Aliasing in signature analysis testing with multiple-input shift-registers

Signature analysis with multiple-input shift registers (MISRs) is often used to realize efficient built-in self-test of digital VLSI circuits. The authors present a statistical theory that explains the dependence of aliasing probability on the main MISR features, such as length and feedback network, and thus makes it possible to prove criteria for the MISR design. The assumption of independent errors at the register inputs is used to model the register behavior as a Markov process, whose equations are then solved to obtain the exact dependence of aliasing probability as a function of test length, input error probabilities, and feedback structure.<
>