Measuring device, measuring method and measuring program thereof

Abstract

The object is to provide a measuring device and a measuring method and a measuring program thereof that can effectively reduce the influence of noise. The measuring device (1) comprises: N (N≥2) sensors (S0, S1, S2); a selection unit (4) that selects a given combination of sensors for each measurement and outputs a detection value to each of M (M<N) sensor terminals based on the measurement value of the selected sensor; a detection unit (5) that obtains the detection value output to each of the M sensor terminals for each measurement; and a correction unit (8) that, after performing L measurements, assumes that each of the M×L detection values ​​obtained by the detection unit (5) contains a time-dependent noise component and corrects each of the M×L detection values ​​so that the time-dependent noise component of each detection value becomes a common noise component, which is a value obtained by averaging the L time-dependent noise components in the time axis direction.

Description

Technical Field

[0001] The present disclosure relates to an assay device, an assay method, and an assay program thereof. Background Art

[0002] Sensors that digitize physical phenomena occurring within a specific space, such as capacitive touch sensors used in touch panels, image sensors used in digital cameras, and more complex three-dimensional image sensors, are widely used in various technical fields. Because these sensors typically perform measurements continuously, the measured values ​​obtained from each measurement contain noise inherent to the measurement (hereinafter referred to as "time-dependent noise").

[0003] In order to reduce the influence of such time-dependent noise, for example, Patent Document 1 discloses a method for normalizing the time-dependent noise contained in each measurement value by making the time-dependent noise contained in each measurement value consistent with the time-dependent noise contained in the measurement value when a certain benchmark is measured.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: U.S. Patent No. 8,976,145 Summary of the Invention

[0007] -Problems to be solved by the invention-

[0008] However, according to the method described in Patent Document 1, depending on the magnitude of the normalized noise component, there is a possibility that each measurement value will uniformly include a large noise component, and the influence of the noise on the detection result will increase.

[0009] The present disclosure has been made in view of such circumstances, and an object of the present disclosure is to provide a measurement device, a measurement method, and a measurement program thereof that can effectively reduce the influence of time-dependent noise.

[0010] -Methods for solving the problem-

[0011] The first embodiment of the present disclosure is a measuring device comprising: N (N≥2) sensors; a selection unit which selects a given combination of sensors from the N sensors for each measurement, and outputs a detection value to each of M (M<N) sensor terminals based on the measurement value of the selected sensor; an acquisition unit which acquires the detection value output to each of the M sensor terminals for each measurement; and a correction unit which, after performing L measurements, assumes that each of the M×L detection values ​​acquired by the acquisition unit contains a time-dependent noise component, and corrects each of the M×L detection values ​​so that the time-dependent noise component of each of the detection values ​​becomes a value obtained by averaging the L time-dependent noise components in the time axis direction, that is, a common noise component.

[0012] The second embodiment of the present disclosure is a measuring method for a measuring device having N (N≥2) sensors, comprising the following steps: selecting a given combination of sensors from the N sensors for each measurement, and outputting a detection value to each of M (M<N) sensor terminals based on the measurement value of the selected sensor; obtaining a detection value output to each of the M sensor terminals for each measurement; and after performing L measurements, assuming that each of the M×L detection values ​​obtained contains a time-dependent noise component, correcting each of the M×L detection values ​​so that the time-dependent noise component of each of the detection values ​​becomes a value obtained by averaging the L time-dependent noise components in the time axis direction, that is, a common noise component.

[0013] The third method disclosed in the present invention is a measurement program for a measurement device having N (N≥2) sensors, which is used to enable a computer to perform the following processing: selecting a given combination of sensors from the N sensors for each measurement, and outputting a detection value to each of M (M<N) sensor terminals based on the measurement value of the selected sensor; obtaining the detection value output to each of the M sensor terminals for each measurement; and after performing L measurements, assuming that each of the M×L detection values ​​obtained contains a time-dependent noise component, correcting each of the M×L detection values ​​so that the time-dependent noise component of each of the detection values ​​becomes a value obtained by averaging the L time-dependent noise components in the time axis direction, that is, a common noise component.

[0014] -Effects of the Invention-

[0015] According to the present disclosure, there is an effect that the influence of noise can be effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a diagram showing the system configuration of the measurement device according to the first embodiment of the present disclosure.

[0017] Figure 2This is a diagram showing an example of the hardware configuration of the measurement device according to the first embodiment of the present disclosure.

[0018] Figure 3 This is a flowchart showing an example of a procedure of processing by the measurement device according to the first embodiment of the present disclosure.

[0019] Figure 4 This is a diagram showing an example of a scan matrix according to the first embodiment of the present disclosure.

[0020] Figure 5 It is a diagram showing another example of the system configuration of the measurement device according to the first embodiment of the present disclosure.

[0021] Figure 6 It is a diagram showing a system configuration of a measuring device according to a second embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] [First embodiment]

[0023] Hereinafter, the first embodiment of the measuring device, the measuring method and the measuring procedure thereof involved in the present disclosure will be described with reference to the accompanying drawings. The measuring device 1 in this embodiment is a measuring device for measuring a physical phenomenon generated in a given space, and can use a sensor that measures the physical phenomenon and digitizes it. As an example, a sensor used in an input device, etc. can be cited. More specifically, for example, the measuring device 1 can be used in an electrostatic capacitive touch sensor used in a touch panel, a sensor for sliding input, an image sensor such as a digital camera, a three-dimensional image sensor, etc. In the following description, the case of using the measuring device of the present disclosure in an electrostatic capacitive touch sensor used in a touch panel to detect the position of a detection object (such as a finger) will be described as an example.

