JP2020118684A - Arithmetic device and program - Google Patents

Abstract

To easily find the fundamental period of an input wave.SOLUTION: A periodic filter 50 includes: sampling circuits 5a1, 5a2, ..., 5aV that output in parallel consecutive discrete points included in sections with different time lengths allocated in advance from an input discrete time signal; adders 5b1, 5b2, ..., 5bV that integrate by adding values of signals output by the sampling circuits 5a1, 5a2, ..., 5aV for each extraction point; and a first arithmetic unit 5d and a second arithmetic unit 5e that perform arithmetic for deriving a fundamental period based on a difference in inclination caused by a locus of vertices of accumulated values for each fundamental period candidate.SELECTED DRAWING: Figure 13

Description

The present invention relates to an arithmetic device and a program.

As a technique for performing frequency analysis, a method such as DFT (Discrete Fourier Transform) or FFT (Fast Fourier Transform) that executes it at high speed is generally known.

JP, 2017-119285, A JP, 2005-207870, A JP, 2003-232674, A JP, 7-311227, A

In frequency analysis using DFT or FFT, there is a problem that the result cannot be guaranteed unless the fundamental frequency of the waveform input beforehand is known. There is also a problem that the calculation cost is high.

Therefore, it is necessary to input the information of a certain long period to minimize the error, apply an empirical value, check the waveform in advance, and estimate the frequency that seems to be the fundamental frequency. Attempts have been made to minimize the error by the method.
In one aspect, the present invention aims to easily grasp the fundamental period of an input wave.

In order to achieve the above objective, the disclosed periodic filter is provided. The periodic filter includes a plurality of signal output units that output, from an input discrete-time signal, consecutive discrete points included in basic period candidate sections having different pre-assigned time lengths to an output port in parallel, and It is generated by an adder that calculates the cumulative value for each basic cycle candidate section by adding the values of the signals output by the signal output section for each output port, and the locus of the vertex of the cumulative value for each basic cycle candidate section. And a calculation unit that executes a calculation for deriving a basic period based on the difference in the slope.

In one aspect, the fundamental period of the input wave can be easily grasped.

It is a figure which shows the integrated circuit of 1st Embodiment. It is a figure which shows an example of the periodic filter of 1st Embodiment. It is a figure explaining the process of an adder. It is a figure explaining the process of an adder. 6 is a flowchart illustrating processing of the periodic filter according to the first embodiment. 6 is a flowchart illustrating processing of the periodic filter according to the first embodiment. It is a figure explaining that a fundamental period can be correctly identified. It is a figure explaining that a fundamental period can be correctly identified. It is a figure explaining that a fundamental period can be correctly identified. It is a figure explaining that a fundamental period can be correctly identified. It is a figure explaining that a fundamental period can be correctly identified. It is a figure explaining the inclined ARS spectrum. It is a figure which shows an example of the periodic filter of 2nd Embodiment. It is a figure explaining an outline of processing of the 1st operation part and the 2nd operation part of a 2nd embodiment. It is a figure explaining the 2nd calculation. It is a figure explaining the 2nd calculation. It is a figure explaining the 2nd calculation. It is a figure explaining the example of application of the 2nd operation. It is a figure explaining the example of application of the 2nd operation. It is a flow chart explaining processing of a periodic filter of a 2nd embodiment. It is a flow chart explaining processing of a periodic filter of a 2nd embodiment. It is a figure explaining the 1st example of processing of the periodic filter of a 2nd embodiment. It is a figure explaining the 1st example of processing of the periodic filter of a 2nd embodiment. It is a figure explaining the 2nd example of processing of the periodic filter of a 2nd embodiment. It is a figure explaining the 3rd example of processing of the periodic filter of a 2nd embodiment. It is a flow chart explaining processing of a periodic filter of a 3rd embodiment. It is a flow chart explaining processing of a periodic filter of a 3rd embodiment. It is a figure explaining an example of processing of a periodic filter of a 3rd embodiment. It is a flowchart explaining the process of the modification of the periodic filter of 3rd Embodiment. It is a flowchart explaining the process of the modification of the periodic filter of 3rd Embodiment.

Hereinafter, an integrated circuit according to an embodiment will be described in detail with reference to the drawings.

The position, size, shape, range, etc. of each configuration shown in the following drawings and the like may not represent the actual position, size, shape, range, etc. in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings and the like.
The elements expressed in the singular in the embodiments include the plural unless explicitly stated in the text.
<First Embodiment>
FIG. 1 is a diagram showing an integrated circuit according to the first embodiment.

