Research On Multi-Target Tracking By Using Passive Bistatic Radar System Of Near-Space Platform

The requests of stealth and strong survivability in motherland security or military applications alsocall for new radar platforms other than satellite and airplane; fortunately, these requirements can befilled by using near-space platforms. Inspired by recent advances in near-space, this thesis proposes anew passive bistatic radar system using opportunistic transmitter as an illuminator that exploits TVand FM radio, to illuminate potential targets and near-space platform as a receiver.This present project looks at the use of Doppler shift capacities from multiple receivers to solve thetarget tracking and association problem. A nonlinear Least Squares Error (NLSE) approximationtechnique, called the Levenberg-Marquardt (L-M) algorithm, is used to define a state of target(position and velocity). Whereas the NLSE method tends to give a precise solution in one quadrant,three other solutions will proportionally exist in each of the remaining three quadrants. The additionof either another receiver or another measurement (such as DOA) is desirable to break this quadrantambiguity. This thesis ponders adding multiple receivers to associate and follow trail multiple targets.Two target association approaches (sequential and simultaneous) are settled, and theircomputational requests and accuracy are associated. Statistical investigation of each method isshowed via Monte Carlo models. A grid-aided L-M search technique is explored in an effort toprovide a better initial target condition to these target association and tracking algorithms. Thisimproved initial estimate can help the NLSE algorithm come together to a more precise state estimate.The exploration and simulation results suggest it is feasible to perform multi-target association andtracking using Doppler shift as the sole measurement. Both of the proposed methods gave optimaltarget association and converged to reasonably accurate state estimates in most of the Monte Carloruns. The target associations and state estimates acquired from these measures can be used to adjust areal-time tracking algorithm, such as an extended Kalman filter (EKF), to realize track maintenance.Passive radar has a precise value for homeland security applications, rather than emitting pulses,this type of near-space passive radar tracking and association problem is considered for multipletargets and multiple receiver/transmitter pairs using only Doppler shift measurements, the number oftargets is assumed to be fixed and it is assumed that all targets produce responses at all receivers.