Multiple-objective backtrace for solving test generation constraints

Current paper presents a new method for solving hierarchical test generation constraints. Unlike most of the previous solvers, the proposed approach relies on the low-level model description of the constraints. The low-level model used is structurally synthesized BDDs, which are generated directly from the gate-level descriptions of the hierarchical modules. The method has been implemented and verified on realistic test generation constraints. Preliminary experiments show that both, the proposed constraint solving engine and the decision diagram based hierarchical test generator achieve a very fast performance in comparison to well-known test tools.