The integrated circuit 1 includes a transmitter (Transmitter) 2, a receiver (Receiver) 3, an ADC (Analog Digital Converter) 4, and a periodic filter (period determination unit) 5.

The transmission unit 2 transmits a periodic signal (electromagnetic wave or the like) such as an electromagnetic wave, a voice, or an ultrasonic wave to an object (measurement object) whose cycle is to be measured, via the antenna 21. The combination of the transmitting unit 2 and the receiving unit 3 is not limited to the one shown in the figure, and for example, the transmitting side may be a speaker and the receiving side may be a microphone.
Examples of the measurement target include body surface deviation caused by the heartbeat and respiration of the living body. The number of measurement objects may be one or may be plural as shown in FIG.
The receiver 3 detects the deviation of the measurement target frequency. Noise (η) is added to the waveform in which the measurement target frequency is detected.
The ADC 4 converts the analog value received by the receiving unit 3 into a digital value (discrete time signal).
Note that some or all of the functions of the integrated circuit 1 may be realized by software.
When a part of the function is realized by software, the software may be operated on a computer.

The receiving unit 3 may have a function of separating a real number and an imaginary number. In that case, the ADC 4 may convert each of the real number and the imaginary number into a digital value.

The periodic filter 5 of the embodiment separates overlapping waves due to the difference in fundamental period and outputs a periodic spectrum. This periodic filter 5 can separate the waves without any prior information about the incoming waves.

The periodic filter 5 may be a dedicated circuit such as an FPGA (Field-Programmable gate array) or a DSP (Digital Signal Processor).
FIG. 2 is a diagram illustrating an example of the periodic filter according to the first embodiment.

The periodic filter 5 may be described as k (k=1... V) sampling circuits (v (v=1, 2,... V) SPC: Serial Parallel Converter in the drawing. ) 5a1, 5a2,..., 5aV and functions as v (v=1, 2,... V) integrators (Accumulators) corresponding to the sampling circuits 5a1, 5a2,. It has adders 5b1, 5b2,..., 5bV and a maximum value detector 5c.

The sampling circuits 5a1, 5a2,..., 5aV sample the discrete time signals respectively input and output in parallel. More specifically, the sampling circuits 5a1, 5a2,..., 5aV are consecutive from the input discrete-time signal and included in preallocated intervals (fundamental cycle candidate intervals) having different time lengths. Extracts discrete points and outputs them in parallel.

That is, the number of discrete points (the number of samples) extracted in the basic period candidate section is different for each of the sampling circuits 5a1, 5a2,..., 5aV. More specifically, these sampling circuits 5a1, 5a2,..., 5aV repeatedly output different sample numbers extracted according to the range of the period assigned in advance from the output port.

For example, when the output of the ADC 4 is 144 samples and the range of the predicted period set in the sampling circuit 5a1 is 3 samples, continuous 3 samples are repeatedly output 48 times. When the range of the predicted period set in the sampling circuit 5a2 is 4 samples, continuous 4 samples are repeatedly output 36 times. The sampling circuit 5a1, 5a2,..., 5aV may have the same number of repeated outputs.

In the present embodiment, of the sampling circuits 5a1, 5a2,..., 5aV, the sampling circuit 5a1 has the smallest number of samples, and the sampling circuits 5a2, 5a3 are sequentially increased in number by one. There is.

The adders 5b1, 5b2,..., 5bV add the values of the respective samples input from the output ports of the sampling circuits 5a1, 5a2,.
The adders 5b1, 5b2,..., 5bV add the values stored therein and the newly input values to update the addition result.

The maximum value detection unit 5c compares the absolute values of the operation results of the adders 5b1, 5b2,..., 5bV, and outputs the largest absolute value of the addition result to the adders 5b1, 5b2,..., 5bV. The distance determined by the length of the section (between the first discrete point and the last discrete point to be sampled) set in the sampling circuits 5a1, 5a2,... Judge that there is and output.

The method of comparing the absolute values of the calculation results is not particularly limited. For example, the absolute value of the average value obtained by dividing the calculation results of the adders 5b1, 5b2,... You may make it compare. In another example, all the calculation results of the adders 5b1, 5b2,..., 5bV may be normalized and compared.
3 and 4 are diagrams for explaining the processing of the adder.

FIG. 3 illustrates the case where the range of the predicted cycle is 8 samples, and the number of output ports of the sampling circuit 5aX is “8”. At this time, each output in the case of parameter g (the order of the corresponding section counted from the beginning of the input waveform when the input waveform is divided with the predicted cycle as the length of one section)=0, 1, 2, 3 The value (output value) of each port is almost equal, and the same waveform appears. Therefore, by adding the values of the signals output from the same output port, the amplitude value becomes a large waveform. Therefore, it can be determined that 8 samples are one cycle.

