Simulation of mid-infrared clutter rejection. 1: One-dimensional LMS spatial filter and adaptive threshold algorithms

Area of Science:

• * Remote sensing
• * Signal processing
• * Atmospheric optics

Background:

• * Cloud clutter presents a significant challenge for remote sensing systems.
• * High-spatial-resolution data is crucial for detecting faint signals.
• * The 4.0-4.8 micrometer spectral band is utilized for atmospheric observation.

Purpose of the Study:

• * To evaluate 1-D signal processing techniques for cloud clutter rejection.
• * To compare the effectiveness of different spatial filters and threshold algorithms.
• * To determine optimal sensor parameters for reliable signal detection.

Main Methods:

• * Digital computer simulations were performed on high-spatial-resolution noise data.
• * Data was collected from back-lit clouds and uniform sky using a scanning system.
• * Evaluated techniques included bandpass filters, a least-mean-square (LMS) spatial filter, and adaptive threshold algorithms.

Main Results:

• * A 1-D LMS filter combined with a 1-D adaptive threshold sensor effectively rejected cloud clutter.
• * This combination provided nearly equal signal detection in clear and cluttered skies.
• * Performance was validated for systems with noise equivalent irradiance (NEI) > 1.5 x 10(-13) W/cm(2) and spatial resolution < 0.15 x 0.36 mrad.

Conclusions:

• * The 1-D LMS filter and adaptive threshold sensor offer a robust solution for cloud clutter mitigation.
• * These techniques enhance the reliability of remote sensing systems in challenging atmospheric conditions.
• * The findings are applicable to systems requiring high spatial resolution and sensitive detection.