Enhanced LFM Detection in Digital ESM Receivers via Oversampled Fractional Fourier Transform

The fractional Fourier transform (FrFT) extends the conventional Fourier transform by providing a hybrid representation between the time and frequency domains. Its distinctive properties make it suitable for detecting and analyzing nonstationary signals, particularly linear frequency modulated (LFM) waveforms. However, in practice, FrFT-based detection may suffer from a peak-loss effect when the projected fractional-domain location of the input signal is not perfectly aligned with the sampled fractional-domain grid. This situation frequently arises in radar electronic warfare (EW) scenarios due to their noncooperative and wideband nature, potentially preventing optimal performance even under proper FrFT order matching. This letter analytically characterizes the resulting peak-loss as a function of system parameters and proposes a general mitigation strategy that exploits oversampled intermediate computations of a widely used numerical FrFT implementation. The technique is applied to two existing FrFT-based detectors and validated through Monte Carlo simulations, demonstrating improved performance without additional computational cost in the FrFT core.