Estimation of linear distortions in telecommunication lines

• Main Authors: Желєзаров, І., Ілларіонов, Р., Прокофьєв, М., Дворський, В.
• Format: Article
• Language: Ukrainian
• Published: Інститут проблем реєстрації інформації НАН України 2021
• Subjects: теорія передачі даних; ймовірність правильного виявлення сигналу; амплітудні та фазові спотворення; data transmission theory; probability of correct signal detection; amplitude and phase distortions
• Online Access: http://drsp.ipri.kiev.ua/article/view/265674

Linear distortions of signals in telecommunication lines used in local area networks of information systems cause deterioration of the signal-to-noise ratio at the inputs of receivers located on individual sections of such systems, which reduces the probability of correct detection and reliable reproduction of transmitted signals. The method of estimating linear distortions of signals proposed by the authors is based on the definition of such distortions as the distances between the functions that describe the transmitted and received signals in the Hilbert space. It is shown that the coefficient of linear distortions determined by the criterion of least squares in the time domain, with small deviations of amplitudefrequency and phase-frequency characteristics from the linear law and choosing a constant value of the transmission coefficient of an ideal path is explicitly determined by the sum of relative amplitude coefficients, determined by the same criterion in the frequency band. Determining the linear distortion in this way makes it possible to use the frequency characteristics of the path to determine the delay time of signal transmission, namely the value of the delay time of the signal can be represented as the algebraic sum of two components, one of which determines the delay due to phase cutoff and the delay due to linear parts of the phase-frequency characteristic.It is shown that the estimate of linear distortions can also be given in the form of a geometric sum of amplitude and phase distortions due to phase clipping and distortions due to nonlinearity of phasefrequency characteristics. Estimation of linear distortions, determined by the least squares criterion, can be given explicitly through the frequency characteristics of the studied system and it can be determined from measurements of deviations of amplitude-frequency and phase-frequency characteristics of telecommunication lines for all signals with known spectral functions. This estimate can be effectively used to calculate the signal-to-noise deterioration caused by linear distortions in both correlation processing and consistent signal filtering.