[0024] (Structure of the measuring device 1)

[0025] Figure 1 1 is a diagram showing the system configuration of the measuring device 1 according to the first embodiment of the present disclosure. Figure 1 As shown, the measuring device 1 according to this embodiment mainly includes a sensor unit 2 and a circuit unit 3. The circuit unit 3 mainly includes a selection unit 4, a control unit 6, a detection unit (acquisition unit) 5, a correction unit 8, and an output conversion unit (demodulation unit) 7.

[0026] The sensor unit 2 includes N (N ≥ 2) sensors. Since this embodiment assumes a touch panel, each sensor is, for example, an electrostatic capacitance type proximity sensor (electrostatic capacitance sensor). Figure 1, a simplified spatial distribution sensor is shown, and as an example, a case where three sensors (sensor S0, sensor S1, sensor S2) are provided (the number of sensors N = 3) will be described. The number of sensors is not limited to three.

[0027] Each sensor outputs the degree of proximity as a measurement value. In the sensor's detection space, the closer an object (e.g., a human finger, a stylus) is to the sensor, the higher the value output as the measurement result.

[0028] The selection unit 4 selects a given combination of sensors from N sensors for each measurement, and outputs a detection value to each of the M (M < N) sensor terminals based on the measurement value of the selected sensor. A large number of sensors are provided in touch sensors, etc., but on the other hand, from the perspective of circuit area, cost, etc., it is not realistic to provide the detection circuits p1 and p2 described later in the same number as the number of sensors. For this reason, the number of detection circuits p1 and p2 is less than the number of sensors (N = 3). Therefore, the selection unit 4 selects sensors in a given combination and outputs the detection value of the selected sensor to the detection circuits p1 and p2. The combination of the selected sensors is pre-set in multiple measurements. In particular, when performing a measurement, the selection unit 4 selects a sensor that includes a combination of sensors that is different from that of the previous measurement.

[0029] exist Figure 1 In the example, the number of sensors is 3 (S0 to S2), whereas there are only 2 detection circuits p1 and p2. Therefore, at a certain measurement timing, 2 measurement values ​​of the sensors are selected and input to the detection circuits p1 and p2.

[0030] The control unit 6 controls the selection unit 4 so that a given combination of sensors is selected for each measurement. The combination of sensors selected is pre-set in the control unit 6. For example, the number of measurements L is set to 2. At time t1, the detection value from sensor S0 is input to detection circuit p1, and the detection value from sensor S2 is input to detection circuit p2. Subsequently, at time t2, the detection value from sensor S1 is input to detection circuit p1, and the detection value from sensor S2 is input to detection circuit p2.

[0031] The measurement is performed at a given time interval or at a given timing. Specifically, the control unit 6 performs L measurements at a given time interval or at a given timing by controlling the selection unit 4 and the detection unit (acquisition unit) 5. That is, the control unit 6 controls so that each measurement is performed at a predetermined measurement interval. The measurement interval is preset and is set short, thereby expecting to suppress changes in the true value. Specifically, it is preferable to end all measurements earlier than the environmental changes assumed for L detections. For example, the count value of the sensor changes from 0 to 100 due to the approach of a finger. In addition, it is assumed that the count value changes from 0 to 100 in the fastest 1 second. In the case where L measurements are performed in 1 second in this sensor, a change from 0 to 100 counts occurs during the period until the L measurements are performed. However, if L measurements can be performed within 0.01 seconds, the change during the period until the L measurements are performed can be reduced to the amount of 1 count. In this way, it is desired that the given number of measurements be completed within the shortest possible period.

[0032] The detection unit (acquisition unit) 5 acquires the detection value outputted from each of the M sensor terminals for each measurement. Specifically, the detection unit 5 has a scanning matrix Z, which is a matrix of M rows and N columns and has the same number as the number of measurements L. i (0≤i≤L), in the i-th detection, obtain the Z-based i The measured value of the selected sensor is used as the detection value. The detection unit 5 has M detection circuits connected to each sensor terminal. In the case where the detection circuit j is connected to the sensor k in the i-th detection, the detection circuit j is connected to the sensor k in the matrix Z. i 1 is stored in the jth row and kth column of the matrix Z. If the sensor k is not connected, then i 0 is stored in the jth row and kth column of the matrix. In this embodiment, the detection unit 5 includes detection circuits p1 and p2 (M = 2), and measures the physical quantities of three sensors S0, S1, and S2 (N = 3) in two scans (L = 2). Therefore, the matrices Z1 and Z2 have two rows and three columns. Ideally, the true value (a value without noise components) of each sensor is input to the detection circuits p1 and p2. However, in reality, the values ​​detected by the detection circuits p1 and p2 contain noise components (time-dependent noise components).

[0033] Specifically, in the measurement at time t1, the detection value V input to the detection circuit p1 is p1 The noise found in (t1) is set to N p1 (t1) The detection value V is input to the detection circuit p2. p2 The noise found in (t1) is set to N p2 In the case of (t1), if the time-dependent noise component commonly found in the detection circuit p1 and the detection circuit p2 is N(t1), then N p1 (t1) = N(t1) and Np2 (t1) = N(t1). That is, in measurements at the same timing, the noise components input to detection circuits p1 and p2 are equal. Because the sensors are located close to each other, the noise detected in detection circuits p1 and p2, if sufficiently small compared to the noise component N(t1), is omitted.

[0034] On the other hand, since the noise component has time dependency, if the timing of measurement is different, for example, at time t2, the noise component becomes N(t2), which is different from N(t1).

