$\mathcal k$-branching uio sequences for partially specified observable non-deterministic fsms

In black-box testing, test sequences may be constructed from systems modelled as deterministic finite-state machines (DFSMs) or, more generally, observable non-deterministic finite state (ONFSMs). Test usually contain identification sequences, with unique input output ( UIOs ) often being used DFSMs. This paper extends the notion of to ONFSMs. One challenge is that, a result non-determinism, application an sequence can lead exponentially many expected sequences. To address this scalability problem, we introduce ${\mathcal K}$ - : that at most states notation="LaTeX">$M$ . We show checking UIO existence PSPACE-Complete if problem suitably bounded; otherwise it in EXPSPACE and PSPACE-Hard. provide massively parallel algorithm for constructing results experiments on randomly generated real FSM specifications. The proposed was able construct cases where existing generation could not FSMs 38K 400K transitions.