JPS62142402A - frequency modulator - Google Patents

Abstract

PURPOSE:To contrive to compact the scale of the circuit by providing an integra tor and a phase amplitude converter that outputs sine values or cosine values making output of the integrator input and lengthen a word length of the integra tor longer by at least more than 1 bit than an input word length of phase angle of the phase amplitude converter in lower direction. CONSTITUTION:A constant C (product of carrier angle frequency omegaC and sam pling period T) is added by an adder 2 to digital modulated signals inputted to a terminal 1, and the result of addition is inputted to an integrator 5 consisting of an adder 3 and a delay unit 4. Phase angle theta is calculated by the integrator 5, and the output of the integrator 5 is inputted to an amplitude phase converter 6. here, the word length of the integrator 5 is made N bits and the word length of the phase amplitude converter 6 is made N-1 bits made by reducing the lowermost bit of the integrator 5 by 1 bit. Thus, when FM modulating and demodulating inputted modulated signals expressed by proper number of bits at gradation of N bits, if frequency of a modulation signal is low, the amplitude of signals of unnecessary frequency is small, and the specific frequency component is attenuated by a low-pass filter, and accord ingly, the gradation of N bits is obtained in demodulating side.

Description

[Detailed description of the invention] Industrial applications The present invention relates to a frequency modulation device.

BACKGROUND OF THE INVENTION An example of a conventional frequency modulation device (FM modulation device) is shown below.

Generally, the modulation signal is V(t), the center angular frequency of the carrier wave is 2,
,.

ω. Then, the FM signal f (t) is f (t)
=Mu sin (ωOt +m /V(t)dt+θ0)
(m, θ0 are constants) When the above is realized by calculating digital signals, the sampling period is T and the sample value of the modulation signal is V.
(nT), the FM signal can be written as f(nT) (n≧2).

Here, let us consider a case where the phase component of the FM signal f(nT) is θ(nT), and θ(nT) is quantized with N bits. θ(nT) varies from 0 to 2π by N-bit quantization.
Up to 2) is divided into 11 parts, and the quantized Stella becomes θ(
nT) with a precision exceeding N bits, the lower bits exceeding N bits are discarded, so
If the word length of the phase input of the phase-amplitude converter is N bits, it is sufficient if the word length of the integrator for calculating θ(nT) is also N bits. In other words, in the conventional configuration, the integrator word length and the phase input word length of the phase/amplitude converter are made equal.

Problems to be Solved by the Invention However, the above configuration method has the following problems.

When attempting to modulate a modulated signal with an appropriate word length with N-bit gradations, the phase inputs of the integrator and phase/amplitude converter both require a word length of N bits. Here, the phase/amplitude converter is often constructed using a ROM or the like, but in that case, the circuit scale becomes larger in proportion to 2N rather than N. Therefore, when N is large, for example 8 bits or 10 bits, the circuit scale of the phase/amplitude converter becomes much larger than that of the integrator, and there are many problems even when attempting to further increase the gradation level.

Means for Solving the Problems In order to solve the above problems, the present invention provides a first method of adding a constant C to a digital modulation signal sampled at a sampling period T.
an integrator that has a second adder and a delay device inside and performs integration using the output of the first adder as input; and a sine value or and a phase/amplitude converter that outputs a cosine value, and the word length of the integrator is made longer by at least one bit in the lower direction than the input word length of the phase angle of the phase/amplitude converter.

Effect of the Invention With the above-described configuration, the present invention eliminates the possibility that when an input modulation signal with an appropriate number of pitches is FM modulated and demodulated with N-bit gradation, unnecessary frequency components other than the modulation signal are generated and subjected to modulation. , when the frequency of the modulation signal is low, the amplitude of the above-mentioned unnecessary frequency signal is small, and the specific frequency component is attenuated by the low-frequency p-wave generator on the demodulation side, so the N-bit gradation is can get.

