JPH02207639A - Demodulation circuit system - Google Patents

Abstract

PURPOSE:To realize a demodulation circuit system with a digital electronic circuit by calculating peculiar change quantity by graphically handling a wave to be modulated such as a frequency modulated wave, etc., or calculating the modulation wave from relation between the change quantity and modulation wave by selecting and extracting a numerical value. CONSTITUTION:The wave to be modulated, for which an unnecessary frequency component is removed with a band pass filter BPF, is adjusted to a suitable signal level with an automatic gain control amplifier AGC and digitized with an analog/digital converter AD. For the output of the AD, an Hm total value is calculated at every half cycle, namely, during a cycle that an (m) is 1-M, with an accumulation adder ALU and defined as an Sn value. In the ALU, the output value of the AD is simultaneously controlled. The Sn value output of the ALU is sent to a computing element CPU together with the M value output of an M value counter m.CNT and the N value output of an N value counter n.CNT. In the CPU, the modulation wave, etc., is calculated and restored by information processing while being tentatively stored at every suitable quantity of N couples. Thus, the demodulation circuit system can be realized by the digital electronic circuit.

Description

DETAILED DESCRIPTION OF THE INVENTION In the conventional method 1, the method of detecting the amount of change in the energy of the modulated wave using an analog electronic circuit is inherited.

Therefore, it consists of a 9'-inch electronic circuit.It is inconvenient to integrate it with information processing equipment.

The present invention is a well-known technique. By focusing on the fact that the modulated wave can be separated and restored from the modulated wave using a waveform analyzer, and by improving the 2-figure silent approximation analysis method, which is one of the waveform analysis methods, This is a demodulation circuit system that performs [actual].

Although it is possible that the present invention can be implemented in an internal analogue circuit, the main purpose of the present invention is to connect with information processing equipment, so the following functions can be provided as an intelligent function. This will be explained as a demodulation circuit system implemented using a digital electronic circuit.

b) In Figures 3 to 5, modulation was performed every half cycle in synchronization with the o11 phase. Various modulated waves are shown as Del-like waveforms with feature emphasis.

A) Figure 3 shows FM wave-e island. In one transmission wave with a constant amplitude value with the maximum value gm, the instantaneous frequency value changes corresponding to the amplitude value of the modulated wave, or the cheetah value. Therefore, in terms of two figures, the waveform area an for each half cycle changes. So-e When A is not modulated, in particular, if A is taken as SO, the difference value Xn between Sn and 5O(2) changes in accordance with the amplitude value of the modulated wave, etc.

Shimafu an -s □ --------------=-1 formula.

In addition, the 'L'irM value also changes as h changes, so 2M
Value too. Modulated wave like Xn! ! 41 value etc. (See Figure 2 n, described below).

B) The 41st FI is Enokihahahen! ! 1 (hereinafter referred to as AM) wave.Fn is a constant value that does not change, and the maximum value when no modulation is 0 is the amplitude value of the carrier wave, and the instantaneous width value corresponds to the amplitude value of the modulated wave. Therefore, like the gate wave, Xn changes.

C) Figure 5 shows a phase modulated (hereinafter referred to as P) wave.

- and h remain constant and do not change, but the 9 phase changes in response to the amplitude value of the variable lI wave. Therefore, the waveform and the zero equation point change in a complicated manner, making it appear that Ffl is changing. In addition, it is difficult to extract the point of change in the order m for each half cycle, so it is difficult to accurately calculate 8H except for 2-8 value PBX etc. Similar to the FM wave, the value changes depending on the amplitude of the modulated wave, etc.

D) Even if the deviation is %N, it can be said that Xn has a variation of 1-1" which is characteristic of various types of f4 waves.

Therefore, it can be used as a demodulator if Xn can be calculated one after another, an appropriate amount can be stored, and enough information can be extracted to restore the modulated wave.

0) 21st! l shows an example of the calculation method of output.

