Performance analysis of cumulant-based detection of non-gaussian signals

✍ Scribed by Boaz Porat; Benjamin Friedlander

• Book ID: 102658997
• Publisher: John Wiley and Sons
• Year: 1996
• Tongue: English
• Weight: 747 KB
• Volume: 10
• Category: Article
• ISSN: 0890-6327
• DOI: 10.1002/(sici)1099-1115(199603)10:2/3<99::aid-acs343>3.0.co;2-#

✦ Synopsis

We consider the problem of detecting a linear stationary non-Gaussian signal in additive Gaussian noise. Detection is assumed to be done in two steps: first a selected set of sample cumulants and/or moments of the data is computed; next a detection statistic based on these sample moments/cumulants is evaluated. Three types of detectors are considered: a likelihood ratio (optimal) detector, a weighted energy detector and a simple energy detector. The paper provides asymptotic performance analysis of these detectors for MA and ARMA signal and noise models. A simple example is worked out in detail.