JPS5886637A - Digital mixing circuit - Google Patents

Abstract

PURPOSE:To simplify the constitution of a circuit which performs such mixing control that the sum of mixing rates of input signals is one, by using a code-free multiplying means of handling only one positive number. CONSTITUTION:Digital input signals A and B are both supplied to an absolute value circuit 1, which outputs a positive/negative polarity sigal simultaneously with their absolute-value difference signal¦A-B¦. The absolute-value signal ¦A-B¦is led to one input of a multiplier 2, and a mixing ratio controlling signal C is supplied to the other input. The multiplier 2 multiplies two positive numbers by each other to round the product to an adequate bit length, and then supplies the product signal¦A-B¦XC to one input of an ALU3. A signal B is supplied to the other input of the ALU3. The ALU3 performs arithmetic B+¦A-B¦XC when the signal A is greater than the signal B, or B-¦A- B¦XC when A is less than B.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a digital mixing circuit that performs mixing control so that the sum of the mixing ratios of two digital signals PCM (pulse code modulated) into natural binary numbers is always 1.

This type of mixing control is shown in the following equation.

AxC・+BX(1-C) (o(Cku1)...
1 is one digital input signal, B is the other digital input signal, and C is a mixing ratio control signal. To simply implement this control, two multiplication means 'ft must be provided for the first and second terms of Equation 1. Digital multipliers are complicated, such as requiring a dedicated LS i′ft,
In addition, the unit cost is high, so it is preferable to reduce the number of units used.

Therefore, it is also possible to expand Equation 1 into the following equation and use only one multiplication means.

B+(A-B)xC...2 This requires a subtracter, a multiplier,
Further, an adder may be provided for adding the first term and the s2 term. However, the sign in the parentheses of the second term depends on the input signal A, and the multiplier must be treated as a complement (signed operation). It is undesirable to use such a limited multiplier. In particular, when the bit length may be small, a multiplier iP-ROM is often used as a substitute, but the increase of one bit by the sign bit, which is a signed operation, cannot be ignored.

The present invention aims to solve such conventional problems and realize a circuit that performs mixing control such that the sum of the mixing ratios of input signals becomes 1 by using 1- unsigned multiplication means. The present invention will be explained below.

If Equation 2 above is further decomposed, it becomes as follows.

If A-13kuO, then B+ l A-B l xC----
(3-1) A-B (for Q, B -l A-B l xC
---(a-2) As is clear from the above equation, the difference between the two input signals A and H is determined in advance to be positive or negative. Depending on the result, if the signal processing result of either equation (3-1) or equation (3-2) is output, the second term is the absolute value IA-Bl of the difference between the two input signals and the mixing ratio C. , all result in multiplication of positive signals, eliminating the need for signed multiplication means.

The figure is a diagram showing the configuration of an embodiment to which the present invention is applied.
The digital input signals A and B are controlled by a mixing ratio control signal C to obtain an output signal.

In the same figure, 1 is an absolute value circuit, 2 is a multiplier, and 3 is an A L
U (Arithmetic Logic Uni
ts).

In this embodiment, digital input signals A and B are both fed to an absolute value circuit 1, which outputs a difference absolute value signal IA-
It outputs positive and negative polarity signals at the same time as Bl. Absolute value signal I
A-Bl- is led to one input of the multiplier 2, and the other input is supplied with the mixing ratio control signal C. The multiplier 2 multiplies these two positive numbers together, rounds them to an appropriate bit length, and then guides the product signal IA-BlxC to one input of the ALU 3. Digital input signal B is supplied to the other input of fcALU3. For example, ALUa is 5N74
It is possible to switch between addition and subtraction by a switching signal such as 8381, and this switching is performed by the positive and negative polarity signals obtained by the absolute value signal IA-alt of the difference in the absolute value circuit 1. . This operation is based on the magnitude relationship of the digital input signal, and if A is greater than B, then B+1A-Blx
Perform addition with C, and if A is less than B, then B-IA-Bl
It performs subtraction with xC. The output signal of the ALU 3 is a mixed signal obtained by controlling the input digital signals A and B with the mixing ratio control signal C so that the sum of the mixing ratios becomes 1. The absolute value circuit used here is generally well known, and includes an adder and an Eialusiue-OR circuit.
Either it is composed of one set of circuits, or it is similar to ALUa, and the subtracted value and the subtracted value are reversed, and both A-B and B-A are performed, and the one that yields a positive result is selected. There are many ways to do this, and in practice, for reasons of processing speed, each processing block may be sandwiched between registers to create a pipeline system.

As explained above, according to the present invention, one multiplier that only needs to handle positive numbers is provided, and it is possible to perform mixing processing equivalent to the conventional one. Also, as mentioned above, an absolute value circuit and a circuit for switching between addition and subtraction are required, but both are AL
It is effective because it can be easily processed by U, and the circuit configuration is simplified to the extent that a signed multiplier is not required.

[Brief explanation of the drawing]

The figure is a block diagram of a digital mixing circuit according to an embodiment of the present invention. 1... Absolute value circuit, 2... Multiplier, 3...
...A-LU.

Claims (1)

[Claims] a circuit that calculates the absolute value of the difference between two digital input signals pulse code modulated into natural binary numbers and determines whether the signal is positive or negative;
a multiplication circuit that multiplies the absolute value signal and the mixing ratio control signal; and a circuit that outputs a calculation that switches whether to add or subtract the output signal of the multiplication circuit from one of the digital input signals, depending on the result of the positive/negative determination. A digital mixing circuit characterized by being provided with.