TWI390848B - Digital wave generator and method of generating a digital wave - Google Patents

Description

Digital waveform generator and method for generating digital waveform

The present invention relates to a waveform generator and a method of generating a waveform, and more particularly to a digital waveform generator for generating a digital waveform using a phase-to-phase conversion method and a method of generating a digital waveform.

In the fields of electronic engineering, communication engineering, automatic control, telemetry control, measuring instruments, meters and computers, it is often necessary to use a wide variety of waveform generators. A waveform generator generally refers to a circuit or instrument that produces a specific periodic time function waveform such as a sine wave, a cosine wave, a square wave, and a triangle wave.

Digital waveform generators for positive/cosine waveforms are commonly used in communication systems, and the resulting positive and cosine waveforms are often used to provide modulation and demodulation of the signal. Most of the general digital waveform generators use the Mapping Method to implement the sine/cosine function and output its waveform. The function table comparison method is to insert the entire positive/cosine function comparison table into a memory (Read Only Memory, ROM), and output the positive/cosine function value of the stored memory by using the address decoding method to generate positive/ Cosine waveform.

However, the address decoding method not only consumes a large amount of memory and address decoding circuits in the digital waveform generator, but also increases the implementation area and manufacturing cost of the wafer. In addition, the large circuit area leads to disadvantages such as reduced circuit operation speed and long reaction time, which causes the overall performance of the digital waveform generator to be low.

The present invention provides a digital waveform generator that can perform a waveform function operation without using a waveform function look-up table and output a periodic digital wave accordingly.

The present invention further provides a method of generating a digital waveform that can perform a waveform function operation without using a waveform function look-up table and thereby generate a periodic digital wave.

The invention provides a digital waveform generator, comprising a frequency processing unit, a phase region discriminating unit, a phase region converting unit, a function value calculating unit and a polarity converting unit. When the frequency processing unit is based on the prototype digital wave, the value of the angle data is limited or amplified by 2^P times according to the frequency selection signal, and the reference angle data is generated accordingly. Where P is an integer and P≧0.

The phase region discriminating unit generates a phase region discrimination signal according to the R most significant bits and the S waveform indication signals in the reference angle data, to point to the (2^R)*S phase regions extended by the prototype digital wave. Specific phase area. Where R and S are positive integers. The phase region converting unit selects a shift relationship corresponding to the specific phase region from the (2^R)*S shift relationship according to the phase region discrimination signal, and shifts the value of the reference angle data to the first phase region of the prototype digital wave. Inside, and according to the phase area data.

The function value calculation unit provides a plurality of function expressions for synthesizing the prototype digit wave in the first phase region, and brings the phase region angle data to one of the function expressions to calculate the phase angle data. The corresponding phase function value. The polarity conversion unit determines the opposite of the output phase function value according to the phase region discrimination signal or directly outputs the phase function value as the waveform output value, and uses the waveform output value to generate the frequency as 1/(2^P) times of the prototype digit wave or It is a 2 × P times periodic digital wave.

In an embodiment of the present invention, the frequency processing unit causes the angle data to be limited or enlarged by 2^P times by shifting the angle data to the right or by shifting the P bit to the left.

In an embodiment of the invention, the frequency processing unit further outputs a DC output driving signal according to the frequency selection signal to drive a circuit external to the digital waveform generator.

In an embodiment of the present invention, the period of the prototype digital wave is 2^N sampling points, n is used to represent the value of the reference angle data, and N is a positive integer, and when K is a positive number, the prototype digital wave is The function is:

In an embodiment of the invention, when R and S are respectively equal to 2, the phase region discrimination signal is used to identify 8 phase regions extended by the prototype digital wave, and the first to fourth phases of the prototype digital wave are The region is used to synthesize a sinusoidal digital wave, and the fifth to eighth phase regions of the prototype digital wave are used to synthesize a cosine digital wave.

In an embodiment of the present invention, N is equal to 8, K is equal to 128, and z is used to represent the value of the phase angle data, and 8 shifting relations and 8 phase regions are stored in the phase converting unit. Corresponding to each other, and the eight shifting relations are as follows: Z=n, where n=0~63; Z=128-n, where n=64~127; Z=n-128, where n=128~191 ;Z=256-n, where n=192~255; Z=64-n, where n=0~63; Z=n-64, where n=64~127; Z=192-n, where n=128 ~191; and Z=n-192, where n=192~255.

In an embodiment of the present invention, the function expressions provided by the function value calculation unit are as follows: F _MP (z)=2*z+z, where z=0~3, 20~23; F _MP (z)=(2*z+1)+z, where z=4~19; F _MP (z)=2*z+24, where z=24~27; F _MP (z)=2*z +26, where z=28~31, 40~43; F _MP (z)=2*z+27, where z=32~39; F _MP (z)=z+70, where z=44~55; F _MP (z)=126, where z=56~59; and F _MP (z)=127, where z=60~64.

In an embodiment of the invention, the digital waveform generator further includes a data input unit, a continuous angle generating unit, and a data selecting unit. The data input unit decodes the input address to treat the digital data as a single angle data or an operation instruction according to the decoded input address. The continuous angle generating unit receives the data selection signal to determine whether to generate continuous angle data. The data selection unit reads the single angle data or the continuous angle data according to the data selection signal to generate the angle data.

The invention provides a method for generating a digital waveform, comprising the following steps: firstly, based on a prototype digital wave, the value of an angle data is limited or amplified by 2^P times according to a frequency selection signal, and accordingly A reference angle data, P is an integer and P ≧ 0. Then, a phase discrimination signal is generated according to the R most significant bits and the S waveform indication signals in the reference angle data, to point to one of (2^R)*S phase regions extended by the prototype digital wave. For a specific phase region, R and S are positive integers.

Then, according to the phase region discrimination signal, the shift relationship corresponding to the specific phase region is selected from the (2^R)*S shift relationship to shift the value of the reference angle data to the first phase region of the prototype digital wave. And according to the data of the phase of the phase. Then, a plurality of functional expressions for synthesizing the prototype digital wave located in the first phase region are provided, and the phase angle data is brought into one of the functional expressions to calculate the phase angle data. The corresponding phase function value.

Finally, depending on the phase region discrimination signal, the inverse of the phase function value is determined or the phase function value is directly provided as a waveform output value, and the waveform output value is used to generate a frequency of 1/(2^M) times the prototype digital wave or It is a periodic digital wave of 2^M times.

Based on the above, the present invention mainly utilizes the symmetry of the prototype digital wave itself, divides the prototype digital wave into a plurality of phase regions, and shifts the reference angle data to a specific phase region of one of them, and then calculates with a plurality of function expressions. The phase function value corresponding to each reference angle data, thereby generating periodic digital waves of different frequencies. Therefore, the digital waveform generator can generate periodic digital waves without a large number of decoding circuits and memories, thereby saving the area of the circuit and increasing the overall operation speed of the circuit.

The above described features and advantages of the present invention will be more apparent from the following description.

The digital waveform generator of the present invention will be enumerated below, and the generation of the positive/cosine wave will be specifically described for convenience of explanation. It is to be noted that the following embodiments are exemplified by the generation of digital sine/cosine waves, but are not intended to limit the present invention. Those skilled in the art can modify the following embodiments in accordance with the spirit of the present invention. However, it still falls within the scope of the invention.

FIG. 1A is a waveform diagram of a sinusoidal function of a single period, and FIG. 1B is a waveform diagram of a cosine function of a single period. Referring to FIG. 1A and FIG. 1B, the horizontal axis and the vertical axis of the two waveform diagrams respectively represent an input angle θ and a sine function sin θ and a cosine function cos θ, wherein the input angle θ is a function of time t, that is, θ=ωt, and ω is an angular velocity. . As shown in FIG. 1A and FIG. 1B, the waveforms of the positive/cosine function can be divided into four phase regions according to their symmetrical characteristics, which are phase regions Ph1 to Ph4 and phase regions Ph5 to Ph8, respectively.

The phase regions Ph1 and Ph5 correspond to the angular range of the input angle θ=0~90 degrees; the phase regions Ph2 and Ph6 correspond to the angular range of the input angle θ=90-180 degrees; the phase regions Ph3 and Ph7 correspond to the input angle θ=90~ The angle range of 270 degrees, and the phase areas Ph4 and Ph8 correspond to the angle range of the input angle θ=270~360 degrees. In other words, the phase regions corresponding to the input angles of the positive/cosine functions can be arranged as shown in Table 1:

It is worth noting that the sinusoidal function sin θ or the cosine function cos θ corresponding to each input angle θ of each phase region has a symmetrical property on the waveform. For example, as shown in FIG. 1A, when the input angle θ=0~180 degrees, the corresponding sinusoidal function sin θ is the symmetry axis of the input angle θ=90 degrees, that is, the waveform of the phase region Ph1 and the phase region Ph2. Symmetrical to the vertical axis of the input angle θ = 90 degrees. When the input angle θ=180~360°, the corresponding sine function sinθ is the symmetry axis with the input angle θ=270 degrees, that is, the waveform of the phase region Ph3 and the phase region Ph4 is symmetrical with respect to the input angle θ=270 degrees. axis.

In addition, when the input angle θ=90~270 degrees, the corresponding sine function sinθ is the antisymmetric axis with the input angle θ=180 degrees, that is, the waveform of the phase region Ph2 and the phase region Ph3 is opposed to θ=180 degrees. The vertical axis. Therefore, as far as the sinusoidal function sin θ is concerned, as long as the function expressions of the input angle θ and the sine function sin θ in any one of the phase regions Ph1 to Ph4 are known, the function expression can be used and the waveform itself has Symmetrical properties, the values of the sinusoidal functions sin θ corresponding to the input angles θ of other phase regions are derived, and the waveforms of the sinusoidal functions of the complete periods (ie, phase regions Ph1 to Ph4) are completed.

Please continue to refer to FIG. 1A and FIG. 1B to see the correlation between the sine function and the cosine function. Here, the value of the cosine function cos θ corresponding to the input angle θ=0 to 90 degrees of the phase region Ph5 in FIG. 1B is equivalent to the sine function sin θ corresponding to the input angle θ=90-180 of the phase region Ph2 in FIG. 1A. Value. The value of the cosine function cos θ corresponding to the input angle θ=90-180 of the phase region Ph6 is equivalent to the value of the sine function sin θ corresponding to the input angle θ=180 to 270 degrees of the phase region Ph3 in FIG. 1A.

The value of the cosine function cos θ corresponding to the input angle θ of the phase region Ph=180 to 270 degrees is equivalent to the value of the sine function sin θ corresponding to the input angle θ=270 to 360 degrees of the phase region Ph4 in FIG. 1A. The value of the cosine function cos θ corresponding to the input angle θ=270 to 360 degrees of the phase region Ph8 is equivalent to the value of the sine function sin θ corresponding to the input angle θ=0 to 90 degrees of the phase region Ph1 in FIG. 1A. That is to say, the value of the cosine function cos θ corresponding to each phase region Ph5~Ph8 of the cosine function can be derived from the numerical value of the sine function sin θ.

In other words, as long as the function expressions of the input angle θ and the sine function sin θ of any one of the phase regions Ph1 to Ph4 are known, then the symmetry properties of the waveforms of the function and the sinusoidal cosine function are used. Not only can the value of the sine function sin θ corresponding to each input angle θ of the other phase regions of the sinusoid function be derived, but also the value of the cosine function cos θ corresponding to the input angle θ of the cosine function in the phase regions Ph5 to Ph8 can be derived. Then complete the waveform diagram of the cosine function of the complete cycle (ie, phase regions Ph5~Ph8). In this way, the waveform of the sine/cosine function of the complete cycle can be obtained without the function comparison table, and the circuit area and the memory consumption are greatly saved.