[0035] In this embodiment, the first measurement is performed at time t1. The detection value from sensor S0 is input to detection circuit p1 at time t1, and the detection value from sensor S2 is input to detection circuit p2 through the selection unit 4 that refers to the scanning matrix Z1. Then, at time t2, the detection value from sensor S1 is input to detection circuit p1, and the detection value from sensor S2 is input to detection circuit p2. Therefore, the detection value V p1 (t1) is expressed as the following equation (1), the detection value of the detection circuit p2 at time t1, that is, V p2 (t1) becomes the following formula (2).

[0036] [Mathematical formula 1]

[0037] V p1 (t1)=V S0 +N(t1) (1)

[0038] V p2 (t1)=V S2 +N(t1) (2)

[0039] Here, V S0 is the true value of the detection value of sensor S0, V S2 The true value of the detection value of sensor S2 is denoted by . The true value is the detection value of an ideal sensor without any noise component.

[0040] Similarly, at time t2, the detection value V of the detection circuit p1 at time t2 is obtained by the selection unit 4 referring to the scanning matrix Z2. p1 (t2) is expressed as the following equation (3), the detection value of the detection circuit p2 at time t2, that is, V p2 (t2) becomes the following formula (4).

[0041] [Mathematical formula 2]

[0042] V p1 (t2) = V S1 +N(t2) (3)

[0043] V p2 (t2) = V S2 +N(t2) (4)

[0044] Here, V S1 = is the true value of the detection value of sensor S1. In this way, the detection values ​​corresponding to the combination of sensors for each measurement are input to each detection circuit p1 and p2. After the number of measurements is performed, the detected information is output to the correction unit 8 described below.

[0045] The correction unit 8 assumes that after performing L measurements, each of the M×L detection values ​​obtained by the detection unit (acquisition unit) 5 contains a time-dependent noise component. The correction unit 8 then corrects each of the M×L detection values ​​so that the time-dependent noise component of each detection value becomes a common noise component, a value obtained by averaging the L time-dependent noise components along the time axis. Thus, the common noise component is the average value of the time-dependent noise components contained in each detection value. Specifically, the correction unit 8 assumes that each detection value detected through multiple measurements contains the time-dependent noise component at each measurement time, and calculates a corrected value (corrected detection value) by adding a predetermined number of time-dependent noise components to the true value.

[0046] As a specific process, the correction unit 8 sets each detection value as a column vector V in ={V p1 (t1), V p2 (t1), V p1 (t2), V p2 (t2)} T , multiplied by the transformation matrix M decode To perform correction related to the noise component. in Based on the number of measurements (L) and the number of detection values ​​obtained in one measurement (i.e., the number of detection circuits (M),), it is represented by an L×M dimensional column vector (a matrix with L×M elements arranged vertically). In this embodiment, it is a column vector with a dimension of 2×2=4. Since the transformation matrix transforms a four-dimensional column vector into a column vector of the same four dimensions, it becomes a 4×4 matrix as shown in the following equation (5).

[0047] [Mathematical formula 3]

[0048]

[0049] The transformation matrix of formula (5) represents an example corresponding to the column vector of this embodiment. The derivation of formula (5) will be described later. in Through the 4-row and 4-column matrix Mdecode Corrected to 4 rows of column vector V′ in If V′ is set in ={V′ p1 (t1), V′ p2 (t1), V′ p1 (t2), V′ p2 (t2)} T , then the correction is performed as shown in the following formula (6).

[0050] [Formula 4]

[0051]

[0052] By using M as in formula (6) decode Column vector V in After transformation, the corrected column vector V′ in Each element in is corrected to a value obtained by adding the average value of the time-dependent noise to the true value corresponding to each sensor as the offset noise. In other words, a common noise component of the same value is added to each detection value. In other words, the correction unit 8 obtains the common noise component by calculating the average value of L time-dependent noise components. It can be said that the correction unit 8 corrects each of the M×L detection values ​​to form a correction matrix M with M×L rows and M×L columns. decode The corrected detection values ​​are output to the output conversion unit 7 described later.

[0053] The output conversion unit 7 calculates the detection results corresponding to each of the N sensors based on the M×L detection values ​​corrected by the correction unit 8. That is, the output conversion unit 7 uses the corrected detection values ​​to calculate the detection results corresponding to each of the sensors S0, S1, and S2. Specifically, the column vector V′ obtained by equation (6) is in For example, if the detection results corresponding to sensors S0, S1, and S2 are respectively set as {V′ S0 , V′ S1 , V′ S2} T , then as shown in the following formula (7), using M as the transformation matrix out to perform the transformation.

[0054] [Formula 5]

[0055]

[0056] That is, the output conversion unit 7 uses the demodulation matrix M with N rows and M columns to the corrected M×L detection values. out , to output the detection results corresponding to each of the N sensors. outThe value of each matrix element affects the size of the final offset noise. For example, if the sum in the row direction is different in each row, it will become uneven. out In the example, the sum in the row direction is preferably set so that it is equal in each row. The smaller the sum in the row direction, the smaller the offset can be. The method of canceling noise by making the sum in each row direction zero is described in the second embodiment.

[0057] In this way, V′ is obtained as the detection result corresponding to each sensor S0, S1, and S2. S0 , V′ S1 , V′ S2 Since these detection results are obtained by adding the average value of the noise component to the true value measured by each sensor S0, S1, and S2 as the common noise component, the noise component for each true value is normalized, suppressing the deviation caused by the influence of the time-dependent noise component.

