Polynomial Model-Based Eye Diagram Estimation Methods for LFSR-Based Bit Streams in PRBS Test and Scrambling

This paper proposes eye diagram estimation methods for linear feedback shift register (LFSR)-based bit streams in pseudorandom binary sequence (PRBS) test and scrambling. The PRBS test uses the LFSR as a random data source; scrambling suppresses the radiated EMI by exclusive-OR (XOR) between the data and the LFSR. Both cases include the LFSR as the last block, thus, the LFSR dominantly determines the eye diagram. This paper introduced the deterministic and statistical eye diagram for the PRBS test and scrambling, respectively. The deterministic eye diagram was verified by comparing to the measurement; the statistical eye diagram was verified by comparing to the transient simulation. We also compared the voltage bathtub curves for the scrambling. The measurement and proposed method were correlated up to the BER of 10^-10.