Example Embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 shows an example in which the present invention is implemented in an FM modulation device, in which 1 is an input terminal, 2.3 is an adder, and 4 is an input terminal.
is a delay device, 6 is an integrator consisting of an adder 3 and a delay device 6/4, and 6 is an integrator with the phase angle θ as input.
A phase amplitude converter outputs inθ, and 7 is a D/A converter. FIG. 2 shows an FM demodulator, in which 8 is an input terminal, 9 is an A/D converter, 1o is a 9C3 phase shifter that receives sin θ as input and outputs sin (θ+9o), and 11 is an amplitude Phase converter, 12° 13 is a subtracter, 14
16 is a delay device, 16 is a low-frequency p-wave device, and 16 is a D/A converter.

(Frequency modulation) The digital modulation signal input to terminal 1 is added with a constant C (product of carrier angular frequency ωC and sampling period) in adder 2, and the addition result is composed of adder 3 and delay device 4. is input to the integrator 6. The phase angle θ is calculated by the integrator 5,
The output of the integrator 6 is input to an amplitude phase converter 6. Here, the word length of the integrator 6 is N bits, and the word length of the phase/amplitude converter 6 is N bits, which is obtained by reducing the least significant bit of the integrator 6 by 1 bit.
-1 bit. In this case, if the frequency of the input modulation signal is low, the change in the least significant bit of the output of the integrator 5 is twice the sampling period as the least significant bit of the phase angle input of the phase amplitude converter 6 Bates 6. Hail to you.

Therefore, modulation is performed in such a way that the signal N(nT) having a frequency of one half of the sampling frequency is superimposed on the modulation signal V(nT), and the output signal f of the phase amplitude converter 6 is
(nT) is expressed as follows.

The FM signal output obtained from the series of operations described in section 2 is converted into an analog signal by the D/A converter 7, but at this time, signals with a frequency exceeding half of the sampling frequency are folded back within the band. In the case of this embodiment, the frequencies of the folding from above and the folding from the bottom match, and if the FM signal frequency is fa and the sampling frequency is f8, the folding frequency f is B f+ = f.

is given by

(Frequency demodulation) The analog FM signal is demodulated by the following method after passing through an appropriate transmission system. The analog signal inputted from the terminal 8 is converted into a digital signal by the A/D converter 9 and inputted to the 9d phase shifter 1o. The 9♂ phase shifter 10 outputs a signal in which the phase of the input signal is shifted by 9♂, and amplitude phase conversion is performed based on the sign, amplitude, and magnitude relationship between the two signals, the input signal and the 90'' phase shifter output signal. The phase information is outputted by the device 11.

A subtracter 12 takes the difference between the output of the amplitude phase converter 11 and the output of the delay device 14, and a subtracter 13 removes the carrier frequency component from the subtraction result, thereby demodulating the modulated signal. The output of the amplitude-phase converter 11 is stored in the delay unit 14 for calculation after one sampling period. Here, the folded frequency component described above is also demodulated at the same time, and the frequency component of frequency f2-111-fol is superimposed on the modulated signal. When the frequency of the modulation signal is low, f2 changes depending on the carrier frequency fa, but since this unnecessary frequency component has a small amplitude and a specific frequency component is attenuated by the low-frequency r wave generator 15, D/ If the word length of the A converter 16 is selected to be N bits or more, a modulated signal of N pit gradations from the demodulated signal is converted into an analog signal.

As described above, according to this embodiment, when FM modulating and demodulating a relatively low frequency digital modulation signal with N-bit gradation, the phase angle input word length of the phase amplitude converter in FM modulation is determined from the integrator word length. Even if the number of bits is reduced by one bit to N-1 bits, N-bit gradation can be obtained on the demodulation side.

Note that the phase amplitude converter 6 outputs a sine value, but since the sine function is a periodic function, the sine value for one period of the phase angle may be stored in the ROM and read out when necessary. . In addition, the amplitude phase converter 11 calculates the sine value from the FM signal and the signal obtained by shifting the FM signal by 90', stores the phase angle for the tangent value for one-quarter period in the ROM, reads out the sine value, and calculates the sine value from the FM signal and the signal obtained by shifting the FM signal by 90'. The value of the phase angle for one cycle can be found from the magnitude relationship between the signs and amplitudes of the two signals.