FIG. 4 illustrates the case where the range of the predicted cycle is 7 samples, and the number of output ports of the sampling circuit 5aY is “7”. At this time, the values (output values) of the respective output ports in the case of the parameter g=0, 1, 2, 3 are different and have different values. Therefore, when the values are added for the same port, they cancel each other out, resulting in a waveform with a small amplitude value.

In this way, when the discrete values of waves with the same period are added, the absolute value gradually increases. Even if the discrete values of waves whose cycles do not match are added, the absolute values do not increase because they cancel each other out. Therefore, the fundamental period can be easily found by detecting the length of the sampling section of the sampling circuit corresponding to the adder having the largest absolute value.

Further, the maximum value detection unit 5c normalizes the absolute values of all integration results so that the largest integration result becomes 1, and compares the normalized absolute value of the integration result with a threshold value of 0 or more and 1 or less. Then, it is possible to determine the number of signals included in the input discrete-time signal and the respective periods.

Next, the processing of the periodic filter 5 will be described using a flowchart. Note that the processing described below is an example, and some processing may be omitted. Some processing may be replaced with other processing. Other processes may be executed between the processes. You may change the order of some processing.
5 and 6 are flowcharts for explaining the process of the periodic filter.
[Step S1] The periodic filter 5 sets the number V of sampling circuits used in the following calculation. After that, the processing makes a transition to Step S2.
[Step S2] The periodic filter 5 sets the parameter v=1. After that, the processing makes a transition to Step S3.
[Step S3] The periodic filter 5 determines the number of output ports of the v-th sampling circuit as u(v). After that, the processing makes a transition to Step S4.
[Step S4] The periodic filter 5 creates an array w(v) of one row u(v) in which all elements are 0. After that, the processing makes a transition to Step S5.
[Step S5] The periodic filter 5 sets parameters g=1 and k=1. After that, the processing makes a transition to Step S6.
[Step S6] The periodic filter 5 determines the number G of cumulative sampling circuit outputs. After that, the processing makes a transition to Step S7.

[Step S7] The adders 5b1, 5b2,..., 5bV input the input samples x[k] to x[k+u _v −1] to the v-th sampling circuit. The obtained output is an array x _v (g) of one row u(v). After that, the processing makes a transition to Step S8.

[Step S8] The adders 5b1, 5b2,..., 5bV perform the following calculations. W(v)=W(v)+ _xv (g), k=g.u(v). After that, the processing makes a transition to Step S9.
[Step S9] The periodic filter 5 performs the following calculation. g=g+1. After that, the processing makes a transition to Step S10.

[Step S10] The periodic filter 5 determines whether the parameter g is equal to or less than the number G. When the parameter g is the number G or less (Yes in step S10), the process proceeds to step S7. When the parameter g is larger than the number G (No in step S10), the process proceeds to step S11.
[Step S11] The periodic filter 5 performs the following calculation. W(v)=W(v)/g. After that, the processing makes a transition to Step S12.

[Step S12] The maximum value detection unit 5c identifies the element having the largest absolute value among the elements of the array W(v), and sets the value of the element as p(v). After that, the processing makes a transition to Step S13.
[Step S13] The periodic filter 5 performs the following calculation. v=v+1. After that, the processing makes a transition to Step S14.

[Step S14] The periodic filter 5 determines whether the parameter v is equal to or less than the number V. When the parameter v is the number V or less (Yes in step S14), the process proceeds to step S3. When the parameter v is larger than the number V (No in step S14), the process proceeds to step S15.

[Step S15] The periodic filter 5 specifies the maximum value from p(1) to p(V) and sets the value as p1. Normalize p(1) to p(V) so that p1 becomes 1, and obtain the obtained values as P(1) to P(V), respectively, and form a matrix of 1 row and V columns P=[P( 1), P(2),..., P(V)]. After that, the processing makes a transition to Step S16.
[Step S16] The periodic filter 5 sets a threshold value Q (0<Q≦1). After that, the processing makes a transition to Step S17.
[Step S17] The periodic filter 5 selects all of the elements of the matrix P that exceed the threshold value Q. Let N be the number of selected elements.
[Step S18] Among the elements of the matrix P, the periodic filter 5 records the number of the element that exceeds the threshold value Q as V _1, 1,... V _N 1.

