Frequency detection device, frequency detection method, and frequency detection circuit

Abstract

A frequency detection device (8) is configured to comprise: a frequency phase calculation unit (11) that converts first quantized data, which is a signal obtained by quantizing a signal to be measured, into a signal in the frequency domain, and calculates the frequency of the first quantized data and the phase of the first quantized data from a first frequency domain signal, which is the signal in the frequency domain; and a phase difference calculation unit (12) that converts second quantized data, which is data delayed from the first quantized data, into a signal in the frequency domain, calculates the phase of the second quantized data from a second frequency domain signal, which is the signal in the frequency domain, and calculates a phase difference between the phase of the first quantized data and the phase of the second quantized data. The frequency detection device (8) is also provided with: a folding-order calculation unit (13) for calculating the Nyquist folding order of the first quantized data on the basis of the first frequency domain signal, the second frequency domain signal, and the phase difference calculated by the phase difference calculation unit (12); and a frequency calculation unit (16) for calculating the frequency of the signal to be measured using the frequency calculated by the frequency phase calculation unit (11) and the Nyquist folding order calculated by the folding-order calculation unit (13).

Description

Frequency detection device, frequency detection method, and frequency detection circuit

[0001] This disclosure relates to a frequency detection device, a frequency detection method, and a frequency detection circuit.

[0002] There is a frequency detection circuit that calculates the frequency of a signal to be measured. As such a frequency detection circuit, for example, Patent Document 1 discloses a frequency detection circuit comprising a first sample-and-hold circuit, a second sample-and-hold circuit, and a frequency calculation circuit. The first sample-and-hold circuit undersamples the signal to be measured using a first clock signal and outputs the signal to be measured after undersampling. The second sample-and-hold circuit undersamples the signal to be measured using a second clock signal that has the same frequency as the first clock signal but a different phase from the first clock signal, and outputs the signal to be measured after undersampling. The frequency calculation circuit comprises a first quantizer that quantizes the output signal of the first sample-and-hold circuit and outputs first quantized data, which is the signal after quantization, and a second quantizer that quantizes the output signal of the second sample-and-hold circuit and outputs second quantized data, which is the signal after quantization. The frequency calculation circuit calculates the frequency of the first quantized data, calculates the order of undersampling for the signal under test from the phase difference between the phase of the first quantized data and the phase of the second quantized data, and calculates the frequency of the signal under test from the frequency of the first quantized data and the order of undersampling.

[0003] International Publication No. 2020-152764

[0004] In addition to the signal under measurement, other signals such as noise (hereinafter referred to as "interference signals") may be mixed into the frequency detection circuit. The phase difference calculated by the frequency calculation circuit of the frequency detection circuit disclosed in Patent Document 1 differs depending on whether or not interference signals are present. Therefore, when interference signals are present, there has been a problem in that the accuracy of calculating the frequency of the signal under measurement in the frequency calculation circuit may deteriorate.

[0005] This disclosure was made to solve the above-mentioned problems, and aims to provide a frequency detection device that can prevent deterioration of the frequency calculation accuracy of the signal under measurement even when interfering signals are mixed in with the signal under measurement.

[0006] The frequency detection device according to this disclosure includes a frequency-phase calculation unit that converts a first quantized data, which is a quantized signal of the signal to be measured, into a frequency domain signal, and calculates the frequency and phase of the first quantized data from the first frequency domain signal, which is a frequency domain signal; and a phase difference calculation unit that converts a second quantized data, which is data delayed compared to the first quantized data, into a frequency domain signal, calculates the phase of the second quantized data from the second frequency domain signal, which is a frequency domain signal, and calculates the phase difference between the phase of the first quantized data and the phase of the second quantized data. The frequency detection device also includes an aliasing order calculation unit that calculates the Nyquist aliasing order of the first quantized data based on the first frequency domain signal, the second frequency domain signal, and the phase difference calculated by the phase difference calculation unit; and a frequency calculation unit that calculates the frequency of the signal to be measured using the frequency calculated by the frequency-phase calculation unit and the Nyquist aliasing order calculated by the aliasing order calculation unit.

[0007] According to this disclosure, even if an interfering signal is mixed in with the signal under measurement, it is possible to prevent deterioration in the accuracy of frequency calculation in the signal under measurement.

[0008] This is a diagram showing a frequency detection circuit including a frequency detection device 8 according to Embodiment 1. This is a hardware diagram showing the hardware of the frequency detection device 8 according to Embodiment 1. This is a hardware diagram of a computer when the frequency detection device 8 is implemented by software or firmware, etc. This is a flowchart showing the frequency detection method, which is the processing procedure of the frequency detection device 8. This is an explanatory diagram showing the frequency, Nyquist frequency, and Nyquist zone of a clock signal. This is an explanatory diagram showing an example in which the frequency of the signal under test, which is present in the kth Nyquist zone, appears in the first Nyquist zone due to Nyquist aliasing by sampling. This is an explanatory diagram showing the phase rotation of the signal under test. This is the frequency f of the signal under test when an interfering signal is present. ＲＦ This is an explanatory diagram showing the estimated failure rate.

