Abstract:Traditional waveform fitting algorithms are not robust to noise. Moreover, these algorithms cannot detect water column scatter or weak signal and complex waveform shapes accurately. To overcome these problems, a method based on generalized Gaussian model is proposed. First, the algorithm estimates noise by calculating the difference between the last part of the original waveform and the filtered waveform. Then, it uses the generalized Gauss model to extract surface and bottom reflection signals, and to simulate the residual signal by iterating. At last, the parameters are optimized using the LM algorithm. In order to avoid redundant components, this paper gives some constraints. This algorithm is tested by measured data from the South China Sea, and it can detect weak signals and complex shapes. Whether in shallow or deep water, the fitting precision of this algorithm is better than traditional algorithms and it presents robustness.
WANG Xiankun,YANG Fanlin,ZHANG Hande et al. An Algorithm of Airborne LiDAR Bathymetric Waveform Simulation Based on Generalized Gaussian Model[J]. jgg, 2018, 38(11): 1180-1185.