Weak NQR signal detection based on generalized matched filter

Nuclear Quadrupole Resonance (NQR) is a nuclear physical phenomenon. It is a solid-state radio frequency (RF) spectroscopic technique that can be used to detect narcotics and highly explosive substances. Unfortunately, there are limitations to NQR. The NQR signal is restricted by the low signal-to-noise ratio (SNR) and the presence of strong RF interference (RFI). The RFI can be reduced dramatically by shielding the search coil indoors. In this paper, numerical accumulation method is used to improve SNR of the weak NQR signal. According to the frequency characteristic of the background signal, AR method is used to estimate the power spectrum of the experimental background data. Then, the generalized matched filter is studied to enhance explosive detection. The experimental results show good performance of the proposed algorithm.