Mixed L-2-L infinity/H infinity filtering for uncertain linear systems: a linear matrix inequality approach

The problem of full-order mixed L-2-L-infinity/H-infinity performance rltering design for uncertain linear systems is addressed in this paper. The uncertain parameters are assumed to belong to convex bounded domains (i.e. polytope-type uncertainty). The main purpose is to design a stable linear filter such that the rltering error output signal remains bounded, using a mixed L-2-L-infinity/H-infinity performance index as criterion. In the mixed L2-L-infinity/H-infinity rltering design, an upper bound to the peak value of the rltering error over all the energy bounded noise signals is minimized while an H-infinity constraint is assured. The conditions for the existence of such robust filter are provided in terms of linear matrix inequalities, allowing the use of standard convex optimization procedures to solve the problem. Both continuous- and discrete-time systems are considered. The formulation presented is illustrated by one example.3191091109