WO2018163364A1

WO2018163364A1 - Filtering device and filtering method

Info

Publication number: WO2018163364A1
Application number: PCT/JP2017/009539
Authority: WO
Inventors: 裕幸三田村
Original assignee: 国立大学法人東京大学
Priority date: 2017-03-09
Filing date: 2017-03-09
Publication date: 2018-09-13
Also published as: JPWO2018163364A1; JP7062302B2

Abstract

The present invention provides technology capable of efficiently removing noise while performing detection in a short time using numerical phase detection with respect to an alternating-current signal. A detection unit 1 performs numeric phase detection with respect to an alternating-current signal. Consequently, a plurality of detection results corresponding to an accumulated number of vibrations in the numerical phase detection can be acquired. A linear combination unit 2 performs a linear combination of the plurality of detection results using a linear coefficient corresponding to a filter characteristic. This makes it possible to filter the alternating-current signal.

Description

Filtering apparatus and filtering method

The present invention relates to a technique for filtering an AC signal.

Analog phase detection and numerical phase detection (also referred to as digital phase detection) are known as techniques for performing phase detection (also referred to as synchronous detection) on AC signals.

The analog phase detection, when the respective frequencies of the signal to be demodulated and omega _0, multiplied by 2cosω ₀ t or 2sinω ₀ t to the original signal, removing the double vibration component (i.e. cos2ω ₀ t and sin2ω ₀ t) By doing so, constant components that do not depend on time are extracted. Here, in the analog phase detection, since the double vibration component is removed by the low-pass filter, there is a problem that integration for a long time is required.

On the other hand, numerical phase detection has a feature that a signal component can be taken out in a short time because a component of double oscillation can be removed by using an integral integral or half integral multiple of a reference period.

By the way, in the numerical phase detection, there is a problem that the noise removal performance is remarkably deteriorated when the cumulative number of vibrations (in short, the number of periods used for the interval integration) is small. On the other hand, when the cumulative number of vibrations is increased, there is a problem that it takes time to extract the signal component. That is, in the numerical phase detection, the reduction in processing time and the noise removal performance are in a trade-off relationship.

Therefore, in order to perform low-noise signal extraction processing in a short time while using numerical phase detection, another technique for removing noise is required.

The present invention has been made based on the above situation. The main object of the present invention is to provide a technique capable of efficiently removing noise while performing detection in a short time using numerical phase detection.

The means for solving the above-described problems can be described as the following items.

(Item 1)
A filtering device for filtering an AC signal,
It has a detection unit and a linear combination unit,
The detection unit is configured to obtain a plurality of detection results according to the cumulative number of vibrations by performing numerical phase detection on the AC signal.
The linear combination unit is configured to perform linear combination of the plurality of detection results using a linear coefficient corresponding to a filter characteristic.

(Item 2)
In addition, it has an output unit,
The filtering device according to item 1, wherein the output unit is configured to output a result obtained by linear combination in the linear combination unit.

(Item 3)
Furthermore, a linear coefficient determination unit is provided,
The linear coefficient determination unit is configured to calculate the linear coefficient using a relationship between the linear coefficients, an overall pass gain in the numerical phase detection, and a target filter characteristic. 2. The filtering device according to 2.

(Item 4)
A filtering method for filtering an AC signal,
Obtaining a plurality of detection results according to the cumulative number of vibrations by performing numerical phase detection on the AC signal; and
Filtering the AC signal by performing a linear combination of the plurality of detection results using a linear coefficient corresponding to a filter characteristic.

(Item 5)
A linear coefficient determination method used for the filtering method according to item 4, wherein the linear coefficient is determined using a relationship between the linear counts, an overall pass gain in the numerical phase detection, and a target filter characteristic. A linear coefficient determination method comprising a step of calculating a coefficient.

(Item 6)
A computer program for causing a computer to execute each step according to

item

4 or 5.

This computer program is stored in an appropriate recording medium (for example, an optical recording medium such as a CD-ROM or a DVD disk, a magnetic recording medium such as a hard disk or a flexible disk, or a magneto-optical recording medium such as an MO disk). Can be stored. This computer program can be transmitted via a communication line such as the Internet.

