estimated using a quadratic phase function (QPF), the transformation order of the fractional Fourier transform (FRFT) is then estimated based on the chirp rate, and the center frequency and amplitude of the chirp signal are obtained by the FRFT.

[0004] Xuelian Liu, Jun Han, Chunyang Wang, Bo Xiao, “Parameters estimation for chirp signal based on QPF-FRFT,” Optik, vol.182, pp.529-537, 2019.

[0005] The technology disclosed in Non-Patent Document 1 is based on the premise that the incoming LFM wave is a single wave. Furthermore, Non-Patent Document 1 does not mention cases where the LFM wave is periodic. When multiple LFM waves arrive as incoming waves or when the LFM wave is periodic, the technology disclosed in Non-Patent Document 1 has the problem that it may not be possible to estimate the parameters of the LFM wave.

[0006] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a signal identification device that can estimate parameters of an LFM wave even when multiple LFM waves arrive as incoming waves and when the LFM wave is periodic.

[0007] The signal identification device according to the present disclosure includes a sweep period estimation unit that estimates a sweep period in which a sweep of a linear frequency modulated (LFM) wave is repeated in a digital received signal obtained by converting an incoming wave consisting of an LFM wave into a digital signal, a chirp rate estimation unit that estimates a chirp rate that is the rate of change of frequency of the LFM wave in the digital received signal, and an initial frequency estimation unit that estimates an initial frequency of the LFM wave in the digital received signal.

[0008] According to the present disclosure, it is possible to estimate parameters of an LFM wave even when a plurality of LFM waves arrive as incoming waves and when the LFM waves are periodic.

[0009] FIG. 1 is a configuration diagram showing a signal identification device according to a first embodiment. FIG. 2 is an explanatory diagram showing definitions of parameters of an LFM wave. FIG. 3 is a hardware configuration diagram showing hardware of the signal identification device according to the first embodiment. FIG. 4 is a flowchart showing a processing procedure for signal identification in the signal identification device according to the first embodiment. FIG. 5 is a configuration diagram showing a signal identification and suppression device according to a second embodiment. FIG. 6 is a flowchart showing a processing procedure for signal identification in the signal identification and suppression device according to the second embodiment. FIG. 7 is a configuration diagram showing a signal identification and suppression device according to a third embodiment. FIG. 8 is a flowchart showing a processing procedure for signal identification in the signal identification and suppression device according to the third embodiment.

[0010] A signal identification device 1 according to a first embodiment will be described with reference to Figures 1 to 4. The signal identification device 1 according to the first embodiment is a signal identification device that identifies parameters of LFM (Linear Frequency Modulation) waves used in applications such as wireless communication, radar, and sonar. The signal identification device 1 according to the first embodiment includes a reception processing unit 11, a signal identification unit 12, a storage unit 13, and an identification result output unit 14.

[0011] Before describing the signal identification device 1 according to the first embodiment, the LFM wave, which is the incoming wave received by the signal identification device 1, will be described with reference to Fig. 2. Fig. 2 shows a case where the signal identification device 1 receives a transmitted wave consisting of an LFM wave, which is a chirp signal whose frequency increases over time, as an incoming wave Rx consisting of an LFM wave. １ ~Rx ３ Here is an example of receiving the message periodically.

[0012] In the description of the embodiment, the parameters of the LFM wave are defined as follows: １ ~Rx ３ Each of these is a chirp signal whose frequency increases over time, and the slope, that is, the rate of change of frequency over time, is defined as the chirp rate. １ ~Rx ３ The reception interval, that is, the period during which the frequency sweep is repeated is defined as the sweep period.

[0013] Arrival wave Rx １ ~Rx ３ The instantaneous frequency at the beginning of the processed data is defined as the initial frequency. １ ~Rx ３ Three are listed for reference, but the number is not limited to three.

[0014] The receiving processing unit 11 receives an LFM wave or a sound wave transmitted from a transmitter (not shown) as an incoming wave Rx. １ ~Rx ３ The incoming wave Rx from the receiving circuit 100 (see FIG. 1) is received using a sensor. １ ~Rx ３ The receiving circuit 100 is configured by a generally known receiving circuit. The digital receiving signal output from the receiving processing unit 11 may contain, in addition to the desired IQ signal and noise, an IQ signal due to an undesired LFM wave, and the IQ signal due to the undesired LFM wave may cause interference with the desired IQ signal.

[0015] The signal identification unit 12 estimates the sweep period, chirp rate, and initial frequency of the LFM wave in the digital received signal from the reception processing unit 11. The signal identification unit 12 has a sweep period estimation unit 12a, a chirp rate estimation unit 12b, and an initial frequency estimation unit 12c.

[0016] The sweep period estimation unit 12a estimates, by one-dimensional search, the sweep period at which the sweep of the LFM wave is repeated in the digital received signal from the reception processing unit 11, and acquires an estimated sweep period value. The sweep period estimation unit 12a also acquires an estimated wave number value of the LFM wave in the digital received signal from the reception processing unit 11. The sweep period estimation unit 12a acquires the estimated sweep period value, that is, estimates the sweep period, as follows.

