Abstract
Shipboard coherent microwave radar is a promising approach for wave measurement during vessel navigation. However, in complex multi-directional sea states, the radial velocities measured by the radar result from the superposition of waves propagating both toward and away from the antenna. This superposition causes overlapping wave spectra and degrades the accuracy of wave parameter inversion. To address this issue, a novel method based on Two-Dimensional Variational Mode Decomposition (2D-VMD) is proposed for separating and reconstructing the radial velocities corresponding to waves propagating toward and away from the radar. First, spatiotemporal radial velocities are extracted from raw radar sea echoes and decomposed into multiple quasi-orthogonal submodes via 2D-VMD. Based on the spectral distributions of these sub-modes, “non-wave velocities” are eliminated, while the toward and away wave velocity components are further separated and reconstructed. Finally, non-overlapping directional wave spectra are derived from the reconstructed radial velocities, from which wave parameters are estimated. The proposed method is validated using two shipboard experiments conducted in the South China Sea: one in 2020 under directional bimodal sea states and another in 2019 under unimodal sea states. The wave spectra and parameters estimated by the radar show excellent consistency with buoy measurements in both scenarios, thus validating the effectiveness and robustness of the proposed method in separating velocity superposition and inverting directional wave spectra precisely.
Funding Information
  • National Natural Science Foundation of China (42276190)
  • Research Program (WDZC70202010206)
  • Science Foundation of Donghai Laboratory (DH-2022KF01020)
  • China Postdoctoral Science Foundation (2025M783518)
  • Open Fund of the State Key Laboratory of Spatial Target Perception (STA2025ZCC0303)
  • National Natural Science Foundation of China (GKZD010091)

This publication has 41 references indexed in Scilit: