Search academic texts

Information × Registration Number 0215U009142, 0115U004145 , R & D reports Title Development of adaptive methods for moving targets detection using the decision statistics of plots popup.stage_title Head Zhuk Serhii Yakovych, Доктор технічних наук Registration Date 28-10-2015 Organization National Technical University of Ukraine "Kiev Polytechnic Institute". popup.description2 Wald test finds widespread occurrence for optimal targets detection, because it provides a significant gain in time at the given decisions probabilities. However, it loses its optimum properties if the parameter that characterizes the actual condition of the test process takes an intermediate value between the values specified in the test hypotheses. Therefore, if the actual signal-to-noise ratio (SNR) below a predetermined target SNR, the efficiency of target track detection is significantly reduced. To improve the target track detection efficiency proceeds to multialternative decision making problem using simple additions sequential test, which is a further development two-alternative Wald test. According to the research developed adaptive sequential methods of moving targets detection using the decision statistics of plots, having the opportunity along with the decision of the main tasks to perform SNR estimation. Based on the simple additions sequential test an adaptive multialternative sequential target track detection method is developed, which in contrast to the known algorithms sequential hypothesis testing on an unknown SNR is performed by the upper threshold. At the same time as an estimate of SNR value is consistent with that hypothesis. On the basis of the simple additions sequential test an adaptive multialternative sequential target track detection method is developed, in which decision in favor of a simple hypothesis is accepted after discarding competing hypotheses. Thus, a value corresponding to the remaining simple hypothesis is used as an SNR estimate. The adaptive two alternative sequential target track detection method is developed, which in contrast to the known sequential hypothesis testing on absence target track is performed against complex alternatives on presence target track with an unknown SNR estimate by the criterion of minimum mean squared error. Product Description popup.authors Неуймін Олександр Станіславович popup.nrat_date 2020-04-02 Close