[0017] The sweep period estimation unit 12a estimates the number of samples N from the digital received signal (chirp signal) from the reception processing unit 11. ０ The digital signal is extracted and the number of samples is N ０ The sweep period estimation unit 12a performs a one-dimensional search for the peak of the autocorrelation function, and calculates the wave number estimate of the LFM wave included in the digital received signal (the symbol shown in the following (1). In the following description, the wave number estimate will be referred to as "Nw^") and its sweep period estimate (the symbols shown in the following (2) and (3). In the following description, the sweep period estimate will be referred to as "τ ｉ ^").

[0018]

[0019] The wave number estimate Nw^ and the sweep period estimate τ obtained by the sweep period estimation unit 12a ｉ is stored in the memory unit 13. If the wave number estimation value Nw^ of the LFM wave acquired by the sweep period estimation unit 12a is zero, it is assumed that multiple LFM waves are not included as arriving waves and that the LFM waves do not have periodicity.

[0020] The chirp rate estimator 12b estimates a chirp rate, which is the rate of change of the frequency of the LFM wave in the digital received signal from the reception processor 11. The chirp rate estimator 12b estimates the sweep period estimated value τ ｉ ^ and the sweep period estimate τ ｉ ^ to obtain an estimated chirp rate of the LFM wave contained in the digital received signal from the reception processing unit 11. The chirp rate estimator 12b obtains an estimated chirp rate for each wave number i (= 1, 2, ... Nw^) for the number of waves in the wave number estimate value Nw^ obtained by the sweep period estimator 12a.

[0021] The chirp rate estimator 12b estimates the sweep period estimated value τ ｉ The chirp rate is estimated by using an extended autocorrelation function to which a phase correction term using ^ has been added.

[0022] The chirp rate estimator 12b first calculates the sweep period estimate τ ｉ The extended autocorrelation function Z incorporating Ｌ Obtain [・].

[0023]

[0024] In the above formula (4), N １ is the number of samples extracted from the digital received signal from the receiving processing unit 11, L is the number of lag samples, μ ｊ (j=1, 2,..., N μ ) is the candidate chirp rate, T Ｓ is the sampling period, (・) ＊ represents the complex conjugate.

[0025] The chirp rate estimation unit 12b estimates candidate chirp rates μ ｊ (j=1, 2,..., N μ ) and calculate the calculated extended autocorrelation function Z Ｌ The peaks of [j] are searched for one-dimensionally, and the chirp rate estimates of each LFM wave (symbols shown in the following (5) and (6) are used. In the following explanation, the chirp rate estimates are referred to as "μ ｉThe chirp rate estimation unit 12b acquires the chirp rate estimate value μ ｉ ^ is stored in the storage unit 13.

[0026]

[0027] The initial frequency estimator 12c estimates an initial frequency, which is the instantaneous frequency at the beginning of the LFM wave in the digital received signal. The initial frequency estimator 12c estimates the sweep period estimated value τ ｉ ^ and the chirp rate estimate value μ obtained by the chirp rate estimation unit 12b ｉ ^ and the sweep period estimate τ ｉ ^ and chirp rate estimate μ ｉ The initial frequency is estimated by using a one-dimensional fractional Fourier transform (FRFT) using ^.

[0028] The initial frequency estimation unit 12c first calculates the chirp rate estimate μ ｉ ^ is the FRFT order α ｉ Convert to.

[0029]

[0030] In the above formula (7), N ２ is the number of samples extracted from the digital received signal from the receiving processing unit 11, and the number of samples N １ Next, the initial frequency estimation unit 12c calculates the FRFT represented by the following equation (8).

[0031]

[0032] However, u in the above equation (8) is an argument in the Fractional Fourier Domain (FRFD), and the following equation (9) holds between u and the frequency f.

[0033]

[0034] The initial frequency estimation unit 12c performs a one-dimensional search for a peak in the FRFT expressed by the above formula (8), and finds the argument corresponding to the peak (symbols shown in the following (10) and (11). In the following explanation, the argument is referred to as "u"). ｉ ^").

[0035]

[0036] Based on the above formula (9), the argument u ｉ ^ is the initial frequency estimate (the symbol shown in the following (12). In the following explanation, the initial frequency estimate is referred to as "f ｉ ^), and the initial frequency estimate f ｉ The initial frequency estimation value f obtained by the initial frequency estimation unit 12c is obtained. ｉ The initial frequency estimate f ｉ ^ may be temporarily stored in the storage unit 13.

[0037] On the other hand, the number of samples N extracted from the digital received signal ２ If multiple sweep periods are included in the ｉ ^ is obtained. That is, the number of samples N ２ If multiple sweep periods are included in the FRFT, multiple peaks will be obtained in the FRFT.

[0038] When the peaks appearing in the FRFT are converted into frequency f based on the above equation (12), the estimated frequency difference between the first peak and the nth peak from the top (symbol shown by the following (13)). In the following explanation, the estimated frequency difference is referred to as "Δf ｉ ^(n)) is expressed by the following equation (14).

[0039]

[0040] Therefore, the frequency difference estimate Δf expressed by the above equation (14) can be calculated from the frequency estimates corresponding to the peaks that appear in the multiple FRFTs. ｉ ^(n) is subtracted, and the ensemble average of the subtracted values ​​is taken as the initial frequency estimate f ｉ Let's say ^.

[0041] The identification result output unit 14 outputs the sweep period estimated value τ ｉ ^, and the chirp rate estimation value μ obtained by the chirp rate estimation unit 12b ｉ ^ is acquired from the storage unit 13, and the initial frequency estimation value f acquired by the initial frequency estimation unit 12c is ｉ ^ is acquired from the initial frequency estimation unit 12c, and the sweep period estimation value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ ^ is output to the display device 200 and the signal processing device 300 at the subsequent stage.

