B-Spline Based Progressive Decomposition of LiDAR Waveform with Low SNR

Chang Liu,Xiaolu Li,Jianbin Huang,Lijun Xu

DOI: https://doi.org/10.1109/tim.2022.3157008

IF: 5.6

2022-01-01

IEEE Transactions on Instrumentation and Measurement

Abstract:Waveform decomposition is the key step in acquiring targets’ 3-D information using full-waveform light detection and ranging (LiDAR). With an attempt to effectively detecting targets with low signal-to-noise ratio (SNR), a B-spline based progressive decomposition method was proposed. This method uses B-spline to precisely model waveforms and uses progressive decomposition method to suppress noise influences. Simulation and ranging experiment were implemented to verify the proposed method at a sampling rate of 1 GSa/s with SNR varies from 10 to 30 dB and target interval varies from 1 to 2 m. Ranging experiment results show that the target extraction rate of the proposed method is the most accurate. The proposed method has strong target detection ability that it can effectively detect 90% targets since 14 dB even when the two targets are as close as 1 m with lowest underfitting rate. Robustness against noise of the proposed method is proven by its overfitting rate, which maintains no more than 5.5% at all tested conditions. Further comparison shows that the ranging accuracy of the proposed method is no more than 11.82 cm and the ranging precision of the proposed method is 13.71%–40.68% smaller than the deconvolution methods and Gaussian decomposition when target interval ranges from 1 to 1.25 m. Comparison results demonstrate the high ranging trueness and repeatability of the proposed method. The proposed method can be effectively applied to target detection from waveforms with low SNR for improving long ranging performances.