A) The time n-axis is divided at equal intervals by a minute time Δt, and the height to the intersection of the waveform and a vertical line set at each division point is continuously measured in rF.

B) Therefore, 8n can be approximated by equation 2-e.

This means that △ in each half cycle indicates the 11th break point, and 2m-0 indicates the starting point of each half cycle -I#, and 2m-1 indicates △ as the 1st) 3rd pass. The point is 2m, 2 is Δ, the second elapsed point is -111 and below, and in the same way, the point just before the fire's zero point is 5 at the M-th elapsed point. Also, in order to make the electronic circuit practical, the error l! I! With the purpose of lte management,
If Δ is started to be separated in synchronization with the m m Q point, the worst error at the ms+++M point in FIG. 2 can be managed to be less than Δ.

Therefore, the operation divided by Δ below is 2m every half period!
It shall be started in synchronization with O.

C) Next, h can be approximated by equation 3 from FIG.

Fn-1/2△t　M　-1---3 formula.

Here, if h at the time of one modulation is specifically FO, if it is an undistorted FM wave, the difference value Yn between Fn and FQ changes in proportion to the amplitude value of the modulated wave, so Sn should be used. It is possible to calculate the modulated wave by Yn. Especially in the case of FS Ko'jlt of about 2 to 64 values,
High distortion rate F! '1iia! However, the 9-ra value can be restored using only Yn.

c) Fig. 1 is a functional diagram for explaining the operation of a wave-resistant AM wave dual-purpose demodulator, which is an example of the implementation of the present invention. I am using it.

A) In Figure 1, * BPF is the appropriate band-wave filter.
The modulated wave from which unnecessary frequency components have been removed (adjusted to an appropriate signal level by automatic gain adjustment + quadrupler AGC, and converted into -iI digital values by analog/digital converter AD).

D in this example is synchronized by the detection pulse of the 0-type point detector Q Det. Driven by the output of the ΔtA pulse oscillator Ptc, digitization is performed and the In value is continuously measured.

B) The AD ratio output cumulative m calculator ALU calculates the total Hm value every half cycle, that is, when m is from 1 to M, and uses it as the Sn value.

That is, from equation 2 as Δt-l1l.

C) The ALU also manages the beam output value at the same time. That is, by limiting the input modulated wave of the AD to a sine wave,
When the measured Hm value exceeds the specified amount of change, the Hm value is discarded by automatic fishing, and l! Replace with the previous Hm value. Therefore, waveform distortion mixed within a half cycle is removed as a result.

The an value output of the re-ALU is m- which is a 9M value counter.
CNTy) M value output and N(ilW number counter n-C
With the N value output of N'I', i! ! Send to calculator CPυ.

E) The CPU calculates each of the following, temporarily stores it for each appropriate IIN fraction, and calculates and restores the modulated wave etc. by the following information processing. From the expression.

Fn Gourd 1/2M F) CPIJ So, Nll's X is temporarily stored.
Search for the state of change from nm. First, if Xn is changing, it means that there is modulation, and if it is 1, it means that there is no modulation, and usually the wave is islanded by the carrier wave. Therefore, we get 8Q at this time.

When Xn changes, search for h, and if Fh changes at the same time, it is an FM wave, and if not, it is an AM wave, so the display next to it lights up.

G) To simplify the explanation, if we take binary FSK as an example, Figure 6 shows an island and the mark corresponding value Xn
The difference between m, space corresponding value, e, Xns, Fz, and carrier wave corresponding value, XnQ, is 1, resulting in a dispersion similar to a normal distribution.

Needless to say, the same tendency is shown for Fn and binary AM waves.

H) From the above, when restoring the modulated wave etc.,
is the demodulation reference level of
Only Xn that falls within the dotted line of Q--1Φ°.

Xnm, that is, as a mark value, all other Xn are Xns
In other words, it is possible to process 0, etc., which is converted into a space value.

