JP2008072411A - Coefficient interpolation equipment - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide coefficient interpolation equipment interpolating coefficients while preventing the occurrence of edges. <P>SOLUTION: When the coefficient interpolation equipment is stopped while a VALUE 16 is "1" and a difference (Difference) is not "0", a control portion 100 turns on switches 110, 111, and 112 to the (a) side and subsequently has a connection state of the switches 110, 111, and 112 as shown in a figure. After adding a delta 62 to a previous value 64 and setting the difference to the delta 62, the control portion 100 clears the difference and sets the VALUE 16 to "-1+RATE" and starts up coefficient interpolation. Since the interpolating processing for finding a coefficient between two values is performed based on an S-shaped function, the occurrence of edges and noise can be prevented. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an improvement of a coefficient interpolation processing technique between two coefficients moving back and forth on the time axis.

  As a conventional coefficient interpolation technique, a method of interpolating between two points with a straight line, a technique of smoothing coefficient changes using a low-pass filter, and the like have been proposed. For example, when switching the electronic volume, when the attenuation coefficient input to the coefficient interpolation unit is switched by an operation from the operator, the coefficient interpolation unit sequentially supplies the attenuation coefficient before and after the switching to the attenuator 22. Thus, there has been proposed a coefficient interpolation process for preventing the occurrence of switching noise (see, for example, Patent Document 1). This is to smooth the coefficient change by an IIR type filter or the like.

JP 2000-341786 (page 4-7, Fig. 6)

  As described above, various techniques have been disclosed for coefficient interpolation processing. However, if the coefficient interpolation process is simply performed using a straight line or a multidimensional function as in the prior art, it is necessary to store a large number of coefficients in a memory. In addition, the coefficient changes abruptly in the vicinity of the current value or the target value, and an edge (a portion in which the coefficient value changes abruptly) occurs, resulting in noise. In order to cope with this, smoothing processing using a low-pass filter has been performed in order to avoid increasing the number of samples in the coefficient table and discontinuous differential coefficients near the target value. This will be described with reference to FIG.

  The dotted line in FIG. 3A is obtained by linear interpolation, and the curve is obtained by smoothing processing. For example, an edge is generated in the code A part by both interpolation methods, and an edge is also generated in the other parts B to E. Such an edge has a problem that noise is inevitably generated. That is, the linear interpolation process and the smoothing process are not satisfactory for preventing the occurrence of noise.

  The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide a coefficient interpolation apparatus that performs coefficient interpolation processing while preventing the occurrence of edges.

  In order to achieve the above object, as a result of earnest research, the present inventor conducted a coefficient interpolation between two points that fluctuate in time, and the differential coefficient is “0” at these two points. It was found that the generation of edges can be prevented if the function is monotonic.

That is, the present invention provides an apparatus for interpolating a coefficient between a current value that is a current coefficient value and a target value that is a target coefficient value that exists later on the time axis.
Based on a function in which the differential value is “0” at a position on the time axis with respect to each of the current value and the target value, and changes monotonically in the time interval between the current value and the target value. Interpolation processing means for interpolating coefficients between them is provided. According to this configuration, if the differential coefficient is “0” at the position on the time axis with respect to each of the current value and the target value, and if the function is a monotonous function in the time interval between these two points, the occurrence of an edge can be prevented and noise can be prevented. Can be prevented.

  The function is preferably “f (X) = 2 · X−X · abs (X): (abs (X) is an absolute value of X, −1 ≦ X ≦ 1)”. Any function that satisfies the above conditions may be used. For example, in a trigonometric function such as the number of SINs or the number of COSs, a part corresponding to a predetermined phase may be used, or a part of a higher order function of the third or higher order may be adopted. Furthermore, it is also preferable that the interpolation coefficient generated sequentially by the coefficient interpolation process can be output to another apparatus so that the other apparatus can use the interpolation coefficient.