[Step S19] The periodic filter 5 outputs the number of elements N as the number of signals. Further, the periodic filter 5 outputs V _1, 1,..., V _N 1 as each period. Further, the periodic filter 5 sets W(V ₁ 1),... W(V _N 1) as the estimated basic waveform. After that, the processing of FIGS. 5 and 6 ends.
7 to 11 are diagrams for explaining that the basic period can be accurately identified.
As shown in FIG. 7, a case where one period is assumed to be 31 samples and a case where one 0 is added and one period is assumed to be 32 samples will be illustrated.
As shown in FIG. 8, the cycle slightly shifts between the case where one cycle is assumed to be 31 samples and the case where it is assumed to be 32 samples.

FIG. 9 shows a matrix P=[P1P(2)... P(39)] of 1 row and V columns in step S15 of the periodic filter 5 for the analog waveform of FIG. The absolute value is the largest when one period is assumed to be 31 samples. Therefore, the periodic filter 5 determines that the distance determined by the length of 31 samples is the basic period and outputs it.
FIG. 10 shows an example of the analog waveform received by the receiver 3.
The analog waveform shown in FIG. 10 has more noise than the analog waveform shown in FIG. 8, and at first glance it is difficult to specify the basic period.
FIG. 11 shows the matrix P=[P1P(2)... P(39)] of 1 row and V column in step S15 of the periodic filter 5 for the analog waveform.

The case where one cycle is assumed to be 31 samples and the case where it is assumed to be 32 samples are significantly outstanding, but the value is largest when one cycle is assumed to be 32 samples. Therefore, the periodic filter 5 determines that the distance determined by the length of 32 samples is the basic period and outputs it. Thus, according to the periodic filter 5, the basic period can be easily detected even with a slight difference in period. That is, it is usually very difficult to distinguish adjacent waves and to separate highly correlated waves with similar waveforms, and at the same time, even when the detected and undetected waves are adjacent, It is difficult to make a clear distinction.

On the other hand, as shown in FIG. 11, according to the periodic filter 5 of the embodiment, it is shown that the period can be correctly detected when the signal shown in FIG. It is shown that even if the waves of 31 samples and the waves of 32 samples are input at the same time, they can be clearly distinguished, and the other waves are not clearly detected.

As described above, according to the integrated circuit 1, the periodic filter 5 having the sampling circuits 5a1, 5a2,..., 5aV, the adders 5b1, 5b2,..., 5bV, and the maximum value detection unit 5c. And using a discrete-time signal as input, the period of one or more waveforms that make up the input waveform can be determined.

Therefore, it is possible to analyze one or a plurality of input waveforms by using a method based on a combination of simple sampling and addition without acquiring the waveform information in advance. In addition, when waveforms of different fundamental frequencies having a high correlation exist at the same time, they can be output separately.

Further, the periodic filter 5 has a particularly high resolution in the low frequency range as compared with a general frequency analysis method such as FFT or DFT. Therefore, the basic period can be obtained with higher accuracy.
In addition, since fixed-point arithmetic is possible and the arithmetic processing is performed mainly using addition processing, high-speed and high-accuracy periodic analysis can be realized with low computational resources.
Further, it is possible to analyze the input waveform in which a plurality of waveforms having different fundamental periods are combined, and to obtain the fundamental period of each waveform and the period of the harmonic of 1/n times the period.
Further, the waveforms of the different basic periods may be waveforms with high correlation.
Further, the waveforms of the different fundamental periods may be waveforms in which the frequency bands are adjacent to each other or overlap each other.
Also. FFT analysis can also be performed using the basic period specified by the periodic filter 5.

The processing performed by the periodic filter 5 may be distributed to a plurality of devices. For example, one device may perform the sampling process and another device may specify the basic period using the output parallel value.
<Second Embodiment>
Next, the integrated circuit of the second embodiment will be described.
Hereinafter, the integrated circuit of the second embodiment will be described focusing on the differences from the above-described first embodiment, and the description of the same matters will be omitted.

The periodic filter 5 according to the first embodiment extracts a relatively large number of samples from a sampling circuit (hereinafter, also referred to as a small SPC) that extracts a relatively small number of samples and outputs them in parallel. A large number of sampling circuits (hereinafter, also referred to as multi-SPC) can be mounted. The cumulative value of addition results of the samples output from the output port of the small SPC is larger than the cumulative value of addition results of the samples output from the output port of the multi SPC. Therefore, if the outputs of the sampling circuits are accumulated as they are, a large number of small SPCs are accumulated, and thus a sloped ARS spectrum (cumulative value of the values output from the output port of each sampling circuit with respect to the period) is obtained.
FIG. 12 is a diagram illustrating a tilted ARS spectrum.
The vertical axis represents the cumulative value of the values output from the output port of each sampling circuit. The horizontal axis indicates the cycle.
Since the sampling circuit having the smaller number of output ports has the larger number of outputs, the cumulative value of the outputs of the sampling circuit increases toward the left side.