[0042] The signal identification device 1 according to the first embodiment is realized by a computer-based hardware configuration, and as shown in FIG. 3 , includes a CPU (Central Processing Unit) 1A including a microprocessor or a DSP (Digital Signal Processor), an auxiliary storage device 1B consisting of a large-capacity semiconductor memory (RAM: Random Access Memory), a main storage device 1C consisting of a storage device (ROM: Read Only Memory) such as a hard disk device or a non-volatile storage device such as an SSD device, an input interface 1D, an output interface 1E, and a signal path (bus) 1F.

[0043] The CPU 1A controls and manages the auxiliary storage device 1B, the main storage device 1C, the input interface 1D, and the output interface 1E. The CPU 1A loads a program stored in the main storage device 1C into the auxiliary storage device 1B, and executes various processes based on the program loaded into the auxiliary storage device 1B. The CPU 1A calculates a sweep period estimate value τ based on the program loaded into the auxiliary storage device 1B. ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ The signal path 1F is a bus that interconnects the CPU 1A, the auxiliary storage device 1B, the main storage device 1C, the input interface 1D, and the output interface 1E.

[0044] Although the signal identification device 1 is realized by a computer hardware configuration, the reception processing unit 11, signal identification unit 12, memory unit 13, and identification result output unit 14 that constitute the signal identification device 1 may each be realized by dedicated hardware.

[0045] Next, the operation of the signal identification device 1 according to the first embodiment will be described with reference to Fig. 4. In particular, the sweep period, chirp rate, and initial frequency of the LFM wave are estimated, and the estimated sweep period value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ ^The method for identifying the parameters of the LFM waves used to obtain each of these will be mainly explained.

[0046] In step ST11, the sweep period estimation unit 12a estimates the sweep period and wave number of the LFM wave in the digital received signal from the reception processing unit 11, and calculates the sweep period estimated value τ ｉ In other words, the sweep period estimation unit 12a obtains the sample number N from the digital received signal (chirp signal). ０ The digital signal is extracted and the number of samples is N ０ The autocorrelation function of the digital signal is calculated. The sweep period estimation unit 12a performs a one-dimensional search for a peak of the autocorrelation function using the calculated autocorrelation function to obtain a sweep period estimate value τ ｉ ^ and the wavenumber estimate Nw^ are obtained.

[0047] Step ST12 is a step in which the chirp rate estimation unit 12b estimates a chirp rate, which is a frequency change rate of an LFM wave in a digital received signal. In step ST12, for the wave number of the wave number estimate value Nw^ acquired by the sweep period estimation unit 12a, the chirp rate estimation unit 12b estimates a sweep period estimate value τ ｉ ^ is used to estimate the chirp rate, and the chirp rate estimate μ ｉ This is the step to obtain ^.

[0048] That is, first, the sweep period estimation value τ where the wave number is 1, which is acquired by the sweep period estimation unit 12a, is １The extended autocorrelation function Z Ｌ The chirp rate estimation unit 12b obtains the acquired extended autocorrelation function Z Ｌ Candidate chirp rate μ for [j] １ , μ ２ , ..., μ ｊ Calculate the expanded autocorrelation function Z for each Ｌ The peak of [j] is searched for one-dimensionally, and the chirp rate estimate μ １ Get ^.

[0049] Step ST13 is a step in which the initial frequency estimator 12c estimates the initial frequency of the LFM wave in the digital received signal. ｉ ^ and the chirp rate estimate value μ obtained by the chirp rate estimation unit 12b ｉ ^ and the sweep period estimate τ ｉ ^ and chirp rate estimate μ ｉ The initial frequency is estimated by using a one-dimensional FRFT using ^.

[0050] In step ST13, first, for the wave number acquired by the sweep period estimation unit 12a being 1, the initial frequency estimation unit 12c calculates the chirp rate estimation value μ １ ^ is the FRFT order α １ and calculates the FRFT expressed by the above formula (8). The initial frequency estimation unit 12c performs a one-dimensional search for a peak in the FRFT expressed by the above formula (8), and calculates the argument u corresponding to the peak. １ Based on the above formula (9), the argument u １ ^ is the initial frequency estimate f １ ^ and calculate the initial frequency estimate f １ Get ^.

[0051] In step ST13, the number of samples N extracted from the digital received signal is ２ When a plurality of sweep periods are included in the FRFT, the frequency difference estimated value Δf expressed by the above equation (14) is calculated from the frequency estimated values ​​corresponding to the peaks that appear in the plurality of FRFT.１ ^(n) is subtracted, and the ensemble average of the subtracted values ​​is taken as the initial frequency estimate f １ Let's say ^.

[0052] In step ST14, the initial frequency estimation unit 12c calculates the initial frequency estimate f ｉ Now, it is determined whether the initial frequency estimation unit 12c has acquired the initial frequency estimate f for the wave number 1. １ ^ is acquired, the process returns to step ST12, and the initial frequency estimate f for wave number 2 is obtained through step ST13. ２ Get ^.

