On the receiver operating characteristics of adaptive radar detectors / by Christoph H. Gierull.: D68-6/117-2013-1E-PDF

Aim of this technical memorandum is to analytically derive the probability of detection (vrs the probability of false alarms) of two variants of a square-law based adaptive radar detector when the target signal is buried in colored interference and when the filter vector to suppress the interference is not constant (i.e. unknown asymptotic covariance matrix) but a random vector itself. The filter or weight vector is assumed to be estimated via either Sample Matrix Inversion (SMI), Loaded SMI (LSMI) or Projection Techniques, such as Eigen-Vector Projection (EVP), Hung-Turner Projection (HTP) or Matrix Transformation based Projections (MTP), based on a limited number of statistical independent secondary training data. The analysis is based on a homogeneous, i.e. Gaussian, stationary clutter plus noise assumption. Target signals comprise different fluctuating RCS models as well as a constant deterministic model. The test statistic provides for the summation of several independent cells (‘looks’) to improve detection performance whereby the target must not necessarily be present in each cell. Although the analyzed detectors may not directly be applicable, the provided analytical analysis permits an analytical comparison of the various adaptive weights estimation techniques.

Permanent link to this Catalogue record:
publications.gc.ca/pub?id=9.821300&sl=0

• MARC XML format
• MARC HTML format

Request alternate formats