JPS58121802A - Digital fm demodulator - Google Patents

Abstract

PURPOSE:To improve the linearity of FM detection, to decrease the distortion, and to attain the reproduction of signals with high fidelity, by A-D converting FM analog signals directly, operating them at an operation circuit, D-A converting and demodulating the FM signals. CONSTITUTION:The FM signals applied to an input terminal 1 is directly converted at an A-D converter 2 and digitally operated to attain FM demodulation. A figure shows the block diagram for the operation and sn+1, sn and sn-1 are obtained from shift registers 6, 7. The sn+1, sn-1 are summed at an adder 8, the output is divided with the sn at a divider 9, and the result is given to an operating device 10 for the operation given in equation. Since the value DELTAT is known, this operation can be performed by a microcomputer or the results of computation for all inputs are writen in an ROM and the data are read out with the required address designation. DELTAf is obtained by subtracting f0 from the output at a subtractor 11 and the DELTAf becomes a digital audio signal.

Description

[Detailed Description of the Invention] The present invention directly transmits FM signals sent as analog signals to h
An object of the present invention is to provide a digital FM demodulator that performs D/M conversion, performs calculation processing, and demodulates an FM signal.

With recent advances in IC technology, various types of signal processing have come to be digitally processed. In particular, high-speed A/D converters are used to process video signals for television, and digital processing is performed, bringing various advantages. However, TVs contain various signals other than video signals, and in particular, in the case of signals sent by FM, for example, audio signals, subchannels of audio multiplexing, color signals of SECAM system, etc. After demodulating in the area of

It is customary to perform digital processing in addition to an A/D converter, which is extremely inconvenient and rare.

The present invention aims to provide a digital FM demodulator that digitally detects the FM by digitally converting the FM as it is.Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a digital FM demodulator according to an embodiment of the present invention. M'T
There are many variations in the number sozg'gD2//'-ta2.

The converted signal is applied to a digital FM detection circuit 3 for detection, applied to an audio processing circuit 4, and then converted by a D/MU converter to obtain an audio signal.

FIG. 3 is a waveform diagram for explaining the digital FM detection circuit 3, where the FM input S=mu5in2rr (fo +
Δf) t and the state of sibling are shown. The sibling interval is ΔT, and the input is S=M5In2π(fo
+Δf)t, and Sn7 i of the sample sequence, *s
It is sufficient if n and sn+1 can be determined and Δf to be demodulated of the FM modulated wave can be determined from these values. A theoretical explanation is given below.

However, let T be the time when s sn is sampled.

From the above, A is deleted.

Decompose equation (2) to find T5.

8 stones 2π(,fo+Δf)(T-ΔT)-ad+2π(
fo+Δf)T=0−×癲〔2π(fo+Δf)T+
ψ) =O...mark-------=-(41(4)) The condition for satisfying (4) is the following equation.

2π(fa +Δf)? +ψ= 2 N rrHowever, U
=.

Substituting equation (6) into equation (3) yields.

Breaking this down and organizing it The following equation is obtained from equation (6) and equation (1).

When you organize Therefore From the above, if we perform the calculation of equation (8) digitally.

FM demodulation is possible.

The calculation of equation (8) can be easily performed using a microcomputer, but it is also possible with hardware.0 Figure 4 is a block diagram for this calculation, and shows Sn+1+sn, 5n-
1 can be obtained in shift register 6.7 o Sn
+4 and 81-1 are added by adder 8, the output is divided by Sn by divider 9, and the result is a known value, so this operation can be performed by a microcomputer, but all It is also possible to write the calculation result in the case of the input into the ROM and read it out by specifying the necessary address.

By subtracting fo from the output using the subtracter 11, Δf is obtained, which becomes a digital audio signal.

When used for television, the composite video signal, which also includes the audio FM signal, is directly subjected to MU/D conversion, and then digitally filtered to convert it into a video signal, a color signal, and an audio signal.

Since it is possible to separate the synchronization signal and process it digitally, it is possible to use only one MU/D converter as shown in Figure 4.

Next, the accuracy of demodulated output will be explained. Since equation (8) is a subtraction, the number of bits of Δf will be small in the case of modulation with a small frequency displacement. Now 4.5 M of audio
Let us consider the case of detecting Ilz. Maximum frequency is ±2
Since the frequency is 6 kHz, if the first term of equation 8 is expressed with 8 bits using T of 4.5M2, only the lower 2 bits will change. In other words, instantaneously, Δf has an accuracy of only about 2 bits. Next, a method of increasing the number of bits by 0 will be explained. (Sn-z+Sn-+)/2Sn is obtained from three sample points.

If this is taken N times and averaged, the accuracy should improve. , Now suppose that the audio signal is extracted at a sampling frequency of 35KHz, and the A/D sampling frequency is 14.4M.
Hz, the number of sample points that can be used to calculate one piece of audio data is approximately 400, and the accuracy is improved by approximately 9 bits. If it was originally 2 bits, the total would be 11 bits, resulting in a significant improvement in accuracy. An embodiment that realizes the above is shown in FIG. In this case, the following calculation is performed to minimize the number of bits on the hardware when averaging. In FIG. 5, 12 is a subtracter, 13 is an averaging circuit, 1
4 is a computing unit.

As described above, according to the present invention, including the averaging operation, FM detection is performed by digital calculation, so the linearity of FM detection is good, there is little distortion, and high-fidelity signal reproduction is possible. Furthermore, an SIF limiter is not required. Furthermore, it can be made smaller by using IC technology. In addition, there are no adjustment points, allowing for rationalization of production and no changes over time.

Furthermore, if used for television, it can be used in common with an A/D converter for video without using a dedicated A/D converter for audio, simplifying one circuit.

[Brief explanation of the drawing]

FIG. 1 is a basic block diagram of a digital FM demodulator of the present invention, FIG. 2 is a waveform diagram for explaining the device, and FIG. 3 is a block diagram of a digital FM demodulator in an embodiment of the present invention. 4 is a block diagram of an example of use in the same place, and FIG. 6 is a block diagram of a digital FM demodulator in an embodiment of the same place. 7
...Shift register, 8...Adder,
9...Divider, 10.14...Arithmetic unit, 11.12...Subtractor, 13...Averaging circuit. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure 3 Figure 4

Claims (2)

[Claims] (1) A digital FM demodulator characterized in that the FM signal, which has become a digital signal, is added to an arithmetic circuit that calculates the following one or two equations, calculates it, and demodulates it. Sn++, Sn, Sn-+: Data sequence of FM signal sampled at sampling interval ΔT fo: Center frequency of FM signal (2) The FM signal that has become a digital signal is obtained by A/D converting a composite video signal including an audio FM signal and adding the converted output to a digital filter. The digital FM demodulator according to item 1.