This paper deals with secrecy, i.e. how an information-flow property of system can be kept secret from observers. Recently, a discrete-event system based model has been developed to study secrecy. With such a model, secrecy preservation of a property is modeled as the impossibility for observers to determine whether executed event sequences belong or not to a given language. In the present paper, we adopt this model of secrecy and study it by using a Unifying Decision-Making Framework (UDMF) which has been recently developed. UDMF is a generic decision-making framework for discrete-event systems, which has been shown to generalize supervisory control, diagnosis and prognosis of discrete-event systems, which are thus particular instances of UDMF. Following the same idea, we express preservation of secrecy as a particular case of UDMF. We take advantage of results of UDMF to obtain results in secrecy preservation.

 A non-deterministic labeled finite automaton is initial-state opaque if the membership of its true initial state to a set of secret states S remains opaque (i.e., uncertain) to an intruder who observes system activity through some natural projection map. The verification of initial-state opacity has been shown to be a PSPACE-complete problem by establishing that it is equivalent to the language containment problem. In this paper we take a slightly different viewpoint and try to assess the ability of a user (who is dictating the activity in the system and indirectly the observations generated) to avoid revealing to the outside observer that the initial state of the system lied within the set of secret states S. A system that does not allow the user to act indefinitely in such a way is said to posses the property of resolution of initial state with respect to S. We show that in discrete event systems that can be modeled as non-deterministic labeled finite automata, this property can be verified with polynomial complexity in a way that resembles the verification of diagnosability.

 This paper reports a novel approach for the supervisory control of discrete event systems. Based on components, the approach provides principles of object-oriented software design to be used in the framework of Supervisory Control Theory. The use of abstract and concrete components allows the modeling of complex systems at a high level of abstraction, making specification and control design easier. Besides aiming modularity and reusability, the proposed framework allows to introduce concepts of composition, polymorphism and inheritance into the design of supervisory controllers.

 The problem of the practical p-th mean stability of a class of stochastic nonlinear and time dependent linear hybrid systems with practically p-mean stable and unstable structures is considered. Sufficient conditions for the practical p-th mean stability under the stabilizing switching signal using Lyapunov techniques, the practical average dwell time and the practical dwell time approach, are derived. Three cases including the existence of the single Lyapunov function, the multiple Lyapunov function and the single practical Lyapunov-like function, for the practical p-th mean stability are discussed. The obtained results are illustrated by an example.

 By considering arbitrary Holder p-norms in the state space, the current paper studies the existence of exponentially contractive sets that are invariant with respect to the trajectories of discrete-time switching linear systems. We first provide necessary and sufficient conditions for testing if state-space sets (characterized by certain attributes such as shape, scaling factor and decreasing rate) are invariant with respect to the trajectories of a given switching system. Next, we synthesize state feedbacks so as the sets characterized by certain attributes are invariant with respect to the trajectories of the closed-loop switching system. For the usual Holder p-norms corresponding to $p in { 1, 2, infty } $ we also discuss the numerical tractability of the theoretical results. A numerical example is included for practical illustration.