System identification using chirp signals and time-variant filters in the joint timefrequency domain

In this paper, we propose a novel method to identify an unknown linear time invariant (LTI) system in a low signal-to-noise ratio (SNR) environment. The method is based on transmitting chirp signals for the transmitter and using linear time-variant filters in the joint time-frequency (TF) domain for the receiver to reduce noise before identification. Due to the TF localization property of chirp signals, a large amount of additive white noise can be reduced, and therefore SNR before identification can be significantly increased. This, however, cannot be achieved in the conventional methods, where pseudo-random signals are used, and therefore, noise-reduction techniques do not apply. Our simulation results indicate that the method proposed in this paper outperforms the conventional methods significantly in low SNR environment. This paper provides a good application of time-frequency analysis and synthesis.