JP4012950B2 - Waveform editing method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a waveform editing method for generating edited waveform data by editing original waveform data stored in a memory.
[0002]
[Prior art]
The arbitrary waveform generator stores waveform data in a waveform memory, and outputs an analog waveform signal by performing digital / analog conversion on the waveform data read from the waveform memory. An arbitrary output waveform signal can be generated by arbitrarily adjusting the contents of the waveform data stored in the memory. In this case, the operator displays the waveform data stored in the waveform memory on a display screen such as a CRT (cathode ray tube) and operates the operation panel to arbitrarily edit the shape of the displayed waveform. Waveforms can be generated.
[0003]
When executing waveform editing, an operation of specifying a part of a display waveform, shifting the specific partial waveform in the horizontal axis direction or the vertical axis direction, and displaying the enlarged or reduced image is frequently performed. In that case, a discontinuity or a blank portion of the waveform may occur due to the specific partial waveform that is shifted, enlarged or reduced, and the original waveform being overlapped or separated. FIG. 4 is a waveform diagram for explaining an example of a conventional waveform editing method. The horizontal axis is a time axis (each time value corresponds to each address value in the memory), and the vertical axis is a voltage axis representing the voltage value at each point. One address on the time axis corresponds to one point of data, and one waveform data is composed of a data group such as 512 points or 1024 points. This shows a case where the waveform from point a (address value: m) to point b (address value: n) of the original waveform is shifted by ΔT in the right direction along the time axis. The portion of the original waveform from point a to point b is denoted as partial original waveform (ab), and after shifting rightward by ΔT to point A (address value: m + ΔT) to point B (address value: n + ΔT) ) Are represented as edited partial waveforms (A to B). In FIG. 4, since all the points of the partial original waveforms (a to b) are shifted by ΔT, the edited partial waveforms (A to B) are represented by waveforms having the same shape. Note that the waveforms of the partial source waveforms (ab) are not displayed for convenience.
[0004]
As described above, when the partial original waveform (ab) is moved to the edited partial waveform (AB), the point a and the point A are separated on the time axis, and the point b and the point B are separated. The interval will overlap on the time axis. Therefore, according to this conventional method, in order to smoothly connect the original waveform and the edited partial waveform (A + B), the waveform data from the point a to the point c of the original waveform is used and the point A to the point Waveform data that smoothly connects up to c is automatically calculated. Similarly, waveform data that smoothly connects point B to point d is automatically calculated using waveform data from point b to point d of the original waveform. As a result, smooth curve portions (c to A) and (B to d) as shown by bold lines in FIG. 4 are automatically obtained, so that the operator selects an arbitrary portion of the original waveform, By simply shifting in the time axis direction, both ends of the shifted waveform portion and the original waveform are automatically and smoothly connected, so waveform editing can be performed very easily and quickly.
[0005]
In FIG. 4, the time axis region from the point c (address value: m−sm) to the point A (address value: m + ΔT) is referred to as a first smoothing region. The time axis region from the point B (address value: n + ΔT) to the point d (n + ΔT + sm) is referred to as a second smoothing region. The range of the first and second smoothing regions can be arbitrarily set by selecting the value of sm. The partial waveforms (c to A) and (B to d) of the first and second smoothing regions indicated by bold lines in FIG. 4 are obtained from the partial waveform data of (c to a) and (b to d) of the original waveform, respectively. These are obtained by automatic calculation and are called a first smoothed edited waveform and a second smoothed edited waveform, respectively. For example, the data of the point f on the first smoothed edited waveform (c to A) is obtained by calculating the data of an arbitrary point e on the partial waveform (c to a) of the original waveform with a predetermined function. . That is, the vertical axis coordinate values of the point e and the point f are the same, and when the difference of the horizontal axis coordinate value (address value) between the point e and the point f is represented by the variable y, the variable y is A predetermined cubic function of the axis coordinate (address value) x is assumed.
