In this paper, we present an indirect model reference adaptive control for SISO non-minimum phase systems with unknown or time-varying time delay. The parameter estimation scheme is a combined adaptive data filtering with a recursive least-squares algorithm with parameter projection and signal normalization. The problem of minimum phase of the plant is handled by an adaptive input-output data filtering. Hence, the zeros of the system-estimated model are relocated inside the unit circle. This estimated model, which is minimum phased, is then used for the control synthesis. It is shown that the adaptive input-output data filtering permits also to solve the difficult problem of model reference adaptive control in the case of unknown time-varying delay. The scheme robustness with respect to unmodelled dynamics and additive noise is also simultaneously improved. Finally, the results are illustrated by umerical examples.