According to the present invention, it is possible to provide a technique for efficiently removing noise while performing detection in a short time using numerical phase detection.

It is a block diagram which shows the outline of the filtering apparatus which concerns on one Embodiment of this invention. It is a flowchart for demonstrating the filtering method performed using the filtering apparatus of FIG. It is a graph which shows the passage gain of numerical phase detection. FIGS. 3A and 3B show the real part and imaginary part of the gain corresponding to the integer period n = 1, 2, 3, 4, 5. FIG. FIG. 3C and FIG. 3D show the real part and the imaginary part of the gain corresponding to the half-integer period n = 1/2, 3/2, 5/2, 7/2, 9/2. 5 is a graph showing a real part (FIG. 4A) and an imaginary part (FIG. 4B) of an actual passing waveform when ω / ω ₀ = 0.1 and amplitude 1; In FIG. 4, “average” indicates an average value of integer cycles n = 1 and n = 2. It is a graph which shows the passage gain of the real part (Drawing 5 (a)) and the imaginary part (Drawing 5 (b)) of the multifrequency notch filter (example 1) in numerical phase detection by an integer period. FIGS. 6A and 6B show the real part and the imaginary part of a high-pass filter (Example 2) in which numerical phase detection with a period up to an integer m is linearly combined and the coefficients are optimized. 6 (c) and 6 (d) show the real and imaginary pass gains of the result of combining n = 1/2 and n = 3/2 (m = 3/2, μ = 2). Indicates. It is a graph which shows the passage gain of the real part (Drawing 7 (a)) and the imaginary part (Drawing 7 (b)) of the high order notch filter (example 3) in numerical phase detection by an integer period. FIG. 8A is a graph for explaining the time required to separately detect signals of 5 [Hz] and 8 [Hz], and FIG. It is a graph for demonstrating time required in order to carry out the separation detection of the signal of 1 [Hz]. FIGS. 8C and 8D are graphs showing the real part and the imaginary part of the pass gain in the case of FIG.

Hereinafter, a filtering device according to an embodiment of the present invention will be described with reference to FIG. This filtering device is for filtering AC signals.

(Configuration of the filtering device of the present embodiment)
The filtering device of this embodiment includes a detection unit 1 and a linear combination unit 2 as main components (see FIG. 1). Furthermore, the filtering device of this embodiment includes an output unit 3 and a linear coefficient determination unit 4 as additional elements.

(Detection part)
The detection unit 1 is configured to acquire a plurality of detection results corresponding to the cumulative number of vibrations by performing numerical phase detection on the AC signal. Details of the operation of the detector 1 will be described later.

(Linear combination part)
The linear combination unit 2 is configured to linearly combine a plurality of detection results obtained by the detection unit 1 using a linear coefficient corresponding to the filter characteristics. Details of the operation of the linear combination unit 2 will also be described later.

(Output part)
The output unit 3 is configured to output the result obtained by the linear combination in the linear combination unit 2. The output destination of the output unit 3 is, for example, a display or a printer, but other storage means or a remote device may be used. In the present embodiment, sending data to some device is also included in the concept of output.

(Linear coefficient determination unit)
The linear coefficient determination unit 4 is configured to calculate the linear coefficient used in the linear combination unit 2 using the relationship between the linear counts, the overall pass gain in numerical phase detection, and the target filter characteristics. Yes. A linear coefficient determination method will also be described later.

(Procedure of the filtering method of this embodiment)
Next, a filtering method using the above-described filtering device will be described with further reference to FIG.

(Step SA-1 in FIG. 2)
First, the linear coefficient determination unit 4 determines a linear coefficient corresponding to the target filtering characteristic. The linear coefficient is determined using the relationship between the linear counts, the overall pass gain in numerical phase detection, and the target filter characteristics. A specific example of determining the linear coefficient will be described later.

(Step SA-2 in FIG. 2)
Before or after the above-described step SA-1, or at the same time, the detection unit 1 performs numerical phase detection on the input AC signal. Thereby, in this embodiment, the some detection result according to the frequency | count of integration vibration can be acquired. Details of this detection method will also be described later.

