Development of a model to study the characteristics of an air traffic control radar

Abstract

'Die availability of digital computers has stipulated the use of digital signal processing in many diwrse fields covering engineeIing, medicine and economics. The aim of this thesis is to develop a model for detection and tracking pmposes of Air Traffic Control Radar. Neyman-Pearson criterion has been used for detection problem and Kalman filtering has been used for the estimation of random signals to extract the pertinent infonnation. Kalman and Bucy proposed. an extremely powerful recursive state estimation teclmique, conUllOnly descIibed today as Kalman filtering. Tlus thesis is concerned with the application of Kalman lilteIing tecluuque to radar signal processing. 'Ole selection of appropriate states to configure the algorithm for use in radar signal processing is also considered. A new practical approach is presented 10 aid the evaluation of radar systems. Various tecluuques arc available for the estimation of random signals in the presence of noise and in doing so the need for solving sds of algebraic equations simultaneously arises. Dus conespomts to inverting a matrix whose order is that of the number of simultaneous equations involved. For the problem of tlus category, a conveluent tecluuque is one in which previously determined estimates are simply updated as new data come in, rather than solving the problem all over again. The recursive estimation teclmique (Kalman filtering) is exactly such a scheme where simultaneous estimates (filtered or predicted) of a number of signal components by mininuzing the mean-square cnor of each signal component simultaneously are looked for. In radar tracking problem one wants to estimate the range, range rate, bearing angle and bearing rate at cach time the radar measurement is available. These signal variables will be arranged in a colullln to be defined as the signal vector. Actually all practical signal processing problems are multidimensional and involve the collection of several signals together. 111e other basic materials that covers the nature of radar, the simple radar equation and th(, propagation of radar waves and how it is contaminated by the atmosphere and other deleteriow; cffects. The thesis deals with the filtering of noisy data in order to extract the signal from nois(: in an optimum (minimum mean-square error) sense. Initially the tracking problem is highlighted and existing stmctllres are discussed. TIle random signal and purely additive noise componenl!: are assumed to be statistically independent. To show the perfomlance of the proposed algorithm real d,lta has been l~~ed that obtained from the Radars at Zia International Airport, Dhaka. Be1,)re using the real data, the model was tested by known available data and fOlmd to be working well. Graphical representation of complete results are also included.