SU1728974A1 - Method of forming of group signal by pulse-code modulation - Google Patents

Abstract

This invention relates to a digital communication technique. The goal is to increase the noise immunity. The goal is achieved by the fact that the signals of each channel are memorized before transmission and the maximum signal count is determined. If this maximum sample is zero, this channel sample is not transmitted, and a conditional code combination is transmitted instead, saying that a signal is being compressed on the transmission. Then the higher-order bits of all samples of all channels are transmitted using the error-correcting code. . The transfer of this additional information becomes possible due to the fact that the samples, the highest bits of which are equal to zero, are not transmitted. A device for implementing the method contains in each of the L channels ADC 1, a comparison unit 2, a drive 3 and an encoder 4, as well as an analyzer 5, a control unit 6, a code combination sensor 7. 1 il.

Description

The invention relates to electrical communication technology and can be used in multi-channel telephone messaging systems.

The purpose of the invention is to improve the noise immunity.

The drawing shows a block diagram of an apparatus for implementing a method for generating a group signal of a pulse code modulation (PCM).

A device that implements the method of forming a group PCM signal consists of L identical channels, each of which includes an analog-to-digital converter 1, a comparison unit 2 and a drive 3 connected in series, and an encoder 4, whose input is connected to the output of block 2, and group equipment consisting of an analyzer 5 whose L inputs are connected to the outputs of all the comparison devices, and the output to the control device 6, the outputs of which are connected to the control inputs of the accumulators and cyclic encoders of all channels and the sensor 7 code combinations, the outputs of which are connected in parallel.

The device works as follows.

The digital signal from the output of the analog-digital converter 1 does not flow directly into the group path, but is preliminarily stored during the cycle of operation of a circuit with duration T, in drive 3. As the drive cells are filled, unit 2 of each channel finds the maximum segment number of the stored code combinations and transmits it to the cyclic encoder, which forms a noise-proof code combination containing information about the segment number of the maximum count of each packet. At the same time, information from the outputs of the comparison devices of all channels goes to the analyzer 6, which, depending on the presence or absence of at least one value of the maximum segment number equal to zero, switches the control device to the required mode of operation, which involves switching the clock pulses in the right order to the sensor inputs code combinations, drives and cyclic coders. First, the line is transferred conditionally.

code combination, then the number of segments of the maximum samples of each channel, then the contents of the accumulators, and without repeating the leading zero bits (in

those channels where they are zero). Then the linear signal shaper converts the signal into a quasi-trivial code and inserts sync information at the corresponding positions.

At the reception, the values of the segments of the maximum samples of each packet are remembered and each next received sample is checked to ensure that the corresponding maximum value is not exceeded. When

By failing to meet this condition, the readout is adjusted. If all channels are high, the drives transmit information to the linear path without changes, except for the time delay T.

Claims (1)

Claim Method A method for generating a baseband signal by pulse code modulation, comprising: speech signals each channel is sampled, amplitude quantized, non-linear coding, and transmitted to a linear path, characterized in that, in order to increase noise immunity, after nonlinear coding code combinations of each channel are divided into packets of a given duration, memorize them, for each channel determine the signal of the maximum amplitude within the stored packet, compare with zero the specified number of higher bits of the maximum count and upon detection of channels in which all of the specified bits are zero, erase their memory older bits of all samples of the channels within the stored packet, form an additional code combination, transfer it to the linear path, then transfer the values of the higher bits of the maximum count of each channel using the error-correcting code, and then transfer the packets of all channels from the memory, and If the highest bits of the maximum count in none the channel is not zero, the code combinations of each channel are transferred from the memory to the linear path without changes.