To better understand how to obtain the sine function sin θ and the cosine function cos θ of the remaining seven phase regions by the functional expression of the sine function sin θ in any phase region of the phase region Ph1~Ph4, the following will be for this part. For a more detailed explanation, and for convenience of explanation, the following method is simply referred to as a phase region conversion method.

It is worth noting that since the values of the sinusoidal function sin θ in the phase region Ph1 are both positive and the angular range of the input angle is the smallest (θ = 0 to 90 degrees), the input angle θ corresponding to the sine function in the phase region Ph1 is used in this embodiment. The function of sin θ is used as the basis for the complete waveform. Before using the phase region conversion method, the sine function must be sampled to generate a digital sine wave function.

2A is a timing diagram of the sine wave function F(t) of a single period of FIG. 1A versus time t, where F(t)=sinωt=sinθ, the period T and the frequency f are reciprocal with each other, and the angular velocity ω=2πf, so the input angle θ = ωt = 2πt / T. Here, when t=0, θ=2πt/T=0; when t=T/2, θ=2πt/T=π; when t=T, θ=2πt/T=2π, and due to π弪 = 180 degrees, so t = 0, T / 2, T correspond to the input angle θ = 0 degrees, 180 degrees and 360 degrees, respectively.

As shown in FIG. 2A, a single-cycle sine wave function F(t) has a plurality of sampling points S, and the time interval T _S of each sampling point _S depends on the number of sampling points S. Here, it is assumed that a single-cycle sine wave function has 2^N sample points S, where N is a positive integer and is used to represent the number of bits required for data input. When the number of sampling points S is larger, that is, 2^N=1024, 2048..., the smaller the time interval T _{S of} each sampling point S is, and the waveform formed is also close to the waveform of the original sine wave function.

However, as the number of sampling points S increases, the amount of data to be processed will be larger, and the number of bits N to be used will be larger, requiring a larger circuit area and more memory for the purpose. Operation. Therefore, in this embodiment, the number of sampling points is set to 2^N=256, that is, the number of bits required for data input is N=8, so as to reduce the overall circuit area and increase the operation speed of the memory. In this way, the sine wave function F(t) of the present example has 256 sampling points S in the period T to evenly distribute the time T, wherein each sampling point S corresponds to a different time and corresponding time. The input angle.

Therefore, in the case of the number of samples 2^N=256, the digital sine wave function can be expressed as: sin(X)=sin(2nπ/256)=F1(n), where X=2nπ/256, and n= 0~255, π=3.14159.

Here, X is used to represent the input angle θ expressed in radiance. n is a digital angle input variable. Since π弪=180 degrees, 1弪=180 degrees/3.14159=57.296 degrees, when n=1, X=2π/256=0.024544弪=1.40625 degrees, that is, when the digital angle input variable n increases by 1, then Indicates that the input angle θ is increased by 1.40625 degrees. By analogy, when the digital angle input variable n=64, X=2π*64/256=π/2弪=1.40625 degrees*64=90 degrees. Therefore, after this conversion, the sine wave function F(t) can be represented by a digital angle input variable n to complete the digital sine wave function, that is, F1(n). Thus, the sine wave function F(t) plotted against time in FIG. 2A can be represented as a digital sine wave function F1(n) plotted as a digital angle input variable in FIG. 2B.

2B is a plot of a single-cycle digital sine wave function F1(n) versus a digital angle input variable, where the digital sine wave function F1(n)=sin(2nπ/256). Referring to FIG. 2B at the same time, FIG. 2A is similar to FIG. 2B, and the main difference is that the horizontal axis of FIG. 2B is a digital angle input variable n, which corresponds to the time t of FIG. 2A. As shown in FIG. 2B, it can be divided into 256 sampling points in the time period T1, that is, the writing period T1=256 can be schematically written, and each of the sampling points corresponds to the input angle θ at different times. For example, when the digital angle input variables n=0, 64, 128, 192, 256, respectively, the corresponding input angles θ=0, 90, 180, 270, and 360 degrees, and the numerical values of the digital sine wave function are respectively F1 (0) = 0, F1 (64) = 1, F1 (128) = 0, F1 (192) = -1, and F (256) = 0.

2C is a plot of another single cycle digital sine wave function F2(m) versus a digital angle input variable m. Please compare FIG. 2B with FIG. 2C. It can be found from the graph that the period T2=1/2*T1=128, that is, the frequency f1 of the digital sine wave function F1(n) is two of the frequency f2 of the digital sine wave function F2(m). Times. In addition, it can be clearly seen from FIG. 2B and FIG. 2C that the values of the digital sine wave function are F2(32)=F1(64), F2(64)=F1(128), . . . , that is, F2(m)=F1. (n=2*m).

That is to say, when the frequency f2 of the digital sine wave function F2(m) is twice the frequency f1 of the digital sine wave function F1(n), if the digital sine wave function F2(m) is to be obtained, the variable m is input at the digital angle. The value at =64, that is, F2(64), can be calculated by taking 64 times 2 and taking the result 128 as a digital angle input variable n, and taking it to the digital sine wave function F1(n) to calculate the value. F2 (64). It can be seen that the value of the digital sine wave function F2(m) can be obtained from the digital sine wave function F1(n), so the frequency f1 of the digital sine wave function F1(n) is defined as the fundamental frequency (Base). Frequency).

Therefore, when the frequency f2 of the digital sine wave function F2(m) is 1/2 times the frequency f1 of the digital sine wave function F1(n), the digital angle of the digital sine wave function F2(m) can be input. The variable m is multiplied by 1/2, and the result 1/2*m is regarded as the digital angle input variable n=1/2*m, and is brought to the digital sine wave function F1(n) to calculate, and the digital sine wave function can be obtained. F2(m). Similarly, the digital sine wave function F2(m) whose frequency f2 is 4 times, 8 times, ... of the fundamental frequency f1 can be similarly generated.

Further, in order to make the digital sine wave function F1(n) more convenient to perform analysis in a digitized manner, the present embodiment further integerizes the numerical value of the digital sine wave function F1(n). Here, in the present embodiment, the numerical amplification principle is used to amplify the numerical value of the digital sine wave function F1(n) by K times, and then the enlarged numerical value is integerized to generate a prototype digital sine wave function F _{ST .} (n), where K is a positive number. In the present embodiment, assuming that K = 128, the prototype digital sine wave function F _ST (n) is expressed as:

F _ST (n)=Int(128*sin(X))=Int(128*F1(n))=Int(128*sin(2nπ/256))=Int(Y)

Where Y=128*sin(2nπ/256) and n=0~255. In addition, n=0~64 corresponds to the input angle θ=0~90 degrees, that is, the angle input range of the phase region Ph1. n=64~128 corresponds to the input angle θ=90~180 degrees, which is the angle input range of the phase zone Ph2. n=128~192 corresponds to the input angle θ=180~270 degrees, which is the angle input range of the phase zone Ph3. n=192~255 corresponds to the input angle θ=270~360 degrees, which is the angle input range of the phase zone Ph4.

Similarly, the digital cosine wave function can also obtain the function of the prototype digital cosine wave as shown below by using the above method:

F _CT (n)=Int(128*cos(X))=Int(128*cos(2nπ/256))=Int(Y')

Where Y' = 128 * cos (2nπ / 256) and n = 0 to 255. In addition, n=0~64 corresponds to the input angle θ=0~90 degrees, that is, the angle input range of the phase region Ph5. n=64~128 corresponds to the input angle θ=90~180 degrees, which is the angle input range of the phase zone Ph6. n=128~192 corresponds to the input angle θ=180~270 degrees, which is the angle input range of phase phase Ph7. n=192~255 corresponds to the input angle θ=270~360 degrees, that is, the angle input range of the phase area Ph8.

It is worth noting that due to the symmetry of the positive/cosine wave function itself, the prototype digital positive/cosine wave function F _ST (n) or F _CT (n) corresponding to each input angle X in the phase regions Ph2 to Ph8 The values can be derived from the functional formula F _ST (n) of the prototype digital sine wave located in the phase region Ph1. Therefore, in this embodiment, the functional formula F _ST (n) of the prototype digital sine wave is set as a function of a prototype digital wave, thereby synthesizing the sine wave and the cosine wave at the same time.

For the example of a digital sine wave function, when n=32, X=π/4=45 degrees, which is located in the phase region Ph1, at which time the function value of the prototype digital sine wave is equivalent to F _ST (32)=+ 91; when n=96, X=3π/4=135 degrees, which is located in the phase region Ph2, at which time the function value of the prototype digital sine wave is equivalent to F _ST (96)=+91=F _ST (32); When n=160, X=5π-4=225 degrees, which is located in the phase region Ph3. At this time, the function value of the prototype sinusoidal bit wave is equivalent to F _ST (160)=-91=-F _ST (32); When n=224, X=7π/4=315 degrees, which is located in the phase region Ph4, at which time the function value of the prototype digital sine wave is equivalent to F _ST (224)=-91=-F _ST (32).

Therefore, the digital angle input variable n (ie, n=64~255) located in the phase regions Ph2~Ph4 is shifted to the value z of the phase angle data of the phase region Ph1 (ie, the digital angle input ranging from 0 to 64) The variable n), and then the function value F _ST (n) of the prototype digital wave can be used to obtain the function value of the prototype digital sine wave corresponding to each digital input variable n of the phase region Ph2~Ph4.

Taking the example of the digital cosine wave function, when n=32, X=π/4=45 degrees, which is located in the phase region Ph5, at which time the function value of the prototype digital cosine wave is equivalent to F _CT (32)= +91=+F _ST (32); when n=96, X=3π/4=135 degrees, which is located in the phase region Ph6, at which time the function value of the prototype digital cosine wave is equivalent to F _CT (96)=- 91=-F _ST (32); when n=160, X=5π/4=225 degrees, which is located in the phase region Ph7, at which time the function value of the prototype digital cosine wave is equivalent to F _CT (160)=-91 =-F _ST (32); When n=224, X=7π/4=315 degrees, which is located in the phase region Ph8. At this time, the function value of the prototype digital cosine wave is equivalent to F _CT (224)=-91= -F _ST (32).

Therefore, the digital angle input variable n (ie, n=0~256) located in the phase regions Ph5~Ph8 is shifted to the digital angle input variable z of the phase region Ph1 (ie, the digital angle input variable n ranging from 0 to 64) Then, using the function F _ST (n) of the prototype digital wave, the function value of the prototype digital cosine wave corresponding to each digital input variable n of the phase region Ph5~Ph8 can be obtained. Among them, the conversion principle of the digital angle input variable n and the digital angle input variable z is derived from the trigonometric function relationship.

3 is a circuit block diagram of a digital waveform generator in accordance with an embodiment of the present invention. As shown in FIG. 3, the digital waveform generator 100 includes a frequency processing unit 110, a phase region identifying unit 120, a phase region converting unit 130, a function value calculating unit 140, and a polarity converting unit 150.

