Research on Weak Chirp Signal Detection Based on Cubic Spline Interpolation

In response to the problem of weak chirp signal detection, a method based on Short-time Fourier transform (STFT) and cubic spline interpolation is proposed. The scheme mainly exploits the continuous time-frequency characteristic of chirp signals to extract them from background noise. To be specific, we short-time Fourier transform the input signals into coherent integration of frequency-shift sample sequences which are with complex envelopes, then achieve a time-frequency curve and numerically analyze it with cubic spline interpolation, in order to eliminate the noise to the utmost extent as well as to detect chirp signals effectively. The algorithm remains available in quite a low signal-to-noise (SNR) with low complexity. Simulation results show its fine detection performance in -18dB condition, which can meet the demand of electronic reconnaissance.