This paper deals with iterative learning control design for multiple-input multiple-output (MIMO), linear time-invariant (LTI) systems. Two particular ILC schemes are considered and analyzed in both frequency and time domains. Some remarks on the convergence, implementation, robustness with respect to disturbances and reinitialization errors, as well as positive realness issues related to both schemes are provided. (C) 2007 Elsevier Ltd. All rights reserved.