SU1319288A1 - Digital device for controlling dynamic range of audio signal - Google Patents

Abstract

Invention m. used in digital signal processing devices. The purpose of the invention is to reduce non-linear distortions by improving the signal-to-noise ratio in the output signal while simplifying the device. The device contains a logarithm 1, a low-pass filter 2, a level 3 regulator, an exponentiator 4. A new delay 5 element, an exponentiator 6, a multiplier 7, and a divider 8 are introduced again. 2 Il. § (L Figure 1 co; o HCHO 00 00

Description

15

The invention relates to broadcasting and communication technology, can be used in digital audio signal processing devices and is intended to compress or expand the dynamic range of audio signals.

The purpose of the invention is to reduce nonlinear distortion by improving the signal-to-noise ratio in the output signal while simultaneously simplifying the device.

FIG. 1 shows a block diagram of the device; in fig. 2 is a timing diagram of the operation of the digital adjustment units for the dynamic range of the audio signal.

The digital device for adjusting the dynamic range of the audio signal contains a logarithm 1, a low pass filter 2 (VLF), a level 3 controller, a first exponentiator 4, and a delay element 5, a second exponential 6, a multiplier 7 20 leading and a divider 8.

The device also contains an external clock generator, which is connected to the control inputs of blocks 1-8 through a control circuit (not shown). The external oscillator has a frequency of 25 sync pulses equal to the sampling frequency of the Fg signal (for the given example, 48.0 kHz).

The device works following P1, im image.

The input signal S (pT) V (pT) E (pT), represented by the product of the carrier V (nT) and the envelope E (pT), is fed to the input of the logarithm 1 and to the first input of the divider 8. The logarithm 1 converts the original multiplicative combination of the component V ( nT) and E (nT) to the additive sum of their logarithms as In IS (nT) 1 1E (nT) + In 1 V (nT) j.

The output of the logarithm 1 is connected to the input of the low pass filter 3. The output signal of the low pass filter 2 is equal to E (pT) InE (nT). The exponential 6 connected to the output of the low-pass filter 2 is assigned its output to the output of the circuit, i.e. in the proposed device, by dividing the input signal into a carrier frequency signal V (nT), the non-existent frequency signal passes through blocks 5 and 8, the non-linear elements mine: the logarithm and the exponentiator causing the spread of the signal spectrum, and the E signal of the accuracy E ( pt), whose spectrum occupies a narrow region of the low-frequency range from

10 Hertz units up to 20-30 Hz with a spectral power density much higher than the carrier power spectral density, signal processing is carried out without an increase in the sampling rate, and therefore the logarithm 1, low-pass filter 2 and the exponentiator 4 can be used (about two times less ).

Thus, the high-frequency component (carrier) V (nT) is extracted using a linear multiplication operation, not to expanding its spectrum. The restoration of the original signal S (nT) is also performed using a linear multiplication operation, which in total does not require an increase in the sampling rate Fg.

thirty

envelope f (. Е (пТ)) and feeds it to the second input of divider 8. At the output of the divider, the carrier V (nT) S (nT) / Е (пТ) (V (nT) Е (пТ)) / Е (пТ ), which is fed to the first input of the multiplier 7. The output of the low-pass filter 2 is connected to the second input of the multiplier through the regulator 3 and the exponential 4. The adjustment is carried out manually by the operator immediately before setting up the device or

The control unit generates at the time ti-te the starting pulses of blocks 1-8, addresses for reading the coefficients of the impulse response of the digital filter (block 2) when calculating its output sample, as well as addresses for recording and reading the next input sample of the audio signal to delay element 5 (performed on random access RAM). With the first starting pulse, the control circuit at time ti produces a - -. starting the logarithm 1 and generates an address for recording the reference of the audio signal to the delay element 5. The second start pulse at the moment of time starts the digital filter 2, the operation of which continues the time T2.