[0058] In this embodiment, the correction unit 8 performs the correction based on M decode The processing is performed in the output conversion unit 7 based on M out However, it is also possible not to separate the processing and set the transformation matrix to M out ·M decode , aggregated into one matrix and calculated simultaneously.

[0059] In this embodiment, the offset noise is equalized using the average value of each time-dependent noise component. However, as long as a common noise component is calculated by statistically processing each time-dependent noise component, the average value is not limited to the average value. However, in order to effectively suppress the influence of the magnitude relationship between each time-dependent noise component, it is preferable to use the average value.

[0060] (Hardware Configuration Diagram of Measurement Device 1)

[0061] Figure 2 It is a diagram showing an example of the hardware configuration of the measurement device 1 according to this embodiment.

[0062] like Figure 2 As shown, the measuring device 1 includes a processor (computer system). For example, the measuring device 1 includes a CPU 11; a ROM (Read Only Memory) 12 for storing programs executed by the CPU 11; a RAM (Random Access Memory) 13 that functions as a workspace during program execution; a hard disk drive (HDD) 14 as a mass storage device; and a communication unit 15 for connecting to a network or the like. A solid-state drive (SSD) can also be used as the mass storage device. These components are connected via a bus 18.

[0063] The measuring device 1 may include an input unit including a keyboard, a mouse, and the like, a display unit including a liquid crystal display device for displaying data, and the like.

[0064] The storage medium for storing the program etc. executed by the CPU 11 is not limited to the ROM 12. For example, it may be another auxiliary storage device such as a magnetic disk, a magneto-optical disk, or a semiconductor memory.

[0065] (Processing Flow of the Measurement Device 1)

[0066] Next, refer to Figure 3 An example of the processing of the above-mentioned measuring device 1 will be described. Figure 3 This is a flowchart showing an example of a procedure of processing by the measurement device 1 according to this embodiment. Figure 3 The illustrated flow is executed, for example, when a measurement is started. Figure 3 The illustrated process is repeated at given time intervals to continuously perform measurements.

[0067] The series of processing described below is recorded in the form of a program on the hard disk drive 14 (see Figure 2 ), etc., and the CPU 11 reads the program into the RAM 13 and executes information processing and calculations to realize the various functions described below. The program can be used in a state pre-installed in the ROM 12 or other storage media, in a state stored in a computer-readable storage medium, or in a state distributed via a wired or wireless communication means. Computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, semiconductor memories, etc.

[0068] First, the selection unit 4 selects a sensor combination corresponding to the Ln-th number of measurements (S101). Ln is initialized to 1, and when S101 is executed for the first time, a combination of Ln=1 is adopted.

[0069] Next, the detection unit 5 detects detection values ​​from each selected sensor ( S102 ).

[0070] Next, it is determined whether the number of measurements has reached a given value ( S103 ). In this embodiment, the given value is set to 2. The determination process of S103 is performed by the control unit 6 , for example.

[0071] If the number of measurements has not reached the predetermined value (NO determination in S103), Ln→Ln+1 is set, the number of measurements is incremented by 1 (S104), and S101 is executed again. The determination process in S104 is performed by the control unit 6, for example.

[0072] When the number of measurements reaches a predetermined value (YES determination in S103), the correction unit 8 performs correction to equalize the noise components in each detection value detected in each measurement (S105). Specifically, S105 performs conversion according to equation (6).

[0073] Next, the output conversion unit 7 performs output conversion based on each corrected detection value, and calculates the detection result corresponding to each sensor (S106). Specifically, S106 performs conversion according to equation (7).

[0074] In this manner, the processing in the measurement device 1 is executed to equalize the offset of the noise.

[0075] (Effect of equalizing offset noise)

[0076] Next, the effect of equalizing offset noise by the above-described processing will be described.

[0077] If the above column vector V in Expressed through a matrix, it is the transformation matrix M scan and the column vector V ideal The product of is represented as follows (8).

[0078] [Formula 6]

[0079]

[0080] Therefore, the formula (6) is represented as the following formula (9).

[0081] [Formula 7]

[0082]

[0083] As is clear from equation (9) (and other examples, such as equation (7)), all noise is reduced to 1 / 2 and added to the true value. In other words, the offset noise is equalized by being the average of N(t1) and N(t2).

[0084] For example, V S0 100, V S1 40, V S2 This is an example of the detection result when the finger is closest to the sensor S0. In addition, N(t1) is 100 and N(t2) is -20. In this situation, without performing the above-mentioned noise equalization, V S0 +N(t1)=100+100=200,V S1 +N(t2)=40-20=20,V S2+N(t1)=10+100=110. That is, although the finger is originally closest to the sensor S0, there is a possibility that the finger is erroneously detected as being near each of the sensors S0 and S2 due to the influence of noise.

[0085] In this situation, when the noise is made to match either N(t1) or N(t2) (assuming that it matches N(t1)), V S0 +N(t1)=100+100=200,V S1 +N(t1)=40+100=140,V S2 +N(t1)=10+100=110. That is, when N(t1) is large, a large offset is applied, and there is a possibility that the true value is buried in the noise.

[0086] However, in this embodiment, the offset noise included in the detection value is calculated as the average of multiple noise components obtained during different measurements. In other words, the offset noise included in each measurement value is (N(t1) + N(t2)) / 2 = 40, which can reduce the absolute value of the offset noise.