As described above, the demodulator to which the present invention is applied temporarily stores the Ni portion of Data repair 2 using the results of
etc. can be introduced. Therefore, a large number of means can be obtained to protect against various causes of typographical errors that may be mixed into the input modulated wave.

=) With conventional demodulators, the process was completed by restoring the modulated wave from the modulated wave.

However, in the present invention, the transmission code is regenerated from the restored modulated wave and 2i1! Data. Furthermore, typo check every 1f-2,
We believe that we can complete the demodulator by checking for typographical errors for each block and accurately restoring the information that is actually to be transmitted.Therefore, we have incorporated an ARQ function into the demodulation circuit.

DTE (Data Terminal Equipment)
ent ) (It is a normal embodiment to be aware of the Q burden. However, needless to say, not incorporating ARQ cannot be a reason for departing from the scope of the present invention.

A) Figure 7 shows an example of restored modulated waves, where h is a start signal, bn----b'n is one character of the transmission code, an----a''n1 is a one-character typo. Inspection signal.

Luster is stop signal, R is 0? Head display signal.

r---r'' is a professional typographical error check signal 1w is an end-of-block indicating signal, and -e is.

B) Previously, if even a single character was found to have a typo, the part containing the typo was retransmitted for each literature or block to be corrected.

The L or b original 11ll-e logically attempts to repair typographical errors, and when the typographical errors are successfully repaired, does not retransmit, thereby implementing reasonable typographical error processing.

c> fil: To the main station, t≠tens: Transmission is executed as a modulated wave etc. in Fig. 7.

* is determined to be a typo.

a) In CP (J), for example, words or phrases that have been classified in advance, normalized, and edited as a series of significant character combinations are stored internally.

b) t-1 among them! Similar beginnings, seven-ate, t-
(Put out ten, t Latin, -----, etc. and guess that one of the letters in the position of * is the correct letter, ``, 5 ---'', and successively repeat each character twice * Assign to .

C) Therefore, every time an assignment is made, a block typo check is performed, and if a correct character is determined, the substitution character is treated as a success in repairing the typo. Therefore.

It is assumed that normal and accurate communication has been carried out without any retransmission request.

Needless to say, the time required for two or more operations is 111.0 times longer than the conventional typographical error repair method that requires a retransmission request for the typographical part. L, 100-10000th of a liter.

d) With the above, it is possible to increase the efficiency of using the 9-pass aS and improve the through meter.

Water) In the conventional demodulator, the base P! is used as the demodulation reference level. !
! l? ll! There was one fixed criterion for 8-R Ware cautery, such as pressure value. Therefore, the modulated wave is amplified fl densely.

In addition to requiring processing, it lacks versatility as a demodulator,
It became necessary to provide a separate demodulator for each type of modulation.

In the present invention, by handling the modulated wave graphically and rationally extracting the characteristic amount of change, the modulated wave, etc., is calculated and restored by calculation, so the level within the relative level is always maintained. Jl + - difference is used as the demodulation reference level b), therefore, the conventional 1! In the preparation.

This corresponds to a demodulation reference voltage value, threshold value f value, etc.

and without fixing the fI quasi-level value.

Freely controlled numerically I-e4! obtain.

A) In Figure 6, the waveform m is the characteristic change amount of the modulated wave.
This shows the distribution state of xn at 2 [FSK - when t*t-dense. Although the number of multivalued FM waves and AM waves changes in -throw, it becomes a normal distribution. alike. li'
It becomes the distribution state of ilQ.

B) There 1# Xn etc. -I! If you temporarily store as many data as possible and perform statistical numerical processing, you will be able to extract or estimate all the results.

Therefore, there is a possibility that even a wave modulated by an unknown modulation method can be demodulated.

In general, it uses a common method published by CX::ITX' etc., so if you store it in the CPU and use it,
This is where most modulated waves can be restored.