[0009] To provide a more detailed explanation of this disclosure, the forms for implementing this disclosure will be described below with reference to the attached drawings.

[0010] Embodiment 1. Figure 1 is a configuration diagram showing a frequency detection circuit including a frequency detection device 8 according to Embodiment 1. Figure 2 is a hardware configuration diagram showing the hardware of the frequency detection device 8 according to Embodiment 1. The frequency detection circuit shown in Figure 1 comprises a clock signal source 1, a distributor 2, a quantization unit 3, and a frequency detection device 8. The quantization unit 3 comprises delay circuits 4 and 5 and quantizers 6 and 7.

[0011] The clock signal source 1 is implemented by, for example, a voltage-controlled oscillator, a crystal oscillator, or a general-purpose signal generator. The clock signal source 1 has a frequency f ｃｌｋ The device generates a clock signal and outputs the clock signal to the quantization unit 3. The distributor 2 divides the signal to be measured, which is the signal to be frequency detected, into two. The distributor 2 outputs one of the divided signals to be measured to the delay circuit 4 of the quantization unit 3, and outputs the other divided signal to be measured to the delay circuit 5 of the quantization unit 3.

[0012] The quantization unit 3 quantizes one of the measured signals after distribution, and generates the first quantized data QD, which is the measured signal after quantization. １is output to the frequency detection device 8. The quantization unit 3 quantizes the other distributed signal under test, and the second quantized data QD, which is the quantized signal under test ２ is output to the frequency detection device 8. The second quantized data QD ２ is data delayed relative to the first quantized data QD １ .

[0013] The delay circuit 4 delays one of the distributed signals under test by a delay time τ １ , and outputs the delayed signal under test to the quantizer 6. The delay circuit 5 delays the other distributed signal under test by a delay time τ ２ , and outputs the delayed signal under test to the quantizer 7. The delay time τ ２ is longer than the delay time τ １ . When the delay circuit 4 is implemented by a line with a line length L １ and the delay circuit 5 is implemented by a line with a line length L ２ , the line length L ２ is longer than the line length L １ , and the delay time τ １ and the delay time τ ２ have a delay time difference Δτ expressed by (L ２ −L １ ) / v. v is the propagation speed of the signal under test in the delay circuits 4 and 5. In the frequency detection circuit shown in FIG. 1, the quantization unit 3 includes two delay circuits 4 and 5. However, this is merely an example, and the quantization unit 3 may, for example, be provided with the delay circuit 5 without the delay circuit 4. The delay time difference Δτ in this case is expressed by L ２ / v.

[0014] The quantizers 6 and 7 are implemented by, for example, an ADC (Analog to Digital Converter). The quantizer 6 uses a clock signal of frequency f output from the clock signal source 1 ｃｌｋ to quantize the signal under test delayed by the delay circuit 4. The quantizer 6 outputs the first quantized data QD, which is the quantized signal under test １ , to the frequency detection device 8. The quantizer 7 uses a clock signal of frequency f output from the clock signal source 1 ｃｌｋThe clock signal is used to quantize the signal under test after delay by the delay circuit 5. The quantizer 7 then processes the second quantized data QD, which is the signal under test after quantization. ２ This is output to the frequency detection device 8.

[0015] The frequency detection device 8 includes a frequency phase calculation unit 11, a phase difference calculation unit 12, an aliasing order calculation unit 13, and a frequency calculation unit 16. The frequency phase calculation unit 11 is implemented, for example, by the frequency phase calculation circuit 21 shown in Figure 2. The frequency phase calculation unit 11 receives the first quantized data QD from the quantizer 6. １ The frequency phase calculation unit 11 obtains the first quantization data QD １ The signal in the frequency domain (hereinafter referred to as "the first frequency domain signal y") １ It converts to ").

[0016] The frequency phase calculation unit 11 calculates the first frequency domain signal y １ From there, the first quantized data QD １ frequency f ｏｕｔ and the first quantized data QD １ Phase θ ｏｕｔ１ Each of these is calculated. Specifically, the frequency phase calculation unit 11 calculates the first frequency domain signal y １ From the frequency bins of the peak frequency components at, the frequency f of the first quantized data ｏｕｔ and the first quantized data QD １ Phase θ ｏｕｔ１ Each of these is identified. The frequency phase calculation unit 11 calculates the first frequency domain signal y １ The first quantization data QD is output to the folding order calculation unit 13. １ frequency f ｏｕｔ The frequency calculation unit 16 outputs information indicating the first quantization data QD １ Phase θ ｏｕｔ１ Information indicating this is output to the phase difference calculation unit 12.