In this embodiment, the frequency processing unit 110 limits or enlarges the value of the received angle data IndexM according to the frequency selection signal Freq_sel based on the prototype digital wave, and generates a reference angle data according to the reference signal. IndexN. Wherein, the functional formula of the prototype digital wave corresponds to the above F _ST (n), and n is the value of the reference angle data IndexN, that is, the above-mentioned digital angle input variable n. The value of the angle data IndexM corresponds to the above-mentioned digital angle input variable m. Further, since the single-cycle sine wave function of the present embodiment has 2^N sampling points, and N=8. Therefore, the angle data IndexM and the reference angle data IndexN are all represented by 8 bits, and the values thereof are all in the range of 0 to 255.

In this embodiment, the frequency processing unit 110 limits or enlarges the angle data by performing a right shift or a left shift on each bit of the angle data IndexM, that is, corresponding to the previous FIG. 2B to FIG. 2C: The digital angle input variable m produces a new digital angle input variable n. In detail, when the number of bits of the angle data IndexM is shifted to the left by 1 bit to generate the reference angle data IndexN, the value m representing the angle data IndexM is enlarged by 2^1=2 times (corresponding to P=1 above) And the newly generated value 2m is the value n which is regarded as the reference angle data IndexN, that is, n=2m, and is output to the phase area converting unit 130 in the form of 8 bits, and finally the frequency is 2 times the frequency of the prototype digital wave. Periodic digital waves.

Similarly, when the number of bits of the angle data IndexM is shifted to the left by 2 bits, the value m of the angle data IndexM is enlarged by 2^2=4 times (corresponding to P=2 above), and the newly generated value 4m is regarded as The value n of the reference angle data IndexN, that is, n = 4 m, is output to the phase area converting unit 130 in the form of 8 bits, and finally a periodic digital wave having a frequency of 4 times the frequency of the prototype digital wave is finally generated. In addition, in this embodiment, the frequency processing unit 110 is further configured to receive the clock signal CLK and the reset signal Reset, and have a function of outputting the DC signal DCout. The DC signal DCout is used to drive a circuit external to the digital waveform generator. When the reset signal Reset is at a logic high level, the DC signal DCout exhibits a logic low level, the reference angle data IndexN=00000000, and the frequency selection signal Freq_sel=00000000.

Here, the frequency selection signal Freq_sel is used to determine whether the value m of the angle data IndexM is limited or multiplied. Table 2 is a truth table of the frequency selection processing unit 110, which will be described in more detail below for different frequency selection signals Freq_sel.

As shown in Table 2, each frequency selection signal Freq_sel corresponds to a different number of left shifting elements. When the reset signal Reset is at a logic low level and the frequency selection signal Freq_sel=00000001, the reference angle data IndexN=00000000 and the logic high level DC signal DCout are output to drive the device to be controlled by the external circuit. When the reset signal Reset is at a logic low level and the frequency selection signal Freq_sel=00000010, the angle data IndexM does not shift to the left of the bit, so the output reference angle data IndexN is equivalent to the input angle data IndexM. That is, the value n of the reference angle data IndexN is the same as the value m of the input angle data IndexM, that is, n=m, so the frequency f2 of the last generated periodic digital wave is equivalent to the frequency f1 of the prototype digital wave.

When the frequency selection signal Freq_sel is 00000100, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 1 bit. The value n of the output reference angle data IndexN is twice the value m of the input angle data IndexM, that is, n=2m, so the frequency f2 of the last generated periodic digital wave is twice the frequency f1 of the prototype digital wave. .

When the frequency selection signal Freq_sel is 00001000, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 2 bits. The value n of the output reference angle data IndexN is 2^2 times the value m of the input angle data IndexM, that is, n=4m, so the frequency f2 of the last generated periodic digital wave will be the frequency f1 of the prototype digital wave. 4 times.

When the frequency selection signal Freq_sel is 00010000, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 3 bits. The value n of the output reference angle data IndexN is 2^3 times the value m of the input angle data IndexM, that is, n=8m, so the frequency f2 of the last generated periodic digital wave will be the frequency f1 of the prototype digital wave. 8 times.

When the frequency selection signal Freq_sel is 00100000, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 4 bits. The value n of the output reference angle data IndexN is 2^4 times the value m of the input angle data IndexM, that is, n=16m, so the frequency f2 of the last generated periodic digital wave will be the frequency f1 of the prototype digital wave. 16 times.

When the frequency selection signal Freq_sel is 01000000, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 5 bits. The value n of the output reference angle data IndexN is 2^5 times the value m of the input angle data IndexM, that is, n=32m, so the frequency f2 of the last generated periodic digital wave will be the frequency f1 of the prototype digital wave. 32 times.

When the frequency selection signal Freq_sel is 10000000, and the reset signal Reset is at a logic low level, the output DC signal DCout exhibits a logic low level, and each bit of the angle data IndexM is shifted to the left by 6 bits. The value n of the output reference angle data IndexN is 2^6 times the value m of the input angle data IndexM, that is, n=64m, so the frequency f2 of the last generated periodic digital wave will be the frequency f1 of the prototype digital wave. 64 times.

Referring to FIG. 3, as shown in FIG. 3, the frequency processing unit 110 also captures the 2 most significant bits in the reference angle data IndexN to generate the phase region signal xdela of the phase region identifying unit 120. The two most significant bits in this embodiment are the 8th bit and the 7th bit in the reference angle data IndexN, respectively. For convenience of explanation, the following describes the 8th and 7th bits in the reference angle data IndexN by IndexN[7] and IndexN[6]. Among them, IndexN[7] and IndexN[6] can be used to indicate that the prototype digital wave can be divided into 2^2 phase regions according to its symmetrical characteristics.

In addition, since the sine wave and the cosine wave can be divided into four phase regions according to their symmetrical characteristics, an additional signal is needed to determine whether the desired periodic digital wave is a sinusoidal bit wave or a cosine digital bit wave. Therefore, in addition to IndexN[7], IndexN[6], the phase region discriminating unit 120 needs two waveform indicating signals xSin and xCos to generate the phase region discriminating signal Ph_discri to point to the extension of the prototype digit wave (2^2). * A specific phase zone of 2 phase zones.

In detail, when the phase region signals xdela=00, 01, 10, 11, respectively, they can represent the value n of the reference angle data IndexN, respectively, in the phase region Ph1 or the phase region Ph5, the phase region Ph2 or the phase region Ph6, and the phase region Ph3. Or phase zone Ph7, phase zone Ph4 or phase zone Ph8. That is to say, in order to judge that the value n of the reference angle data IndexN is located in the phase region Ph1 or the phase region Ph5, the phase signal xdela of the light is insufficient, and must be determined according to the waveform indication signals xSin and xCos.

Further, when the phase region signal xdela=00 and the waveform indication signal xSin=1, it can be determined that the value n of the reference angle data IndexN is located in the phase region Ph1. On the other hand, if the waveform indication signal xCos=1, it can be determined that the value n of the reference angle data IndexN is located in the phase region Ph5. Table 3 is a comparison table of input and output signals and feature descriptions of the phase region discriminating unit 120.

As shown in Table 3, the different input signals correspond to different phase region discrimination signals Ph_discri, which respectively represent the phase regions in which the value n of the reference angle data IndexN is located. In summary, the phase identification signal Ph_discri is used to identify the eight phase regions extended by the prototype digital wave. The phase regions Ph1~Ph4 of the prototype digital wave are used to synthesize the sinusoidal digital wave, and the phase regions Ph5~Ph8 of the prototype digital wave are used to synthesize the cosine digital wave.

Referring to FIG. 3, as shown in FIG. 3, the phase area converting unit 130 receives the reference angle data IndexN from the frequency processing unit 110, and the phase area discrimination signal Ph_discri from the phase area identifying unit 120. After receiving the signals, the phase region converting unit 130 selects a shifting relationship corresponding to the specific phase region from the eight shifting relationships, and shifts the value n of the reference angle data IndexN to the first phase region of the prototype digit wave to Phase phase angle data PhaseZ is generated. The eight shifting relationships respectively correspond to the phase regions Ph1~Ph8. It is to be noted that the function performed by the phase area converting unit 130 is the previously mentioned phase area converting method. This section will be described in more detail below.

Figure 4 shows a trigonometric function on a unit circle. As shown in Fig. 4, P(x, y) is a moving point on the unit circle, with O(0, 0) as the center and coordinates (1, 0) as the starting point, and moving at an angular rate ω counterclockwise. Line segment The angle θ with the x-axis is a function of time t (θ = ωt), which represents the input angle of the trigonometric function. The coordinates (x, y) are defined as cosine function cos θ and sine function sin θ, respectively, ie x=cos θ and y=sin θ are unit circles =1) the natural dynamics, and the coordinates (x, y) also correspond to the numerical values of the prototype digital cosine wave function F _CT (n) and the numerical value of the prototype digital sine wave function F _ST (n), respectively. .

It is worth noting that the basic properties of the sine/cosine function can be expressed by the geometric properties of the circle (mainly symmetry), including: (1) Pythagorean theorem: cos ² θ + sin ² θ = =1; (2) periodicity: circumference circumference = 2π, cos(2π + θ) = cos θ and sin(2π + θ) = sin θ; (3) symmetry: cos(-θ) = cos(θ), Sin(-θ)=-sinθ; cos(π-θ)=-cosθ, sin(π-θ)=sinθ; sin(π/2-θ)=cosθ, cos(π/2-θ)=sinθ.

Wherein, the coordinates (1, 0), (0, 1), (-1, 0), (0, -1) respectively correspond to the input angles of the trigonometric functions θ = 0 degrees, 90 degrees, 180 degrees, and 270 degrees, that is, In the embodiment, the value of the reference angle data IndexN is n=0, 64, 128, 192, and the values of the reference angle data IndexN are n=0~63, 64~127, 128~191 and 192~255 respectively corresponding to the phase regions Ph1~Ph4. In addition, for convenience of explanation, the following reference angle data IndexN located in the phase region Ph1, that is, the value n=0~63, is defined as the phase angle data PhaseZ, that is, the value Z=0~63, and will be located in the phase region Ph1. The function F _ST (n) of the prototype digital wave is defined as the function expression F _MP (z).

Therefore, referring to FIG. 4 and using the symmetry of the above-described trigonometric function, the correspondence relationship between the reference angle data IndexN of the phase regions Ph1 to Ph8 and the phase angle data PhaseZ can be respectively found, and the eight shift relations are summarized. A more detailed description of this part will be made.

When n=0~63 and corresponding to the phase region Ph1, since the reference angle data IndexN is located in the phase region Ph1, that is, the phase region in which the phase angle data PhaseZ is located. Therefore, the value n of the reference angle data IndexN is equivalent to the value Z of the phase angle data PhaseZ. Table 4A is a table of the shift relationship of the value n of the reference angle data IndexN in the phase regions Ph1, n and z.

When n=64~127 and corresponds to the phase region Ph2, since the reference angle data IndexN is located in the phase region Ph2, the symmetry sin(π-θ)=sinθ using the trigonometric function (where π corresponds to the value n of the reference angle data IndexN) =128, θ corresponds to the value of the reference angle data IndexN n=64~127), and the comparison between the value of the function F _ST (n) of the following prototype digital wave and the value of the function expression F _MP (z) can be obtained:

F _ST (64)=F _MP (64), F _ST (65)=F _MP (63), F _ST (66)=F _MP (62), F _ST (67)=F _MP (61),... F _ST (124) = F _MP (4), F _ST (125) = F _MP (3), F _ST (126) = F _MP (2), F _ST (127) = F _MP (1).

That is to say, if the value of the functional formula F _MP (z) is used to obtain the value of the functional formula F _ST (n) of the prototype digital wave located in the phase region Ph2, the value z of the phase angle data PhaseZ is equivalent to 128. Subtract the value n of the reference angle data IndexN, ie z=128-n. Table 4B is a table for shifting the value n of the reference angle data IndexN in the phase region Ph2, n and z.