(Step SA-3 in FIG. 2)
Next, the linear combination unit 2 performs linear combination of a plurality of detection results using a linear coefficient corresponding to the filter characteristic. Thereby, desired filtering with respect to an alternating current signal can be performed.

(Step SA-4 in FIG. 2)
Next, the output unit 3 outputs the obtained filtering result.

(Principle of filtering device)
Hereinafter, the principle of the filtering device in the present embodiment will be described in detail, and then specific examples will be described.

The basic idea of phase detection is that if the angular frequency of the signal to be demodulated is ω ₀ , the original signal is multiplied by 2 cos ω ₀ t or 2 sin ω ₀ t and smoothed, and a constant component independent of time is obtained. To extract. The original signal f (t)

If

Each

It becomes. If removing the section twice a vibration cos2ω _{0 t} and sin2ω _{0 t,} (corresponding to the real and imaginary parts) are constant component as a component which does not depend on time A, it is possible to obtain the B. The synchronous detection of a conventional analog type, so had to remove the double vibration component by a low-pass filter was required integration of long time with respect to the reference period of _{_{ω 0 (= 2π / ω 0}} ). In contrast, the recently proposed numerical phase detection method (digital method)

Using this relational expression, the vibration component can be removed by using interval integration at an integral multiple or half integer multiple of the reference period, and detection can be performed in a very short time.

Here, a pass gain (G _n ^CC : real part gain, G _n ^SS : imaginary part gain, G _n ^SC , with respect to a general angular frequency ω in phase detection in the n-fold period of the demodulated angular frequency ω ₀ . ₍ ^GnCS : Gain of orthogonal component)

It becomes. Here, the angular frequency is made dimensionless and set as x = ω / ω0. Since G _n ^SC (x) and G _n ^CS (x) are always zero, G _n ^CC (x) and G _n ^SS (x) when n = 1, 2, 3, 4, 5 are shown in FIG. ) and (b) show those when n = 1/2, 3/2, 5/2, 7/2 and 9/2, respectively. The characteristics of the gain differ depending on the cumulative number of vibrations (that is, the number of periods in the interval integration) n. The characteristics are different between the real part and the imaginary part of the gain. The actual amplitude of the passing waveform with respect to a specific angular frequency is a value that oscillates between the two values (that is, between the envelopes) (see FIGS. 4A and 4B). According to FIG. 4, it can be seen that the waveform varies depending on the cumulative vibration frequency n. Since x = 0.1 in the example of FIG. 4, the amplitude of the envelope oscillates between G _n ^CC (0.1) and G _n ^SS (0.1). If this is the case, there will be no significant attenuation effect, but when the results of n = 1 and n = 2 are added and divided by 2, they almost cancel each other. This is also a specific example of a third-order high-pass filter described later.

As can be seen from FIG. 3 and FIG. 4, the noise removal performance, which is said to be the advantage of phase detection, is extremely deteriorated due to the trade-off relationship between the shortening of the integrated vibration frequency n. Therefore, when the numerical phase detection method is used and n is small, a noise removal method must be considered separately as described above. Here, when an analog filter or a normal numerical filter is used, since data at a time far away is required, the advantage of being able to integrate in a short time is sacrificed. The cause of noise generally varies, but it is often the case that the noise is localized at a specific frequency. Therefore, the present embodiment proposes a method for efficiently removing a specific frequency component using data in a limited time range in numerical phase detection.

In the present embodiment, by utilizing the fact that the characteristics of the pass gain differ depending on the difference in the number of integrated vibrations n, and by taking a linear combination of the detection results based on a plurality of integrated vibrations, it is superior to the detection based on a single integrated number of vibrations. New knowledge that it can have a filter function. Note that in this linear combination, no extra phase rotation in the signal component occurs.

Here, when n is limited to an integer, a plurality of (μ (= 2, 3, 4,...)) Integrated vibration times n (where 1 ≦ n ≦ m, μ ≦ m = 2, 3, 4,...). The overall pass gain (G _total ^CC (x) and G _total ^SS (x)) in the case of taking a linear combination of the detection results is as shown in the following relational expression.

(If the inequality sign μ <m, it means that n is missing.)