[0017] The phase difference calculation unit 12 is implemented, for example, by the phase difference calculation circuit 22 shown in Figure 2. The phase difference calculation unit 12 receives the second quantization data QD from the quantizer 7. ２The frequency phase calculation unit 11 obtains the first quantization data QD １ Phase θ ｏｕｔ１ Information indicating the second quantization data QD is obtained. The phase difference calculation unit 12 obtains the second quantization data QD ２ The signal in the frequency domain (hereinafter referred to as "the second frequency domain signal y") ２ The phase difference calculation unit 12 converts the second frequency domain signal y ２ From there, the second quantized data QD ２ Phase θ ｏｕｔ２ Specifically, the phase difference calculation unit 12 calculates the second frequency domain signal y ２ From the frequency bins of the peak frequency components in the second quantized data QD ２ Phase θ ｏｕｔ２ The phase difference calculation unit 12 identifies the first quantization data QD １ Phase θ ｏｕｔ１ and the second quantized data QD ２ Phase θ ｏｕｔ２ The phase difference Δθ is calculated from the second frequency domain signal y. ２ The information indicating the phase difference Δθ is output to the folding order calculation unit 13.

[0018] The folding order calculation unit 13 is implemented, for example, by the folding order calculation circuit 23 shown in Figure 2. The folding order calculation unit 13 includes a matrix calculation unit 14 and an order calculation processing unit 15. The folding order calculation unit 13 receives the first frequency domain signal y from the frequency phase calculation unit 11. １ The phase difference calculation unit 12 obtains the second frequency domain signal y ２ The first frequency domain signal y １ and the second frequency domain signal y ２ Based on the phase difference Δθ, the first quantized data QD １ The Nyquist folding order Fo is calculated. The folding order calculation unit 13 calculates the first quantized data QD １ The Nyquist aliasing order Fo is output to the frequency calculation unit 16.

[0019] The matrix calculation unit 14 obtains a first frequency domain signal y from the frequency phase calculation unit 11, and a second frequency domain signal y from the phase difference calculation unit 12. ２ The matrix calculation unit 14 obtains information indicating the phase difference Δθ and the k-th order Nyquist component x based on the phase difference Δθ calculated by the phase difference calculation unit 12. ｋ (k=1, ..., K) is the first frequency domain signal y １ and a second frequency domain signal y ２ A matrix W is calculated for each of the following conversions. K is an integer greater than or equal to 2. The matrix calculation unit 14 calculates the second frequency domain signal y ２ Each of the matrix W is output to the degree calculation processing unit 15.

[0020] The order calculation processing unit 15 receives the first frequency domain signal y from the frequency phase calculation unit 11. １ The order calculation processing unit 15 obtains the second frequency domain signal y from the matrix calculation unit 14. ２ The matrix W is obtained. The order calculation processing unit 15 obtains the matrix W and the Nyquist component x. ｋ The result of multiplying by (k=1, ..., K) is the first frequency domain signal y １ and a second frequency domain signal y ２ Under the condition that each of the following conditions is met, the 1st to Kth order Nyquist component x １ ~x Ｋ Based on the Manhattan distance, the 1st to Kth order Nyquist component x １ ~x Ｋ Among them, the first quantized data QD １ The Nyquist component x corresponds to the Nyquist folding order Fo. ｋ The order calculation processing unit 15 identifies the Nyquist component x of order 1 to K. １ ~x Ｋ Among them, the 1st to Kth order Nyquist component x １ ~x Ｋ The Nyquist component that minimizes the Manhattan distance is obtained using the first quantized data QD. １ The Nyquist component x corresponds to the Nyquist folding order Fo. ｋ It is identified as such. The order calculation processing unit 15 calculates the identified Nyquist component x ｋto obtain the Nyquist aliasing order Fo. The order calculation processing unit 15 outputs the Nyquist aliasing order Fo to the frequency calculation unit 16.

[0021] The frequency calculation unit 16 is implemented by, for example, the frequency calculation circuit 24 shown in FIG. 2. The frequency calculation unit 16 receives first quantized data QD from the frequency and phase calculation unit 11 １ with a frequency f ｏｕｔ , and acquires information indicating the Nyquist aliasing order Fo from the aliasing order calculation unit 13. The frequency calculation unit 16 uses the first quantized data QD １ with a frequency f ｏｕｔ and the Nyquist aliasing order Fo to calculate the frequency f of the signal under measurement ＲＦ .

[0022] In FIG. 1, it is assumed that each of the frequency and phase calculation unit 11, the phase difference calculation unit 12, the aliasing order calculation unit 13, and the frequency calculation unit 16, which are constituent elements of the frequency detection device 8, is implemented by dedicated hardware as shown in FIG. 2. That is, it is assumed that the frequency detection device 8 is implemented by a frequency and phase calculation circuit 21, a phase difference calculation circuit 22, an aliasing order calculation circuit 23, and a frequency calculation circuit 24. Each of the frequency and phase calculation circuit 21, the phase difference calculation circuit 22, the aliasing order calculation circuit 23, and the frequency calculation circuit 24 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.