[0053] Steps ST12, ST13 and ST14 are repeated until the wave number becomes Nw^, and when the wave number becomes Nw^, the process proceeds to step ST15. In step ST15, the identification result output unit 14 outputs the sweep period estimated value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ ^ is output to the display device 200 and the downstream signal processing device 300, and the series of processes is completed.

[0054] In step ST11, if the wave number estimation value Nw^ estimated by the sweep period estimation unit 12a for the LFM waves in the digital received signal from the receiving processing unit 11 is zero, steps ST12 to ST14 are skipped, and the series of processes is terminated in step ST15.

[0055] The method for identifying LFM wave parameters in steps ST11 to ST15 is performed by the CPU 1A executing processing in accordance with a program stored in the main storage device 1C. That is, the program stored in the main storage device 1C includes a procedure for estimating the sweep period at which the sweep of the LFM wave is repeated in a digital reception signal obtained by converting an incoming wave made up of LFM waves into a digital signal, a procedure for estimating the chirp rate, which is the rate of change of the frequency of the LFM wave in the digital reception signal, and a procedure for estimating the initial frequency of the LFM wave in the digital reception signal.

[0056] The signal identification device 1 according to the first embodiment includes a sweep period estimation unit 12a that estimates a sweep period in which a sweep of a linear frequency modulated (LFM) wave is repeated in a digital received signal obtained by converting an incoming wave consisting of an LFM wave into a digital signal, and a sweep period estimation value τ ｉ a chirp rate estimator 12b that estimates a chirp rate, which is the frequency change rate of the LFM wave in the digital received signal, using ^; and a sweep period estimate value τ ｉ ^ and the chirp rate estimated value μ estimated by the chirp rate estimator 12b ｉ Since the system is equipped with an initial frequency estimation unit 12c that estimates the initial frequency of the LFM wave in the digital received signal using ^, even if the arriving wave includes LFM waves with multiple periodicities, the sweep period, chirp rate, and initial frequency, which are parameters of the LFM wave, can be estimated with a low amount of calculation by combining one-dimensional searches.

[0057] Embodiment 2 A signal identification and suppression device 3 according to embodiment 2 will be described using Figures 5 and 6. The signal identification and suppression device 3 according to embodiment 2 differs from the signal identification device 1 according to embodiment 1 in that a signal suppression device 2 that performs processing to suppress unwanted incoming waves is added, but is otherwise similar. Note that in Figures 5 and 6, the same reference numerals as those used in Figures 1 to 4 indicate the same or corresponding parts.

[0058] The signal identification and suppression device 3 according to the second embodiment uses the sweep period estimate value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ The signal suppression device 2 generates an incoming wave replica using ^, and subtracts the generated incoming wave replica from the digital received signal from the reception processing unit 11 to output a digital received signal in which the LFM waves in the unwanted incoming waves have been suppressed. That is, the signal discrimination and suppression device 3 according to the second embodiment outputs a digital received signal in which the unwanted incoming waves have been suppressed from the digital received signal from the reception processing unit 11.

[0059] A signal identification and suppression device 3 according to the second embodiment includes a signal identification device 1 and a signal suppression device 2. The signal identification device 1 includes a reception processing unit 11, a signal identification unit 12, and a storage unit 13. The reception processing unit 11, the signal identification unit 12, and the storage unit 13 of the signal identification device 1 are the same as the reception processing unit 11, the signal identification unit 12, and the storage unit 13 in the signal identification device 1 according to the first embodiment.

[0060] The signal discrimination and suppression device 3 is configured to obtain the sweep period estimated value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ Since the parameters of ^ are not output to the display device 200 and the downstream signal processing device 300, the identification result output unit 14 in the signal identification device 1 according to embodiment 1 is not required.

[0061] Also, the initial frequency estimate f obtained by the initial frequency estimator 12c in the signal discriminator 12 ｉ is stored in the storage unit 13. The sweep period estimate value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ If it is necessary to output the parameters of {circumflex over (^)} to the display device 200 and the signal processing device 300 at the subsequent stage, a classification result output unit 14 is provided.

[0062] The signal suppression device 2 includes a suppression unit 21 and a signal output unit 22. The suppression unit 21 includes a replica restoration unit 21a and a difference processing unit 21b. The replica restoration unit 21a restores the sweep period estimate value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ Based on this, the complex amplitude of the LFM wave in the digital received signal from the receiving processing unit 11 is estimated for the number of waves indicated by the wave number estimation value Nw^, a complex amplitude vector corresponding to the estimated complex amplitude estimation value is obtained, and a replica of the LFM wave is restored.

[0063] The complex amplitude of the LFM wave is estimated by, for example, a sparse restoration method. When Orthogonal Matching Pursuit (OMP) is used, the estimation is performed as follows. In the replica restoration unit 21a, OMP uses a greedy method to solve the optimization problem shown in the following equation (16), and obtains a complex amplitude vector (symbol shown in the following equation (15); in the following explanation, the obtained complex amplitude vector will be referred to as "x^").

[0064]

[0065] y is an observation vector, Φ is a dictionary matrix, and x is an unknown complex amplitude vector. Note that the observation vector y is the number of samples N extracted from the digital received signal from the reception processing unit 11. ４ is the vector represented by the digital signal.

[0066] The dictionary matrix Φ is the sweep period estimate τ obtained by the signal identification device 1 shown in the following equation (17): ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ It is a matrix using ^.

