Abstract:
In this PhD thesis, we investigated the problem of interference cancellation, i.e.,
the adaptive filtering in the processing radar chain, in PBRs (Passive Bistatic Radars) known to exploit existing signals such as radiobroadcast, telecommunications or radio navigation emissions as their transmitting sources. These systems are able to detect and track objects using reflections from, for instance, FM (Frequency Modulation) and DVB-T (Digital Video Broadcasting-Terrestrial) signals, known as illuminators of opportunity. The most important feature of PBRs is that these systems are completely passive; which makes them undetectable. To this end, first, we simulated a PBR via Matlab/Simulink based on the broadcasting FM signals received through the RTL-SDR (RealTek Label-Software Defined Radio) dongle. Then, we also simulated a PBR using DVB-T signals. Finally, for this last type of emitter of opportunity, we contributed to resolving one of the major problems encountered in PBR systems; that is, the masking effect induced by the existence of interferences such as the DPI (Direct Path Interference) and the MPI (Multiple Path Interferences), whose effects are due to the dynamic and static reflectors. To overcome
this troublesome phenomenon, we developed a novel algorithm called RD-FBLMS (Range Doppler-Fast Block Least Mean Square) whose primary function is the cancellation of these two kinds of interferences. In doing this, we showed through Matlab/Simulink simulations that the proposed algorithm not only offers a faster convergence rate and a shorter processing time, but also yields a better detection performance than the paradigm algorithms found in the literature.