[0023] The components of the frequency detection device 8 are not limited to those implemented by dedicated hardware; the frequency detection device 8 may also be implemented by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the computer's memory. A computer refers to hardware that executes programs, and includes, for example, a CPU (Central Processing Unit), GPU (Graphics Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor).

[0024] Figure 3 is a hardware configuration diagram of a computer when the frequency detection device 8 is implemented by software or firmware. When the frequency detection device 8 is implemented by software or firmware, a program that causes the computer to execute the respective processing procedures in the frequency phase calculation unit 11, the phase difference calculation unit 12, the aliasing order calculation unit 13, and the frequency calculation unit 16 is stored in memory 31. The computer's processor 32 then executes the program stored in memory 31.

[0025] Furthermore, Figure 2 shows an example in which each component of the frequency detection device 8 is implemented by dedicated hardware, and Figure 3 shows an example in which the frequency detection device 8 is implemented by software or firmware, etc. However, this is merely one example, and some components of the frequency detection device 8 may be implemented by dedicated hardware, while the remaining components may be implemented by software or firmware, etc.

[0026] Next, the operation of the frequency detection circuit shown in FIG. 1 will be described. FIG. 4 is a flowchart showing a frequency detection method that is a processing procedure of the frequency detection device 8. When a signal under measurement is supplied as a signal whose frequency is to be detected, the distributor 2 distributes the signal under measurement into two parts. The distributor 2 outputs one of the distributed signals under measurement to the delay circuit 4 of the quantization unit 3, and outputs the other of the distributed signals under measurement to the delay circuit 5 of the quantization unit 3.

[0027] The delay circuit 4 delays the one distributed signal under measurement by a delay time τ １ , and outputs the delayed signal under measurement to the quantizer 6. The delay circuit 5 delays the other distributed signal under measurement by a delay time τ ２ , and outputs the delayed signal under measurement to the quantizer 7. The delay time τ １ and the delay time τ ２ have a delay time difference Δτ, which is expressed as (L ２ -L １ ) / v, as described above.

[0028] The quantizer 6 quantizes the delayed signal under measurement output from the delay circuit 4 using a clock signal of frequency f ｃｌｋ output from the clock signal source 1. The quantizer 6 outputs first quantization data QD, which is the quantized signal under measurement, to the frequency and phase calculation unit 11. １ At this time, when the frequency f ＲＦ of the signal under measurement is higher than the Nyquist frequency f ｃｌｋ which is half of the frequency f ｃｌｋ / 2, a phenomenon called aliasing occurs. As a result, the calculation result of the frequency f １ of the first quantization data QD ｏｕｔ by the frequency and phase calculation unit 11 indicates a frequency present in the first Nyquist zone lower than the Nyquist frequency f ｃｌｋ / 2.

[0029] FIG. 5 is an explanatory diagram showing the frequency of the clock signal, the Nyquist frequency, and Nyquist zones. In the example of FIG. 5, since the frequency f ｃｌｋ of the clock signal is α [MHz], the Nyquist frequency f ｃｌｋ / 2 is α / 2 [MHz]. The first Nyquist zone is the bandwidth from DC (direct current) to α / 2 [MHz], the second Nyquist zone is the bandwidth from α / 2 [MHz] to α [MHz], the third Nyquist zone is the bandwidth from α [MHz] to 3α / 2 [MHz], the fourth Nyquist zone is the bandwidth from 3α / 2 [MHz] to 2α [MHz], and the fifth Nyquist zone is the bandwidth from 2α [MHz] to 5α / 2 [MHz]. The frequency f of the signal under measurement. ＲＦ However, if the frequency is in the second to fifth Nyquist zones, the first quantization data QD is calculated by the frequency phase calculation unit 11. １ frequency f ｏｕｔ The calculation result indicates the frequencies present in the first Nyquist zone.

[0030] The quantizer 7 receives the frequency f output from the clock signal source 1. ｃｌｋ The clock signal is used to quantize the signal under test after delay by the delay circuit 5. The quantizer 7 then processes the second quantized data QD, which is the signal under test after quantization. ２ The second quantization data QD is output to the phase difference calculation unit 12. ２ This is the first quantized data QD １ This data is more delayed than the above.

[0031] The frequency phase calculation unit 11 receives the first quantized data QD from the quantizer 6. １ The frequency phase calculation unit 11 obtains, for example, the first quantization data QD １ By performing FFT (Fast Fourier Transform) on it, the first quantized data QD １ The first frequency domain signal y １ Convert to the first frequency domain signal y. １ As shown in equation (1) below, the k-th Nyquist component x ｋ It has (k=1, ..., K).