[0067]

[0068] In the above formula (17), N ３ is the number of samples extracted from the digital received signal from the receiving processing unit 11, and ω ｉ (n) is expressed by the following equation (18).

[0069]

[0070] In the above equation (18), the symbols shown in the following (19) and (20) are relative initial frequency estimates within the chirp band, fIF is the intermediate frequency, mod(·) in the second item on the right side is a modulo operation, and [·] in the third item on the right side is a floor function. In the following explanation, the relative initial frequency estimate within the chirp band will be referred to as "f' ｉ It is written as "^".

[0071]

[0072] Relative initial frequency estimate f' within the chirp band ｉ ^ is the initial frequency estimate f ｉ ^ is calculated using the following equation (21).

[0073]

[0074] The replica restoration unit 21a performs OMP iterations for each wave number i (= 1, 2, ... Nw^) for the number of waves in the wave number estimate value Nw^ acquired by the sweep period estimation unit 12, and obtains a complex amplitude vector x^ corresponding to the complex amplitude estimate value of each LFM wave. The replica restoration unit 21a restores the LFM wave replica using the dictionary matrix Φ and the complex amplitude vector x^ according to the following equation (23), and generates a restored replica (symbol shown in the following (22); in the following explanation, the restored replica will be referred to as "γ^") for the arriving wave.

[0075]

[0076] Even if the dictionary matrix Φ contains LFM wave components that are not contained in the digital received signal from the reception processing unit 11, the complex amplitude estimates corresponding to those components will have very small values, and therefore their influence can be suppressed. Here, a threshold may be set for the complex amplitude estimates, and replicas with complex amplitude estimates below the threshold may be deleted from the dictionary matrix Φ.

[0077] The differential processing unit 21b subtracts the restored replica γ^ for the incoming wave generated by the replica restoration unit 21a from the digital received signal from the reception processing unit 11 to obtain a digital received signal in which LFM waves in the unwanted incoming waves have been suppressed from the digital received signal. The digital received signal obtained by the differential processing unit 21b is a digital received signal in which undesired LFM waves contained in the digital received signal from the reception processing unit 11, i.e., LFM waves in the unwanted incoming waves, have been suppressed, so interference from undesired LFM waves can be suppressed. The signal output unit 22 outputs the digital received signal in which LFM waves in the unwanted incoming waves have been suppressed obtained by the differential processing unit 21b to the signal processing device 400.

[0078] The signal suppression device 2 in the signal identification and suppression device 3 according to the second embodiment is also realized by the hardware configuration of a computer shown in Fig. 3 together with the signal identification device 1 according to the first embodiment. That is, the signal identification and suppression device 3 according to the second embodiment is realized by the hardware configuration of a computer shown in Fig. 3.

[0079] The CPU 1A calculates the sweep period estimate τ based on the program loaded in the auxiliary storage device 1B. ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ It functions as a signal discrimination device 1 that acquires the ^, and as a signal suppression device 2 that acquires the digital received signal after suppressing the LFM wave.

[0080] Although the signal identification / suppression device 3 is realized by a computer hardware configuration, the receiving processing unit 11, signal identification unit 12, memory unit 13, and identification result output unit 14 that constitute the signal identification device 1, as well as the suppression unit 21 and signal output unit 22 that constitute the signal suppression device 2, may each be realized by dedicated hardware.

[0081] Next, the operation of the signal identification and suppression device 3 according to embodiment 2 will be described with reference to Fig. 6. Steps ST11 to ST14 are operations performed by the signal identification device 1, and are the same as the operations performed by the signal identification device 1 according to embodiment 1, that is, steps ST11 to ST14, which are the parameter identification method.

[0082] The sweep period estimate value τ for wave numbers 1 to Nŵ acquired in step ST14 and stored in the storage unit 13 ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ In step ST21, the replica restoration unit 21a estimates the complex amplitude vector x^ by the above equation (16) using complex amplitude estimation using the sparse restoration method.

[0083] In step ST21, before estimating the complex amplitude vector x̂, the relative initial frequency estimate f′ shown in the above equation (20) is calculated. ｉ^, ω shown in the above equation (18) ｉ (n), and the sweep period estimate τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ The dictionary matrix Φ shown in the above equation (17) using ^ is obtained, and the number of samples N extracted from the digital received signal from the receiving processing unit 11 is ４ The complex amplitude vector x̂ is estimated using the observation vector y, which is the vector indicated by the digital signal of and the dictionary matrix Φ.

[0084] Step ST21 is executed for each wave number i (= 1, 2, ... Nw^) for the number of waves of the wave number estimate value Nw^ acquired by the sweep period estimation unit 12, to obtain a complex amplitude vector x^ corresponding to the complex amplitude estimate value of each LFM wave. Step ST21 is a step in which the replica restoration unit 21a estimates the complex amplitude vector x^.

[0085] In step ST22, the replica restoration unit 21a restores the LFM wave replica using the dictionary matrix Φ and the complex amplitude vector x^ according to the above equation (23), thereby generating a restored replica γ^ for the arriving wave. Step ST22 is a step in which the replica restoration unit 21a generates the restored replica γ^.

[0086] In step ST23, the differential processing unit 21b subtracts the restored replica γ^ for the incoming wave generated by the replica restoration unit 21 from the digital received signal from the reception processing unit 11 to obtain a digital received signal in which the LFM waves in the unwanted incoming waves have been suppressed from the digital received signal. The digital received signal obtained by the differential processing unit 21b is a digital received signal in which the undesired LFM waves contained in the digital received signal from the reception processing unit 11, i.e., the LFM waves in the unwanted incoming waves, have been suppressed, so that interference from the undesired LFM waves can be suppressed.

