Gain, spatial and temporal noise, pattern, and non-linearity in QWIP focal plane array

Abstract

Signal-to-noise ratio (SNR) depends on how the signal is generated above the noise level at the detector. For QWIP (Quantum Well Infrard Photoconductor) photoconductive gain, noise gain and quantum efficiency are important properties that determine the operating parameters in a background-limited operation. At high bias the photoconductive and noise gain becomes comparable, and this allows for the experimental extraction of its value and quantum efficiency. It is important that signal and noise at the detector are the dominant values that determine the SNR of the entire system. In addition, the amplifier gain on Read Out Integrated Circuit (ROIC) is required to be large enough in order for input referred noises from subsequent amplifiers to be insignificant. Temporal and spatial noise limits the performance of the Focal plane Array (FPA). It is assumed that the time-independent noises can be calibrated out using linear two-point non-uniformity correction (NUC). The origin of spatial noise is ROIC's fixed pattern, Cosine4 (aperture effect), dark current, pixel response variation, etc. Unfortunately, gain and offset matrices from a finite time data capture do not represent detector array and ROIC behavior at time scale much larger than the gain and offset data capture time, and therefore 1/f noises can corrupt the images at larger time scale. Thus, an investigation of spatial and temporal noise has shown dependence on the number of frames collected. Non-linearity on some QWIP FPAs makes two-point NUC inapplicable, but it is not observed on some FPAs. The origin of this is not completely understood.© (2008) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.