CORRELATION METHOD FOR FORMING THE TRAINING SAMPLE FOR ADAPTATION OF THE SPATIAL FILTERстатья из журнала
Аннотация: Contex. To ensure noise immunity of modern radar stations, two-stage space-time signal processing is used. However, withsimultaneous exposure to active and passive interference, the latter decorrelates the active interference. This significantly limits thenoise immunity of coherent-impulse radar stations. Therefore, the formation of a classified training sample, generated only by activenoise interference, is quite important task.Objective. The aim of the work is to investigate the correlation method for the formation of a classified learning sample in realtime by the current analysis of the interchannel correlation coefficient module over distance in each period of repeating the radarsignal.The method is realized by forming an interval synchronous evaluation of both the interchannel correlation coefficient modulusand the spatial filter weight coefficients. In this case, the interval with the maximum value of the interchannel correlation coefficientmodulus is defined as the classified training sample, and the weight coefficient formed on this interval is used to adapt the spatialfilter weight factor when the active noise interference is compensated in the next period of repeating the signals of the radar.Results. The theoretical and practical aspects of the formation of the classified training sample are considered. It was developed ablock diagram of the spatial filter with a correlation analysis of the current passive interference distribution over distance by estimatingthe interchannel correlation coefficient modulus of the signals acting in the channels of the spatial filter. A mathematical modelof the adaptive spatial filter was developed and tested. The possibility of working a spatial filter with the formation of a classifiedtraining sample in real time has been confirmed.Conclusions. The scientific novelty of the study is to further develop a new method for the formation of a classified training sample for the adaptation of a spatial filter. It is shown that the current analysis of the module of the interchannel correlation coefficient allows in real time to determine the interval on which there is no passive interference and to form a training sample for adaptation of the weight coefficients of the spatial filter.
Год издания: 2018
Авторы: Д. М. Пиза, S. N. Romanenko, D. S. Semenov
Издательство: National University "Zaporizhzhia Polytechnic"
Источник: Radio Electronics Computer Science Control
Ключевые слова: Advanced Control and Stabilization in Aerospace Systems, Military Technology and Strategies, Industrial and Mining Safety
Открытый доступ: diamond
Том: 0
Выпуск: 3