Comments on "On minimal realization of transfer function matrices using Markov parameters and time moments"

It is pointed out that the algorithm proposed in the above letter was published by the authors in 1976. Some important features of this algorithm, which have been missed in the above letter, are also described. In the above paper Shrikhande et al. have presented a method of minimal normalization based on a modified Hankel matrix containing time moments as well as Markov parameters Several comments are in order.