Effect of Non-Appropriate Sampling on Radar Detection Algorithm

Many detection algorithms for received radar signals are based on Fast Fourier Transform (FFT) processing. The detection scheme for radar signals is affected by the prior knowledge of the Power Spectral Density (PSD), which is affected by non appropriate sampling. A performance comparison of two detectors based on frequency domain and changing place of sample namely periodogram and quadratic detectors, is presented, and analytic evaluation of the performance of these detectors supported by computer simulation is conducted. When the desired frequency is not in a bin of the discrete PSD, one standard technique to recover the peak is to add zeros (zero-fill). The performance measure utilized in this work is the Receiver Operating Characteristic (ROC), which measures the probability of detection with the probability of false alarm at different values of Signal to Noise Ratio (SNR). It is obvious that the performance will be degraded if there are estimation errors.