[0032]

[0033] The frequency phase calculation unit 11 calculates the first frequency domain signal y １ From the frequency bins of the peak frequency components in the first quantized data QD１ frequency f ｏｕｔ and the first quantized data QD １ Phase θ ｏｕｔ１ Identify each of them (step ST1 in Figure 4). From the frequency bins of the peak frequency components, obtain the first quantized data QD １ frequency f ｏｕｔ and the first quantized data QD １ Phase θ ｏｕｔ１ The process of identifying each of these is a well-known technique, so a detailed explanation will be omitted. The frequency phase calculation unit 11 calculates the first frequency domain signal y １ The first quantization data QD is output to the folding order calculation unit 13. １ frequency f ｏｕｔ The frequency calculation unit 16 outputs information indicating the first quantization data QD １ Phase θ ｏｕｔ１ Information indicating this is output to the phase difference calculation unit 12.

[0034] Figure 6 is an explanatory diagram illustrating an example where the frequency of the signal under test, which is located in the k-th Nyquist zone, appears in the first Nyquist zone due to Nyquist aliasing during sampling. In Figure 6, the horizontal axis represents frequency, and the vertical axis represents signal intensity. In the example in Figure 6, the frequency of the signal under test is located in the k-th Nyquist zone, and signals such as noise (hereinafter referred to as "interference signals") are located in the third Nyquist zone. When interference signals are present in addition to the signal under test, as shown in Figure 6, not only the frequency of the signal under test but also the frequency of the interference signals may appear in the first Nyquist zone due to Nyquist aliasing during sampling. The first quantized data QD is calculated by the frequency phase calculation unit 11. １ frequency f ｏｕｔ This is the frequency that appears in the first Nyquist zone due to Nyquist folding, and is the actual frequency f of the signal under measurement. ＲＦ It is different.

[0035] The phase difference calculation unit 12 receives the second quantization data QD from the quantizer 7. ２ The frequency phase calculation unit 11 obtains the first quantization data QD １ Phase θｏｕｔ１ Information indicating the following is obtained. The phase difference calculation unit 12 obtains, for example, the second quantization data QD ２ By performing FFT on this, the second quantized data QD ２ to the second frequency domain signal y ２ Convert to the second frequency domain signal y. ２ The k-th Nyquist component x is as shown in equation (2) below. ｋ It has (k=1, ..., K).

[0036] In equation (2), iφ ｋ is the first frequency domain signal y １ The k-th Nyquist component x that it possesses ｋ The phase of and the second frequency domain signal y ２ The k-th Nyquist component x that it possesses ｋ This is the phase difference with respect to the phase of [the other element].

[0037] The phase difference calculation unit 12 calculates the second frequency domain signal y ２ From the frequency bins of the peak frequency components in the second quantized data QD ２ Phase θ ｏｕｔ２ Identify the second quantization data QD from the frequency bins of the peak frequency components. ２ Phase θ ｏｕｔ２ The process of identifying the first quantized data QD is a well-known technique, so a detailed explanation will be omitted. The phase difference calculation unit 12 calculates the first quantized data QD as shown in equation (3) below. １ Phase θ ｏｕｔ１ and the second quantized data QD ２ Phase θ ｏｕｔ２ The phase difference Δθ (= iφ) ｋ Calculate (Step ST2 in Figure 4). Δθ = iφ ｋ = θ ｏｕｔ２ -θ ｏｕｔ１　　　　　 (3) The phase difference calculation unit 12 calculates the second frequency domain signal y ２ and phase difference iφ ｋ Each piece of information indicating this is output to the line-folding order calculation unit 13.

[0038] The matrix calculation unit 14 receives the first frequency domain signal y from the frequency phase calculation unit 11. １The matrix calculation unit 14 obtains the second frequency domain signal y from the phase difference calculation unit 12. ２ and phase difference iφ ｋ The matrix calculation unit 14 obtains each of the pieces of information that indicate the phase difference iφ ｋ Based on this, the k-th Nyquist component x ｋ (k=1, ..., K) is the first frequency domain signal y １ and a second frequency domain signal y ２ Calculate the matrix W to transform each of them (step ST3 in Figure 4).

[0039] The matrix W calculated by the matrix calculation unit 14 has a phase difference iφ as shown in the following equation (4). １ ~iφ Ｋ This is a matrix that includes terms relating to x. Equation (5) below is the k-th Nyquist component x ｋ (k=1, ..., K) is the first frequency domain signal y １ and a second frequency domain signal y ２ The formulas for converting to each of these are shown. First frequency domain signal y １ The amplitude and phase θ ｏｕｔ１ It contains information about the second frequency domain signal y ２ The amplitude and phase θ ｏｕｔ２ It contains the information of the second frequency domain signal y ２ Each of the matrix W is output to the degree calculation processing unit 15.

