A wavelet de-noising signal processing method for overall noise-to-signal (NSR) profile extraction, characterization and comparison of 3?m-5?m MWIR strained-layer super-lattice (SLS) photo-detectors enhanced with microsphere lenses of different material s

MWIR detectors have received a lot of attention due to their importance and high viability in many civilian and military applications, such as target identification, bio-imaging, medical field, and defense. One crucial component in achieving high-sensitivity performance in MWIR photo-detectors, is obtaining reasonable MWIR high detector signal-to-noise ratio (SNR) profiles. The classical IR photo-detector SNR model is limited as it does not reflect modeling of all physical factors, including detector environment sources causing undesired noise. In general, there is a need for better empirical methods to accurately estimate and extract the detector integrated-sources of imperfection noise. In this paper we present and discuss a non-parametric wavelet-based de-noising signal processing method for accurate extraction, characterization and comparison of the overall combined effects of detector sources of noise, including spectral fabrication imperfection noise, electronics testing equipment imperfection noise, and microsphere lens material fabrication/alignment imperfection noise, among others, of eleven (11) 3μm-5μm MWIR InAS/GaSb strained-layer Superlattice (SLS) single photodetectors enhanced with microsphere-lenses of different material structures/sizes. With the collected spectral FTIR data considered and the detector wavelet-denoising-based extracted NSR, the results show that there is a decrease (improvement) of about 71 % (difference between lens-enhanced and no-lens-enhanced NSR values divided by no-lens-enhanced NSR) in the noise-to-signal ratio (NSR) of the SLS photodetector of size 40μm and enhanced with a microsphere-lens sapphire of size 300μm in the targeted 3μm-5μm MWIR wavelength-band considered.