C) It is well known that in k-level multilevel modulation, the larger k is, the larger the data transmission amount is, but the more sensitive it is to noise and the more typographical errors are, and the opposite is true for small k. But the actual communication! Then, there is an optimal k depending on various conditions such as weather, time, traffic, etc., and it changes randomly from moment to moment.

Therefore, in the present invention, by utilizing the above-mentioned features, it is possible to perform communication variably.

a) When the cheetah value is d, the corresponding communication level value is g
It is. In multi-value modulation of levels that can be defined by functions according to 4 lengths.

tll) Send in advance 1. The reception was agreed upon by both parties.

A communication level value g corresponding to one or more specific d values is transmitted in a certain appropriate amount prior to communication of 9 texts.

C) On the receiving side, the following conditions are met: the initial received communication level value is g″e, the cheater value at that time is d′, the relationship of Equation 4 is valid, and the multilevel modulation of the level is satisfied. All i-signal level values corresponding to all data values are calculated from , and this is the level of the four bases.Actually, one major formula is defined for the four formulas, and two values at both ends of the level are defined. Take the d value.

It is often implemented using the first method of first specifying a value in the binary FSK using the two values.

Table 1 shows an example of the transmission code among the restored modulated waves and the like, and the transmission status is serially numbered from left to right and from the top to the bottom.

[Brief explanation of the drawing]

Figure 1 is an example of implementation of the present invention-1! be. FM waves and AM
FIG. 2 is a functional block diagram for explaining the operation of the wave dual-purpose demodulator. FIG. 2 is an explanatory diagram showing an example of a method for calculating the characteristic change amount of the modulated wave, Sn, which is 1-11'' when the present invention is implemented. There is a 2°C Del waveform chart-e in which the characteristics of FM waves, AM waves, and PM waves are extremely emphasized, and modulation is applied every half cycle in synchronization with the 0 degree phase point. Figure 5 shows an example of the dispersion of Xn in a binary wind, where the horizontal axis shows the Xn value and the vertical axis shows the number of outputs.

Claims (1)

[Claims] B) Graphically handling a modulated wave such as a frequency modulated (hereinafter referred to as FM) wave, A) calculating a characteristic amount of change, and B) or selecting and extracting numerical values. C) Calculate the modulated wave from the relationship between the amount of change and the modulated wave. Or create a numerical conversion. As described above, this demodulation circuit method has the feature of calculating and restoring the included modulated wave from the modulated wave. (b) The input modulated wave is limited to a specific standardized waveform such as a sine wave, and the waveform portions that protrude from the standard value are reshaped by automatic deletion or automatic appropriate repair. The demodulation circuit system according to claim 1), which reduces nonlinear noise such as pulsed mixed noise as a result. C) In a communication system that has an automatic error detection/automatic retransmission request (hereinafter referred to as ARQ) function for demodulated modulated waves, A) Temporarily stores an appropriate amount of restored transmission codes from demodulated modulated waves. , B) Edit them as a meaningful series of characters, such as properly connected words and phrases (hereinafter referred to as words, etc.). C) Next, the items that have been classified, standardized, and stored in advance,
Search for similar words by comparing and contrasting with standard words, etc. D) Estimate the correct characters of the transmission code that has been determined to be a typo, and perform another typo check. When the spelling is determined to be correct, it is considered that the typo has been repaired successfully. E) In other words, the demodulation circuit system according to claim A), which uses the estimated correct characters to replace the erroneous characters and omit the request for retransmission of the erroneous characters, thereby increasing communication line usage efficiency and improving through foot. d) In the demodulation of a modulated wave by multilevel modulation related to digital code transmission, A) a communication level value corresponding to a characteristic data value is transmitted prior to main text communication, and B) on the demodulation side, Based on the characteristic numerical values received,
Calculate the demodulation reference level corresponding to all numerical values. C) According to the above, the demodulation circuit system according to claim 1) performs demodulation using the calculated demodulation reference level without fixing the demodulation reference level on the demodulation side.