Study on the spectral characteristics of IF signals and its influence on the performance of heterodyne Φ-OTDR

The heterodyne phase-sensitive optical time-domain reflectometer (Φ-OTDR) is one of the most popular distributed optical fiber sensing (DOFS) technologies due to its high sensitivity, wide frequency response bandwidth, large dynamic range, and resistance to electromagnetic interference. In the signal processing workflow, like the I/Q algorithm, digital filtering of intermediate frequency (IF) signals is unavoidable. This study combines the advantages of the scattering point superposition model and the phase-modulated signal model. Three methods for quasi-matched IF filters are proposed. A signal evaluation system combining Parseval signal-to-noise ratio (Pa-SNR), peak-to-peak value, and signal-to-noise ratio (SNR) is employed. For CIC-type IF filters, it is recommended to cascade no more than 3 low-tap filters. For the variant CIC filter, its demodulation rate is faster than that of the other two filters. But resulted in an SNR of 10 dB lower than the CIC filter, and 15 dB lower than FIR filters because of the asymmetry of IF signals we demonstrated. For FIR-type IF filters, we recommend that the passband cutoff frequency should be at least 2.43 times the reciprocal of the pulse width. This factor decreases with increasing pulse width and narrowing of the transition band, but should not be lower than 1.08. Additionally, it is important to note that within approximately 0.8 times the transition band range of the cutoff frequency, half of the demodulation performance may degrade due to the IF filter passband ripple characteristics.