[0087] Generally, random values ​​such as noise are often described as normal distribution. As a characteristic of normal distribution, it is known that it follows the mean 0 and variance σ 2 The result of extracting K samples from the normal distribution and adding them together is a mean of 0 and a variance of K×σ 2 Similarly, it is known that it will follow a normal distribution with mean 0 and variance σ 2 The result of multiplying the samples of the normal distribution by 1 / K is 0 in average and (σ / K) in variance. 2 For this reason, if we assume that the noise component N(t) follows a normal distribution with mean 0 and variance σ 2 If the normal distribution is set, then when the noise component is multiplied by 1 / K times and summed up K times, it is considered that its value follows the mean 0 and variance (σ) 2 / K. That is, the variance of the offset is reduced to 1 / K times.

[0088] Therefore, in this embodiment, the influence of noise can be effectively reduced, and it can be said that the measurement accuracy can be improved.

[0089] (Scan matrix Z i and the transformation matrix M decode Description)

[0090] Next, the scanning matrix Z that can perform the function of the present disclosure, namely, equalization of offset noise, is described. i And the transformation matrix M decode .

[0091] (Z iconditions)

[0092] An arbitrary scanning matrix Z constituting the sensor circuit of the present disclosure i and Z j (0≤i<j<L) satisfies the following condition a or condition b.

[0093] <Condition a>

[0094] Existence Z i The p-th row component and Z j The qth row of the same components p, q (0≤p,q<M)

[0095] <Condition b>

[0096] Existence Z i The p0th row component and Z k0 The q0th row has the same components p0, q0, k0 (0≤p0, q0<M, 0≤k0<L),

[0097] Existence Z k0 The p1th row component and Z k1 The q1th row has the same components as p1, q1, k0, k1 (0≤p1, q1<M, 0≤k0, k1<L),

[0098] Existence Z k1 The p2th row component and Z k2 The q2th row has the same components as p1, q2, k1, k2 (0≤p2, q2<M, 0≤k1, k2<L),

[0099] ···

[0100] Existence Z kx-1 The pth x-1 Row components and Z kx The qth x-1 P with the same row components x-1 ,q x-1 、k x-1 、k x (0≤p x-1 ,q x-1 <M, 0≤k x-1 、k x <L)、

[0101] Existence Z kx The pth x Row components and Z j The qth x p with the same row components x ,q x 、k x (0≤p x ,q x <M, 0≤kx 、 j < L, 0 < x < L - 1).

[0102] For example, Figure 4 as shown in the upper layer of, Z1 to Z4 respectively with respect to Z i and Z j Satisfy condition a or condition b. Specifically, when i = 1 and j = 2, condition a is satisfied. That is, like R1 in Z1 and Z2 in the upper layer of UR, the first row of Z1 and the first row of Z2 are the same. When i = 1 and j = 4, condition a is not satisfied (the rows of Z1 and Z4 in the upper layer are all different), but condition b is satisfied. That is, like R2 in Z1 and Z3 in the upper layer, the second row of Z1 and the second row of Z3 are the same, and like R3 in Z3 and Z4 in the upper layer, the first row of Z3 and the first row of Z4 are the same. Like R4 in Z2 and Z4 in the upper layer, the second row of Z2 and the second row of Z4.

[0103] On the other hand, since Figure 4 as shown in the lower layer of, for the scanning matrices Z1 to Z4, there are i and j that do not satisfy the conditions respectively, so they cannot be used as any scanning matrix constituting the sensor circuit of the present disclosure. In Figure 4 the lower layer of, the corresponding rows are shown as R5 to R8 respectively.

[0104] If the scanning matrix Z as described above is prepared i , then the following proposition M1 holds.

[0105] <Proposition M1>

[0106] When the scanning matrix Z i satisfies condition a or condition b, for any i, j, there must exist D(i, j) that satisfies the following formula (10).

[0107] [Mathematical formula 8]

[0108] D(i, j) V in = N(t i ) - N(t j ) (10)

[0109] <Proof of Proposition M1>

[0110] According to D(j, i) = -D(i, j), when Proposition M1 holds in the case of i < j, the proposition also holds in the case of i > j. Therefore, only the case of (i < j) will be proven hereafter. First, the case of satisfying condition a will be proven.

[0111] When a certain i, j (i < j) satisfies condition a, according to the definition of condition a, since the detection circuit p at time t i and the detection circuit p at time tj The detection circuit q at the same scanning setting becomes the same, so the detection value is V p (t i ) and V q (t j ) are consistent except for the noise component. Therefore, if D(i, j) is set to be equivalent to V p (t i ) in the p+M*i column, fill in 1, in the column corresponding to V q (t j ) is filled with -1 in the q+M*j column and 0 in the rest of the column, then D(j, j)V in The value becomes from V p (t i ) minus V q (t j ), equation (11) holds.

[0112] [Formula 9]

[0113] D(i, j)V in =V p (t i )-V q (t j )=N(t i )-N(t j ) (11)

[0114] Therefore, when i and j satisfy condition a, proposition M1 becomes true.

[0115] Next, we will prove the case where condition b is satisfied. When certain i and j satisfy condition b, the following equation (12) holds true according to the definition.

[0116] [Formula 10]

[0117]

[0118] When the left and right sides of equation (12) are added together, the following equation (13) is obtained.

[0119] [Mathematical formula 11]

[0120]

[0121] Therefore, as long as D(i, j) is multiplied by V in Then the left side of equation (13) is obtained, and proposition M1 is presented. That is, if D(i, j) is set to be equivalent to V p0 Fill in 1 in the column p0+M*i of (ti) and in the column corresponding to V p(u+1) (t ku ) (u+1) +M*ku Fill in 1 in the column, which is equivalent to V qu (t ku ) u +M*k u Fill in -1 in the column, which is equivalent to V qx (t j ) x +M*j columns are filled with -1, and other rows are filled with 0, then D(j, j)V in The value of becomes the left side of formula (13), and as a result, the following formula (14) holds.