[0040]

[0041] The order calculation processing unit 15 receives the first frequency domain signal y from the frequency phase calculation unit 11. １ The order calculation processing unit 15 obtains the second frequency domain signal y from the matrix calculation unit 14. ２ The order calculation processing unit 15 obtains the matrix W and the Nyquist component x as shown in the following equation (6). ｋ The result of multiplying by (k=1, ..., K) is the first frequency domain signal y １ and a second frequency domain signal y ２ Under the condition that each of the following conditions is met, the 1st to Kth order Nyquist component x １ ~xＫ Among them, the 1st to Kth order Nyquist component x １ ~x Ｋ The Nyquist component x that minimizes the Manhattan distance ｋ Identify the Nyquist component x that minimizes the Manhattan distance. ｋ This is the first quantized data QD １ This is the Nyquist component corresponding to the Nyquist folding order Fo.

[0042]

[0043] In addition to the signal being measured, the presence of an interfering signal can result in, for example, the Nyquist component x of the signal being measured, as shown in Figure 6. ｋ In addition, the Nyquist component x of the contamination signal ３ This may appear, but the Nikest ingredient x １ ~x Ｋ Most of it becomes approximately 0. That is, the Nyquist component x １ ~x Ｋ Among them, the Nyquist component x of the signal under measurement ｋ and the Nyquist component of the contamination signal x ３ Excluding the above, the Nyquist component x is approximately 0. Therefore, the Nyquist component x １ ~x Ｋ The solution that minimizes the Manhattan distance is spars, and therefore, according to equation (6), the first quantized data QD １ The Nyquist component x corresponds to the Nyquist folding order Fo. ｋ It is possible to explore it.

[0044] The search for the solution that minimizes the Manhattan distance can generally be performed using the LASSO (Least Absolute Shrinkage and Selection Operator) optimization algorithm. Therefore, the order calculation processing unit 15 calculates the Nyquist components x of order 1 to K. １ ~x Ｋ Under the condition that they are independent of each other, the Nyquist component x that minimizes the error function E(x) shown in equation (9) below ｋ The Nyquist component x that minimizes the Manhattan distance. ｋIt may also be used to search as follows. Since the error function E(x) is convex, the order calculation processing unit 15 calculates the Nyquist component x with less computation. ｋ It is possible to explore it.

[0045] In equation (9), λ is a hyperparameter.

[0046] The order calculation unit 15 calculates the Nyquist component x that minimizes the Manhattan distance. ｋ Once identified, the identified Nyquist component x is used as the Nyquist folding order Fo. ｋ The k corresponding to this is determined (step ST4 in Figure 4). The order calculation processing unit 15 determines the identified Nyquist component x ｋ The corresponding k is output to the frequency calculation unit 16 as the Nyquist folding order Fo.

[0047] The frequency calculation unit 16 receives the first quantization data QD from the frequency phase calculation unit 11. １ frequency f ｏｕｔ Information indicating the frequency is obtained. The frequency calculation unit 16 obtains the Nyquist aliasing order Fo from the aliasing order calculation unit 13. The frequency f of the clock signal oscillated by the clock signal source 1 ｃｌｋ Information indicating this is stored, for example, in the internal memory of the frequency calculation unit 16. Alternatively, the frequency f ｃｌｋ Information indicating this is provided, for example, to the frequency calculation unit 16 from outside the frequency detection device 8.

[0048] If the Nyquist aliasing order Fo is 0 or greater (in step ST5: YES in Figure 4), the frequency calculation unit 16 calculates the frequency f of the clock signal as shown in the following equation (10): the Nyquist aliasing order Fo and the frequency f of the clock signal. ｃｌｋ The frequency f of the first quantized data is obtained by multiplying the result of the first quantization. ｏｕｔ By adding this, the frequency f of the signal under measurement is obtained. ＲＦ The frequency calculation unit 16 calculates the frequency f of the clock signal, as shown in equation (11) below, if the Nyquist aliasing order Fo is negative (step ST5 in Figure 4: NO). ｃｌｋFrom the result of multiplication, the frequency f of the first quantized data is obtained. ｏｕｔ By subtracting the frequency f of the signal under measurement, ＲＦ The frequency f of the signal under test is calculated (step ST7 in Figure 4). ＲＦ Information indicating this is provided, for example, to a signal processing device (not shown) or a display device (not shown). The frequency f of the signal under measurement is provided to the display device (not shown). ＲＦ If this is displayed, the user can determine the frequency f of the signal being measured. ＲＦ This can be confirmed.

[0049]

[0050] As described above, the frequency detection device 8 shown in Figure 1 can detect the first quantized data QD even if an interfering signal, as shown in Figure 6, is mixed into the signal being measured. １ The Nyquist component x corresponds to the Nyquist folding order Fo. ｋ It is possible to search for the frequency f of the signal under measurement. Therefore, the frequency detection device 8 shown in Figure 1 can detect the frequency f of the signal under measurement even if there is an interfering signal as shown in Figure 6. ＲＦ It is possible to calculate the first quantization data QD, rather than determining the Nyquist aliasing order Fo based on the Manhattan distance, as in the frequency detection device 8. １ Phase θ ｏｕｔ１ and the second quantized data QD ２ Phase θ ｏｕｔ２ The Nyquist folding order Fo can be determined using only the phase difference Δθ. Below, the first quantized data QD １ Phase θ ｏｕｔ１ and the second quantized data QD ２ Phase θ ｏｕｔ２ A method that uses only the phase difference Δθ is called a phase determination method.