In order to solve this problem, for example, a method of accumulating the outputs of the sampling circuit, dividing by the cumulative number, and averaging can be considered. This eliminates the difference in cumulative numbers and flattens the ARS spectrum.

However, if the same number of divisions as the number of sampling circuits are executed, the amount of calculation will increase significantly. Further, even if the average is taken, when the S/N ratio is low, this time the ARS spectrum rises to the right. If the ARS spectrum is inclined in this way, peak detection may not be successful in some cases. The periodic filter according to the second embodiment is intended to solve this problem.
FIG. 13 is a diagram illustrating an example of the periodic filter according to the second embodiment.

The periodic filter 50 according to the second embodiment has a first calculation unit 5d and a second calculation unit 5e. The first calculation unit 5d and the second calculation unit 5e execute the calculation for deriving the basic cycle based on the difference in the slopes caused by the loci of the vertices of the cumulative value for each basic cycle candidate section.
FIG. 14 is a diagram for explaining the outline of the processing of the first calculation unit and the second calculation unit of the second embodiment.

The first calculator 5d calculates the difference between the right side value and the left side value of the value of the ARS spectrum (hereinafter referred to as the left difference). In addition, the first calculation unit 5d calculates (first calculation) the difference between the values of the values of the ARS spectrum from the left side to the right side (hereinafter referred to as right difference).

In the example shown in FIG. 14, the symbols (1) to (6) for identifying the values of the ARS spectrum are attached. Then, the calculation formulas for the left difference and the right difference are shown. In the example shown in FIG. 14, the difference between adjacent values is calculated, but the left difference and the right difference may be calculated at intervals of 1 or more (this example will be described later).

Then, the second calculation unit 5e performs the second calculation using the obtained left difference and right difference. Specifically, the second calculation unit 5e performs a calculation that shifts the left difference and the right difference by the interval calculated in the first calculation and adds the left difference and the right difference. For example, when the right difference and the left difference are calculated between the values of the adjacent ARS spectra, the second calculation unit 5e shifts the left difference and the right difference by one and performs the calculation of adding the left difference and the right difference. To do. This offsets the slope of the ARS spectrum, raising the peak and making it easier to understand.
15 to 17 are diagrams for explaining the second calculation.

The ARS spectrum shown in FIG. 15 has one peak. The value of the ARS spectrum indicated by an asterisk in FIG. 15 is the peak value. The peak can be detected by performing the second calculation using the left difference and the right difference.

The ARS spectrum shown in FIG. 16 has two peaks. The value of the ARS spectrum indicated by an asterisk in FIG. 16 is the peak value. Also in this case, the peak can be detected by performing the second calculation using the left difference and the right difference as in the case shown in FIG.

The ARS spectrum shown in FIG. 17 has three peaks. The value of the ARS spectrum indicated by an asterisk in FIG. 17 is the peak value. Also in this case, the peak can be detected by performing the second calculation using the left difference and the right difference as in the case shown in FIG.
18 and 19 are diagrams illustrating application examples of the second calculation.
The ARS spectrum shown in FIG. 18 has three peaks, like the ARS spectrum shown in FIG.

As described above, the second calculator 5e may calculate the left difference and the right difference at intervals of 1 or more. In the example illustrated in FIG. 18, the second calculation unit 5e performs the first calculation on the left difference and the right difference at one interval. When such a first operation is performed, the second operation unit 5e performs an operation of shifting the left difference and the right difference by two and adding the left difference and the right difference.
Further, by taking the difference between a plurality of spectra, it is possible to detect adjacent signals.
20 and 21 are flowcharts illustrating the process of the periodic filter according to the second embodiment.
[Step S21] The periodic filter 50 sets the number V of sampling circuits used in the following calculation. After that, the processing makes a transition to Step S22.
[Step S22] The periodic filter 50 sets the parameter v=1. After that, the processing makes a transition to Step S23.
[Step S23] The periodic filter 50 determines the number of output ports of the v-th sampling circuit as u(v). After that, the processing makes a transition to Step S24.
[Step S24] The periodic filter 50 creates an array W(v) of one row u(v) in which all the elements are 0. After that, the processing makes a transition to Step S25.
[Step S25] The periodic filter 50 sets parameters g=1 and k=1. After that, the processing makes a transition to Step S26.
[Step S26] The periodic filter 50 determines the number G of cumulative sampling circuit outputs. After that, the processing makes a transition to Step S27.