[0122] [Mathematical formula 12]

[0123]

[0124] Therefore, proposition M1 regarding condition b holds.

[0125] Through the above, it can be proved that when the scanning matrix meets condition a or condition b, there must be a in Extract the difference N(t i )-N(t j ) such a matrix D(i, j).

[0126] (M decode conditions)

[0127] The matrix M of M*L rows and M*L columns constituting the sensor circuit of this patent decode Created to satisfy the following condition c.

[0128] <Condition c>

[0129] The matrix M with M*L rows and M*L columns decode As shown in the following formula (15), it is decomposed into M*L row vectors (length M*L). When the p+M*i-th row vector is set to H(p, i), H(p, i) can be represented by D(i, k) as shown in the following formula (16).

[0130] [Mathematical formula 13]

[0131]

[0132]

[0133] Here, One(p, i) refers to a row vector in which 1 is filled in only the p+M*i-th column and 0 is filled in the other columns.

[0134] <Proposition M2>

[0135] Set time t iThe true value (ideal detection value) of the detection circuit p at is set to W(p, t i ) (Only detect the value of the sensor physical quantity without noise). decode When condition c is satisfied, as in formula (17), when input V in Multiply by M decode The corrected input V′ is obtained in Each element of becomes the truth value W(p, t i ) and the average value of all L measurement noises.

[0136] [Mathematical formula 14]

[0137]

[0138] <Proof of Proposition M2>

[0139] Since the following formula (18) can be proved to be true with any p and i, then the proposition M2 is also true. Therefore, in M decode When condition c is satisfied, prove the following for arbitrary p and i.

[0140] [Mathematical formula 15]

[0141]

[0142] According to condition c, since H(p, i) is replaced by the sum of the One(p, i) vector and the D(i, k) vector, the following equation (19) holds.

[0143] [Mathematical formula 16]

[0144]

[0145] According to the definition of One(p, i) (a vector with 1 in the column p+L*i and 0 in the rest), One(p, i) and V in Since only the p+L*i-th element is selected, the product becomes the following formula (20).

[0146] [Mathematical formula 17]

[0147] One(p,i)V in =W(p,i)+N(t i ) (20)

[0148] According to the definition of D(i, k), that is, the following formula (21), it becomes formula (22).

[0149] [Mathematical formula 18]

[0150] D(i, k)V in =N(t i)-N(t k ) (twenty one)

[0151]

[0152] Therefore, the following formula (23) is obtained.

[0153] [Mathematical formula 19]

[0154]

[0155] If N(t i ) are summarized, and the following formulas (24) and (25) are obtained.

[0156] [Mathematical formula 20]

[0157]

[0158]

[0159] Since the equation to be proved can be obtained by transforming the equation, Proposition M2 is positive. decode When condition c is met, by inputting V in Multiply by M decode The corrected input V′ is obtained in Becomes the hold input V in The true value of remains unchanged and only the value of the noise (the average value of all L noises) is corrected.

[0160] The proof of Proposition M1 and Proposition M2 shows that by preparing a scanning matrix Z that satisfies conditions a or b i and the correction matrix M that satisfies condition c decode , can get only input V in All noise components are corrected to the same value (the average value of all noise) correction input V' in .

[0161] Next, we will explain M decode Specific example. When L=2 as in this embodiment, it is necessary to satisfy condition a when i=t0 and j=t1. Specifically, there exists Z t0 The p-th row component and Z t1 The condition a is that p and q (0≤p, q<2) are the same as the qth row component of . If one detection circuit (for example, p1) is set to detect the same sensor S2 at any time, then Z t0 The p1 component (row 1) and Z t1 The p1 component (row 1) is the same, satisfying condition a.

[0162] In this case, the detection value of the detection circuit p1 at time t0, that is, V p1 (t0) and the detection value of the detection circuit p1 at time t1, that is, V p1 (t1) are the same except for the noise. Therefore, D(t0, t1) is obtained by converting V p1 (t0) minus V p1 (t1) is obtained by: D(t1, t0) = -D(t0, t1).

[0163] [Mathematical formula 21]

[0164]

[0165] Here, the value V detected at time t0 p0 (t0) as V p0 (t0) = V S0 +N(t0), there is a noise component N(t0), but since only the noise component N(t0) is replaced by the average value of two noises ((N(t0)+N(t1)) / 2), the following equation (27) and this V p0 (t0) are added together.

[0166] [Mathematical formula 22]

[0167]

[0168] The result of adding them together becomes the following formula (28).

[0169] [Mathematical formula 23]

[0170]

[0171] That is, if we use the following equation (29) such that the row vector T p0，t0 , then as only V is replaced p0 The value of the noise component at (t0) can be obtained as V′ p0 (t0).

[0172] [Mathematical formula 24]

[0173]

[0174] By using V p1 (t0), V p0 (t1), V p1 (t1) is also processed in the same way, and the row vectors T obtained for them are p、t In the row direction, we can get 4×4 M decodeThe above description is for the case where the detection circuits are p0 and p1, but M can also be derived in the same manner when the detection circuits are p1 and p2 as in this embodiment. decode , specifically represented by formula (5), and calculated as formula (6).

[0175] In this embodiment, the case where N=3, M=2, and L=2 is described, but the present invention is not limited thereto. It is preferable to set N<M×L. For example, Figure 5 As shown, when N=9, M=3, and L=4, the same processing is performed, and the offsets are equalized as the average values ​​of the four time-dependent noise components.

