ESTIMATION OF THE RADIOTECHNICAL CHARACTERISTICS OF THE RADIATOR

• Main Author: A. A. Moiseev
• Format: Article
• Language: English
• Published: CRI «Electronics» 2017-11-01
• Series: Радиопромышленность
• Subjects: radiotechnical characteristics; carrier frequency; bandwidth; frequency deviation; pulse width; pulse duty cycle; monte carlo method; direct scanning; woodward uncertainty function; euclidean metric; manhattan metric
• Online Access: https://www.radioprom.org/jour/article/view/256

Along with the kinematic characteristics of the receiving and radiating antennas, the measurement model is also determined by the radiotechnical characteristics of the receiver and the radiator. These characteristics are mutually unambiguously determined and, consequently, are determined by the probability of detection by frequency, introduced and evaluated in the kinematic model. Considering the aforementioned, the parameters of the radiotechnical component of the model are estimated on the condition of the minimal deviation in the detection probability by frequency from the proportion of crossing between functions (cross-sections of) uncertainty of Woodward’s receiver and radiator in the cross-section of the receiver. Estimation of radiotechnical parameters of the radiator is carried out by the Monte Carlo method or by direct scanning. Subsequently, these estimates are interpreted as estimates of the radiotechnical parameters of the radiator for the pre-set characteristics of the receiver. The conducted analysis shows that combination of the reduction method with the use of the Woodward uncertainty functions allows us to estimate the parameters of the detected object using the multidimensional optimization. In which case, the object is displayed as a point in the multidimensional space of parameters. A similar point corresponds to the standard or the center of the corresponding cluster in the presence of a spread in the parameters of the  reference.  Comparison  of  objects  to  standards  is  carried  out  by  the  criterion  of  minimum  Euclidean  or  Manhattan distance  in  the  space  of  normalized  parameters.  The  probability  of  correct  recognition  is  evaluated  in  this  case  by  the integer vector of the objects’ numbers that are compared to the standards.