[0087] In step ST23, the differential processing unit 21b obtains a digital received signal in which the LFM waves in the unwanted incoming waves have been suppressed from the digital received signal. The digital received signal in which the LFM waves have been suppressed obtained by the differential processing unit 21b is output from the signal output unit 22 to the signal processing device 400.

[0088] Steps ST21 to ST23 constitute a method for suppressing LFM waves in unwanted incoming waves. In combination with the parameter identification method shown in steps ST11 to ST14, steps ST11 to ST14 and steps ST21 to ST23 constitute a received signal acquisition method for acquiring a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

[0089] The method of suppressing LFM waves in unwanted incoming waves in steps ST21 to ST23 is performed by the CPU 1A executing processing in accordance with an LFM wave suppression program stored in the main storage device 1C. That is, the LFM wave suppression program stored in the main storage device 1C includes the steps of generating a restored replica for the incoming wave by complex amplitude estimation using a sparse restoration method based on a sweep period estimate that estimates the sweep period of a linear frequency modulated (LFM) wave sweep in a digital received signal obtained by converting an incoming wave consisting of an LFM wave into a digital signal, a chirp rate estimate that estimates the chirp rate which is the frequency change rate of the LFM wave in the digital received signal, and an initial frequency estimate that estimates the initial frequency of the LFM wave in the digital received signal, and a step of subtracting the generated restored replica for the incoming wave from the digital received signal to obtain a digital received signal in which the LFM waves in the unwanted incoming waves are suppressed.

[0090] The signal suppression device 2 in the signal identification and suppression device 3 according to the second embodiment includes a replica restoration unit 21a that generates a restored replica for an arriving wave by complex amplitude estimation using a sparse restoration method, based on the sweep period estimate, chirp rate estimate, and initial frequency estimate obtained by the signal identification device 1, and a differential processing unit 21b that subtracts the restored replica for the arriving wave generated by the replica restoration unit 21a from the digital received signal from the reception processing unit 11 to obtain a digital received signal in which LFM waves in unwanted arriving waves are suppressed. Therefore, even if the arriving waves include LFM waves with multiple periodicities, it is possible, with a low amount of calculation, to suppress distortion of the desired digital received signal in the digital received signal from the reception processing unit 11 while suppressing undesired LFM waves included in the digital received signal from the reception processing unit 11, in other words, it is possible to obtain a digital received signal in which LFM waves in unwanted arriving waves are suppressed, and interference from undesired LFM waves can be suppressed.

[0091] Embodiment 3. A signal identification and suppression device 3A according to embodiment 3 will be described using Figures 7 and 8. The signal identification and suppression device 3A according to embodiment 3 differs from the signal identification device 1 according to embodiment 1 in that a signal suppression device 2A that performs processing to suppress unwanted incoming waves is added, but the remaining points are the same. Note that in Figures 7 and 8, the same reference numerals as those in Figures 1 to 4 indicate the same or corresponding parts.

[0092] The signal identification and suppression device 3A according to the third embodiment uses the sweep period estimated value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ The signal suppression device 2A calculates a filter using the formula ^, performs filter processing on the digital received signal from the reception processing unit 11, and outputs a digital received signal in which the LFM waves in the unwanted incoming waves are suppressed. That is, the signal discrimination and suppression device 3A according to the third embodiment outputs a digital received signal in which the unwanted incoming waves are suppressed from the digital received signal from the reception processing unit 11.

[0093] A signal identification and suppression device 3A according to the third embodiment includes a signal identification device 1 and a signal suppression device 2A. The signal identification device 1 includes a reception processing unit 11, a signal identification unit 12, and a storage unit 13. The reception processing unit 11, the signal identification unit 12, and the storage unit 13 of the signal identification device 1 are the same as the reception processing unit 11, the signal identification unit 12, and the storage unit 13 in the signal identification device 1 according to the first embodiment.

[0094] The sweep period estimate value τ acquired by the signal identification device 1 as the signal identification and suppression device 3A is ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ Since the parameters of ^ are not output to the display device 200 and the downstream signal processing device 300, the identification result output unit 14 in the signal identification device 1 according to embodiment 1 is not required.

[0095] Also, the initial frequency estimate f obtained by the initial frequency estimator 12c in the signal discriminator 12 ｉ is stored in the storage unit 13. The sweep period estimate value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ If it is necessary to output the parameters of {circumflex over (^)} to the display device 200 and the signal processing device 300 at the subsequent stage, a classification result output unit 14 is provided.

[0096] The signal suppression device 2A includes a suppression unit 21A and a signal output unit 22. The suppression unit 21A includes a filter calculation unit 21Aa and a filter processing unit 21Ab. The filter calculation unit 21Aa calculates the sweep period estimate value τ ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ Based on the wavenumber estimate Nw^, a filter in the fractional Fourier domain is calculated for the number of waves indicated by the wavenumber estimate Nw^.