During operation (2) of the digital filter, the control circuit sequentially generates addresses for reading the impulse response coefficients of the filter stored in the ROM. At time ts, the control circuit generates an address for reading the audio sample from the delay element 5. The next start pulse at time t4 the control circuit includes a level regulator block and an additional exponentiator 6. Then at time

40

45

the pause time, thus being one-time ts, is triggered by blocks by an exponential operation. Regulator 3 has an adjustable transmission coefficient K, the magnitude of which ultimately determines the mode of operation of the device, since the output signals of controller 3 and exponential 4 are equal to К1пЕ (пТ) and Е (пТ) | respectively. The multiplier 7, multiplying the carrier V (nT) with the envelope), restores the audio signal 5 (pT), with

55

torus 4 and divider 8. And at the conclusion of the operation cycle of the device for adjusting the dynamic range of the audio signal, equal to the sampling period of the input audio signal Tg, the last start pulse at time ta includes a multiplier 7, the output of which after a time takes a fully processed audio sample.

five

0 driving

five

this is its output is the output of the circuit, i.e. in the proposed device due to the separation of the input signal into a carrier frequency signal V (nT), the carrier frequency signal passes through blocks 5 and 8, the nonlinear elements mine: the logarithm and the exponentiator causing the spread of the signal spectrum, and the envelope signal E ), whose spectrum occupies a narrow region of the low-frequency range from

0 Hertz to 20-30 Hz with a spectral power density much higher than the carrier power spectral density, signal processing is carried out without an increase in the sampling rate, and therefore the logarithm 1, low-pass filter 2 and the exponentiator 4 can be used (approximately two times less ).

Thus, the high-frequency component (carrier) V (nT) is extracted using a linear multiplication operation, not to expanding its spectrum. The restoration of the original signal S (nT) is also performed using a linear multiplication operation, which in total does not require an increase in the sampling rate Fg.

0

The control unit generates at the time ti-te the starting pulses of blocks 1-8, addresses for reading the coefficients of the impulse response of the digital filter (block 2) when calculating its output sample, as well as addresses for recording and reading the next input sample of the audio signal to delay element 5 (performed on random access RAM). With the first starting pulse, the control circuit at time ti starts the logarithm 1 and generates an address for recording the reference of the sound signal to the delay element 5. The second start pulse at the moment of time starts the digital filter 2, the operation of which continues the time T2.

During operation (2) of the digital filter, the control circuit sequentially generates addresses for reading the impulse response coefficients of the filter stored in the ROM. At time ts, the control circuit generates an address for reading the audio sample from the delay element 5. The next start pulse at time t4 the control circuit includes a level regulator block and an additional exponentiator 6. Then at time

0

five

five

torus 4 and divider 8. And at the conclusion of the operation cycle of the device for adjusting the dynamic range of the audio signal, equal to the sampling period of the input audio signal Tg, the last start pulse at time ta includes a multiplier 7, the output of which after a time takes a fully processed audio sample.

At the times of time-t (Fig. 2), the processing of the reference of the sound signal is performed, respectively, in blocks 1-8 of the digital level control. The recording and storage time T5 of reference in delay element 5 is equal to T5 T9- | - Tio-fTu, where T9 and TP are the read and write times of the input signal reading into the element (and from the element) of the delay (random access memory), byte - time storing an audio sample in a delay element.

Claims (1)

Invention Formula containing a serially connected logarithm, a low-pass filter, a level regulator, and a first exponentiator, distinguished by the fact that, in order to reduce nonlinear distortion by improving the signal-to-noise ratio in the output signal while simplifying the device, a second exponentiator and multiplier are introduced, with In this case, the first input of the multiplier 10 is connected to the input of the logarithmizer through a delayed element and a divider introduced in series, and the output of the low-pass filter is connected to The second exponentiator, to the second input of the multiplier, whose Digital control device has a device output, a dynamic range of the audio signal, is connected to the output of the first exponentiator. containing a serially connected logarithm, a low-pass filter, a level regulator, and a first exponentiator, distinguished by the fact that, in order to reduce nonlinear distortion by improving the signal-to-noise ratio in the output signal while simplifying the device, a second exponentiator and multiplier are introduced, with This is connected to the input of the logarithm with the first input of the multiplier through the delayed and divider inputted in series, the divider whose output of the low-pass filter is connected to the second input Roy exponential i / i2 tjU 5 6 us.2 Impulse / start lokob. managements