[Step S27] The adders 5b1, 5b2,..., 5bV input the input samples x[k] to x[k+u _v −1] to the v-th sampling circuit. The obtained output is an array X _v (g) of one row u(v). After that, the processing makes a transition to Step S28.

[Step S28] The adders 5b1, 5b2,..., 5bV perform the following calculations. W(v)=W(v)+X _v (g), k=g·u(v). After that, the processing makes a transition to Step S29.
[Step S29] The periodic filter 50 performs the following calculation. g=g+1. After that, the processing makes a transition to Step S30.

[Step S30] The periodic filter 50 determines whether the parameter g is equal to or less than the number G. When the parameter g is the number G or less (Yes in step S30), the process proceeds to step S27. When the parameter g is larger than the number G (No in step S30), the process proceeds to step S31.
[Step S31] The periodic filter 50 performs the following calculation. v=v+1. After that, the processing makes a transition to Step S32.

[Step S32] The periodic filter 50 determines whether or not the parameter v is equal to or less than the number V. When the parameter v is the number V or less (Yes in step S32), the process proceeds to step S23. When the parameter v is larger than the number V (No in step S32), the process proceeds to step S33.

[Step S33] The periodic filter 50 specifies the maximum value from p(1) to p(V) and sets the value as p1. Normalize p(1) to p(V) so that p1 becomes 1, and obtain the obtained values as P(1) to P(V), respectively, and form a matrix of 1 row and V columns P=[P( 1), P(2),..., P(V)] are created. After that, the processing makes a transition to Step S34.

[Step S34] The first calculation unit 5d executes the first calculation on the matrix P. In addition, the second calculation unit 5e executes the second calculation. Then, a matrix P0=[P0(2), P0(3),..., P0(V-1)] of one row (V-2) column is created. After that, the processing makes a transition to Step S35.

[Step S35] In the periodic filter 50, among the elements of the matrix P00, the number (element number) of an element having an S/N ratio of s [dB] (S is an arbitrary real number exceeding 0) or more is V _1, 1,... -Record as V _N 1. After that, the processing makes a transition to Step S36.

[Step S36] The periodic filter 50 outputs the number of elements N as the number of signals. Further, the periodic filter 50 outputs V _1, 1,..., V _N 1 as each period. Further, the periodic filter 50 sets W(V ₁ 1),... W(V _N 1) as the estimated basic waveform. Then, the processing of FIGS. 20 and 21 is ended.
The processing of the periodic filter 50 according to the second embodiment will be described below using actual measurement results.
22 and 23 are diagrams illustrating a first example of the process of the periodic filter according to the second embodiment.

FIG. 22A shows the ARS spectrum of the first embodiment. The case where one period is 51 samples, 52 samples, 102 samples, and 104 samples is significantly outstanding.

FIG. 22B shows the calculation result of the periodic filter 50 performing the first calculation on the ARS spectrum shown in FIG. The slope is removed and the overhanging sample is more emphasized.
FIG. 23 shows a calculation result obtained by executing the second calculation on the calculation result of the first calculation shown in FIG.
FIG. 24 is a diagram illustrating a second example of the process of the periodic filter according to the second embodiment.

FIG. 24A shows the ARS spectrum of the second embodiment. In the second embodiment, there are a plurality of consecutive samples (101 to 105) whose cumulative values are significantly protruding. Further, there is a sample (108) that largely projects at a position slightly apart from the continuous sample.

FIG. 24B shows the calculation result of the periodic filter 50 performing the first calculation on the ARS spectrum shown in FIG. The slope is removed also for the ARS spectrum as in the second embodiment, and the sample having a large protrusion is more emphasized.
According to the integrated circuit of the second embodiment, the peak can be detected easily and surely by the relatively simple calculation.
<Third Embodiment>
Next, an integrated circuit according to the third embodiment will be described.

Hereinafter, the integrated circuit of the third embodiment will be described with a focus on the differences from the first embodiment and the second embodiment described above, and the description of the same matters will be omitted.
In the processing of the integrated circuit according to the second embodiment, the sample existing in a place slightly isolated by noise has a relatively large cumulative value.
FIG. 25 is a diagram illustrating a third example of the process of the periodic filter according to the second embodiment.
The third embodiment shows a case where the periodic filter 50 processes an input having a low S/N ratio (an analog value received by the receiving unit 3).