In general

So,

Automatically imposes the condition

It becomes. _a _1, a 2, ..., in order to determine the _{a m} is a μ-1 or necessary after equations. Or to have any function in numeric filter of the present _{_{embodiment, a 1, a 2, ...}} , are conceivable various depending on the conditions imposed on a _m.

The method of this embodiment is the same when n is a half integer. Below, * is attached to the formula in the case of a half integer.

Multiple (μ (= 2, 3, 4, ...)) half integers

The _total passing gain (G _total ^CC (x) and G _total ^SS (x)) in the case of taking a linear combination of the detection results obtained by (1) is as shown in the following relational expression.

(Here, if the inequality sign μ−1 / 2 <m, it means that n is missing.)

also

So,

Automatically imposes the condition

It becomes. _a _1/2, a 3/2, ..., is the equation in order to determine the _{a m} is required after μ-1. It is also _a 1/2 to have what kind of function to this value _{filter, a 3/2, ...,} can be considered a variety depending on the conditions imposed on _{a m.}

The method of this embodiment is effective for removing noise of a specific angular frequency under a time constant as short as possible in numerical phase detection. Hereinafter, as an example, a configuration method of a multi-frequency notch filter, a high-pass filter, and a high-order notch filter will be described. An embodiment of a detection method that can be used as an alternative to the OFDM method will also be described. For the sake of simplicity, let us consider a case where there is no missing number in n. That is, in the case of an integer period, μ = m and in the case of a half integer period, μ−1 / 2 = m.

In the above description, acquisition of G _n ^CC or G _n ^SS used for filtering corresponds to an example of “acquisition of a plurality of detection results according to the number of integrated vibrations”. Moreover, calculating the coefficient of the linear combination of the detection results according to the purpose of filtering (by a computer or artificially) corresponds to an example of “determination of linear coefficient”.

Hereinafter, a more specific example of a linear coefficient determination method according to the purpose of filtering will be described as an example.

(Example 1: Multi-frequency notch filter)
Angular frequency of noise to be removed

Is discrete and finite (μ-1)

You can impose a condition that Generally from equations (7) and (8)

So,

Holds. If the angular frequency of the noise to be removed is ω _i (≠ ω ₀ ), then x _i = ω _i / ω ₀ ,

When n is an integer, only the zero of G _total ^SS (x _i ) _needs to be considered. In this case, m = μ. Similarly, when n is a half integer, only the zero point of G _total ^CC (x _i ) _needs to be considered, and in this case, m = μ−1 / 2. This does not apply when integers and half integers are mixed. When n is an integer and put together in a matrix,

It becomes. Linear coefficients a _1, a _{2, _...,} the system of linear equations for a _m can be solved numerically with ease. According to the first embodiment, the linear combination of detection results is performed using this linear coefficient. This allows arbitrary and known angular frequency

Numerical phase detection having a notch filter function for removing the noise component can be performed using only information of a time width of 2 mπ / ω ₀ at the maximum without rotating the phase of the signal. As an example, a filter design example in which the pass gain of the angular frequency of x = 0.3 is zero for both the real part and the imaginary part, and the pass gain of the angular frequency of x = 0.3 and x = 0.5 are simultaneously FIG. 5 shows a filter design example in which both the real part and the imaginary part are zero. Here, as an example, each of the case where the angular frequency of the noise to be removed is 0.3ω ₀ alone (m = 2) and the case where both 0.3ω ₀ and 0.5ω ₀ are both (m = 3). The parameters of the linear combination were determined. The same processing is not limited to the case where n is an integer, and can be performed in the same manner even when it is a half integer.

(Example 2: High-pass filter)
Expanding G _n ^CC (x) in Equation (7) and G _n ^SS (x) in Equation (8) around x = 0,

It becomes. It should be noted here that the coefficient of each lowest-order term is 2 · (−1) ^{n + 1} , so that even if n → ∞, it does not converge to zero. This means that there is almost no effect even if the cumulative number of vibrations n is simply increased in order to suppress noise at the low frequency limit. However, when m = 2, a ₁ = 1/2, and a ₂ = 1/2, from the equations (11), (12), (26), and (27)

Thus, the lowest order term can be erased while keeping G _total ^CC = G _total ^SS = 1 (where a ₁ + a ₂ = 1). This means that it is far more effective in removing low-frequency noise than simply setting the integrated vibration frequency n to 2 even though the time range required for integration is the same.