Next, an example of 9 pages of actual FM modulation and demodulation using the present invention is shown below. Here, if the input modulation signal is a direct current and the carrier frequency is expressed by an appropriate word length, the gradation of the input signal to the sine value generator can be determined by the word length of the carrier frequency. Therefore, the word length of the carrier frequency is N bits, the word length of the integrator 5 is N bits, and the phase angle input word length of the phase amplitude converter 6 is N−.
FM modulation and demodulation was performed as 1 bit, and the gradation of the demodulated signal was evaluated.

However, the above sine value generator is the adder 3 in FIG.
．． An integrator 6 and a phase/amplitude converter 6 composed of a delay device 4
This refers to equipment including.

FIG. 3 shows a waveform passed through a p-wave modulator after demodulation, in which a frequency component generated by the above modulator configuration is superimposed on a DC signal (modulation signal). 4 and 5 are diagrams showing the amplitude of unnecessary frequency components when the frequency of the modulation signal is changed, that is, the size of a in FIG. 3. The figure shows the case where l=9. In both figures, the horizontal axis is the frequency, and one period (0 to 2π) is expressed as N-
It is quantized by 1 bit, and the vertical axis is the unnecessary frequency acid 1o.
With amplitude equal to be I', flN-+ has a resolution of N-1 bits, 1l
) I is the resolution of N bits. In both FIG. 4 and FIG. 5, the amplitude of the unnecessary frequency component superimposed on the modulated signal is smaller than the resolution of N bits.

In other words, it can be confirmed that when an N-bit signal is input to the sine value generator, an N-bit gradation can be obtained on the demodulation side even if the phase input word length of the phase-amplitude converter is N-1 bits. .

Note that in both Figures 4 and 5, the amplitude differs depending on the frequency, and in Figure 4 in particular, N-bit gradation is not obtained at low frequencies, but this is due to the 9 bits used in the experiment above.
It is thought that this is caused by the frequency characteristics of the 0 phase shifter and the low-pass p-wave shifter. The transfer functions of the 9d phase shifter and the low-pass p-wave shifter are as follows.

9o phase shifter H(z)=(1/12g)(3(1-Z)+a(Z
2-z") + 21 (z'-Z") + 79 (Z' low range p
Wave device 117.

H(z)=(1/266)(-2(1-Z)+7(
Z-Z)-20(Z-Z)-1-79(Z
-Z)+128Z) As described in detail, by using the frequency modulation device according to the present invention, the input of the phase-to-amplitude converter can be configured with a word length smaller than that of the conventional one, so that the circuit of the phase-to-amplitude converter can be The scale will be significantly smaller. In addition, the phase-amplitude converter is the same as the conventional one.
However, if the frequency of the modulation signal is low, the gradation can be increased simply by increasing the word length of the integrator. In other words, it is not effective for parts with small changes, but for relatively flat parts, adding a simple circuit that increases the word length of the integrator by 1 or more bits in the lower direction can increase the level of change. You can raise the tone. - Furthermore, if the input word length of the phase-amplitude converter can be reduced, it is expected that the word length of subsequent outputs of the phase-amplitude converter, D/A converter, etc., can also be reduced.

[Brief explanation of drawings]

FIG. 1 shows a diagram of an embodiment of a frequency modulation device according to the present invention.
Figure 2 is a block diagram of the frequency demodulator, Figure 3 is a demodulated waveform diagram when modulation and demodulation is actually performed using the present invention, and Figures 4 and 6 are diagrams showing the frequency of the modulated signal. FIG. 4 is a waveform diagram showing the amplitude of unnecessary frequency components in the demodulated waveform when the amplitude is changed. 3...Adder, 4...Delay unit, 5...
...Integrator, 6...Phase amplitude converter.

Claims (1)

[Claims] A first adder that adds a constant C to a digital modulation signal sampled at a sampling period T, and a second adder and a delay device inside, and inputs the output of the first adder. an integrator that performs integration, and a phase-amplitude converter that receives the integrator output as input and outputs a sine value or a cosine value,
A frequency modulation device characterized in that the word length of the integrator is made longer by at least 1 bit in the lower direction than the input word length of the phase angle of the phase amplitude converter.