Nonnegative linear systems, which have traditionally been investigated with
in the state-space framework, are in this paper introduced and analyzed by
means of the behavioral approach. Starting from certain definitions and res
ults presented in a recent paper [J.W. Nieuwenhuis, Systems Control Lett. 1
(1982) 283], we have explored the general autonomous case, by deriving an
extended set of necessary and sufficient conditions for an autonomous behav
ior to be nonnegative. In the scalar case, in particular, necessary conditi
ons for nonnegativity, which refer to the set of zeros of any polynomial in
volved in the kernel description of the behavior have been provided. (C) 20
00 Elsevier Science Inc. All rights reserved.