FIG. 25A shows the ARS spectrum of the third embodiment. In the third embodiment, there are a plurality of consecutive samples (101 to 105) whose cumulative values are significantly protruding. Further, there is a sample (108) that largely projects at a position slightly apart from the continuous sample.
FIG. 25B shows the calculation result of the periodic filter 50 performing the first calculation on the ARS spectrum of the third embodiment shown in FIG.

In the case of an input having a low S/N ratio, even if the periodic filter 50 performs the first process, a sample (127) that should not originally exist is detected in addition to the samples (101 to 105 and 108). I understand. In the case of such an input having a low S/N ratio, it may be difficult to determine whether the detected sample is a signal or noise.

One purpose of the integrated circuit of the third embodiment is to reduce the influence of noise so that the peak and the noise can be easily distinguished by executing the following processing.

26 and 27 are flowcharts for explaining the process of the periodic filter according to the third embodiment. In the process of the periodic filter of the third embodiment, the processes of steps S42, S43, S57 and S58 are newly added.
[Step S41] The periodic filter 50 sets the number V of sampling circuits used in the following calculation. After that, the processing makes a transition to Step S42.
[Step S42] The periodic filter 50 sets the parameter m=1. After that, the processing makes a transition to Step S43.
[Step S43] The periodic filter 50 receives an input of the number of repetitions M. After that, the processing makes a transition to Step S44.
[Step S44] The periodic filter 50 sets the parameter v=1. After that, the processing makes a transition to Step S45.
[Step S45] The periodic filter 50 determines the number of output ports of the v-th sampling circuit as u(v). After that, the processing makes a transition to Step S46.
[Step S46] The periodic filter 50 creates an array W(v) of one row u(v) in which all elements are 0. After that, the processing makes a transition to Step S47.
[Step S47] The periodic filter 50 sets parameters g=1 and k=1. After that, the processing makes a transition to Step S48.
[Step S48] The periodic filter 50 determines the number G of accumulated sampling circuit outputs. After that, the processing makes a transition to Step S49.

[Step S49] The adders 5b1, 5b2,..., 5bV input the input samples x[k] to x[k+u _v −1] to the v-th sampling circuit. The obtained output is an array X _v (g) of one row u(v). After that, the processing makes a transition to Step S50.

[Step S50] The adders 5b1, 5b2,..., 5bV perform the following calculations. W(v)=W(v)+X _v (g), k=g·u(v). After that, the processing makes a transition to Step S51.
[Step S51] The periodic filter 50 performs the following calculation. g=g+1. After that, the processing makes a transition to Step S52.

[Step S52] The periodic filter 50 determines whether the parameter g is equal to or less than the number G. When the parameter g is the number G or less (Yes in step S52), the process proceeds to step S49. When the parameter g is larger than the number G (No in step S52), the process proceeds to step S53.
[Step S53] The periodic filter 50 performs the following calculation. v=v+1. After that, the processing makes a transition to Step S54.

[Step S54] The periodic filter 50 determines whether or not the parameter v is equal to or less than the number V. When the parameter v is the number V or less (Yes in step S54), the process proceeds to step S45. When the parameter v is larger than the number V (No in step S54), the process proceeds to step S55.

[Step S55] The periodic filter 50 identifies the maximum value from p(1) to p(V) and sets the value as p1. Normalize p(1) to p(V) so that p1 becomes 1, and obtain the obtained values as P(1) to P(V), respectively, so that the matrix P=[P(P(P 1), P(2),..., P(V)] are created. After that, the processing makes a transition to Step S56.

[Step S56] The first calculation unit 5d executes the first calculation on the matrix P. In addition, the second calculation unit 5e executes the second calculation. Then, a matrix P0=[P0(2), P0(3)..., P0(V)] of one row (V-2) column is created. After that, the processing makes a transition to Step S57.
[Step S57] The periodic filter 50 creates a matrix P00 by integrating the matrix P0. After that, the processing makes a transition to Step S58.

[Step S58] The periodic filter 50 determines whether the parameter m is less than or equal to the number of repetitions M. When the parameter m is equal to or smaller than the number of repetitions M (Yes in step S58), the process proceeds to step S44. When the parameter m is larger than the number of repetitions M (No in step S58), the process proceeds to step S59.

[Step S59] In the periodic filter 50, among the elements of the matrix P00, the number (element number) of an element whose S/N ratio is s[dB] (S is an arbitrary real number exceeding 0) is V _11, 1,. -Record as V _N 1. After that, the processing makes a transition to Step S60.