When n=128~191 and corresponding to the phase region Ph3, since the reference angle data IndexN is located in the phase region Ph3, the symmetry sin(θ-π)=-sinθ of the trigonometric function is used (where π corresponds to the value of the reference angle data IndexN) n=128, θ corresponds to the value of the reference angle data IndexN n=128~191), and the comparison between the value of the function F _ST (n) of the following prototype digit wave and the value of the function expression F _MP (z) can be obtained:

F _ST (128)=-F _MP (0), F _ST (129)=-F _MP (1), F _ST (130)=-F _MP (2), F _ST (131)=-F _MP (3 ), F _ST (132)=-F _MP (4),..., F _ST (188)=-F _MP (60), F _ST (189)=-F _MP (61), F _ST (190)=- F _MP (62), F _ST (191) = -F _MP (63).

That is to say, if the value of the functional formula F _MP (z) is used to obtain the value of the functional formula F _ST (n) of the prototype digital wave located in the phase region Ph3, the value z of the phase angle data PhaseZ is equivalent to the reference. The value n of the angle data IndexN is subtracted by 128, i.e., z = n - 128, and F _S T(n) = -F _MP (z). Table 4C is a table for shifting the value n of the reference angle data IndexN in the phase region Ph3, n and z.

When n=192~256 and corresponding to the phase region Ph4, since the reference angle data IndexN is located in the phase region Ph4, the symmetry Sin(2π-θ)=-sinθ using the trigonometric function (where 2π corresponds to the value of the reference angle data IndexN) n=256, θ corresponds to the value of the reference angle data IndexN=192~255), and the comparison between the value of the function F _ST (n) of the following prototype digital wave and the value of the function expression F _MP (z) can be obtained:

F _ST (192)=-F _MP (64), F _ST (193)=-F _MP (63), F _ST (194)=-F _MP (62), F _ST (195)=-F _MP (61 ),..., F _ST (252)=-F _MP (4), F _ST (253)=-F _MP (3), F _ST (254)=-F _MP (2), F _ST (255)=- F _MP (1), F _ST (256) = -F _MP (0).

That is to say, if the value of the functional formula F _MP (z) is used to obtain the value of the function F _ST (n) of the prototype digital wave located in the phase region Ph4, the value z of the phase angle data PhaseZ is equivalent to 256. Subtract the value n of the reference angle data IndexN, ie z=256-n, and F _ST (n)=-F _MP (z). Table 4D is a table for shifting the value n of the reference angle data IndexN in the phase region Ph4, n and z.

When n=0~63 and corresponds to the phase region Ph5 (the phase region to which the prototype digital cosine wave belongs), the symmetry cos θ=sin(π/2-θ) using the trigonometric function (where π/2 corresponds to the reference angle data IndexN) The value n=64, θ corresponds to the value of the reference angle data IndexN n=0~63), and the comparison between the value of the following prototype digital cosine wave function F _CT (n) and the value of the function expression F _MP (z) can be obtained. result:

F _CT (0)=F _MP (64), F _CT (1)=F _MP (63), F _CT (2)=F _MP (62), F _CT (3)=F _MP (61),... F _CT (60) = F _MP (4), F _CT (61) = F _MP (3), F _CT (62) = F _MP (2), F _CT (63) = F _MP (1).

That is to say, if the value of the function number F _MP (z) is used to obtain the value of the prototype digital cosine wave function F _CT (n) located in the phase region Ph5, the value z of the phase angle data PhaseZ is equivalent to 64. Subtract the value n of the reference angle data IndexN, ie z=64-n, and F _CT (n)=F _MP (z). Table 4E is a table for shifting the relationship between n and z when the value n of the reference angle data IndexN corresponds to the phase region Ph5.

When n=64~127 and corresponding to the phase region Ph6 (ie, the phase region to which the prototype digital cosine wave belongs), the symmetry cos θ=-sin(θ-π/2) of the trigonometric function is used (where π/2 corresponds to the reference angle) The value of the index IndexN is n=64, and θ corresponds to the value of the reference angle data IndexN (n=64~127), and the value of the following prototype digital cosine wave function F _CT (n) and the value of the function expression F _MP (z) can be obtained. Comparison results:

F _CT (64)=-F _MP (0), F _CT (65)=-F _MP (1), F _CT (66)=-F _MP (2), F _CT (67)=-F _MP (3 ),..., F _CT (124)=-F _MP (60), F _CT (125)=-F _MP (61), F _CT (126)=-F _MP (62), F _CT (127)=- F _MP (63).

That is to say, if the value of the function number F _MP (z) is used to obtain the value of the prototype digital cosine wave function F _CT (n) located in the phase region Ph6, the value z of the phase angle data PhaseZ is equivalent to the reference. The value n of the angle data IndexN is subtracted by 64, i.e., z = n - 64, and F _CT (n) = -F _MP (z). Table 4F is a table for shifting the relationship between n and z when the value n of the reference angle data IndexN corresponds to the phase region Ph6.

When n=128~191 and corresponds to the phase region Ph7 (ie, the phase region to which the prototype digital cosine wave belongs), the symmetry cos θ=-sin(3π/2-θ) using the trigonometric function (where 3π/2 corresponds to the reference) The value of the angle data IndexN is n=192, θ corresponds to the value of the reference angle data IndexN n=128~191), and the value of the following prototype digital cosine wave function F _CT (n) and the function expression F _MP (Z) can be obtained. Comparison of values:

F _CT (128)=-F _MP (64), F _CT (129)=-F _MP (63), F _CT (130)=-F _MP (62), F _CT (131)=-F _MP (61 ),..., F _CT (188)=-F _MP (4), F _CT (189)=-F _MP (3), F _CT (190)=-F _MP (2), F _CT (191)=- F _MP (1).

That is to say, if the value of the function number F _MP (z) is used to obtain the value of the prototype digital cosine wave function F _CT (n) located in the phase region Ph7, the value z of the phase angle data PhaseZ is equivalent to 192. Subtract the value n of the reference angle data IndexN, ie z=192-n, and F _CT (n)=-F _MP (z). Table 4G is a table for shifting the relationship between n and z when the value n of the reference angle data IndexN corresponds to the phase region Ph7.

Table 4G

When n=192~256 and corresponds to the phase region Ph8 (ie, the phase region to which the prototype digital cosine wave belongs), the symmetry cos θ=sin(θ-3π/2) using the trigonometric function (where 3π/2 corresponds to the reference angle) The value of the index IndexN is n=192, and θ corresponds to the value of the reference angle data n=192~256), and the value of the following prototype digital cosine wave function F _CT (n) and the value of the function expression F _MP (z) can be obtained. Comparing results:

F _CT (192)=F _MP (0), F _CT (193)=F _MP (1), F _CT (194)=F _MP (2), F _CT (195)=F _MP (3),... F _CT (252) = F _MP (60), F _CT (253) = F _MP (61), F _CT (254) = F _MP (62), F _CT (255) = F _MP (63).

That is to say, if the value of the function number F _MP (z) is used to obtain the value of the prototype digital cosine wave function F _CT (n) located in the phase region Ph8, the value z of the phase angle data PhaseZ is equivalent to the reference. The value n of the angle data IndexN is subtracted by 192, i.e., z = n - 192, and F _CT (n) = F _MP (z). Table 4H is a table for shifting the relationship between n and z when the value n of the reference angle data IndexN corresponds to the phase region Ph8.

In summary, the value of the prototype digital sine wave function F _ST (n) of each phase region Ph1~Ph4 and the value of the prototype digital cosine wave function F _CT (n) of each phase region Ph5~Ph8 can be The value of the function expression F _MP (z) is obtained. Among them, the eight shifting relations of the value n of the reference angle data IndexN corresponding to each phase region and the value z of the phase angle data PhaseZ are as shown in Table 4I:

In addition, Table 4J is a comparison table between the prototype digital sine wave function F _ST (n) and the function expression F _MP (Z) in the phase region Ph2~Ph4, and Table 4K is the prototype in the phase region Ph5~Ph8. Comparison of the relationship between the digital cosine wave function F _CT (n) and the function expression F _MP (Z):

Table 4J

Referring to Table 4J, as shown in Table 4J, F _ST (96) = F _MP (32), F _ST (160) = -F _MP (32), F _ST (224) = -F _MP (32). Therefore, once again, as long as the value of the function expression F _MP (Z) at z=32 is known, the value of the prototype digital sine wave function F _ST (n) at n=96, 160, and 224 can be obtained.

Next, please refer to Table 4K:

As shown in Table 4K, F _CT (32) = F _MP (32), F _CT (96) = -F _MP (32), F _CT (160) = -F _MP (32), F _ST (224) = F _MP (32). Similarly, Table 2J of Table 2 again shows that as long as the value of the function expression F _MP (Z) at Z=32 is known, the prototype digital cosine wave function F _CT (n) can be obtained at n=32. Values at 96, 160 and 224 hours.

Please continue to refer to FIG. 3. Therefore, after the phase region converting unit 130 receives the phase region discriminating signal Ph_discri, the reference angle data IndexN can be moved by selecting one of the eight shifting relations according to the phase region discriminating signal Ph_discri. Action to generate PhaseZ angle data PhaseZ. For example, when the reference angle data IndexN=01000001 and the phase region discrimination signal Ph_discri=00000010, the reference angle data IndexN corresponds to the phase region Ph2, and the phase region conversion unit 130 uses the z=128-n in the shift relationship. The reference angle data IndexN is calculated to obtain the phase angle data PhaseZ.

Among them, the calculation method is a binary calculation system, that is, -n can be expressed as not(IndexN)+0000001. For example, when n=64, -64 can be expressed as not(01000000)+00000001=10111111+00000001=11000000; when n=128, -128 can be expressed as not(10000000)+00000001=01111111+00000001=10000000; when n=192 , then -192 can be expressed as not (11000000) + 00000001 = 00111111 + 00000001 = 01000000. In this way, the eight shifting relationships can be implemented in a binary manner, as shown in Table 5:

The next action is to find the value of the function expression F _MP (z) by the value z of the phase angle data PhaseZ. The function value calculation unit 140 of this embodiment is used to perform this function. The function value calculation unit 140 will provide a plurality of function expressions F _MP (z) for synthesizing the prototype digit waves in the first phase region, and bring the phase angle data PhaseZ to these function expressions F _MP One of (z) to calculate the phase function value Mag_PhZ corresponding to the phase angle data PhaseZ.

Tables 6A to 6J are numerical table of the value z of the phase angle data PhaseZ and the function expression F _MP (z). Where F _MP (z)=Int(128*sin(2zπ/256))=Int(128*sin(X))=Int(Y), z=0~64.

Please refer to Table 6A. As shown in Table 6A, when the value of the phase angle data PhaseZ is z=0~3, the relationship between z and Int(Y) in Table 6A can be summarized. The function expression F _MP (z)= 2*z+z, where z=0~3.

Please refer to Table 6B. As shown in Table 6B, when the value of the phase angle data PhaseZ is z=4~19, the function expression F _MP (z) which can be summarized by the relationship between z and Int(Y) in Table 6B. ) = (2 * z + 1) + z, where z = 4 ~ 19.

Please refer to Table 6C. As shown in Table 6C, when the value of the phase angle data PhaseZ is z=20~23, the function expression F _MP (z) can be summarized from the relationship between z and Int(Y) in Table 6C. =2*z+z, where z=20~23.