[0176] As described above, according to the measurement device, measurement method, and measurement program of this embodiment, even when the detection values ​​contain a time-dependent noise component, each detection value can be corrected using each measurement so that the time-dependent noise component contained in each detection value becomes the common noise component (average value) obtained by statistically processing multiple detection values ​​obtained at different measurement times. This equalizes the noise components of each detection value. This suppresses the influence of the noise component on the true value, thereby improving measurement accuracy.

[0177] By averaging the common noise component, the influence of the noise component can be effectively reduced even when there are large variations in the values ​​of the individual time-dependent noise components. For example, when aligning the time-dependent noise with the time-dependent noise contained in a measurement value used as a reference, if the reference's time-dependent noise is large, even with normalization, there is a possibility that the true value will be buried in the noise. However, averaging the common noise component can reduce the possibility of the true value being buried in the noise.

[0178] During measurement, a different sensor combination is selected from the previous measurement, so the combination can be changed for each measurement, allowing for efficient measurement. By presetting the combination for each measurement, the appropriate transformation matrix can be used for transformation.

[0179] By performing measurements at predetermined time intervals and predetermined timings, each measurement can be performed appropriately. For example, by shortening the time interval, changes in the true value can be suppressed, thereby improving measurement accuracy.

[0180] By calculating the detection results corresponding to the sensors respectively using the corrected detection values, the detection results corresponding to the sensors can be obtained.

[0181] By using an electrostatic capacitance sensor, noise components can be suppressed, and the position of a measurement object such as a finger can be grasped more accurately.

[0182] [Second embodiment]

[0183] Next, let's discuss a measurement device, a measurement method, and a measurement program thereof according to a second embodiment of the present disclosure.

[0184] This embodiment describes a case where offset noise cancellation is performed by equalization in the output conversion unit 7. Hereinafter, regarding the measurement device, its measurement method, and measurement procedure according to this embodiment, descriptions of points common to the first embodiment will be omitted, and differences will be mainly described.

[0185] Figure 6 : This is a diagram showing the system structure of the measuring device 1 in the present embodiment. In the present embodiment, there are a total of 4 sensors consisting of sensors S0 to S3, and further, the selection unit 4 is provided with a composite value calculation unit 9. The selection unit 4 (composite value calculation unit 9) calculates a composite output value based on the measurement value of the selected sensor and outputs it as a detection value. The composite value calculation unit 9 is provided with, for example, an inverting amplifier circuit and an addition circuit. An amplifier circuit that amplifies at a magnification other than 1 times can be provided in the composite value calculation unit 9. The selection unit 4 controls the connection state by a switch, and in the composite value calculation unit 9, the detection values ​​of each selected sensor are added and / or subtracted to generate a composite output value. When the composite output value is calculated in this way, the correction unit 8 corrects each composite output value (detection value).

[0186] Then, the output conversion unit 7 in this embodiment cancels out the common noise components evenly contained in each of the corrected M×L detection values, thereby outputting the detection results corresponding to each of the N sensors. That is, after correction, the output conversion unit 7 cancels out the offset noise (common noise components) evenly contained in each detection value (each composite output value), and uses each detection value after offset noise cancellation to output the detection results corresponding to each sensor. For example, the number of measurements is set to 5, and measurements are performed at times t0 to t5. If the composite output values ​​are set to W0 to W5, the column vectors of each composite output value obtained through the five measurements are as shown in the following equation (30).

[0187] [Mathematical formula 25]

[0188]

[0189] If the noise is equalized for a column vector such as Equation (30) in the same manner as Equation (6) and the like, the calculation becomes as shown in the following Equation (31).

[0190] [Mathematical formula 26]

[0191]

[0192] In this way, the synthesized output value is also equalized between the synthesized output values ​​by taking the average value of the noise component. However, at this stage, since the synthesized output value is a value obtained by synthesizing the detection values ​​of each sensor, it does not represent the detection result corresponding to each sensor. For this reason, the conversion is performed in the output conversion unit 7. For example, each synthesized output value is defined as shown in the following formula (32) using the detection value of each sensor. In the synthesized output value of formula (32), the value obtained by adding the values ​​of the two sensors together (W0=V S0 +V S1 、W1=V S2 +V S3 ) as the reference, prepare the value of subtracting one side (W2=V S0 -V S1 、W3=V S2 -V S3 、W4=-V S0 +V S1 、W5=-V S2 +V S3 ) combination.

[0193] [Mathematical formula 27]

[0194]

[0195] Based on formula (32), it is possible to derive M for restoring the detection value of each sensor from the composite output value. out (Restoration Matrix). At this time, the restoration matrix is ​​set so that out The total of the matrix elements in the row direction (the sum of the column elements included in the same row) becomes 0. Specifically, restoration is performed using the restoration matrix as shown in the following equation (33).

[0196] [Mathematical formula 28]

[0197]

[0198] That is, by preparing each composite output value as shown in equation (32), the result of subtraction of two composite output values ​​is halved as shown in equation (33), thereby indicating the detection value of a certain sensor.

[0199] Here, each composite output value includes an equalized noise component (average value) as shown in equation (31). Therefore, if equation (33) is calculated using the equalized total output value of equation (31), then when a certain composite output value is subtracted from a certain composite output value as shown in equation (33), the equalized noise component is canceled. Therefore, V calculated by equation (33) S0 -V S3Ideally, the true value of the detection value of each sensor is obtained. In other words, the absolute value of the offset noise is zero. Here, the output conversion unit 7 can be said to use the demodulation matrix M in which the sum of the matrix elements included in any row is zero when added in the column direction. out .

