Аналіз впливу зосереджених завад на системи передачі інформації зі згортковим кодуванням
Вантажиться...
Файли
Дата
2019
Автори
Макаришкін, Д.А.
Кисіль, Т.М.
Карпова, Л.В.
Драбинка, О.С.
Makaryshkin, D.
Kysil, T.
Karpova, L.
Drabynka, O.
Назва журналу
Номер ISSN
Назва тому
Видавець
Хмельницький національний університет
Анотація
В статті описаний аналіз впливу зосереджених завад на системи передачі інформації зі згортковим кодуванням. Розглядається структура алгоритму згорткового декодування Вітербі, яка дозволяє здійснювати його внутрішню адаптацію при різних внутрішніх умовах. Проводиться дослідження умов роботи алгоритму згорткового декодування Вітербі, які забезпечують рівень завади, за якого даний алгоритм знову стає працездатним. Встановлено, що в умовах впливу зосереджених по спектру завад великого рівня здійснюється різке погіршення корректуючої здатності згорткових кодів.
The article describes the analysis of the influence of concentrated impediments on information transfer systems with convolutional encoding. The structure of the Viterbi convergent decoding algorithm is considered, which allows for its internal adaptation under different internal conditions. The study of the conditions of the Viterbi rollout algorithm, which provides a level of interruption, at which this algorithm becomes operational again, is carried out. It was established that under conditions of influence of large-scale luminescent spectrum, a sharp deterioration of the correction capacity of convolutional codes is carried out. There are various ways to decode convolutional codes based on the interleaving and de-interleaving of digital characters to deal with error groups as a result of fading signals. As a result of interleaving, consecutively badly distorted characters are spaced apart in time and can be successfully used for convolutional decoding according to the Viterbi algorithm. However, when using interleaving, it is possible to make the wrong choice of the best path on the grid and, accordingly, to incorrectly decode some portion of the transmitted information correspondence. Since each path metric is calculated according to the same rule, this corresponds to the assumption that the average noise level over the entire length of the path remains constant. In other words, the value of the contribution of each character in the formation of the total metric of each path must be different, and in the standard Viterbi algorithm this adjustment is not made. The purpose of the convolutional code decoding method is to reduce the likelihood of decoding error and to increase the noise immunity of the transmitting information. It is known that the basic principle of the "soft" method of decoding according to the Viterbi algorithm is to choose the path on the lattice diagram, which generally has a minimum metric, that is, the minimum sum of Euclidean distances for all accepted characters from the code variant corresponding to each character, to the value of the voltage obtained from the demodulator output when this symbol arrive. Each variant of the path corresponds to a specific variant of the correspondence of the transmitted symbols, the path with the smallest metric corresponds to the most probable transmitted correspondence of the symbols. According to the maximum likelihood method, it is necessary to select the most probable ones from all possible sequences of the received symbols. Suppose that the probability of the qth variant of the sequence is Pq, then we must find the number q with which the maximum value {Pq} is provided.
The article describes the analysis of the influence of concentrated impediments on information transfer systems with convolutional encoding. The structure of the Viterbi convergent decoding algorithm is considered, which allows for its internal adaptation under different internal conditions. The study of the conditions of the Viterbi rollout algorithm, which provides a level of interruption, at which this algorithm becomes operational again, is carried out. It was established that under conditions of influence of large-scale luminescent spectrum, a sharp deterioration of the correction capacity of convolutional codes is carried out. There are various ways to decode convolutional codes based on the interleaving and de-interleaving of digital characters to deal with error groups as a result of fading signals. As a result of interleaving, consecutively badly distorted characters are spaced apart in time and can be successfully used for convolutional decoding according to the Viterbi algorithm. However, when using interleaving, it is possible to make the wrong choice of the best path on the grid and, accordingly, to incorrectly decode some portion of the transmitted information correspondence. Since each path metric is calculated according to the same rule, this corresponds to the assumption that the average noise level over the entire length of the path remains constant. In other words, the value of the contribution of each character in the formation of the total metric of each path must be different, and in the standard Viterbi algorithm this adjustment is not made. The purpose of the convolutional code decoding method is to reduce the likelihood of decoding error and to increase the noise immunity of the transmitting information. It is known that the basic principle of the "soft" method of decoding according to the Viterbi algorithm is to choose the path on the lattice diagram, which generally has a minimum metric, that is, the minimum sum of Euclidean distances for all accepted characters from the code variant corresponding to each character, to the value of the voltage obtained from the demodulator output when this symbol arrive. Each variant of the path corresponds to a specific variant of the correspondence of the transmitted symbols, the path with the smallest metric corresponds to the most probable transmitted correspondence of the symbols. According to the maximum likelihood method, it is necessary to select the most probable ones from all possible sequences of the received symbols. Suppose that the probability of the qth variant of the sequence is Pq, then we must find the number q with which the maximum value {Pq} is provided.
Опис
Ключові слова
завади, алгоритм згорткового декодування Вітербі, ймовірність помилок, перемеження символів, obstacles, algorithm of convolutional decoding of viterbi, probability of errors, interleaving of symbols
Бібліографічний опис
Аналіз впливу зосереджених завад на системи передачі інформації зі згортковим кодуванням [Текст] / Д. А. Макаришкін, Т. М. Кисіль, Л. В. Карпова, О. С. Драбинка // Вимірювальна та обчислювальна техніка в технологічних процесах. – 2019. – № 1. – C. 54-61.