Data-driven control approaches have been proved highly effective in applications where the dynamics of the system are unknown. The reason is that the use of data in control overcomes the challenge of parametric modeling which requires effort and often leads to an insufficient description of the system’s behavior. By contrast, the behavior of the system can be captured more accurately when input-output data is used. In this project a frequency-domain approach is proposed for robust controller design for controllers of pre-specified structure. Specifically, an existing technique is extended so that it guarantees not only nominal, but also robust performance. This is achieved by ensuring performance for multiple frequency-domain models acquired from data. The related computational efficiency in the overall approach is also a fundamental criterion and thus we propose a series of techniques by which it can be improved. Additionally, a major contribution of this work is an iterative algorithm that minimizes a linear approximation of the H infinity norm of the closed-loop system. Last but not least, a method to optimize the linear constraints of the optimization scheme is proposed.