[0200] This embodiment is not limited to the combination of the first embodiment. Specifically, the output conversion unit 7 of this embodiment can be widely used by correcting each composite output value so that the time-dependent noise components contained in each composite output value are equalized (i.e., the offset noise can rise and fall in a common manner). When the time-dependent noise components contained in each composite output value are equalized, the output conversion unit 7 cancels out the time-dependent noise components evenly contained in each composite output value. Each composite output value after the time-dependent noise components have been canceled out can be used to output the detection results corresponding to each sensor.

[0201] As described above, according to the measurement device, measurement method, and measurement program of this embodiment, when the time-dependent noise components contained in each synthesized output value are equalized, the time-dependent noise components uniformly contained in each synthesized output value cancel each other out. Therefore, the noise components can be effectively suppressed, and the detection results (e.g., true values) of each sensor can be obtained. In other words, the measurement accuracy can be effectively improved.

[0202] By calculating a combined output value based on the detection values ​​of selected sensors and correcting each combined output value, it is also possible to cope with combining the detection values ​​of multiple sensors, for example, and to cope with complex circuit configurations.

[0203] The present disclosure is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the invention.

[0204] In the above embodiment, the common noise component is described as an average value. However, other statistical quantities (such as median values) may be used as long as they are obtained by statistically processing a plurality of detection values ​​obtained at different measurement times.

[0205] -Explanation of symbols-

[0206] 1: Measuring device

[0207] 2: Sensor part

[0208] 3: Circuit Department

[0209] 4: Selection Department

[0210] 5: Detection unit (acquisition unit)

[0211] 6: Control Department

[0212] 7: Output conversion unit

[0213] 8: Correction Department

[0214] 9: Composite value calculation unit

[0215] 11: CPU

[0216] 12: ROM

[0217] 13: RAM

[0218] 14: Hard Drive

[0219] 15: Ministry of Communications

[0220] 18: Bus

[0221] S0: Sensor

[0222] S1: Sensor

[0223] S2: Sensor

[0224] S3: Sensor

[0225] p1: detection circuit

[0226] P2: Detection circuit.

Claims

1. A measuring device, characterized in that: have: N sensors, where N ≥ 2; a selection unit that selects a given combination of sensors from the N sensors for each measurement, and outputs a detection value to each of the M sensor terminals based on a measurement value of the selected sensor, where M<N; an acquisition unit that acquires detection values ​​outputted from each of the M sensor terminals for each measurement; and A correction unit, which, after performing L measurements, assumes that each of the M×L detection values ​​obtained by the acquisition unit contains a time-dependent noise component, and corrects each of the M×L detection values ​​so that the time-dependent noise component of each of the detection values ​​becomes a common noise component, that is, a value obtained by averaging the L time-dependent noise components in the time axis direction.

2. The measuring device according to claim 1, wherein The correction unit corrects each of the M×L detection values ​​using a correction matrix of M×L rows and M×L columns so that the values ​​become values ​​obtained by adding the common noise component to a true value.

3. The measuring device according to claim 1, wherein The measurement device includes a control unit configured to perform control so that the selection unit selects a predetermined combination of sensors in each measurement.

4. The measuring device according to claim 3, wherein The control unit controls the selection unit and the acquisition unit to perform the L measurements at predetermined time intervals or at predetermined timings.

5. The measuring device according to any one of claims 1 to 4, wherein The selection unit calculates a combined output value based on the measurement value and outputs the combined output value as the detection value.

6. The measuring device according to any one of claims 1 to 4, wherein The measuring device includes an output conversion unit that calculates a detection result corresponding to each of the N sensors based on the M×L detection values ​​corrected by the correction unit.

7. The measuring device according to claim 6, wherein The output conversion unit cancels out the common noise components evenly included in each of the corrected M×L detection values, thereby outputting a detection result corresponding to each of the N sensors.

8. The measuring device according to claim 6, wherein The output conversion unit outputs detection results corresponding to the N sensors respectively by using a demodulation matrix of N rows and M×L columns for the corrected M×L detection values.

9. The measuring device according to claim 8, wherein The output conversion unit uses the demodulation matrix in which the sum of matrix elements included in an arbitrary row added in the column direction becomes zero.

10. The measuring device according to any one of claims 1 to 4, wherein Each of the N sensors is an electrostatic capacitance sensor.

11. A measurement method using a measurement device having N sensors, wherein N ≥ 2, characterized in that: The measuring method has the following steps: Selecting a given combination of sensors from the N sensors for each measurement, and outputting a detection value to each of the M sensor terminals based on a measurement value of the selected sensor, where M<N; Obtaining detection values ​​outputted from each of the M sensor terminals for each measurement; After performing L measurements, it is assumed that each of the M×L detection values ​​obtained contains a time-dependent noise component, and each of the M×L detection values ​​is corrected so that the time-dependent noise component of each detection value becomes a common noise component, that is, a value obtained by averaging the L time-dependent noise components in the time axis direction.

12. A measurement program product comprising a measurement program for a measurement device having N sensors, wherein N ≥ 2, characterized in that: The measurement program is used to cause the computer to execute the following processing: Selecting a given combination of sensors from the N sensors for each measurement, and outputting a detection value to each of the M sensor terminals based on a measurement value of the selected sensor, where M<N; Obtaining detection values ​​outputted from each of the M sensor terminals for each measurement; After performing L measurements, it is assumed that each of the M×L detection values ​​obtained contains a time-dependent noise component, and each of the M×L detection values ​​is corrected so that the time-dependent noise component of each detection value becomes a common noise component, that is, a value obtained by averaging the L time-dependent noise components in the time axis direction.

Images