Please refer to Table 6D. As shown in Table 6D, when the value of the phase angle data PhaseZ is z=24~27, the function expression F _MP (z)=2 can be summarized from z and Int(Y) in Table 6D. *z+24, where z=24~27.

Refer to Table 6E, as shown in Table 6E, when the value of the phase angle data PhaseZ zone of z = 28 ~ 31, the function can be summarized as calculation formula F _MP (z) = 2 z in Table 6E and Int (Y) *z+26, where z=28~31.

Please refer to Table 6F. As shown in Table 6F, when the value of the phase angle data PhaseZ is z=32~39, the function expression F _MP (z) can be summarized by z and Int(Y) in Table 6F. =2*z+27, where z=32~39.

Please refer to Table 6G. As shown in Table 6G, when the value of the phase angle data PhaseZ is z=40~43, the function expression F _MP (z)=2 can be summarized from z and Int(Y) in Table 6G. *z+26, where z=40~43.

Please refer to Table 6H. As shown in Table 6H, when the value of the phase angle data PhaseZ is z=44~55, the function expression F _MP (z)=z+ can be summarized by z and Int(Y) in Table 6H. 70, where z=44~55.

Please refer to Table 6I. As shown in Table 6I, when the value of the phase angle data Phasez is z=56~59, the function expression F _MP (z)=126 can be summarized by z and Int(Y) in Table 6H. When the value of the phase angle data PhaseZ is z=60~64, the function expression F _MP (z)=127 can be summarized.

In summary, the relationship between the value z of the phase angle data PhaseZ and the function expression F _MP (z) can be summarized as shown in Table 6J:

FIG. 5 is an internal circuit block diagram of the function value operation unit 140 of FIG. As shown in FIG. 5, the function value operation unit 140 includes an operation control unit 142, a first value adjustment unit 144a, a second value adjustment unit 144b, a third value adjustment unit 144c, a first multiplex unit 146a, A second multiplexing unit 146b and an adding unit 148.

The arithmetic control unit 142 encodes the third to seventh bits (b[2] to b[6]) of the phase angle data PhaseZ with reference to the function expression F _MP (z) to generate a corresponding arithmetic control signal Angle_discri(n). ), where n = 1~17, and here the definition of the signal Ph62Z is defined as the signal consisting of the 3rd to 7th bits of PhaseZ. The operation control signal Angle_discri(n) is used to control a plurality of function expressions F _MP (Z), which are organized as shown in Table 6K.

As shown in Table 6K, when the phase angle data PhaseZ=00000011, its third bit to the seventh bit = 00000, so the operation control signal Angle_discri(1) is generated correspondingly, and the function expression F _MP (Z)= 2z+z will be selected. It is worth noting that the value of the phase angle data PhaseZ at this time is z=3, which belongs to the range of the value z of the phase angle data PhaseZ in the function expression F _MP (z)=2z+z. When the phase angle data PhaseZ=00000100, its third bit to the seventh bit=00001, the corresponding operation control signal Angle_discri(2) is generated, and the function expression F _MP (z)=(2z+1)+ z will be selected. It is worth noting that the value of phase phase angle data PhaseZ is z=4, which belongs to the range of the value z of the phase angle data PhaseZ in the function expression F _MP (z)=(2z+1)+z. Therefore, it is sufficient to determine the function expression F _MP (z) corresponding to the phase angle data PhaseZ by the third bit to the seventh bit in the phase angle data PhaseZ.

Referring to FIG. 5, the first value adjusting unit 144a is configured to multiply the value z of the phase angle data PhaseZ to generate the first adjustment data {2Z}. The specific method is to shift the phase angle data PhaseZ to the left by 1 bit. In addition, the second value adjusting unit 144b is configured to multiply the value Z of the phase angle data PhaseZ and add 1 to generate a second adjustment data {2Z+1}. The specific method is to shift the phase angle data PhaseZ to the left by 1 bit and add 00000001. The third value adjustment unit 144c receives the phase angle data PhaseZ and outputs it as the third adjustment data {Z}.

Referring to FIG. 5, the first multiplex unit 146a selects the first adjustment data {2Z}, the second adjustment data {2Z+1}, the first constant data N70, and a second constant data according to the operation control signal Angle_discri(n). Select one of the N100 to generate the first function adjustment data. The value of the first constant data N70 is 70, and the value of the second constant data N100 is 100. In addition, the second multiplex unit 146b selects one of the third adjustment data {Z}, the third constant data N24, the fourth constant data N26, and the fifth constant data N27 according to the operation control signal Angle_discri(n) to generate the first Two function adjustment data. The value of the third constant data N24 is 24, the value of the fourth constant data N26 is 26, and the value of the fifth constant data N27 is 27. The adding unit 148 is configured to add the first function adjustment data and the second function adjustment data to generate the phase function value Mag_PhZ.

Table 6L is a comparison table of the operation control signal Angle_discri(n) and the first function adjustment value and the second function adjustment value.

Referring to FIG. 5 and referring to Table 6L, when the operation control unit 142 receives the phase angle data PhaseZ, it will encode according to the third to seventh bits of the phase angle data PhaseZ to generate a corresponding operation. Control signal Angle_discri(n). Here, the arithmetic control signal Angle_discri(n) is output to the first multiplex unit 146a and the second multiplex unit 146b. At the same time, the first value adjustment unit 144a, the second value adjustment unit 144b, and the third value adjustment unit 144c also perform different operations on the phase angle data PhaseZ to generate the first adjustment data {2Z} and the second adjustment data, respectively. {2Z+1}, and the third adjustment data {Z}.

The first adjustment data {2Z} and the second adjustment data {2Z+1} are output to the first multiplex unit 146a. In addition, the first multiplex unit 146a further receives the first constant data N70 and the second constant data N100. Next, the first multiplex unit 146a selects the input data of one of the above according to the operation control signal Angle_discri(n) to adjust the data as the first function. Similarly, the second multiplex unit 146b also receives the third adjustment data {Z}, the third constant data N24, the fourth constant data N26, and the fifth constant data N27, and according to the operation control signal Angle_discri(n), Select one of the above to adjust the data as a second function. Finally, the adding unit 148 adds the received first function adjustment data and the second function adjustment data. At this point, the operation of the phase function value Mag_PhZ is completed.

For example, when the phase angle data PhaseZ=00010100 (ie, z=20), the operation control unit 142 generates an operation control signal Angle_discri(1) and outputs it to the first multiplex unit 146a and the second multiplex unit 146b. . Next, the first multiplex unit 146a selects the first adjustment data {2Z} generated by the first value adjustment unit 144a as the first function adjustment data, wherein the value of the second adjustment data is 2z=2*20=40 . The second multiplex unit 146b selects the third adjustment data {Z} generated by the third adjustment value unit 146c as the second function adjustment data, wherein the value of the third adjustment data is z=20. Finally, the adding unit 148 adds the first function adjustment data and the second function adjustment data to generate a phase function value Mag_PhZ, where the phase function value = (2z) + z = 40 + 20 = 60.

Referring to FIG. 3, as shown in FIG. 3, the polarity switching unit 150 determines the opposite of the output phase function value Mag_PhZ or directly outputs the phase function value Mag-PhZ according to the phase region discrimination signal xPh_discri(n). Waveform output value Mag_SC. The polarity conversion unit 150 further generates a periodic digital wave whose frequency is 1/(2^P) times or 2^P times the prototype digital wave by using the waveform output value Mag_SC. In the present embodiment, the method of polarity switching employs a method of adding 2 to the complement. The description is as follows: Assuming an octet constant A = 01101100, its negative number B = not A + 1. Since A=01101100, not A=10010011, so its negative number B=not A+1=10010011+00000001=10010100. When A=01101100 and B=10010100, then A+B=01101100+10010100=00000000, so A and B are said to be opposite to each other, that is, B is a negative number of A.

Table 6M is a comparison table of input and output signals and characterization of the polarity conversion unit 150.

As shown in Table 6M, when the phase region discrimination signal Ph_discri=00000001, it indicates that the phase angle data PhaseZ is in the phase region Ph1, that is, in the region of 0 to 90 degrees of the digital sine wave function F _ST (n), and this The corresponding digital angle input value is n=0~63. Since the values of the digital sine wave function F _ST (n) in this range are all positive values, the polarity conversion unit 150 directly outputs the phase function value Mag_PhZ as the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=00000010, it indicates that the phase angle data PhaseZ is in the phase region Ph2, that is, in the region of 90 to 180 degrees of the digital sine wave function F _ST (n), and the corresponding digital angle at this time Enter the value n=64~127. Since the values of the digital sine wave function F _ST (n) in this range are all positive values, the polarity conversion unit 150 directly outputs the phase function value Mag_PhZ as the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=00000100, it indicates that the phase angle data PhaseZ is in the phase region Ph3, that is, in the region of 90 to 180 degrees of the digital sine wave function F _ST (n), and the corresponding digital angle at this time Enter the value n=128~191. Since the values of the digital sine wave function F _ST (n) in this range are all negative values, and the values of the function expression F _MP (z) are opposite to each other, an inverse conversion operation is required. Therefore, the polarity conversion unit 150 adds 1 to the complement of the phase function value Mag_PhZ to generate the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=00001000, it indicates that the phase angle data PhaseZ is in the phase region Ph4, that is, in the region of 180 to 270 degrees of the digital sine wave function F _ST (n), and the corresponding digital angle at this time Enter the value n=192~255. Since the values of the digital sine wave function F _ST (n) in this range are all negative values, and the values of the function expression F _MP (z) are opposite to each other, an inverse conversion operation is required. Therefore, the polarity conversion unit 150 adds 1 to the complement of the phase function value Mag_PhZ to generate the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=00010000, it indicates that the phase angle data PhaseZ is in the phase region Ph5, that is, in the region of 0 to 90 degrees of the digital cosine wave function F _CT (n), and the corresponding digital angle at this time Enter the value n=0~63. Since the values of the digital cosine wave function F _CT (n) in this range are all positive values, the polarity converting unit 150 directly outputs the phase function value Mag_PhZ as the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=00100000, it indicates that the phase angle data PhaseZ is in the phase region Ph6, that is, in the region of 90 to 180 degrees of the digital cosine wave function F _CT (n), and the corresponding digital angle at this time Enter the value n=64~127. Since the values of the digital cosine wave function F _CT (n) in this range are all negative values, and the values of the function expression F _MP (Z) are opposite to each other, an inverse conversion operation is required. Therefore, the polarity conversion unit 150 adds 1 to the complement of the phase function value Mag_PhZ to generate the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=01000000, it indicates that the phase angle data PhaseZ is in the phase region Ph7, that is, in the region of 180 to 270 degrees of the digital cosine wave function F _CT (n), and the corresponding digital angle at this time Enter the value n=128~191. Since the values of the digital cosine wave function F _CT (n) in this range are all negative values, and the values of the function expression F _MP (z) are opposite to each other, an inverse conversion operation is required. Therefore, the polarity conversion unit 150 adds 1 to the complement of the phase function value Mag_PhZ to generate the waveform output value Mag_SC.

When the phase region discrimination signal Ph_discri=10000000, it indicates that the phase angle data PhaseZ is in the phase region Ph8, that is, in the region of 270 to 360 degrees of the digital cosine wave function F _CT (n), and the corresponding digital angle at this time Enter the value n=191~255. Since the values of the digital cosine wave function F _CT (n) in this range are all positive values, the polarity converting unit 150 directly outputs the phase function value Mag_PhZ as the waveform output value Mag_SC.