It is possible to push this idea further and eliminate coefficients of any order. Since equation (23) is established, G _total ^CC (ω) is automatically optimized when G _total ^SS (ω) is considered below. From equations (12) and (27), if m ≧ 3

Because

In combination with equation (14)

It becomes the simultaneous linear equation. By solving this equation, terms up to x ^2m-1 can be dropped. The specific exact solution is combined with the results of (32) and (33).

It becomes. The same applies to G _total ^CC (x). In this case, terms up to x ^{2 m} can be deleted. Since G _total ^SS (x) has a lower order than G _total ^CC (x), the function as a filter is determined by the characteristic of G _total ^SS (x).

As a result, numerical phase detection having a 2m-1 order high-pass filter function can be performed using only information of a time width of at most 2mπ / ω ₀ without rotating the phase of the signal. Specific gains of real part and imaginary part are shown in FIGS. 6 (a) and 6 (b). These figures show a real part and an imaginary part of a high-pass filter in which numerical phase detection with a period up to an integer m is linearly combined and the coefficients are optimized. It can be seen that each time the period used is increased by one, the order of the filter increases by two.

The above description assumes that n is an integer. However, the present invention is not limited to this, and the same processing as described above is possible when n is a half integer. As an example, the passing gains of the real part and the imaginary part when m = 3/2 are shown in FIGS. 6 (c) and 6 (d). These consist of no excellent characteristics that produced by their linear combination in the original _G ⁿ CC (x) and _G ⁿ SS (x). FIGS. 6C and 6D show the pass gains of the real part and the imaginary part of the combination of the results of n = 1/2 and n = 3/2 (m = 3/2). In the case of a half-integer period, each G _n ^CC has a finite value in the direct current limit ω / ω ₀ → 0, so there was a problem that the direct current component could not be dropped, but a plurality of (that is, μ) integrated vibrations This shows that if the coefficient of the linear combination of the detection result corresponding to the number n is taken well, this can be made zero. This is a specific example of a secondary high-pass filter, which is realized by integrating data for 3/2 periods.

It becomes. Here, the condition for dropping the coefficient up to δx2 ^m−1 in equation (12) is the same as in equation (31). Therefore, the filter of the type of Expression (33) has a function as a high-order notch filter for even harmonics (that is, x = 2k, k = 1, 2, 3,. . At this time, the odd multiple wave functions as a primary notch filter.

(Example 3: High-order notch filter)
Consider a high-order notch filter for removing noise distributed in the vicinity of angular frequency ω ₁ (≠ ω ₀ ) with a certain width. here

If this condition is imposed, an l-th order notch filter is realized. When n is an integer, from equation (23)

Because

Always holds.

If n is a half-integer, from formula (24)

The same can be done. This does not apply when integers and half integers are mixed.

From equation (12), for any integer j

So is _{_{true, a 1, a 2, ...}} , the simultaneous linear equations necessary for the determination of _{a m,}

It becomes. Here, l = m−1.

_a _1, a 2, ..., the system of linear equations for _{a m} can be solved numerically with ease. As a result, numerical phase detection with an m−1 order notch filter function that removes noise components of arbitrary and known angular frequency ω ₁ (≠ ω ₀ ) is maximized without rotating the phase of the signal. However, it can be performed using only information of a time width of 2 mπ / ω ₀ . As an example, FIG. 7 shows a design example of a first-order filter and a second-order filter that make the passing gain of the angular frequency of x = 0.3 zero for both the real part and the imaginary part. Here, the angular frequency ω ₁ of the noise to be removed is set to 0.3ω _0, and the parameters are determined so as to be a first-order (m = 2) and second-order (m = 3) filter. This is the same when n is not limited to an integer but a half integer.

