SU1663773A1 - Adaptive regenerator for digital transmission systems - Google Patents

Abstract

The invention relates to telecommunications and can be used in the development of digital linear paths. The aim of the invention is to improve the noise immunity by increasing the accuracy of the digital linear path correction when exposed to destabilizing factors. The regenerator contains a variable equalizer 1, a variable equalizer 2, a correction amplifier 3, a decisive block 4, an output signal shaper 5, a clock oscillator 6, a limiter amplifier 7, a rectifier 8, low-pass filters 9 and 18, a DC amplifier 10, delay elements 11, 12 and 13, a decoder 14, a comparator 15, a multiplier 16, and a switch 17. The goal is achieved by an adaptive correction produced from the evaluation of the standard deviation of intersymbol interference from the nominal value from even instants. 2 Il.

Description

The invention relates to telecommunications and can be used in the development of digital linear paths.

The purpose of the invention is to improve the noise immunity by increasing the accuracy of correction of the digital linear path under the influence of destabilizing factors.

Figure 1 shows the structural electrical circuit of the regenerator; Fig. 2 shows timing diagrams for his work.

The regenerator contains a variable equalizer 1, a variable equalizer 2, a correction amplifier 3, a decisive block 4, an output signal shaper 5, a clock oscillator 6, a limiter amplifier 7, a rectifier 8, a first low-pass filter 9, a DC amplifier 10, the first , second and third elements 11,12 and 13 delays, decoder

14, comparator 15, multiplier 16, switch 17, second low-pass filter 18 (LFC).

The adaptive regenerator for the digital transmission system operates as follows.

With an ideal correction, the path response to a single parcel is described by the expression

9 (O

, 31

sln (-j311

xt

t

cos (:

i

about . + -

1 - () 2

about s-h

with

where T is the duration of the clock interval, p is the smoothing coefficient. In this case, the amplitude of the response at the reference time is zero (see Fig 2a). When the path characteristics deviate from the ideal response selectivity is violated. Two options are possible1 if undercorrection leads to a narrowing of the analog path bandwidth, then the response changes as shown in Figure 2b. Otherwise, the undercorrection leads to a broadening of the band, which corresponds to the response in FIG. Thus, according to FIG. 2, the nature of the undercorrection of the analog path can be unambiguously determined from the value of the response at times that are one main clock from the main one. If the main response count coincides in sign with the first count of the pre-action and after-effect, the path bandwidth is narrowed and the correction needs to be made in the direction of the band extension. If the sign of the main reading does not coincide with the signs of the first readings of the pre- and after-effect, then the line of the path is extended and it is necessary to narrow it. These functions are implemented by an analysis device consisting of blocks 11-H8, whose operation algorithm is described by the following difference equation:.

,. (xn) xsign {D (mn-i + mn + mn + i}.

Here AI is the first difference operator; g 0, n is the estimation of the variable component of the main reference of the impulse response: Kf is the coefficient in the feedback circuit of the evaluation of the parameter g o; slgn (“) is a sign operation; xn are the sample values of the signal at the input of decision block 4; m {1,0, -1} - information symbols regenerated, for example, according to the law of the PRIL; D (-) is the key function defined for the PRF code.

KaKD (m n-t + m n + m n-Hh / 1. t hm + t htgN-1 and C-1, m n-1- | -m rH-1-1 and I 0, in other cases .

The signal from the output of the decision block 4 is fed to the first input of the decoder 14 and through delay elements 11 and 12 with a delay of one and two clock cycles to the second and third inputs of the decoder 14, respectively. The decoder 14 selects the signal combinations on which the response is analyzed and determines the sign of the package. With the input combinations 100 and 001, the output of the decoder 14 sets a signal 1, with the combinations -100 and 00-1, the output -1. In all other embodiments, the input signals at the output of the decoder 14 is set to O.

The signal from the output of the correction amplifier 3 passes through a comparator 15 with a zero threshold, where the sign of the received signal 1 or -1 is determined. Through the delay element 13, the sign signal delayed by one clock cycle is applied to the multiplier 16, to the second input of which

the signal from the output of the decoder 14 is received. The output signal of the multiplier 16 can take on three values: O - if the combination of symbols used to analyze the response form has not been received; 1 - if the allowed combination has been received and the sign of the main reference coincides with the sign of the first reference after or pre-action and, therefore, the analog frequency band

The tract 0 is narrowed and it is necessary to expand it: -1 - the allowed combination arrived and the signs of the main count and the first count after or pre-action do not match, therefore the band is extended it needs to be narrowed. The output of the multiplier 16 is fed to the control input of the switch 17. With the control signal O, the switch 17 is in a passive state, the signal at its output

0 is absent. When entering control input 1 or -1, the input of the low-pass filter 18 switches + or - to the power source, or vice versa, depending on the nature of the adjusting characteristic of the variable equalizer 2. The low-pass filter 18 averages from a predetermined time constant the output of the switch 17 and generates a continuous signal which is fed to the control input of the variable offset 2.

Claims (1)

Claims of the Invention An adaptive regenerator for a digital transmission system comprising a variable equalizer whose input is 5 input of the regenerator, a serially connected correction amplifier, a decision unit and an output shaper, the output of which is the output of the regenerator, a serially connected clock oscillator and an amplifier-limiter whose output is connected to the second inputs of the decision unit and the output signal generator; the first low-pass filter and the DC amplifier, the output of which is connected to the second input of the variable equalizer, and the tiruyuschego amplifier is connected to inputs delineator 0 clock oscillation and rectifier, characterized in that, in order to increase noise immunity by increasing the accuracy of the digital linear path correction under the influence of destabilizing factors, a variable equalizer is inserted into it, the output of which is connected to the input of the correcting amplifier serially connected comparator delay, multiplier, switch and second low pass filter, output which is connected to the control input of the variable offset, the second and third delay elements are connected in series, as well as the decoder, the output of which is connected to the second input of the multiplier, the output of the variable equalizer is connected to the input of the variable offset, the second and third inputs of the switch are connected respectively to the polar0 the positive and negative poles of the power source, the output of the rectifier is connected to the input of the comparator, the output of the decision block is connected to the input of the second delay element and the first input of the decoder, and the outputs of the second and third delay elements are respectively connected to the second and third decoder inputs. but -Y . figure 1 & ill lT