[Step S60] The periodic filter 50 outputs the number of elements N as the number of signals. Further, the periodic filter 50 outputs V _1, 1,..., V _N 1 as each period. Further, the periodic filter 50 sets W(V ₁ 1),... W(V _N 1) as the estimated basic waveform. After that, the processing of FIGS. 26 and 27 ends.
By this processing, it is possible to suppress the influence of noise and detect one or a plurality of signals even if the input has a low S/N ratio.
FIG. 28 is a diagram illustrating an example of processing of the periodic filter according to the third embodiment.
FIG. 28A shows a processing result when the number of repetitions M=5. FIG. 28B shows the processing result when the number of repetitions M=10.

As described above, in the case of an input having a low S/N ratio, even if the first process is performed, in addition to the samples (101 to 105 and 108) in FIG. ) Has been detected, the value of noise with respect to the signal becomes relatively small by performing the process of the periodic filter of the third embodiment, and it is easy for the sample (127) to be noise. I can judge.
<Modification>

29 and 30 are flowcharts for explaining the process of the modification of the periodic filter according to the third embodiment. The process of step S50a differs from the process of the periodic filter according to the third embodiment described above.

[Step S50a] The adders 5b1, 5b2,..., 5bV perform the following calculations. W(v)=W(v)×X _v (g), k=g·u(v). After that, the processing makes a transition to Step S51.

By this processing, it is possible to detect one or a plurality of signals while suppressing the influence of noise by using only addition and subtraction that consume less computational resources. Further, even when it is desired to detect a signal having a large intensity ratio, the signal and the noise can be clearly distinguished.

The arithmetic device and the program of the present invention have been described above based on the illustrated embodiment, but the present invention is not limited to this, and the configuration of each unit is an arbitrary configuration having the same function. Can be replaced with Further, other arbitrary components and steps may be added to the present invention.
Further, the present invention may be a combination of any two or more configurations (features) of the above-described embodiments.

The above processing functions can be realized by a computer. In that case, a program describing the processing contents of the function of the periodic filter 5 is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing content can be recorded in a computer-readable recording medium. Examples of the computer-readable recording medium include a magnetic storage device, an optical disc, a magneto-optical recording medium, and a semiconductor memory. Examples of the magnetic storage device include a hard disk drive, a flexible disk (FD), and a magnetic tape. Examples of the optical disc include a DVD, a DVD-RAM, a CD-ROM/RW and the like. Examples of the magneto-optical recording medium include MO (Magneto-Optical disk).

In order to put the program into the market, for example, a portable recording medium such as a DVD or a CD-ROM in which the program is recorded is sold. It is also possible to store the program in the storage device of the server computer and transfer the program from the server computer to another computer via a network.

The computer that executes the program stores, for example, the program recorded in the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, the computer can also sequentially execute processing according to the received program every time the program is transferred from a server computer connected via a network.

Further, at least a part of the above processing functions can be realized by an electronic circuit such as DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device).

1 integrated circuit 2 transmitter 3 receiver 4 ADC
5, 50 Periodic filter 5a1, 5a2,..., 5aV Sampling circuit 5b1, 5b2,..., 5bV Adder 5c Maximum value detection unit 5d First calculation unit 5e Second calculation unit

Claims (4)

From the input discrete-time signal, a plurality of signal output unit for outputting to the output port in parallel the continuous discrete points included in the basic period candidate sections having different pre-assigned time lengths,
An adder that calculates a cumulative value for each of the basic period candidate sections by adding the values of the signals output by the signal output units for each of the output ports;
A calculation unit that executes a calculation for deriving a basic cycle based on a difference in inclination caused by a locus of vertices of the cumulative value for each of the basic cycle candidate sections;
An arithmetic unit characterized by having. The calculation unit calculates a first difference by subtracting the cumulative value of the other one from the cumulative value of each of the pair of basic cycle candidate sections in a combination that is assigned in advance, and calculates the cumulative value of the one from the cumulative value of the other A first calculation unit that calculates a second difference by subtracting the value;
A second calculation unit that adds the first difference to a second difference of another combination;
The arithmetic unit according to claim 1, further comprising: The arithmetic device according to claim 2, wherein the second arithmetic unit adds the first difference using the one cumulative value and the second difference using the one cumulative value. On the computer,
From the input discrete-time signal, the continuous discrete points included in the basic period candidate sections having different pre-assigned time lengths are output in parallel to the output port,
The value of the signal output by each of the signal output unit, to calculate the cumulative value for each of the basic period candidate section by adding each of the output port,
Performing a calculation for deriving a basic cycle based on a difference in inclination caused by a locus of vertices of the cumulative value for each of the basic cycle candidate sections
A program that causes processing to be executed.

Images