NLFM Radar Waveform Generation Using a Neural Network Approach to Rapidly Predict Bezier Curve Shape

Nonlinear frequency modulated (NLFM) pulse compression waveforms have become a mainstream methodology for radars across multiple sectors, including weather observation. NLFM affords the ability to generate a low-sidelobe autocorrelation function while avoiding windowing, resulting in more power incident on the target. This capability can lead to significantly lower system design costs due to the possibility of sensitivity gains on the order of 3 dB or more compared with traditional, windowed linear frequency modulated (LFM) waveforms. Generation of an optimal NLFM waveform, however, may involve complex optimization and non-closed-form solutions that take a considerable amount of computational power and time. For a multi-mission phased array radar, which may utilize a wide combination of frequencies, pulse lengths, and windows (among other factors), this could lead to an extremely large waveform table for selection that would be difficult to store and generate. This study takes a neural network approach to this problem by optimizing a set of waveforms spanning a wide space and using the results to interpolate the waveform possibilities to a higher resolution. A modified form of a previous NLFM method (based on Bezier curves) is combined with a four-hidden-layer neural network to show the integrated and peak range sidelobes of the generated waveforms across the model training space. The results are applicable to multi-mission radars and weather radars that need precise waveform specifications in rapid succession. The expected waveform generation times are addressed and quantified, and the potential applicability to multi-mission and weather radars is discussed. The addition of more training waveforms compared to previous work is discussed, and the difference in results is quantified.