Waveform-Diverse Stretch Processing

Abstract

Stretch processing with the use of a wideband LFM transmit waveform is a commonly used technique, and its popularity is in large part due to the large time-bandwidth product that provides fine range resolution capabilities for applications that require it. It allows pulse compression of echoes at a much lower sampling bandwidth without sacrificing any range resolution. Previously, this technique has been restrictive in terms of waveform diversity because the literature shows that the LFM is the only type of waveform that will result in a tone after stretch processing. However, there are also many examples in the literature that demonstrate an ability to compensate for distortions from an ideal LFM waveform structure caused by various hardware components in the transmitter and receiver. This idea of compensating for variations is borrowed here, and the use of nonlinear FM (NLFM) waveforms is proposed to facilitate more variety in wideband waveforms that are usable with stretch processing. A compensation transform that permits the use of these proposed NLFM waveforms replaces the final fast Fourier transform (FFT) stage of the stretch processing configuration, but the rest of the RF receive chain remains the same. This modification to the receive processing structure makes possible the use of waveform diversity for legacy radar systems that already employ stretch processing. Similarly, using the same concept of compensating for distortions to the LFM structure along with the notion that a Fourier transform is essentially the matched filter bank for an LFM waveform mixed with an LFM reference, a least-squares based mismatched filtering (MMF) scheme is proposed. This MMF could likewise be used to replace thefinal FFT stage, and can also facilitate the application of NLFM waveforms to legacy radar systems. The efficacy of these filtering approaches (compensation transform and least-squares based MMF) are demonstrated in simulation and experimentally using open-air measurements and are applied to different scenarios of NLFM waveform to assess the results and provide a means of comparison between the two techniques.