Addressing control-related problems in probabilistic Boolean networks may reveal new insights into the intrinsic functions of complex biological systems and enable us to develop strategies for manipulating biological systems using exogenous inputs. This paper investigates the problem of global stabilization of a probabilistic Boolean network to a given set of states using a state feedback law. The discussion is based on the recently developed technique of the semi-tensor product of matrices. A simple necessary and sufficient condition is derived for the existence of such a stabilizing feedback law, and a control design method is proposed when global stabilization via state feedback is feasible.