(Example 4: Alternative to orthogonal frequency division division method (OFDM method))
Using the example shown in the first embodiment, it is possible to prevent interference when performing phase detection at high speed in parallel using a plurality of frequencies. Conventionally, when phase detection is performed at _two frequencies ν ₁ and ν ₂ and ν ₂ = (p ₂ / p ₁ ) ν ₁ (p and q are relatively prime natural numbers), in order to prevent interference, ν ₁ p ₁ times for, for the [nu ₂ was necessary cumulative number of oscillations _twice p. However, in the method of this example, the maximum number of integrated vibrations is sufficient for both, and in most cases, detection can be performed in a shorter time than in the past.

For example, if ν ₁ = 5 [Hz] and ν ₂ = 8 [Hz], p ₁ = 5 and p ₂ = 8 in the conventional method (“Ordinary method” in FIG. 8), and each accumulated time (p ₁ / (ν ₁ , p ₂ / ν ₂ ) are both 1 [sec]. On the other hand, in the method of this example (“New method” in FIG. 8), the respective integration times (2 / ν ₁ , 2 / ν ₂ ) are 0.4 [sec] and 0.25 [sec], respectively. It can be shorter than the conventional method (see FIG. 8 (a)).

Next, the case where ν ₂ = 8.1 [Hz] is considered. In the conventional method, p ₁ = 50 and p ₂ = 81, and the respective integration times are both 10 [sec], which is much longer than that in the case of ν ₂ = 8 [Hz]. On the other hand, in the method of this example, each integration time is 0.4 [sec], 0.24. . . [Sec], which is not much different from the case of ν ₂ = 8 [Hz] (see FIG. 8B). Therefore, in both cases of ν ₂ = 8 [Hz] and ν ₂ = 8.1 [Hz], the method of this example can reduce the integration time.

Fig. 8 (c) and (d) show the passing gains of the individual real and imaginary parts designed in the former case. In the case of passing 5 [Hz], the passing gain is 1 for both the imaginary part and the real part at 5 [Hz], but the passing gain is 0 for both the imaginary part and the real part at 8 [Hz]. On the other hand, in the case of passing 8 [Hz], the pass gain of the real part imaginary part is 1 at 8 [Hz], but the pass gain of the real part imaginary part is 0 at 5 [Hz].

If this idea is advanced for the case where more frequencies are used, detection can be completed in a shorter integrated time than the orthogonal frequency division division method (OFDM method) currently used in broadband communication or the like depending on the conditions. This is considered to contribute to the improvement of the data transmission density of various communications. The comparison with the OFDM method is described below.

First, consider a case where phase detection is performed by the OFDM method using _N frequencies ν ₁ , ν ₂ ,..., Ν _N. in this case

To become a set of integers _p _1, p 2, ..., each frequency ν ₁ to _{p N} is present, ν _2, ..., begins with choosing a ν _N. Ie

It is. At this time, τ is an integration time for all N frequencies. In practice, we want to make the frequency band to be used as narrow as possible.

Is generally. M is an integer such that M ≧ N, and is a parameter that determines the maximum frequency. When M = N

But in general it doesn't have to start from 1.

It is. Therefore

It becomes. Rewriting this for τ

It becomes.

Next, the case where the method according to the present embodiment is used will be considered. Although this method has some degree arbitrariness in-taking frequency, where also N frequency [nu ₁ and OFDM _method, [nu _2, ..., a [nu _N

Take it. According to the consideration in the first embodiment, the total number of vibrations required for detection without interfering with each other at N frequencies is N times at each frequency. Therefore, if the integration time at each frequency is τ _i

It becomes.

Next, let us consider the conditions under which the method of this example shortens the integration time compared to the OFDM method. To make τ _i ≦ τ at all N frequencies,

Should be established. Solving this for N results in N ≦ M / 2 + 1/2. This means that when using a frequency above approximately half of the maximum frequency, the narrower the bandwidth, the more advantageous the method of this embodiment over the OFDM method in terms of data transfer density. To do. In the case of communication using public radio waves, etc., the method of this example is useful because there are restrictions on the frequency band that can be used.

