We consider the problem of minimizing a given positive linear combinat
ion of the l(1) norm and the square of the H-2 norm of the closed loop
over all internally stabilizing controllers. The problem is analysed
for the discrete-time, SISO, linear time-invariant case. It is shown t
hat a unique optimal solution always exists, and can be obtained by so
lving a finite-dimensional convex optimization problem with an a prior
i determined dimension. It is also established that the solution is co
ntinuous with respect to changes in the coefficients of the linear com
bination. (C) 1997 Elsevier Science Ltd.