[0006]
FIG. 5 is a diagram for explaining a method of calculating the first and second smoothed edited waveform portions corresponding to the case of FIG. The horizontal axis represents the address value (variable x) of the original waveform data corresponding to the time axis, and the vertical axis represents the address shift amount (variable y) from the original waveform data to the edited waveform data. That is, all points in the portion of the original waveform from point a (address value: m) to point b (address value: n) are shifted by ΔT, and edited partial waveforms (A to B from point A to point B). ). In this conventional method, it is necessary to obtain a curve that smoothly connects the point c to the point A. Therefore, the curve to be obtained is in contact with the x-axis at the point c and is inclined so as to be smoothly connected to the line segment AB at the point A. Must be the same as line segment AB. In FIG. 5, the line segment AB is a horizontal line segment. However, when a part of the original waveform is enlarged or reduced, the address shift amount at the continuous points of the partial waveform is changed according to the enlargement ratio or the reduction ratio. Therefore, there may occur a case where the line segment AB is inclined or a curve AB. Therefore, the function of the curve that smoothly connects the point c and the point A is that the point c touches the x-axis and the point A has a slope that matches the slope at the point A of the line segment (or curve) AB. Condition. A simple function that satisfies this condition can be uniquely determined by a cubic equation. That is, a cubic equation that touches the x-axis at the x-coordinate point x0 and passes through the point (x1, y1) with a slope K is expressed by the following equation 1.
[Expression 1]
y = p * (x−x0) * (x−x0) * (x−q)
However,
p = {K * (x1-x0) -2 * y1} / (x1-x0) ** 3
q = x1-y1 * (x1-x0) / {K * (x1-x0) -2 * y1}
Here, “*” represents multiplication, “**” represents power, and “/” represents division.
[0007]
Of course, when the line segment AB is a horizontal straight line as shown in FIG. 5, the slope K at the point A is zero. In order to apply the above formula 1 to the case of FIG. 5, it can be easily applied by simply replacing the following formula 2.
[Expression 2]
x0 = (m-sm)
x1 = m
y1 = ΔT
Similarly, the data of each point of the second smoothing waveform from point B to point d in FIG. 5 can be easily obtained from the calculation of the cubic equation.
[0008]
[Problems to be solved by the invention]
In the above-described conventional method, since there is a limit on the density of address values on the horizontal axis of the waveform data, the points of the edited waveform data that can be actually generated deviate from the calculation points. FIG. 6 is an enlarged view for explaining a problem of the conventional method. In FIG. 6, a curve 50 indicates a part of the waveform of the original waveform data, and a point e represents one data point in the original waveform data. As described above, according to the conventional method, the shift amount y of time (address value) is calculated from the value of the x coordinate of the point e. A curve 52 indicated by a broken line represents an ideal edited data curve obtained by such calculation. However, in an actual waveform editing apparatus, ideal edited data for calculation cannot be generated due to the limitation of the address density of the waveform memory. In FIG. 5, the edited data point g calculated from the point e does not necessarily correspond to the address value on the actual waveform memory. Point g merely represents a calculated value, and two points f and g before and after the calculated point correspond to actual address values i and j, respectively. In this example, the point f is selected as the point closest to the point g. As described above, in the conventional method described with reference to FIG. 4, the calculated points are not obtained as edited data as they are, but the edited data is merely generated as a set of approximate points of the calculated points. . In other words, in the above example, the calculated point g is replaced with the point f corresponding to the actual address value, so that the actually obtained edited data inevitably contains an error.
[0009]
Accordingly, an object of the present invention is to provide a waveform editing method capable of automatically generating accurate edited waveform data while eliminating errors of the conventional waveform editing method.
