Non-linear processing based on Dolphin inspired real LFM signals for enhanced delay-Doppler resolution

In this paper, we study two non-linear processing approaches, inspired by biosonars, to enhance the delay-Doppler resolution. We use real linear frequency modulated (LFM) signals to fulfill a model, based on an abstraction and modification of the dolphin signals of certain species. With the help of Hilbert transform, two delay-Doppler images can be generated through single LFM signal. Non-linear algorithms are then applied and analyzed for producing a super delay-Doppler resolution when combining the images together. We also investigate the detection performance of those algorithms. Overall, the developed non-linear processing offers a great improvement in delay-Doppler resolution while the detection performance is close to that of the conventional matched filter receiver without significant increase in system or computational complexity.