A Joint Pre-Compensation and Windowing Framework for Sidelobe Suppression of Linear Frequency Modulated Signal

A linear frequency modulation (LFM) signal is widely used in radar systems. However, its inherently high autocorrelation sidelobes can degrade weak-target detection, while amplitude and phase distortions caused by transmitter systems may further elevate sidelobe levels. To address these issues, a joint pre-compensation and windowing optimization framework is proposed for a transmitter-distorted LFM signal. First, a regularized pre-compensation filter with gain constraints is constructed to compensate for transmitter-induced distortions and restore the waveform. Considering that the system frequency response is difficult to estimate accurately in practice, amplitude and phase perturbations are introduced, and a pre-compensation filter under perturbation is derived to improve robustness. To overcome the limited flexibility of fixed windows, a parameterized cosine-series window is employed, and the firefly algorithm is employed to jointly optimize the window coefficients and width, achieving a better trade-off among peak sidelobe ratio, integral sidelobe ratio, main lobe width, and peak-to-average power ratio. Simulation results demonstrate that the proposed method compensates transmitter distortions, significantly suppresses autocorrelation sidelobes, and maintains favorable performance under perturbations.