The Effect of Transmitter Nonlinearity on Passive Radar Ambiguity Processing

Delay-Doppler processing in passive radar is enhanced by first preprocessing the reference signal to remove noise and impurities, usually by demodulation and reconstruction. However cancellation of the direct path component and stationary clutter returns can be impaired when a reconstructed reference signal is used. This study hypothesises transmitter nonlinearities as a potential cause of poor cancellation. The memory polynomial is chosen as a suitable model for transmitter nonlinearity (NL). Simulation studies demonstrate that neglecting to model nonlinearities on the reconstructed reference signal does not significantly affect ambiguity surfaces directly, but that peak-to-floor metrics are degraded when zero-Doppler cancellation is performed without nonlinear modelling. Analysis of single-channel real data yields a small improvement when nonlinearities are modelled.