In summary, the digital waveform generator 100 can use these waveform output values Mag_SC to generate periodic digital waves having a frequency of 1/(2^P) times or 2^P times the prototype digital wave.

FIG. 6 is a block diagram of a peripheral circuit of the digital waveform generator of the embodiment. As shown in FIG. 6, the waveform generator 100 further includes a data input unit 160, a continuous angle generating unit 170, and a data selecting unit 180.

The data input unit 160 decodes the 8-bit input address ADDRin to treat the 8-bit digital data DATAin as a single-angle data xDatain or an 8-bit operation instruction Command according to the decoded input address ADDRin. In detail, when the reset signal Reset is at a logic high level, the output port for transmitting the single angle data xDatain and the operation command Command, etc., will be reset to zero.

When the reset signal Reset is at a logic low level, the data write signal WR is at a logic high level, and the clock signal CLK is triggered from a logic low level to a logic high level positive edge (Positive Edge): if at the input The address number of E5 appears on the address ADDRin. After being decoded by the data input unit 160, the data input unit 160 regards the digital data DATAin as a single angle data xDatain and outputs it to the data selection unit 180; otherwise, if the input address is ADDRin appears with the address number of E7. After decoding by the data input unit 160, the data input unit 160 regards the digital data DATAin as the operation command Command.

Since the address data of E5 or E7 is input on the input address ADDRin, it can be clearly known that the digital data DATAin is a single angle data xDatain or an operation instruction Command. Therefore, by using the address data decoding method, the data from the processor can be easily and correctly transmitted to the destination, that is, where the data is processed.

Referring to FIG. 6, as shown in FIG. 6, the continuous angle generating unit 170 receives a data selection signal xVW to determine whether to generate the continuous angle data xAngle. When the reset signal Reset is at a logic high level, the continuous angle data xAngle=00000000. When the reset signal Reset is a logic low level data selection signal xVW is a logic high level, and the clock signal CLK is triggered from a logic low level to a logic high level positive edge, the angle signal inside the continuous angle generating unit 170 The generator automatically generates a continuous angle data xAngle=00000000~11111111, that is, a continuous angle data xAngle with a value of 0~255, and also serves as an input signal of the data selection unit 180.

When both the reset signal Reset and the data selection signal xVW are at a logic low level, the angle signal generator inside the continuous angle generating unit 170 stops the generation of the angle data xAngle. Therefore, the data selection unit 180 can read the single angle data xDatain or the continuous angle data xAngle according to the material selection signal xVW to generate the angle data IndexM, and output the angle data IndexM to the frequency processing unit 110.

FIG. 7 is a block diagram of another peripheral circuit of the digital waveform generator of the embodiment. As shown in FIG. 7, the digital waveform generator 100 further includes a state control unit 190, a temporary storage unit 200, a data output unit 210, and an identification unit 220.

The state control unit 190 sequentially encodes a part of the bit of the operation command Command to generate a data selection signal xVW, a frequency selection signal Freq_sel, a waveform indication signal xCos, a waveform indication signal xSin, and a temporary value selection signal Internal_sel. The state control unit 190 is organized by three control circuits such as a single continuous state control circuit, an internal register state control circuit, and a frequency selective state control circuit. Wherein, the single continuous state control circuit encodes the first and second bits of the operation command Command, where Command[1:0] is defined as the first to second bits of the operation command Command, and Command[0] The command [1] represents the first bit and the second bit of the operation command Command, respectively. In addition, when the reset signal Reset is at a logic high level, the data selection signal xVW, the waveform indication signal xCos, and the waveform indication signal xSin are all reset to zero. A more detailed description of the single continuous state control circuit will follow. Table 7A is a truth table for a single continuous state control circuit.

Please refer to FIG. 7 and refer to Table 7A. As shown in Table 7A, when the reset signal Reset is at a logic low level and the clock signal CLK is triggered from the logic low level to the logic high level positive edge, There are three types of output and the relationship between the inputs: the operation command Command[1] is used to represent the data selection signal xVW. The operation command Command[0] is used to represent the waveform indication signal xCos. The operation command Command[0] can be used to represent the waveform indication signal xSin after the inverse operation.

Further, as shown in FIG. 7A, when Command[1]=0, Command[0]=0, it is represented as a single sine value state, and at this time, the digital waveform generator 100 performs a single digital sine wave function F _ST (n). Value operation. When Command[1]=0, Command[0]=1, it is represented as a single cosine value state, and the digital waveform generator 100 performs a single value operation of the digital cosine wave function F _CT (n). When Command[1]=1, Command[0]=0, it is expressed as a continuous sine value state, and the digital waveform generator 100 performs a continuous value operation of the digital sine wave function F _ST (n). When Command[1]=1, Command[0]=1, it is expressed as a continuous cosine value state, and at this time, the digital waveform generator 100 performs a continuous value operation of the digital cosine wave function F _CT (n).

In addition, when the material selection signal xVW is at a logic high level, the continuous angle generating unit 170 and the material selecting unit 180 are activated. The continuous angle generating unit 170 automatically generates a continuous angle data xAngle, and at this time, the data selecting unit 180 reads the continuous angle data xAngle to generate the angle data IndexM and outputs the angle data IndexM to the frequency processing unit 110. In this way, the digital waveform generator 100 can be operated to perform a numerical operation of a continuous prototype digital sine wave function F _ST (n) or a numerical operation of a prototype digital cosine wave function F _CT (n).

When the material selection signal xVW is at a logic low level, the state control unit 190 will turn off the continuous angle generating unit 170 and the material selecting unit 180. At this time, the material selection unit 180 reads the single angle data xDatain from the data input unit 160 to generate the angle data IndexM and outputs the angle data IndexM to the frequency processing unit 110. In this way, the digital waveform generator 100 can be operated to perform a numerical operation of a single prototype digital sine wave function F _ST (n) or a numerical operation of a prototype digital cosine wave function F _CT (n).

In addition, when the waveform indication signal xCos is at a logic high level and the waveform indication signal xSin is at a logic low level, the digital waveform generator 100 only processes the numerical operation of the functional formula F _CT (n) of the prototype digital cosine wave. When the waveform indicating signal xCos is at a logic low level and the waveform indicating signal xSin is at a logic high level, the digital waveform generator 100 only processes the numerical operation of the functional formula F _ST (n) of the prototype digital cosine wave. It should be noted that the waveform indication signal xSin and the waveform indication signal xCos are two signals of exclusive or exclusive, that is, the two signals cannot be logic high level or both logic low level, and the two signals are The indication signal xSC is generated after the mutual exclusion operation.

A further description of the operation of the internal register state control circuit in state control unit 190. The internal register state control circuit encodes the 3rd to 4th bits of the operation command Command to generate a 4-bit temporary value selection signal Internal_sel and outputs it to the temporary storage unit 200. Here, the 3rd and 4th bits of the operation command Command are defined as Command[2] and Command[3], respectively, and Command[3:2] is defined herein as the 3rd to 4th bits of the operation command Command. . Further, when the reset signal Reset is at a logic high level, the temporary value selection signal Internal_sel is reset to zero. The internal register state control circuit will be described in more detail below. Table 7B is a truth table of the register state control circuit. Where b[0]~b[3] represent the first to fourth bits of the temporary value selection signal Intemal_sel, respectively.

Please refer to Table 7B in conjunction with FIG. 7. As shown in Table 7B, when the reset signal Reset is at a logic low level, and the clock signal CLK is triggered from a logic low level to a logic high level positive edge, there are the following The relationship between the four outputs and the inputs:

b[0]=(Command[3])nor(Command[2]).

b[1]=not(Command[3])and(Command[2]).

b[2]=not(Command[2])and(Command[3]).

b[3]=(Command[3])and(Command[2])

As shown in Table 7B, when Command[3]=0 and Command[2]=0, the temporary value selection signal Internal_sel=0001 is indicated. At this time, the register state control circuit starts the temporary storage unit 200, and saves the waveform output value Mag_SC to the internal portion of the temporary storage unit 200 when the logical low level of the clock signal CLK is triggered to the positive edge of the logic high level. Register. When Command[3]=0 and Command[2]=1, it indicates that the temporary value selection signal Internal_sel=0010, the register state control circuit starts the temporary storage unit 200, and cooperates with the clock signal CLK logic low level to When the logic high level is triggered to the positive edge, the phase function value Mag_PhZ is stored in the internal register of the temporary storage unit 200.

When Command[3]=1 and Command[2]=0, it indicates that the temporary value selection signal Internal_sel=0100, the register state control circuit starts the temporary storage unit 200, and cooperates with the clock signal CLK logic low level. When the logic high level is triggered to the positive edge, the phase area angle data PhaseZ is stored in the internal register of the temporary storage unit 200. When Command[3]=1 and Command[2]=1, it indicates that the temporary value selection signal Internal_sel=1000, the register state control circuit starts the temporary storage unit 200, and cooperates with the clock signal CLK logic low level to When the logic high level is triggered to the positive edge, the reference angle data IndexN is stored in the internal register of the temporary storage unit 200.

It can be seen that the temporary storage unit 200 is configured to store the phase function value Mag_PhZ, the waveform output value Mag_SC, the reference angle data IndexN, and the phase angle data PhaseZ. In addition, the temporary storage unit 200 further outputs one of the stored storage phase function value Mag_PhZ, the waveform output value Mag_SC, the reference angle data IndexN, and the phase angle data PhaseZ of the internal temporary register according to the temporary value selection signal Internal_sel. As the internal temporary value Int_reg. Table 7C is a truth table of the temporary storage unit.

As shown in Table 7C, when the reset signal Reset is at a logic high level, the internal temporary storage value Int reg = 00000000, that is, the internal temporary storage value Int reg is reset to zero. When the reset signal Reset is at a logic low level, when the clock signal CLK is triggered from a logic low level to a logic high level positive edge, and the temporary value selection signal Internal sel=0001, the temporary storage unit 200 will The waveform output value Mag_SC in the internal register is output as the internal temporary value Int_reg.

When the reset signal Reset is at a logic low level, the clock signal CLK is triggered from a logic low level to a logic high level positive edge, and the temporary value selection signal Internal sel=0010, then the temporary storage unit 200 will internally The phase function value Mag_PhZ in the register is output as the internal temporary value Int_reg.

When the reset signal Reset is at a logic low level and the clock signal CLK is triggered from a logic low level to a logic high level positive edge, if the temporary value selection signal Internal sel=0100, the temporary storage unit 200 will The phase zone angle data PhaseZ output in the internal register is used as the internal temporary storage value Int_reg.

When the reset signal Reset is at a logic low level, the clock signal CLK is triggered from a logic low level to a logic high level positive edge, and the temporary value selection signal Internal sel=1000, the temporary storage unit 200 will internally The reference angle data IndexN output in the scratchpad is used as the internal temporary storage value Int_reg.

When the reset signal Reset is at a logic low level, the clock signal CLK is triggered from a logic low level to a logic high level positive edge, and if the temporary value selection signal Internal_sel is not the above condition, the temporary storage unit 200 The internal internal storage value Int_reg=00000000 is output.

Therefore, by the internal temporary storage value Int_reg outputted by the temporary storage unit 200, it can be clearly understood whether the phase function value Mag_PhZ, the waveform output value Mag_SC, the reference angle data IndexN, and the phase angle data PhaseZ are correct, thereby detecting Break out where the arithmetic function is wrong.

