Публікація: Selection and recognition of the specified radio emissions based on the autoregression signal model
Завантаження...
Дата
2020
Назва журналу
ISSN журналу
Назва тома
Видавництво
Radio Electronics, Computer Science, Control
Анотація
Context. A solution to the relevance problem of selecting and recognizing specified radio emissions in the presence of unknown
radio emissions in automated radio monitoring is considered. It is proposed to solve the problem in an unconventional method for the
recognition of statistically specified random signals in the presence of a class of unknown signals.
Objective. The goal of the work is іnvestigation of the possibility of using random signal recognition methods in conditions of
increased a priori uncertainty to solve the problem. The features of the signal recognition method are discussed, as well as the results
of a study of the recognition quality indicators of given radio emissions, which are obtained by statistical modeling on samples of the
corresponding signals.
Method. The recognition method is based on the description of signals by a probabilistic model in the form of Gaussian
autoregressive processes. It is proposed to use the new decision rule for the selection and recognition of statistically specified signals
in the presence of unknown signals class. The proposed method of signal selection and recognition can be implemented in a recognition system that operates in training and recognition modes. In the training mode, unknown parameters of the decision rule
are evaluated by classified samples of the given signals.
Results. Research conducted by statistical tests on samples of the corresponding signals characteristic of automated radio
monitoring of radio communications equipment. Practical results of studies of the problem of selection and recognition of specified
radio emissions are presented. Values of indicators of quality of radio emissions recognition acceptable for the practice of radio monitoring are obtained. The dependences of quality indicators on some conditions and recognition parameters are investigated.
Conclusions. Undertaken studies showed possibility of decision of problem by application of an unconventional method
of selection and recognition of specified random signals. The practical significance lies in obtaining recommendations on the
construction of systems for the recognition of radio emissions for specialists in the design of automated radio monitoring complexes.
Such signal recognition systems are implemented by computer technology and is adaptive. The structure and parameters of the systems are set according to the samples of signals that are obtained for the corresponding given radio emissions.
Опис
Ключові слова
recognition system, automated radio monitoring, radio emission, selection, autoregressive model, recognition, decisive rule, statistical tests
Бібліографічний опис
Bezruk V. М., Kaliuznyi N. М., Qiang Guo, Zheng Yu, Nikolaev I. М. Selection and recognition of the specified radio emissions based on the autoregression signal model // Radio Electronics, Computer Science, Control.-2020.–No. 2.-рр. 7-14.