부산대학교 도서관

Although multiple edge responses method has been proven effective to capture the input and output nonlinearity in a channel simulation, the method may not be applicable for memory simulation due to the nonlinear crosstalk effect and also the requirement of a massive number of step responses, which significantly reduces the simulation efficiency. In this work, we present an improved simulation methodology involving multiple edge responses to accurately estimate the worst-case eye diagram, specifically applicable for memory simulation. The proposed method is able to capture the nonlinearity effect due to crosstalk, plus has nearly 20 times simulation acceleration compared with the traditional long random input bit pattern simulation, based on a multiple processors system.