  According to the present invention, there is an effect that it is possible to realize a coefficient interpolation apparatus that performs coefficient interpolation processing while preventing the generation of edges.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  Hereinafter, as an example of the interpolation function, a quadratic function “f (X) = 2 · X−X · abs (X): (abs (X) is an absolute value of X, −1 ≦ X ≦ 1)” is used. It is assumed to be used. FIG. 4 shows the shape of this function with a solid line. As shown in FIG. 4, it is understood that the function is an S character type function (S character type function). When X is “−1” and “+1”, the differential coefficient is “0”, and the function f (X) changes monotonously in the meantime.

(Coefficient interpolation device 1)
FIG. 1 is an explanatory diagram of the configuration and operation of the coefficient interpolation apparatus 1 according to the embodiment of the present invention. The control unit 100 has a function of controlling the switches 110, 111, and 112 in conjunction with each other. Before starting the coefficient interpolation function, the switches 110, 111, and 112 are turned on to the “b” side, When the coefficient interpolation function is activated, the switches 110, 111, and 112 are turned on to the “a” side. In the upper left of the figure, reference numeral 10 denotes an adder, 15 denotes a limiter, 18 denotes a fixed coefficient value, and 16 denotes a generated value value. First, when “Rate (0 <Rate <1)” is set, the coefficient interpolation function is not activated, so the switch 110 is connected to the b side. In this case, the value 16 and the value of the rate are added to the adder 10, and this is limited by the limiter 15 to repeatedly execute a process of setting a new value 16. The limiter 15 limits the value 16 to a value from “−1” to “+1”.

  On the other hand, when the coefficient interpolation is started, the switch 110 is turned on to the a side, so that Rete and the fixed value “−1” of the sign 18 are added to the adder 10, and this value is limited by the limiter 15 to become the VALUE value 16. Then, the switch 110 is turned on to the b side. Thus, the VALUE value 16 changes as shown in the upper right figure of the VALUE value. That is, RATE is a parameter for adjusting the rate of change of the VALUE value.

  Next, on the upper right side of the drawing, the bit shifter 26 shifts the VALUE value 16 by 1 bit to the left, that is, doubles it and sends it to the adder 28. On the other hand, the VALUE value is converted to an absolute value by the absolute value unit 20, multiplied by the multiplier 22, multiplied by “−1” by the multiplier 24, and input to the adder 28. The limiter 30 limits the addition result of the adder 28 to “−1” and “+1”, and the result is input to the adder 40. The waveform at this time is shown in the lower right of the adder 40. The variable A is “−1 ≦ A ≦ 1” and corresponds to the vertical axis of FIG. The adder 40 adds the fixed value “1” and the signal from the limiter 30 and sends the result to the bit shifter 42. The bit shifter 42 is configured to shift a given digital value by one bit in the right direction, that is, multiply it by “½” and send it to the multiplier 50. The waveform at this time is shown on the left side of the bit shifter 42. At this time, the variable A is a variable A ′ of “0 ≦ A ′ ≦ 1”.

  The multiplier 50 multiplies the value of the delta 62 and the value from the bit shifter 42 and sends the multiplication result to the adder 52. The value sent from the adder 52 is K (coefficient). When the coefficient interpolation function is activated, the coefficient K is sequentially generated and output.

  Now, how to give “Difference” in the left middle part of the figure will be described. In FIG. 5, it is assumed that the horizontal direction is a time axis. The vertical direction is a coefficient value. Now, if only the first two points P1 and P2 are described, the value of P1 is the current value, the value of P2 is the target value, and the difference is the difference. When the coefficient interpolation is not started, since all the switches 110, 111, and 112 are turned on to the b side, the delta 52 and the previous value (Previous) 64 store the values as they are.

  Next, the activation operation of the coefficient interpolation function will be described. When the VALUE value 16 remains “1”, the interpolation device is in a stopped state, and the difference (Difference) is not “0”, the control unit 100 turns on each of the switches 110, 111, and 112 to the a side. As a result, the connection states of the switches 110, 111, and 112 are as shown in FIG. Then, the delta 62 is added to the previous value 64, the difference is set to the delta 62, the difference is cleared, the VALUE value 16 is set to “−1 + RATE”, and the coefficient interpolation is activated. , 111 and 112 are turned on to the b side, respectively, to enter the coefficient interpolation state.