Next is a description of the operation of the frequency selection state control circuit in the state control unit 190. The frequency selection state control circuit encodes the 5th to 7th bits of the operation command Command to generate an 8-bit frequency selection signal Freq_sel and outputs it to the frequency selection unit 110. For convenience of explanation, in addition, when the reset signal Reset is at a logic high level, the frequency selection signal Freq_sel is reset to zero. Table 7D is a truth table of the frequency selection state control circuit. Wherein b[0]~b[7] sequentially represent the first to eighth bits of the frequency selection signal Freq_sel, and the operation instruction Command(n) is used to represent the (n+1)th bit in the operation instruction Command. n=0~7, and the Command[6:4] table operates the 5th to 7th bits in the Command.

As shown in Table 7D, when Command(7)=1 is maintained and Command[6:4] is changed, when the reset signal Reset is at a logic low level, and the clock signal CLK is from a logic low level to logic When the high level is triggered on the positive edge, there are 8 kinds of output and the coding relationship between the inputs. Where Command(7)=1 indicates that the bit is preset to a logic high level to reserve the future expansion of the instruction bit.

Please refer to FIG. 3 and FIG. 7 and refer to Table 2 and Table 7D. As shown in Table 7D, when Command(7)=1 and Command[6:4]=000, it indicates that the frequency selection state control circuit is in frequency selection. The status of the letter Freq_sel=00000001. At this time, the state control unit 190 outputs the frequency selection signal Freq_se1=00000001 to the frequency processing unit 110. When the clock signal CLK is at a logic low level to a logic high level, the frequency processing unit 110 outputs a DC signal DCout of a logic high level, indicating that the frequency processing unit 110 is currently in a DC level state, and has not performed a periodic digital sine wave and period. The operation of the generation process of the sexual digital cosine wave.

When Command(7)=1 and Command[6:4]=001, it indicates that the frequency selection state control circuit is in the state of Freq_sel=00000010. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=00000010 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 directly outputs the angle data IndexM as the reference angle data IndexN, thereby generating a periodic digital sine wave or periodicity whose frequency is equivalent to the prototype digital wave. Digital cosine wave.

When Command(7)=1 and Command[6:4]=010, it indicates that the frequency selection state control circuit is in the state of Freq_sel=00000100. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=00000100 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 1 bit as the reference angle data IndexN, thereby generating a periodic digital number twice the frequency of the prototype digital wave. Sine wave or periodic digital cosine wave.

When Command(7)=1 and Command[6:4]=011, it indicates that the frequency selection state control circuit is in the state of Freq_sel=00001000. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=00001000 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 2 bits as the reference angle data IndexN, thereby generating a frequency 4 times the periodic digit of the prototype digital wave. Sine wave or periodic digital cosine wave.

When Command(7)=1 and Command[6:4]=100, it indicates that the frequency selection state control circuit is in the state of Freq_sel=00010000. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=00010000 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 3 bits as the reference angle data IndexN, thereby generating a frequency 8 times the periodic digit of the prototype digital wave. Sine wave or periodic digital cosine wave.

When Command(7)=1 and Command[6:4]=101, it indicates that the frequency selection state control circuit is in the state of Freq_sel=00100000. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=00100000 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 4 bits as the reference angle data IndexN, thereby generating a frequency 16 times the periodic digit of the prototype digital wave. Sine wave or periodic digital cosine wave.

When Command(7)=1 and Command[6:4]=110, it indicates that the frequency selection state control circuit is in the state of Freq_sel=01000000. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=01000000 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 5 bits as the reference angle data IndexN, thereby generating a frequency 32 times the periodic digit of the prototype digital wave. Sine wave or periodic digital cosine wave.

When Command(7)=1 and Command[6:4]=111, it indicates that the frequency selection state control circuit is in the state of Freq_sel=10000000. At this time, the state control unit 190 outputs the frequency selection signal Freq_sel=10000000 to the frequency processing unit 110. When the pulse signal CLK is from the logic low level to the logic high level, the frequency processing unit 110 shifts the angle data IndexM to the left by 6 bits as the reference angle data IndexN, thereby generating a frequency 64 times the periodic digit of the prototype digital wave. Sine wave or periodic digital cosine wave.

Referring to FIG. 7, the data output unit 210 determines whether to temporarily store and output the internal temporary storage value Int_reg as the data output SCdata_out according to the data read signal RD and the decoded input address ADDRin. In detail, when the reset signal Reset is at a logic high level, the data output unit 210 is reset to zero. When the reset signal Reset is at a logic low level and the data readout signal RD is at a logic high level, and the clock signal CLK is triggered from a logic low level to a logic high level positive edge, if an 8-bit input bit is present The address number of E6 appears on the address ADDRin. After decoding by the data output unit 210, the data output unit 210 outputs the internal temporary storage value Int_reg as the data output SCdata_out.

On the other hand, the identification unit 220 is formed according to the indication signal xSC generated by the waveform indication signals xCos and xSine, the data selection signal xVW, part of the bit of the operation instruction Command, and the two most significant bits in the reference angle data IndexN. The phase region signal xdela, the input address ADDRin, and the phase region discrimination signal Ph_discri determine whether or not the operation completion signal Done_out is generated to determine whether the output of the periodic digital wave has been completed. Here, Ph_discri[0, 3, 4, 7] of FIG. 7 schematically shows the first, fourth, fifth, and eighth bits of the phase identification signal Ph_discri.

In addition, when the reset signal Reset is at a logic high level, the operation completion signal Done_out=000, that is, the operation completion signal Done_out is reset to 0. In the present embodiment, the operation completion signal Done_out appears in the form of a pulse wave. Table 7E is a truth table of the identification unit 220. Among them, Ph_discri(3) and Ph_discri(7) respectively represent the 4th and 8th bits of the phase identification signal Ph_discri.

As shown in Table 7E, when the reset signal Reset is at a logic low level and the clock signal CLK is triggered by a logic low level to a positive edge of a logic high level, when the data selection signal xVW=1 , it means that the digital waveform generator 100 needs to input the continuous angle data, that is, the angle is automatically scanned from 000000 (0 degrees) to 11111111 (360 degrees) one by one. Therefore, the first, fourth, third, and eighth bits of the phase identification signal Phdiscri (ie, Ph_discri(0), Ph_discri(3), Ph_discri(4), Ph_discri(7)) and the phase-area signal xdela, And when the angle data IndexM=11111111, the representative waveform digit generator 100 has been completed, so the identification unit 220 outputs a completion signal Done_out indicating that the prototype digital sine wave function F _sT (n) or the continuous angle data xAngle input is executed. The operation of the prototype digital cosine wave function F _CT (n). In addition, the completion signal Done_out (here, a pulse signal) can be used to notify the processor such that the processor obtains the prototype digital sine wave function F _ZT (n) or the prototype digital cosine wave function F _CT (n) has been calculated. News.

When the data selection signal xVW=0, it indicates that the digital waveform generator 100 is to input the single angle data xDatain, that is, the single angle data xDatain is automatically input from the data input unit 160 to the data selection unit 180. At this time, the identification unit 220 cooperates with the operation command Command[7], the input address ADDRin, the indication signal xSC, the internal counter Count_out.., etc., and outputs the completion signal Done_out after the internal counter Count_out is completed to count the number of digits. The device 100 has been operational. Therefore, by completing the signal Done_out, it can be known whether the output of the periodic digital wave is completed.

Viewed from another perspective, FIG. 8 is a flow chart of a method of generating a digital waveform in accordance with an embodiment of the present invention. Referring to FIG. 8 , firstly, based on a prototype digital wave, the value of the angle data is limited or amplified by 2^P times according to a frequency selection signal, and a reference angle data is generated, and P is an integer and P ≧ 0 (step S801). Then, a phase discrimination signal is generated according to the R most significant bits and the S waveform indication signals in the reference angle data, to point to one of (2^R)*S phase regions extended by the prototype digital wave. For a specific phase region, R and S are positive integers (step S802).

Then, according to the phase region discrimination signal, the shift relationship corresponding to the specific phase region is selected from the (2^R)*S shift relationship to shift the value of the reference angle data to the first phase region of the prototype digital wave. And generating a phase region angle data accordingly (step S803). Then, a plurality of functional expressions for synthesizing the prototype digital wave located in the first phase region are provided, and the phase angle data is brought into one of the functional expressions to calculate the phase angle data. Corresponding one phase function value (step S804).

Finally, depending on the phase region discrimination signal, the inverse of the phase function value is determined or the phase function value is directly provided as a waveform output value, and the waveform output value is used to generate a frequency of 1/(2^M) times the prototype digital wave or It is a periodic digital wave of 2^M times (step S805). The detailed flow of the method for generating a digital waveform described in this embodiment is included in the above embodiments, and thus will not be described herein.

In summary, the present invention uses the phase-to-area conversion method, which not only saves the use of the positive/cosine function comparison table, but also uses only eight formulas to generate a complete periodic sine/cosine digital wave, so that it can be improved. The speed of the circuit is reduced and the capacity of the memory is reduced. In addition, the digital waveform generator of the present invention adopts an address decoding method, which can provide a user to modify a micro code on an address parameter register of a circuit source code to have different micro control codes. Digital waveform generator. Therefore, except for the designer, it is difficult for others to crack this specific address code, so that the function of confidentiality and counterfeiting can be achieved.

In addition, the digital waveform generator of the present invention can be applied to navigation, calendar, astronomy, geography, aviation, architecture, engineering, land surveying and the like. The generated periodic digital positive and residual waveforms can also be used in a Communications Systems wafer product for modulation and demodulation.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

100. . . Digital waveform generator

110. . . Frequency processing unit

120. . . Phase identification unit

130. . . Phase conversion unit

140. . . Function value calculation unit

142. . . Operation control unit

144a. . . First value adjustment unit

144b. . . Second value adjustment unit

144c. . . Third value adjustment unit

146a. . . First multiplex unit

146b. . . Second multiplex unit

148. . . Addition unit

150. . . Polarity conversion unit

160. . . Data input unit

170. . . Continuous angle generating unit

180. . . Data selection unit

190. . . State control unit

200. . . Staging unit

210. . . Data output unit

220. . . Identification unit

Ph1~Ph8. . . Phase zone

T, T1, T2. . . cycle

S. . . Sampling point

Freq_sel: frequency selection signal

IndexM: Angle data

IndexN: reference angle data

DCout: DC signal

Xdela: phase zone signal

xSin, xCos: waveform indication signal

Ph_discri, Ph_discri[0,3,4,7]: phase zone discrimination signal

PhaseZ. . . Phase angle data

xSC. . . Indication signal

Mag_PhZ. . . Phase function value

Mag_SC. . . Waveform output value

Reset. . . Reset signal

CLK. . . Clock signal

N70. . . First constant data

N100. . . Second constant data

N24. . . Third constant data

N26. . . Fourth constant data

N27. . . Fifth constant data

DATAin. . . Digital data

ADDRin. . . Input address

WR. . . Data write signal

xDatain. . . Single angle data

Command, Command(7), Command[1:0], Command[3:2], Command[6:4]. . . Operational instruction

xVW. . . Data selection signal

xAngle. . . Continuous angle data

Internal_sel. . . Temporary value selection signal

RD. . . Data readout signal

Int_reg. . . Internal temporary value

SCdata_out. . . Data output

Done_out. . . Operation completion signal

Angle_discri(n). . . Operational control signal

S801~S805. . . step

FIG. 1A is a waveform diagram showing a sine wave function of a single period.