[0010]
[Means for solving the problems]
The present invention is an edited version in which partial original waveform data of selected continuous points of original waveform data displayed as a plurality of consecutive points when a waveform is displayed on a screen having a time axis and a voltage axis are shifted in a certain direction, respectively. The partial waveform data is stored in the memory, and the value (voltage value) on the voltage axis corresponding to the value (time value) on the time axis of the smoothing area in a predetermined range continuous from one end of the edited partial waveform data to the outside. This is a waveform editing method for generating smoothed edited waveform data for smoothly connecting the curve represented by the original waveform data and the curve represented by the edited partial waveform data. According to the waveform editing method, determined on the basis of the virtual time value on the original waveform data corresponding to a voltage value at a particular time value in the smoothing region to a predetermined function, the actual multiple times in the vicinity of the virtual time value obtains a virtual voltage value of the virtual time value by interpolation calculation from the voltage value of the original waveform data of each of the values, by the virtual voltage value and the voltage value at a particular time value in the smoothing region, data smoothing edited waveform Is generated.
[0011]
As a result, accurate edited waveform data can be automatically generated without causing a conventional error.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 3 is a block diagram showing an example of the configuration of an arbitrary waveform generator suitable for applying the present invention. This apparatus is controlled by a microprocessor (μP) 10 via a main bus 12, and various patterns of waveform data, control software, and the like are stored in a system memory 14 including a ROM and a RAM. The operator performs various operations via the operation panel 16. The output waveform memory 18 stores arbitrary waveform data created by the operator. The display device 20 is well known to those skilled in the art, such as the contents of the output waveform memory 18 on the display screen such as a CRT (cathode ray tube) or a liquid crystal display device, and various conditions and operation modes set by the operator. Display status. The output circuit 22 converts the waveform data read from the output waveform memory 18 into an analog waveform signal and outputs it through the output terminal 24. According to the present invention, the waveform data stored in the output waveform memory 18 is read, and the operator executes waveform editing while displaying the waveform on the screen of the display device 20. Therefore, the waveform data being edited is also stored in a predetermined storage area in the output waveform memory 18, and edited waveform data is generated by using the original waveform data. How to use the original waveform data will be described later in detail. Note that an editing waveform memory may be provided as hardware different from the output waveform memory 18, and the waveform data being edited and the edited waveform data may be stored.
[0013]
FIG. 1 is a waveform diagram for explaining the waveform editing method of the present invention. For comparison with the conventional method, the waveform diagram is almost the same as the example of FIG. However, it should be noted that the method for obtaining the waveform data in the smoothing region is different from the conventional method. That is, the method for obtaining the data value of a specific point 62 on the smoothed edited waveform 60 from point c to point A in the first smoothing region is completely different from the conventional method. When the address value on the time axis of the specific point 62 is z (known), the virtual address value x of the virtual point 64 on the original waveform 50 corresponding to the specific point can be calculated using the above-described Equation 1. That is, the above formula 1 may be substituted as follows.
[Equation 3]
z = x + y
That is, z = x + p * (x−x0) * (x−x0) * (x−q)
However,
p = {K * (x1-x0) -2 * y1} / (x1-x0) ** 3
q = x1-y1 * (x1-x0) / {K * (x1-x0) -2 * y1}
[0014]
Of course, as in the case of FIG. 5, when the line segment AB is a horizontal straight line, the slope K at the point A is zero. Similar to the case of Equation 2 above, the above Equation 3 can be easily applied to the case of FIG.
[Expression 4]
x0 = (m-sm)
x1 = m
y1 = ΔT
The virtual address value x can be easily obtained by solving the cubic function of Equation 3. Similarly, the data of each point of the waveform of the second smoothing region from the point B to the point d in FIG. 1 can be easily obtained from the same calculation.
[0015]
In FIG. 1, when the virtual address value x is calculated, the voltage data value Vx of the virtual point 64 on the original waveform 50 at this virtual address value is calculated. The voltage data value Vx obtained as a result is adopted as the data value of the specific point 62 on the smoothed edited waveform 60. Eventually, the data of the specific point 62 is completely determined.
