“…When a system is verified to be non-opaque, one important problem is to enforce opacity via some mechanisms. In general, there are two approaches for enforcing opacity: one is to control the actual behavior of the system so that those secret-revealing behaviors can be avoided [1], [4], [14], [16], [34], [41], [43], [50], [56] the other one is to change the information-flow of the system so that the intruder can be "cheated" or be "confused" [3], [10], [19], [20], [28], [44], [45], [52], [54]. In particular, the first approach is essentially the supervisory control of opacity that aims to find a supervisor that restricts the behavior of the system dynamically such that the closed-loop system is opaque.…”