[0097] The filter calculated by the filter calculation unit 21Aa may be, for example, a notch filter. ｉ ^, chirp rate estimate μ ｉ^, and the initial frequency estimate f ｉ After performing one-dimensional FRFT on the digital received data based on ^, the peak of the FRFT is searched for one-dimensionally, and the argument u corresponding to the peak is ｉ This is obtained by calculating a notch filter that forms a null in

[0098] The filter processing unit 21Ab transforms the digital received signal from the reception processing unit 11 into the fractional Fourier domain, and applies a filter calculated by the filter calculation unit 21Aa to the received signal transformed into the fractional Fourier domain in the FRFD domain to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

[0099] That is, the filter processing unit 21Ab applies the filter calculated by the filter calculation unit 21Aa to the digital received signal from the reception processing unit 11, and then applies an inverse fractional Fourier transform (IFRFT) to obtain the filtered digital received signal. ｉ This corresponds to the process of inverting the sign of and applying the FRFT of the above equation (8).

[0100] The digital received signal acquired by the filter processing unit 21Ab is a digital received signal in which undesired LFM waves contained in the digital received signal from the reception processing unit 11, i.e., LFM waves in unwanted incoming waves, have been suppressed, making it possible to suppress interference from undesired LFM waves. The signal output unit 22 outputs the digital received signal acquired by the filter processing unit 21Ab in which LFM waves in unwanted incoming waves have been suppressed to the signal processing device 400.

[0101] The signal suppression device 2A in the signal identification and suppression device 3 according to the third embodiment is also realized by the hardware configuration of a computer shown in Fig. 3 together with the signal identification device 1 according to the first embodiment. That is, the signal identification and suppression device 3A according to the third embodiment is realized by the hardware configuration of a computer shown in Fig. 3.

[0102] The CPU 1A calculates the sweep period estimate τ based on the program loaded in the auxiliary storage device 1B.ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ It functions as a signal discrimination device 1 that acquires the ^, and functions as a signal suppression device 2A that acquires the digital received signal after suppressing the LFM wave.

[0103] Although the signal identification / suppression device 3A is realized by a computer hardware configuration, the receiving processing unit 11, signal identification unit 12, memory unit 13, and identification result output unit 14 that constitute the signal identification device 1, as well as the suppression unit 21A and signal output unit 22 that constitute the signal suppression device 2, may each be realized by dedicated hardware.

[0104] Next, the operation of the signal identification and suppression device 3A according to the third embodiment will be described with reference to Fig. 8. Steps ST11 to ST14 are operations performed by the signal identification device 1, and are the same as steps ST11 to ST14 performed by the signal identification device 1 according to the first embodiment.

[0105] The sweep period estimate value τ for wave numbers 1 to Nŵ acquired in step ST14 and stored in the storage unit 13 ｉ ^, chirp rate estimate μ ｉ ^, and the initial frequency estimate f ｉ In step ST31, the filter calculation unit 21Aa calculates a filter, for example, a notch filter in the fractional Fourier domain, using {circumflex over (^)}. Step ST31 is a step in which the filter calculation unit 21Aa calculates a filter in the fractional Fourier domain.

[0106] In step ST32, the filter processing unit 21Ab applies the filter calculated by the filter calculation unit 21Aa to the digital received signal from the reception processing unit 11 to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed. In step ST32, in the fractional Fourier domain, the filter calculated by the filter calculation unit 21Aa is applied to the digital received signal from the reception processing unit 11 and an IFRFT is performed to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

[0107] In step ST32, the filter processing unit 21Ab obtains a digital received signal in which the LFM waves in the unwanted incoming waves have been suppressed from the digital received signal. The digital received signal in which the LFM waves have been suppressed and obtained by the filter processing unit 21Ab is output from the signal output unit 22 to the signal processing device 400.

[0108] Steps ST31 and ST32 constitute a method for suppressing LFM waves in unwanted incoming waves. Steps ST11 to ST14 and steps ST31 and ST32, together with the parameter identification method shown in steps ST11 to ST14, constitute a received signal acquisition method for acquiring a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

[0109] The method of suppressing LFM waves in unwanted incoming waves in steps ST31 and ST32 is performed by the CPU 1A executing processing in accordance with an LFM wave suppression program stored in the main storage device 1C. That is, the LFM wave suppression program stored in the main storage device 1C includes the steps of calculating a filter in the fractional Fourier domain based on a sweep period estimate that estimates the sweep period of a linear frequency modulated (LFM) wave sweep in a digital received signal obtained by converting an incoming wave consisting of an LFM wave into a digital signal, a chirp rate estimate that estimates the chirp rate which is the frequency change rate of the LFM wave in the digital received signal, and an initial frequency estimate that estimates the initial frequency of the LFM wave in the digital received signal, and a step of applying a filter to the digital received signal in the fractional Fourier domain to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

[0110] The signal suppression device 2A in the signal identification and suppression device 3A according to the third embodiment includes a filter calculation unit 21Aa that calculates a filter in the fractional Fourier domain based on the sweep period estimate, chirp rate estimate, and initial frequency estimate acquired by the signal identification device 1, and a filter processing unit 21Ab that applies the filter calculated by the filter calculation unit 21Aa to the digital received signal from the reception processing unit 11 to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed. Therefore, even when the incoming waves include LFM waves with a plurality of periodicities, it is possible, with a low amount of calculation, to suppress distortion of the desired digital received signal in the digital received signal from the reception processing unit 11 while suppressing undesired LFM waves included in the digital received signal from the reception processing unit 11, in other words, it is possible to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed, and interference from undesired LFM waves can be suppressed.