  As described above, when a value is sent from the bit shift 42, the multiplier 50 multiplies the value by the delta 62 and sends the result to the adder 52. The adder 52 adds the multiplication result of the multiplier 52 and the previous value 64 to obtain the coefficient K. The coefficient interpolation process is sequentially executed by repeating such an operation.

(Coefficient interpolation device 2)
FIG. 2 is an explanatory diagram of the configuration and operation of the coefficient interpolation device 2 which is another preferred embodiment of the present invention. The same components as those shown in FIG. In this apparatus, the adder 40, the bit shifter 42, and the fixed value “+1” 44 in FIG. 1 are removed. In this apparatus, a target value is set instead of the difference, and a goal 72 is set instead of the delta 62. The value of the goal 72 is a direct “target value”. The control unit 100 transfers the goal 72 to the previous value 64 and sets the target to the goal 72 when the VALUE value 16 remains “1” and the target and the goal 72 are different. , VALUE is set to “−1 + RATE”, and coefficient interpolation is activated.

  The adder 52 multiplies “the multiplication result from the multiplier 50”, “goal value”, “previous value”, “multiply the previous value by the multiplier 74, and further multiply the coefficient“ −1 ”by the multiplier 76. Coefficient interpolation is sequentially performed by repeating the operation of adding the four types of “result” and supplying the coefficient K obtained by multiplying the addition result by “1/2” by the bit shift 73.

  FIG. 3B shows a coefficient interpolation result using the S-shaped function of the present invention, and no edge is generated in the parts A to E. The point that no edge is generated at the part A at the start of coefficient interpolation is particularly characteristic, and it becomes possible to prevent the generation of noise at the start of coefficient interpolation.

  As described above, according to the embodiment of the present invention, the differential coefficient is “0” at the position on the time axis with respect to each of the current value and the target value, and the monotonous function in the time interval between the two points. Then, it is possible to prevent the generation of edges and to prevent the generation of noise. It should be noted that this coefficient interpolation processing can be adopted in all fields as a configuration that can output the coefficient K obtained by both apparatuses 1 and 2 to another apparatus (for example, a music apparatus).

  As described above, according to the coefficient interpolation apparatus of the present invention, for example, coefficient interpolation processing for musical tone signal data in the music field can be realized.

It is explanatory drawing of the coefficient interpolation apparatus. It is explanatory drawing of the coefficient interpolation apparatus. It is explanatory drawing of the coefficient interpolation result by a prior art and this invention. It is explanatory drawing of a S-shaped interpolation function. It is explanatory drawing of operation | movement.

Explanation of symbols

1 Coefficient Interpolator 2 Coefficient Interpolator 10 Adder 15 Limiter 20 Absolute Value Unit 22 Multiplier 24 Multiplier 26 Bit Shifter 28 Adder 30 Limiter 40 Adder 42 Bit Shifter 50 Multiplier 52 Adder 100 Control Unit 110 Switch 111 Switch 112 Switch

Claims (3)

In an apparatus that performs a coefficient interpolation process between a current value that is a current coefficient value and a target value that is a target coefficient value that exists later on the time axis,
The differential value is “0” at the position on the time axis with respect to each of the current value and the target value, and the two values are based on a function that changes monotonically in the time interval between the current value and the target value. A coefficient interpolation apparatus comprising coefficient interpolation processing means for performing a coefficient interpolation process. The apparatus of claim 1.
The function used by the coefficient interpolation processing function for coefficient interpolation processing is:
A coefficient interpolating device characterized in that “f (X) = 2 · X−X · abs (X): (abs (X) is an absolute value of X, −1 ≦ X ≦ 1)”. The coefficient interpolation apparatus according to any one of claims 1 and 2,
A coefficient interpolating apparatus configured to be able to output interpolation coefficients sequentially generated by the coefficient interpolating process to another apparatus.

Images