Research On Passive Detection Based On The Analysis Of The Properties Of DVB-S Signal

With the development of emerging information technology such as navigation and communication, diverse kinds of electromagnetic signals appear in the air. Therefore, the passive radar is attracting more attention. Especially the rise of DVB-S signals in recent years promote the fast development of passive radar as it can provide reliable illuminators. The work of this paper includes the following three aspects:Firstly, DVB-S signals is continuous wave communication signal, and the parameters and waveforms are not specifically designed for radar. The properties of DVB-S signal are analyzed thoroughly based on radar signal theory in order to make better use of this kind of signals. Besides, the passive radar data acquisition system is designed to collect measurement data and validate the properties of DVB-S signal. The simulation and experimental results both demonstrate the feasibility of using DVB-S signal for passive radar application.Secondly, the side lobes of ambiguity function are high, in order to avoid losing the interest target in multi-target area, which is caused by the shading effect of side lobes of strong target echo signal. The weighting technique is applied to suppress the side lobes in Doppler domain of the ambiguity function considering a tradeoff among side lobe suppression level, main lobe spread and signal-to-noise ratio (SNR) loss. Better results are achieved. Afterwards, the side lobe suppression in range domain is achieved by taking advantage of the reciprocal filter. The semi-physical simulation results show the feasibility of the method.Finally, the target echo signal is submerged in the noise. In order to extract the target signal for passive radar detection much better, the introduction on the existing methods for improving the SNR of the target echo signal is made. And the target detection method based on cyclostationary characteristic is introduced to detect moving targets. Simulation results show that the method in this dissertation is more robust to noise than other traditional methods. Finally, a fast calculation method of the cross-ambiguity function is proposed to reduce the computational complexity.