A method for the robust identification of interval Linear Time Invariant (LTI) models is presented in this contribution. The procedure is based on the work of Ploix et al. (1999) and extends the contribution of Blesa et al. (2011) to those cases in which the experimental dataset lies in the frequency domain. The use of the proposed method results in a transfer function with a previously fixed structure and interval parameters. The frequency response of the obtained model includes all the experimantal observed data.