In this paper, the problem of system identification in H-infinity with gene
ral orthonormal basis functions is investigated. A two-stage algorithm is s
how to be robust, provided that the umber of basis elements as a function o
f the amount of data does not increase faster than O (N (2/5)). Worst-case
identification error bounds in the H-infinity norm are derived. The algorit
hm also works on nonuniformly spaced frequency response measurements. An ex
ample is provided to illustrate the application of the algorithm.