In this paper, non-fragile observer-based $${\mathcal {H}}_{\infty }$$ H ∞ controller design is investigated for a class of discrete-time systems. The system under consideration is assumed to have random fluctuations in both the state feedback controller gain and observer gain matrices. The random fluctuations are defined using Bernoulli-distributed white sequences with time-varying probability measures. The probability-dependent controller gains are designed to guarantee the stochastic stability of the system with a prescribed mixed $${\mathcal {H}}_{\infty }$$ H ∞ and passivity performance. Lyapunov stability theory, passivity theory and a linear matrix inequality (LMI) approach are used to derive sufficient conditions for the existence of the state feedback controller and observer gains. The probability-dependent gain-scheduled controllers are designed based on a convex optimization problem using a set of LMIs, which can be easily solved with standard numerical packages. Finally, a practical application is presented as an example to illustrate the effectiveness and potential of the method.