[0111] It should be noted that the embodiments may be freely combined, any of the components of the embodiments may be modified, or any of the components of the embodiments may be omitted.

[0112] In the field of wireless communications, where LFM signals, also known as chirp signals, are used, the present invention is applied to a signal identification device that estimates and identifies the parameters of LFM waves, and a signal identification and suppression device that outputs a digital received signal in which unwanted incoming waves have been suppressed from a digital received signal due to incoming waves.

[0113] 1 signal identification device, 2, 2A signal suppression device, 3, 3A signal identification / suppression device, 11 reception processing unit, 12 signal identification unit, 12a sweep period estimation unit, 12b chirp rate estimation unit, 12c initial frequency estimation unit, 13 memory unit, 14 identification result output unit, 21, 21A suppression unit, 21a replica recovery unit, 21b difference processing unit, 21Aa filter calculation unit, 21Ab filter processing unit, 22 signal output unit.

Claims

1. A signal identification device comprising: a sweep period estimation unit that estimates the sweep period of a linear frequency modulated (LFM) wave in a digital received signal that is a digital signal obtained by converting an incoming wave consisting of a linear frequency modulated (LFM) wave; a chirp rate estimation unit that estimates the chirp rate, which is the rate of change of frequency of the LFM wave in the digital received signal; and an initial frequency estimation unit that estimates the initial frequency of the LFM wave in the digital received signal.

2. A signal identification device according to claim 1, wherein the sweep period estimation section calculates an autocorrelation function of an LFM wave in the digital received signal and estimates the sweep period using the calculated autocorrelation function.

3. A signal identification device according to claim 1, wherein the chirp rate estimation unit estimates the chirp rate using an extended autocorrelation function incorporating the sweep period estimated by the sweep period estimation unit.

4. A signal identification device according to any one of claims 1 to 3, wherein the estimation of the initial frequency by the initial frequency estimation unit is performed by a one-dimensional fractional Fourier transform (FRFT) using the sweep period estimated by the sweep period estimation unit and the chirp rate estimated by the chirp rate estimation unit.

5. A signal suppression device comprising: a replica restoration unit that generates a restored replica for an arriving wave by complex amplitude estimation using a sparse restoration method based on: a sweep period estimate that estimates the sweep period of a linear frequency modulated (LFM) wave sweep repeated in a digital received signal obtained by converting an arriving wave consisting of a linear frequency modulated (LFM) wave into a digital signal; a chirp rate estimate that estimates the chirp rate which is the frequency change rate of the LFM wave in the digital received signal; and an initial frequency estimate that estimates the initial frequency of the LFM wave in the digital received signal; and a differential processing unit that subtracts the restored replica for the arriving wave generated by the replica restoration unit from the digital received signal to obtain a digital received signal in which LFM waves in unwanted arriving waves are suppressed.

6. A signal suppression device according to claim 5, wherein the complex amplitude estimation in the replica restoration unit is performed by orthogonal matching pursuit.

7. A signal suppression device comprising: a filter calculation unit that calculates a filter in the fractional Fourier domain based on a sweep period estimate that estimates the sweep period of a linear frequency modulated (LFM) wave sweep in a digital received signal obtained by converting an incoming wave consisting of a linear frequency modulated (LFM) wave into a digital signal, a chirp rate estimate that estimates the chirp rate which is the frequency change rate of the LFM wave in the digital received signal, and an initial frequency estimate that estimates the initial frequency of the LFM wave in the digital received signal; and a filter processing unit that applies the filter calculated by the filter calculation unit to the digital received signal to obtain a digital received signal in which LFM waves in unwanted incoming waves are suppressed.

8. A signal identification and suppression device comprising: a signal identification device according to any one of claims 1 to 4; and a signal suppression device according to any one of claims 5 to 7.

9. A parameter identification method for identifying parameters of a linear frequency modulated (LFM) wave using a signal identification device having a sweep period estimation unit, a chirp rate estimation unit, and an initial frequency estimation unit, comprising: a step in which the sweep period estimation unit estimates the sweep period at which the sweep of the LFM wave is repeated in a digital received signal obtained by converting an incoming wave consisting of an LFM wave into a digital signal; a step in which the chirp rate estimation unit estimates the chirp rate, which is the rate of change of frequency of the LFM wave in the digital received signal; and a step in which the initial frequency estimation unit estimates the initial frequency of the LFM wave in the digital received signal.

10. A parameter identification program for an LFM wave that causes a computer to execute the following steps: a procedure for estimating the sweep period at which the sweep of an LFM wave is repeated in a digital received signal obtained by converting an incoming wave consisting of a linear frequency modulated (LFM) wave into a digital signal; a procedure for estimating the chirp rate, which is the rate of change of frequency of the LFM wave in the digital received signal; and a procedure for estimating the initial frequency of the LFM wave in the digital received signal.

11. A recording medium storing a program that causes a computer to execute the following steps: a procedure for estimating the sweep period at which an LFM wave sweep is repeated in a digital received signal obtained by converting an incoming wave consisting of a linear frequency modulated (LFM) wave into a digital signal; a procedure for estimating the chirp rate, which is the rate of change of frequency of the LFM wave in the digital received signal; and a procedure for estimating the initial frequency of the LFM wave in the digital received signal.

Images