JPS60240218A - Differential pcm encoding system - Google Patents

Abstract

PURPOSE:To reproduce a correct value at a receiving side while making the scale of a circuit small, by using a value accumulated in a previous value forecasting memory for taking a difference from an input value, as a past input val- ue. CONSTITUTION:When transmitting a signal from a PCM encoder 1, with respect to a present signal X0, X-1 before the signal X0 by one is stored in advance as a forecasting value in a memory 4'', and a difference between the signals X0 and X-1 is derived by a subtracter 2 and sent out as a difference d0. At a receiving side, encoding can be executed by adding this difference d0 to the pre-input value X-1, and the circuit scale can be made small.

Description

DETAILED DESCRIPTION OF THE INVENTION fal Technical Field of the Invention The present invention relates to an improvement in a differential PCM encoding method in which an analog signal 'ff is PCM encoded, band compressed, and transmitted.

fbl Technology background The differential PCM (hereinafter referred to as DPCM) encoding method is often used for signal transmission, such as television signal transmission, which requires a wide band, has a large transmission path cost, and has a large band compression effect. Ru.

(C1 Prior art and problems DPCMPCM encoding system - Figure 1 CM circuit and value X -
1r xQ l xl """ is zn, difference output value do
+d1...... is dn, then dn is calculated based on the previous value prediction series value xn-2' stored in the previous value prediction memory 4, as shown in the following equation α) (2). , and the difference output value dn-1 in the adder 8, and the difference between the previous predicted sequence value Xn-1' and the input sequence value xn is calculated in the subtracter 2.

xn -1' == dn -l+ xn -2'-=
=...-(1) dn= xn -zn -1'...
・・・・・・・・・・・・・・・・・・・・・(2) This mainly means that the differential output value dn may be larger than the maximum value dn that can be transmitted, for example, dn −1) dn When , the previous predicted series value is xn -1' = dn -1+ xa
−z' becomes as shown in the following equation (8), and zn −
1' = dM+Xn-2・・・・・・・・・・・・・
This is because (8) Xn-1'3Xrs-1 cannot be obtained, and a method of calculating the difference with the previous input value cannot be used.

(Of course, if there is no case where dn>dM, it can be replaced with the previous predicted sequence value.) For example, as shown in FIG. 2 (5), the input sequence value Ya + bpc
, d, and the value of b is twice as thick as the maximum transmittable value dM than the value of a (assuming that the values of b, c, and d are the same), the memory 4 stores (xn-1) of the previous input sequence value. If the method is to accumulate Y, take the difference from the input series value If you are reinstated, c, d
This is because the values of b', ', and d' are much larger by dM than the value of a, as shown in FIG. 2 (cl).

However, 71 previous predicted series values (x
n-1'), when b is input, the difference output is 2d
It becomes dM instead of M, but as shown in (B) inputted by C, the value of b becomes b' which is larger than the value of a by dM, and c
, the value of d is the value of c, d which is 2 dM larger than the value of a,
This is because the correct value is not obtained at the time of b, but the correct value is obtained at the following times of c and d.

However, the maximum value dM that can be transmitted so that always dM > dM
In the DPCM encoding method, in which the correct value can be reproduced on the receiving side by increasing the number of bits, the following (4) is obtained from equation (1).
The formula holds, and formula (4)7 is derived from formula (4).

x n-s' = d n-s 1 x n-/...
・・・・・・・・・・・・・・・・・・・・・(4)d n
-1' = xn-1'-Xl! -＊'・・・・・・・・・
・・・・・・・・・・・・・・・(4)′ On the other hand, the following (2)′ is derived from equation (2), and dn 1=)(n t−zB
-Myself...Song......
(2) 'Xn 1':) (n becomes 1, so the previous value predicted series value The DPCM circuit shown in the figure has the disadvantage of a large circuit scale.

In addition, as another DPCM encoding method that allows correct values to be reproduced on the receiving side, when dn dM, the DPCM encoding method is used, and when dII) dMQ, the PCM signal is sent as it is without using the difference, and both values are predicted. The previous predicted code string value (
A method has already been proposed in which the previous input sequence value (x-) is used instead of the previous input sequence value (x,').

This method will be explained using Figure 3 and Figure 4.

In this case, when dfi - dM, switch SWI is D
It is connected to the PCM side, and in the DPCM section 5, the same <
, the difference do is taken between the input value Xo encoded by PCM in the PCM encoder l and the previous value prediction sequence value x-1' stored in the memory 4', and inputted to this difference multiplexing circuit 10. do.

At this time, for example, O is generated in the identification signal generating section 9, inputted to the multiplexing circuit 10, and from the multiplexing circuit 10,
The code indicated by ■ is output.