As described above, in the method of the present embodiment, the frequency ν _{_1,} ν _{_2, ...,} the way of taking the [nu _N, there is some arbitrariness. In the above example, the arithmetic sequence is the same as in the OFDM method. However, in practice, data sampling is performed at discrete times, so it is necessary to consider that the phase returns to the original state by N vibrations at each frequency. is there. Based on this consideration, it is convenient to set the reciprocal frequency ν ₁ ⁻¹ , ν ₂ ⁻¹ ,..., Ν _N ⁻¹ to be an integral multiple of the reciprocal number of the sampling rate ν _0. A method is possible with this approach.

(Advantages of this embodiment)
As described above, the filter characteristics that are newly created by taking the linear combination of the numerical phase detection results based on multiple accumulated vibration counts are not limited to those shown as examples, and combinations of them can be freely created. is there. In general, using the applied method, it is possible to efficiently remove noise at a specific frequency while taking advantage of the feature of numerical phase detection that “integration time can be shortened”.

As described above, the method of the present embodiment uses the fact that the characteristic of the pass gain varies depending on the number of accumulated vibrations in numerical phase detection, and the detection result of a plurality of (ie, μ ≧ 2) accumulated vibrations By taking the linear combination, it is possible to have a characteristic of a pass gain that does not exist in the detection based on the single accumulated number of vibrations.

Demand for numerical phase detection technology for AC signals extends to various industrial fields through measurement and communication. For example, in measurement, built-in phase detection technology is widely used to demodulate the output of various sensors with high sensitivity. Basically, the filtering technique of this embodiment is considered to contribute to the performance improvement of the lock-in amplifier. In the method of this example, even when the modulation frequency cannot be increased for some reason, demodulation including noise removal can be performed with data accumulation in a shorter time than conventional methods. If multiple sensors can be used at the same time to prevent interference and respond quickly, it is expected that all automatic machines and robots can be controlled more accurately and agilely. In communication, it can be expected that the amount of transmission data can be efficiently increased while preventing interference at a predetermined frequency.

The contents of the present invention are not limited to the above embodiments. In the present invention, various modifications can be made to the specific configuration within the scope of the claims.

For example, in the above-described embodiment, the linear coefficient determination unit 4 calculates the linear coefficient. However, for example, the linear coefficient may be a predetermined value calculated in advance according to the purpose of filtering. Or the structure which changes a linear coefficient dynamically according to the use of filtering is also possible.

Further, each of the above-described constituent elements only needs to exist as a functional block, and does not need to exist as independent hardware. As a mounting method, hardware or computer software may be used. Furthermore, one functional element in the present invention may be realized by a set of a plurality of functional elements, and a plurality of functional elements in the present invention may be realized by one functional element.

Furthermore, the functional elements may be arranged at physically separated positions. In this case, the functional elements may be connected by a network. It is also possible to realize functions or configure functional elements by grid computing or cloud computing.

DESCRIPTION OF SYMBOLS 1 Detection part 2 Linear combination part 3 Output part 4 Linear coefficient determination part

Claims

A filtering device for filtering an AC signal,
It has a detection unit and a linear combination unit,
The detection unit is configured to obtain a plurality of detection results according to the cumulative number of vibrations by performing numerical phase detection on the AC signal.
The linear combination unit is configured to perform linear combination of the plurality of detection results using a linear coefficient corresponding to a filter characteristic.
In addition, it has an output unit,
The filtering device according to claim 1, wherein the output unit is configured to output a result obtained by linear combination in the linear combination unit.
Furthermore, a linear coefficient determination unit is provided,
2. The linear coefficient determination unit is configured to calculate the linear coefficient using a relationship between the linear coefficients, an overall pass gain in the numerical phase detection, and a target filter characteristic. Or the filtering apparatus of 2.
A filtering method for filtering an AC signal,
Obtaining a plurality of detection results according to the cumulative number of vibrations by performing numerical phase detection on the AC signal; and
Filtering the AC signal by performing a linear combination of the plurality of detection results using a linear coefficient corresponding to a filter characteristic.
The linear coefficient determination method used for the filtering method according to claim 4, wherein the relationship between the linear counts, the overall pass gain in the numerical phase detection, and the target filter characteristics are used. A linear coefficient determination method comprising a step of calculating a linear coefficient.
A computer program for causing a computer to execute each step according to claim 4 or 5.