Computation of the maximal corank of a covariance matrix by radial search methods
Roberto P. Guidorzi, M. Pierantoni · 1997
This paper proposes a new algorithm to evaluate the maximal corank of a covariance matrix in the context of the Frisch scheme, i.e. the maximal number of linear relations that can be associated with a set of noise-corrupted data under the assumptions of this scheme. This algorithm is simpler than the previously available sliding gradient and exhibits a better efficiency.