Next, when the difference obtained by the subtracter 2 is d+s)dM, the overload detection circuit 6 detects this and switches SWI 'Y
Connect to the PCM side and drive the gate 8 to read the current input sequence value X+IY stored in the memory 7 memory 4
' and sends it to the multiplexing circuit 10, and the identification signal generating section 9 generates, for example, 1, which inputs it to the multiplexing circuit 10. is output. Next, the difference t between the input series value -R+-1 stored in the memory 4' and the input series value x0g is obtained.

However, in this case as well, the difference l is taken when dn<, dM, and the difference t is not taken when dn)dM. Therefore, as explained above, instead of the previous predicted sequence value, the actual previous input sequence value However, since the previous prediction series value Z is used, there is a drawback that the scale of the circuit pair shown in FIG. 3 is large.

OBJECT OF THE INVENTION In view of the above-mentioned drawbacks, the object of the present invention is to provide a DPCM encoding method using MoE input values that can reduce the circuit size in a DPCM encoding method in which correct values can be reproduced on the receiving side.

[el Structure of the Invention The above object is achieved by the structure of the present invention in which, in the DPCM encoding system, the value stored in the previous value prediction memory that takes the difference from the input value is the past input value.

That is, by doing this, there is no need to use the predicted sequence value, so the adder 3 in the case of FIG. 1 and the adder 3, gate 8, and memory in the case of FIG. 3 can be deleted,
The circuit scale can be reduced.

(fl　Embodiments of the invention Embodiments of the present invention will be described below with reference to the drawings.

FIG. 5 is a block diagram of a circuit according to an embodiment of the present invention when the number of bits of the maximum transmittable value dM is increased by 7 so that dn decreases dM.

Components in the figure that have the same functions as those in FIG. 1 are indicated by the same symbols, and 4'' is a memory.

In this case, the previous input value X-1 stored in the memory 4'',
The difference with the current input value XQ is sent out as the difference d0 by the subtracter 2, and this difference mountain is therefore made unnecessary, as is clear from a comparison with FIG. 1, and the circuit scale can be reduced.

FIG. 6 is a block diagram of a circuit according to an embodiment of the present invention in the case of dn)dM, and corresponds to FIG. 3.

Components with the same functions as those in FIG. 3 are indicated by the same symbols, and 4' is a memory.

In the case of dnkudM, the switch SWI is connected to the DPCM side, and as in the case of FIG. 5, the difference Z between the previous input value x-1 stored in the memory 4' and the current input value Xo is
Subtractor 2 makes the difference d. As in the case of FIG. 3, the 0 and 7 multiplexing circuits/f10 generated by the identification signal generator 9 are arranged to output the code shown in FIG. 4.

Next, input value x+1 is input and the difference obtained by subtracter 2 is d+
If l>dM, the overload detection circuit 6 detects this,
The input value x+1 connected to the switch SWI YPCM side and stored in the memory 4M is sent to the multiplexing circuit 0, and the identification signal generating section 9 generates 1 as in the case of Fig. 3, which is input to the multiplexing circuit IO. However, from the multiplexing circuit 10, Fig. 4 ■■
The code shown in is output. Next, the subtracter 2 calculates the difference between the input value x+1 and the input value x+s stored in the memory 4M.

That is, the same band compression as in FIG. 3 is performed in FIG. 6, and as can be seen from the comparison, one memory IJ, adder, and gate circuit can be eliminated, and the circuit scale can be reduced.

(gl Effects of the Invention As explained in detail above, according to the present invention, in the DPCM encoding method in which correct values can be reproduced on the receiving side, there is an effect that the circuit scale can be made smaller than in the conventional method.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a conventional DPCM circuit, Fig. 2 is a diagram explaining why predicted sequence values are used, and Fig. 3 is a block diagram of a conventional DPCM circuit that can reproduce correct values on the receiving side. FIG. 4 is an explanatory diagram showing an example of the encoded output in the case of FIG. 3.
? μl11m1i Figure 5 is a block diagram of a DPCM circuit using human input values that can reproduce the correct value on the receiving side according to an embodiment of the present invention, and Figure 6 is a block diagram of a DPCM circuit according to an embodiment of the present invention in which the correct value can be reproduced on the receiving side. It is a block diagram. In the figure, 1 is a PCM encoder, 2 is a subtracter, 3 is an adder, 4.
4', 4", 4", 7 is memory, 5 is DPCM section,
6 is an overload detection circuit, 8 is a gate, 9 is an identification signal generator, 10 is a multiplexing circuit, and SWI is a switch. Tei 1 Zucha 2 Inginbaku-'114M1- Tea 3 Mecha 4- Su

Claims (1)

[Claims] A differential PCM encoding method is characterized in that a value stored in a previous value prediction memory that takes a difference from an input value is taken as a past input value.