[0051] Figure 7 is an explanatory diagram showing the phase rotation of the signal under test. In the phase determination method, when an interfering signal, as shown in Figure 6, is present, the phase rotation angle of the signal under test is affected by the interfering signal, as shown in Figure 7. That is, when no interfering signal is present, the signal under test after phase rotation will be as shown in Figure 7, but when an interfering signal is present, the signal under test after phase rotation will be the entire signal after phase rotation, as shown in Figure 7. The entire signal is the signal under test with the interfering signal present. For this reason, in the phase determination method, when an interfering signal, as shown in Figure 6, is present, the accuracy of calculating the phase difference Δθ deteriorates. As a result, the frequency f of the signal under test ＲＦ The calculation accuracy deteriorates.

[0052] Figure 8 shows the frequency f of the measured signal when an interfering signal is present. ＲＦ This is an explanatory diagram showing the estimated failure rate. In Figure 8, the horizontal axis represents the signal intensity of the interfering signal, and the vertical axis represents the estimated failure rate. The L1 norm estimation method uses the frequency detection device 8 shown in Figure 1 to detect the frequency f ＲＦ The estimation method is shown. Figure 8 shows that the estimation failure rate of the L1 norm estimation method is lower than that of the phase determination method.

[0053] In the above embodiment 1, the frequency detection device 8 is configured to include a frequency-phase calculation unit 11 that converts the signal to be measured into a first quantized data which is a quantized signal, into a frequency-domain signal, and calculates the frequency and phase of the first quantized data from the first frequency-domain signal, which is a frequency-domain signal; and a phase difference calculation unit 12 that converts the second quantized data which is data delayed compared to the first quantized data into a frequency-domain signal, calculates the phase of the second quantized data from the second frequency-domain signal, which is a frequency-domain signal, and calculates the phase difference between the phase of the first quantized data and the phase of the second quantized data. Furthermore, the frequency detection device 8 includes an aliasing order calculation unit 13 that calculates the Nyquist aliasing order of the first quantized data based on the first frequency domain signal, the second frequency domain signal, and the phase difference calculated by the phase difference calculation unit 12, and a frequency calculation unit 16 that calculates the frequency of the signal under test using the frequency calculated by the frequency phase calculation unit 11 and the Nyquist aliasing order calculated by the aliasing order calculation unit 13. Therefore, the frequency detection device 8 can prevent deterioration of the accuracy of frequency calculation in the signal under test even if an interfering signal is mixed in with the signal under test.

[0054] In Embodiment 1, the frequency detection device 8 is configured such that the order calculation processing unit 15 identifies the Nyquist component that minimizes the Manhattan distance between multiple Nyquist components of multiple orders as the Nyquist component corresponding to the Nyquist aliasing order of the first quantized data. Therefore, the frequency detection device 8 can reduce the failure rate of estimating the frequency of the signal under test compared to the phase determination method, even when an interfering signal is mixed in with the signal under test.

[0055] In Embodiment 1, the frequency detection device 8 is configured such that the order calculation processing unit 15 uses the LASSO optimization algorithm to find the Nyquist component among multiple Nyquist components of multiple orders that minimizes the Manhattan distance between them, and identifies the Nyquist component with the minimum Manhattan distance as the Nyquist component corresponding to the Nyquist folding order of the first quantized data. Therefore, the frequency detection device 8 can reduce the computational cost required to search for the Nyquist component.

[0056] It should be noted that this disclosure allows for modifications of any component of the embodiment, or the omission of any component of the embodiment.

[0057] This disclosure includes a frequency-phase calculation unit that converts a first quantized data, which is a quantized signal of the signal to be measured, into a frequency-domain signal, and calculates the frequency and phase of the first quantized data from the first frequency-domain signal, which is a frequency-domain signal; and a phase difference calculation unit that converts a second quantized data, which is data delayed compared to the first quantized data, into a frequency-domain signal, calculates the phase of the second quantized data from the second frequency-domain signal, which is a frequency-domain signal, and calculates the phase difference between the phase of the first quantized data and the phase of the second quantized data. Furthermore, the frequency detection device includes an aliasing order calculation unit that calculates the Nyquist aliasing order of the first quantized data based on a first frequency domain signal, a second frequency domain signal, and a phase difference calculated by a phase difference calculation unit, and a frequency calculation unit that calculates the frequency of the signal under test using the frequency calculated by the frequency phase calculation unit and the Nyquist aliasing order calculated by the aliasing order calculation unit. This prevents deterioration of the accuracy of frequency calculation in the signal under test even if other signals are mixed in with the signal under test, making it suitable for frequency detection devices, frequency detection methods, and frequency detection circuits.