[0016]
FIG. 2 is an enlarged view of a part of the waveform diagram of FIG. Since every point on the smoothed edited waveform 60 corresponds to a known address value, the address value z of this particular point 62 is also known. However, since the voltage data value of each point on the smoothed edited waveform 60 is unknown, it is necessary to obtain the voltage data value of each address value in order to determine the data of the smoothed edited waveform 60. As described above, the voltage data value of the specific point 62 is calculated based on the mathematical expressions 3 and 4, and the virtual address value x with high accuracy is calculated, and the voltage data of the virtual point 64 on the original waveform 50 at the virtual address value x is calculated. By obtaining the value Vx, the voltage data value at the specific point 62 can be accurately determined. Here, since the virtual address value x of the virtual point 64 generally does not match the address value of the memory, it is merely a virtual point for calculation. Therefore, by accurately obtaining the voltage data value of the virtual point 64, the data value of the specific point 62 on the smoothed edited waveform 60 is accurately determined. For this purpose, interpolation calculation of a plurality of points in the vicinity of the virtual point 64 is executed. In the case of FIG. 2, an accurate data value Vx is easily obtained by performing a linear interpolation calculation between the actual point 66 of the address value s and voltage value V1 and the actual point 68 of the address value r and voltage value V2. be able to.
[0017]
That is, the address values and voltage values of the points 66 and 68 on the original waveform 50 are known, and the virtual address value x is also determined by the above calculation. Therefore, the voltage value of the virtual point 64 can be easily calculated by linear interpolation. Desired. The voltage value Vx of the virtual point 64 is adopted as the voltage value of a specific point 62 on the smoothed edited waveform 60. As a result, the data value of the point 62 is accurately determined. In FIG. 2, linear interpolation of two known points 66 and 68 in the vicinity of the virtual point 64 is calculated, but the interpolation calculation is not limited to linear interpolation. For example, a secondary interpolation calculation may be performed between three points in the vicinity of the virtual point 64, or a cubic interpolation calculation or a well-known sinX / X interpolation calculation may be employed.
[0018]
Although the preferred embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications and corrections can be made without departing from the gist of the present invention. It will be apparent to those skilled in the art.
[Brief description of the drawings]
FIG. 1 is a waveform diagram for illustrating a waveform editing method of the present invention.
FIG. 2 is an enlarged waveform diagram for showing details of a part of FIG. 1;
FIG. 3 is a block diagram showing an example of a configuration of an arbitrary waveform generator suitable for using the present invention.
FIG. 4 is a waveform diagram for explaining an example of a conventional waveform editing method.
FIG. 5 is a diagram for explaining the waveform editing method of FIG. 4;
6 is a diagram for explaining problems of the conventional waveform editing method of FIGS. 4 and 5. FIG.
[Explanation of symbols]
50 Original waveform 60 Smoothed edited waveform 62 Specific point 64 Virtual point x Virtual address value z Specific address value

Claims (1)

Edited partial waveform data obtained by shifting the partial original waveform data of selected continuous points of the original waveform data displayed as a plurality of continuous points when the waveform is displayed on a screen having a time axis and a voltage axis, respectively, in a certain direction. Storing in a memory, obtaining a value (voltage value) on the voltage axis corresponding to a value (time value) on the time axis of a smoothing region of a predetermined range continuous from one end of the edited partial waveform data; A waveform editing method for generating smoothed edited waveform data for smoothly connecting the curve represented by the original waveform data and the curve represented by the edited partial waveform data,
A virtual time value on the original waveform data corresponding to a voltage value at a specific time value in the smoothing region is obtained based on a predetermined function,
Obtains a virtual voltage value at the virtual time value by interpolation calculation from the voltage value of the original waveform data of each of the actual plurality of time values in the vicinity of the virtual time value,
A waveform editing method, wherein the data of the smoothed edited waveform is generated by using the virtual voltage value as a voltage value at the specific time value in the smoothing region.

Images