Backward scattering suppression in an underwater LiDAR signal processing based on CEEMDAN-fast ICA algorithm

Area of Science:

• Underwater optics and acoustics
• Signal processing
• Remote sensing technology

Background:

• Underwater environments present significant challenges for remote sensing due to scattering and attenuation.
• Traditional Independent Component Analysis (ICA) for blind source separation (BSS) requires multiple measurements, limiting efficiency and introducing variability.
• Weak target reflections are often obscured by strong backward scattering in turbid water, hindering accurate ranging.

Purpose of the Study:

• To develop a novel signal processing method for blind source separation (BSS) in underwater lidar-radar systems.
• To improve ranging accuracy by effectively recovering weak target reflections from strong backward scattering.
• To overcome the limitations of traditional ICA by enabling BSS from a single measurement.

Main Methods:

• The study introduces a new method combining Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Independent Component Analysis (ICA).
• CEEMDAN is utilized to construct an observation matrix from a single measurement, enabling ICA to perform BSS on mixed source signals.
• This CEEMDAN-ICA approach mitigates issues related to the number of observations required by standard ICA and reduces uncertainty from changing measurement conditions.

Main Results:

• The CEEMDAN-ICA method significantly improved ranging accuracy in an underwater lidar-radar system.
• For a mirror target, ranging accuracy increased from 12.5 cm to 4.33 cm at 2 m distance in water with a 7.1 m⁻¹ attenuation coefficient.
• For a PVC plate target, ranging errors were reduced from 21.54 cm to 5.01 cm at 3.75 attenuation lengths, demonstrating substantial accuracy enhancement.

Conclusions:

• The developed CEEMDAN-ICA method effectively enhances blind source separation in underwater lidar-radar systems.
• This approach substantially improves ranging accuracy by isolating weak target signals from clutter in challenging aquatic environments.
• The method increases detection efficiency by reducing the number of required measurements, offering a more practical solution for underwater sensing.