[0058] 1 Clock signal source, 2 Distributor, 3 Quantization unit, 4, 5 Delay circuits, 6, 7 Quantizer, 8 Frequency detection device, 11 Frequency phase calculation unit, 12 Phase difference calculation unit, 13 Alignment order calculation unit, 14 Matrix calculation unit, 15 Order calculation processing unit, 16 Frequency calculation unit, 21 Frequency phase calculation circuit, 22 Phase difference calculation circuit, 23 Alignment order calculation circuit, 24 Frequency calculation circuit, 31 Memory, 32 Processor.

Claims

1. A frequency detection device comprising: a frequency-phase calculation unit that converts a first quantized data, which is a quantized signal of the signal to be measured, into a frequency-domain signal, and calculates the frequency and phase of the first quantized data from the first frequency-domain signal; a phase difference calculation unit that converts a second quantized data, which is data delayed compared to the first quantized data, into a frequency-domain signal, calculates the phase of the second quantized data from the second frequency-domain signal, and calculates the phase difference between the phase of the first quantized data and the phase of the second quantized data; an aliasing order calculation unit that calculates the Nyquist aliasing order of the first quantized data based on the first frequency-domain signal, the second frequency-domain signal, and the phase difference calculated by the phase difference calculation unit; and a frequency calculation unit that calculates the frequency of the signal to be measured using the frequency calculated by the frequency-phase calculation unit and the Nyquist aliasing order calculated by the aliasing order calculation unit.

2. The frequency detection device according to claim 1, wherein the aliasing order calculation unit comprises: a matrix calculation unit that calculates a matrix for converting a plurality of Nyquist components of a plurality of orders into the first frequency domain signal and the second frequency domain signal, respectively, based on the phase difference calculated by the phase difference calculation unit; and an order calculation processing unit that, under the condition that the result of multiplying the matrix calculated by the matrix calculation unit by the plurality of Nyquist components of a plurality of orders becomes the first frequency domain signal and the second frequency domain signal, respectively, identifies a Nyquist component among the plurality of Nyquist components of a plurality of orders that corresponds to the Nyquist aliasing order of the first quantized data, based on the Manhattan distance of the plurality of Nyquist components, and determines the Nyquist aliasing order from the identified Nyquist component.

3. The frequency detection device according to claim 2, characterized in that the order calculation processing unit identifies the Nyquist component among the plurality of Nyquist components of order that minimizes the Manhattan distance of the plurality of Nyquist components as the Nyquist component corresponding to the Nyquist aliasing order of the first quantized data.

4. The frequency detection device according to claim 2 or 3, characterized in that the order calculation processing unit uses the LASSO (Least Absolute Shrinkage and Selection Operator) optimization algorithm to find the Nyquist component among the plurality of Nyquist components of the plurality of orders that minimizes the Manhattan distance of the plurality of Nyquist components, and identifies the Nyquist component that minimizes the Manhattan distance as the Nyquist component corresponding to the Nyquist aliasing order of the first quantized data.

5. A frequency detection method comprising: a frequency-phase calculation unit converting a signal under test into a first quantized data signal, which is a quantized signal of the signal under test, into a frequency domain signal, and calculating the frequency and phase of the first quantized data from the first frequency domain signal, which is a frequency domain signal; a phase difference calculation unit converting a second quantized data signal, which is data delayed compared to the first quantized data, into a frequency domain signal, and calculating the phase of the second quantized data from the second frequency domain signal, which is a frequency domain signal, and calculating the phase difference between the phase of the first quantized data and the phase of the second quantized data; an aliasing order calculation unit calculating the Nyquist aliasing order of the first quantized data based on the first frequency domain signal, the second frequency domain signal, and the phase difference calculated by the phase difference calculation unit; and a frequency calculation unit calculating the frequency of the signal under test using the frequency calculated by the frequency-phase calculation unit and the Nyquist aliasing order calculated by the aliasing order calculation unit.

6. A quantization unit that quantizes a signal under test and outputs a first quantized data which is the signal under test after quantization and a second quantized data which is delayed compared to the first quantized data; a frequency-phase calculation unit that converts the first quantized data output from the quantization unit into a frequency domain signal and calculates the frequency and phase of the first quantized data from the first frequency domain signal which is the frequency domain signal; a phase difference calculation unit that converts the second quantized data output from the quantization unit into a frequency domain signal, calculates the phase of the second quantized data from the second frequency domain signal which is the frequency domain signal, and calculates the phase difference between the phase of the first quantized data and the phase of the second quantized data; and an aliasing order calculation unit that calculates the Nyquist aliasing order of the first quantized data based on the first frequency domain signal, the second frequency domain signal, and the phase difference calculated by the phase difference calculation unit. A frequency detection circuit comprising a frequency calculation unit that calculates the frequency of the signal to be measured using the frequency calculated by the frequency phase calculation unit and the Nyquist aliasing order calculated by the aliasing order calculation unit.

Images