This paper presents an approach to the optimal stabilization of a linear time-invariant system via decentralized control, The class of controllers considered does not restrict the controller to be finite dimensional or time invariant, In fact, the controller structure proposed can be regarded as a form of time-varying sampled data control, The main result of the paper presents a scheme for constructing a collection of decentralized controllers which stabilizes the system and is optimal with respect to a quadratic performance index.