Revising the application of cross-spectrum processing in motion parameter estimation for harmonic sources

Abstract

Single-receiver motion parameter estimation is an effective and economical technology for passive source localization and train-bearing fault diagnosis, in which time-consuming time-frequency analysis (TFA) methods are widely used to suppress noise when extracting the continuous Doppler shift of the overhead pass. Cross-spectrum processing is a potential way to improve the computational efficiency of TFA methods, but its application is overshadowed by the phenomena of unknown Doppler shift offset and power spectrum estimation error. In this paper, conventional cross-spectrum processing is proven to be an approximation trick for power spectrum estimation in a small frequency interval, and the two phenomena are fully explained by the frequency aliasing of bandpass sampling and the approximation error. On this basis, an revised framework for applying the cross-spectrum processing is provided. Processing results of the SWellEx-96 experiment data demonstrate that the computational efficiencies of spectrogram and a parameterized TFA method could be improved up to 85% and 88.2%, respectively, without a noticeable impact on the accuracy of parameter estimates.