FIG. 1B is a waveform diagram of a cosine wave function of a single period.

2A is a timing diagram of the sine wave function F(t) of the single cycle of FIG. 1A versus time t.

2B is a plot of a single-cycle digital sine wave function F1(n) versus a digital angle input variable n.

2C is a plot of another single cycle digital sine wave function F2(m) versus a digital angle input variable m.

3 is a circuit block diagram of a digital waveform generator in accordance with an embodiment of the present invention.

Figure 4 shows a trigonometric function on a unit circle.

FIG. 5 is an internal circuit block diagram of the function value operation unit 140 of FIG.

FIG. 6 is a block diagram of a peripheral circuit of one of the digital waveform generators of the embodiment.

FIG. 7 is a block diagram of another peripheral circuit of the digital waveform generator of the embodiment.

FIG. 8 is a flow chart of a method for generating a digital waveform according to an embodiment of the invention.

100. . . Digital waveform generator

110. . . Frequency processing unit

120. . . Phase identification unit

130. . . Phase conversion unit

140. . . Function value calculation unit

150. . . Polarity conversion unit

Freq_sel. . . Frequency selection signal

IndexM. . . Angle data

IndexN. . . Reference angle data

DCout. . . DC signal

Xdela. . . Phase signal

xSin, xCos. . . Waveform indication signal

Ph_discri. . . Phase zone discrimination signal

PhaseZ. . . Phase angle data

xSC. . . Indication signal

Mag_PhZ. . . Phase function value

Mag_SC. . . Waveform output value

Reset. . . Reset signal

CLK. . . Clock signal

Claims (21)

A digital waveform generator includes: a frequency processing unit that limits or enlarges the value of an angle data by 2^P times according to a frequency selection signal based on a prototype digital wave, and generates a reference angle accordingly Data, P is an integer and P≧0; a phase-area identification unit generates a phase-region discrimination signal according to the R most significant bits and the S waveform indication signals in the reference angle data to point to the prototype digital wave Extending out a specific phase region of (2^R)*S phase regions, R and S are positive integers; a phase-to-phase conversion unit discriminates signals from (2^R)*S according to the phase region discrimination signal Selecting a shift relationship corresponding to a specific phase region to shift the value of the reference angle data to the first phase region of the prototype digit wave, and thereby generating a phase region angle data; a function value calculation unit Providing a plurality of function expressions for synthesizing the prototype digit wave located in the first phase region, and bringing the phase region angle data to one of the function expressions to calculate the phase region a phase function value corresponding to the angle data; and a polarity The switching unit determines whether to output the opposite value of the phase function value according to the phase region discrimination signal or directly outputs the phase function value as a waveform output value, and uses the waveform output value to generate a frequency of 1/(the prototype digital wave) 2^M) times or 2^M times a periodic digital wave. The digital waveform generator of claim 1, wherein the frequency processing unit uses the angle data to be shifted right or left by P bits, so that the value of the angle data is limited or enlarged. P times. The digital waveform generator of claim 1, wherein the frequency processing unit further outputs a direct current signal according to the frequency selection signal to drive a circuit external to the digital waveform generator. The digital waveform generator according to claim 1, wherein when the period of the prototype digital wave is 2^N sampling points, n is used to represent the value of the reference angle data, and N is a positive integer, K is In the case of a positive number, the function of the prototype digital wave is: The digital waveform generator of claim 4, wherein when R and S are respectively equal to 2, the phase identification signal is used to identify 8 phase regions extended by the prototype digital wave, and the prototype digit The first to fourth phase regions of the wave are used to synthesize a sinusoidal digital wave, and the fifth to eighth phase regions of the prototype digital wave are used to synthesize a cosine digital wave. The digital waveform generator of claim 5, wherein when N is equal to 8, K is equal to 128, and z is used to represent the value of the phase angle data, 8 are stored in the phase conversion unit. The shift relationship corresponds to the eight phase regions, and the eight shift relations are as follows: z=n, where n=0~63; z=128-n, where n=64~127; z=n- 128, where n=128~191; z=256-n, where n=192~255; z=64-n, where n=0~63; z=n-64, where n=64~127; z= 192-n, where n=128~191; z=n-192, where n=192~255. The digital waveform generator according to claim 6, wherein the function expressions provided by the function value calculation unit are respectively as follows: F _MP (z)=2*z+z, wherein z=0 ~3, 20~23; F _MP (z)=(2*z+1)+z, where z=4~19; F _MP (z)=2*z+24, where z=24~27;F _MP (z)=2*z+26, where z=28~31, 40~43; F _MP (z)=2*z+27, where z=32~39; F _MP (z)=z+70 , where z=44~55; F _MP (z)=126, where z=56~59; and F _MP (z)=127, where z=60~64. The digital waveform generator of claim 7, wherein the function value calculation unit comprises: an operation control unit that encodes a part of the bit data of the phase region with reference to the function expressions to generate An operation control signal; a first value adjustment unit multiplying the value of the phase angle data to generate a first adjustment data; and a second value adjustment unit multiplying the value of the phase angle data by 1 to generate a second adjustment data; a third value adjustment unit receives the phase region angle data, and outputs the data as a third adjustment data; a first multiplex unit, according to the operation control signal, the first adjustment data, The second adjustment data, a first constant data, and a second constant Selecting one of the data to generate a first function adjustment data, wherein the first constant data has a value of 70, and the second constant data has a value of 100; a second multiplex unit according to the operation control signal Selecting one of the third adjustment data, a third constant data, a fourth constant data, and a fifth constant data to generate a second function adjustment data, wherein the third constant data has a value of 24, the fourth The value of the constant data is 26, and the value of the fifth constant data is 27; and an adding unit for adding the first function adjustment data and the second function adjustment data to generate the phase function value. The digital waveform generator of claim 1, further comprising: a data input unit that decodes an input address to treat a digital data as a single angle data according to the decoded input address; Or a operation command; a continuous angle generating unit receiving a data selection signal to determine whether to generate a continuous angle data; and a data selection unit for reading the single angle data or the continuous angle data according to the data selection signal, To generate this angle data. The digital waveform generator of claim 9, further comprising: a state control unit, sequentially encoding a part of the bit of the operation instruction to generate the data selection signal, the waveform indication signals, The frequency selection signal and a temporary value selection signal; a temporary storage unit for storing the phase function value, the waveform output a value, the reference angle data, and the phase angle data, and according to the temporary value selection signal, an internal temporary storage value is output; and a data output unit, according to a data readout signal and the decoded input address And decide whether to temporarily store and output the internal temporary value. The digital waveform generator of claim 9, further comprising: an identification unit, according to the waveform indication signal, the data selection signal, a partial bit of the operation instruction, and R in the reference angle data The most significant bit, the input address, and the phase identification signal determine whether an operation completion signal is generated to determine whether the output of the periodic digital wave has been completed. A method for generating a digital waveform includes: limiting a value of an angle data by a frequency selection signal according to a prototype digital wave by 2^P times, and generating a reference angle data, P is An integer and P≧0; generating a phase region discrimination signal according to the R most significant bits in the reference angle data and the S waveform indication signals to point to (2^R)*S extended by the prototype digital wave a specific phase region in each phase region, R and S are positive integers; according to the phase region discrimination signal, a shift relationship corresponding to a specific phase region is selected from (2^R)*S shift relations to The value of the reference angle data is shifted to the first phase region of the prototype digital wave, and accordingly, a phase angle data is generated; and the prototype digital wave for synthesizing the position in the first phase region is provided. a function expression, and the phase region angle data is brought to one of the function expressions to calculate a phase function value corresponding to the phase region angle data; and the decision is provided according to the phase region discrimination signal The inverse of the phase function value directly provides the phase function value as a waveform output value, and uses the waveform output value to generate a frequency of 1/(2^M) times or 2^M times the prototype digital wave. A periodic digital wave. The method for generating a digital waveform according to claim 12, wherein the step of limiting or enlarging the value of the angle data by 2^P times comprises shifting the angle data to the right or left to the P bit. The method for generating a digital waveform according to claim 12, wherein when the period of the prototype digital wave is 2^N sampling points, n is used to represent the value of the reference angle data, and N is a positive integer, K When it is a positive number, the function of the prototype digit wave is: The method for generating a digital waveform according to claim 14, wherein when R and S are respectively equal to 2, the phase identification signal is used to identify 8 phase regions extended by the prototype digital wave, and the prototype The first to fourth phase regions of the digital wave are used to synthesize a sinusoidal digital wave, and the fifth to eighth phase regions of the prototype digital wave are used to synthesize a cosine digital wave. The method for generating a digital waveform according to claim 15, wherein when N is equal to 8, K is equal to 128, and z is used to represent the value of the phase angle data, 8 shifting relations corresponding to 8 phase regions are used. The formulas are as follows: z=n, where n=0~63; z=128-n, where n=64~127; z=n-128, where n=128~191; z=256-n, where n=192~255; z=64-n, where n=0~63; z=n-64, where n=64~127; z=192-n, where n=128~191; and z=n-192, where n=192 ~255. The method for generating a digital waveform according to claim 16 of the patent application, wherein the function expressions are as follows: F _MP (z)=2*z+z, wherein z=0~3, 20~23; F _MP (z)=(2*z+1)+z, where z=4~19; F _MP (z)=2*z+24, where z=24~27; F _MP (z)=2* z+26, where z=28~31, 40~43; F _MP (z)=2*z+27, where z=32~39; F _MP (z)=z+70, where z=44~55 ; F _MP (z) = 126, where z = 56 to 59; and F _MP (z) = 127, where z = 60 to 64. The method for generating a digital waveform according to claim 17, wherein the step of calculating the phase function value corresponding to the phase angle data of the phase region comprises: referring to the function of the phase angle data of the phase region Bits are encoded to generate an operational control signal; multiplying the value of the phase region angle data to generate a first adjustment Multiplying the value of the angle data of the phase region and adding 1 to generate a second adjustment data; according to the phase region angle data, and transmitting the phase region angle data as a third adjustment data; The signal selects one of the first adjustment data, the second adjustment data, a first constant data, and a second constant data to generate a first function adjustment data, wherein the first constant data has a value of 70, The value of the second constant data is 100; selecting one of the third adjustment data, a third constant data, a fourth constant data, and a fifth constant data according to the operation control signal to generate a second function Adjusting the data, wherein the value of the third constant data is 24, the value of the fourth constant data is 26, the value of the fifth constant data is 27; and the first function adjustment data and the second function adjustment data are performed Addition to generate the phase function value. The method for generating a digital waveform according to claim 12, further comprising: decoding an input address to treat a digital data as a single angle data or an operation according to the decoded input address; An instruction; determining whether to generate a continuous angle data according to a data selection signal; and reading the single angle data or the continuous angle data according to the data selection signal to generate the angle data. The method for generating a digital waveform according to claim 19, further comprising: sequentially encoding a part of the bit of the operation instruction to generate the data selection signal, the waveform indication signal, and the frequency selection signal. And a temporary value selection signal; storing the phase function value, the waveform output value, the reference angle data, and the phase angle data, and selecting a signal according to the temporary value to provide an internal temporary storage value; The data read signal and the decoded input address determine whether to temporarily store and transmit the internal temporary value. The method for generating a digital waveform according to claim 19, further comprising: determining, according to the waveform indication signal, the data selection signal, a part of the bit of the operation instruction, and the R most effective in the reference angle data. The bit, the input address, and the phase identification signal determine whether an operation completion signal is generated to determine whether the generation of the periodic digital wave has been completed.