Covariance analysis of a charge carrier device processing algorithm for stellar sensors

A special charge coupled device is used to process CCD output data for stellar detection. Implementation of the process is analyzed to determine the effect of the signal-to-noise ratio on the probability of stellar detection and the charge transfer efficiency. A covariance matrix is computed to determine information correlations. The problem of modeling a CCD imager array is addressed, and equations are developed for propagating the signal. Covariance equations are also developed to propagate the noise. An example is presented showing how the probability of detection varies as a function of the array signal-to-noise ratio and charge-transfer inefficiency (CTI). Doubling the stellar intensity results in the probability of detection going from unlikely to very likely; performance increases over a range of the CTI, and then decreases, demonstrating that some CTI is desirable in order to provide the minimum probability of nondetection.