Distributed Passive Tracking of a Non-Cooperative Underwater Target Utilizing Temporal Correlation of Line Spectrum

Distributed passive acoustic tracking is an important technique for detecting and localizing a non-cooperative underwater target, in which the frequency difference of arrival (FDOA) is a widely used measurement. FDOA reflects the differences in the target’s radial velocity relative to spatially distributed receiving nodes through Doppler-induced variations in the instantaneous frequencies of line-spectrum components. However, conventional FDOA-based methods rely heavily on the stable and accurate estimation of instantaneous line-spectrum frequencies, and their performance degrades when the line spectrum is affected by frequency fluctuations caused by target operating variations and external disturbances. To address this issue, this paper proposes a new measurement, the cross-correlation frequency difference of arrival (CFDOA), which exploits the temporal correlation of line-spectrum to characterize inter-node radial-velocity differences and reduces the reliance on accurate instantaneous-frequency estimation. To evaluate the effect of the proposed CFDOA measurement on positioning performance, a unified FDOA/CFDOA measurement equation is established within the same target position estimation framework. In addition, for scenarios with a limited number of receiving nodes, a recursive estimation scheme combining constrained initial-state search and particle filtering is developed. The simulation and sea-trial results demonstrate that, in the presence of line-spectrum frequency fluctuations, the proposed CFDOA measurement yields more accurate position estimates than conventional FDOA.