TY - GEN
T1 - Determining worst-case eye height in low BER channels using Bayesian optimization
AU - Dolatsara, Majid Ahadi
AU - Swaminathan, Madhavan
N1 - Funding Information:
ACKNOWLEDGMENT This work has been supported by the DARPA CHIPS project under the award number N00014-17-1-2950.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/2
Y1 - 2020/2
N2 - Eye diagram simulation and bit error rate (BER) estimation is an essential task in signal integrity. A lengthy time domain simulation is required for non-LTI systems where statistical methods are generally inaccurate. However, with the BER reaching less than 10-12, and with exponential increase in bandwidth, this task is expected to become more challenging and exorbitantly time consuming. In particular, the concern is with inter-symbol interference (ISI) effect, which can be caused by the state of several earlier bits. Therefore, this paper suggests an optimization method to find the bit patterns causing the lowest received high symbol, and the highest received low symbol, at the sampling time point. Difference of these values can be used to estimate the worst-case eye height. The proposed approach is based on a mapping method and Bayesian optimization, which provides a significant speedup compared to the traditional transient eye. This optimization technique is capable of solving both non-linear and non-convex problems. A numerical example is provided to show performance of the proposed approach.
AB - Eye diagram simulation and bit error rate (BER) estimation is an essential task in signal integrity. A lengthy time domain simulation is required for non-LTI systems where statistical methods are generally inaccurate. However, with the BER reaching less than 10-12, and with exponential increase in bandwidth, this task is expected to become more challenging and exorbitantly time consuming. In particular, the concern is with inter-symbol interference (ISI) effect, which can be caused by the state of several earlier bits. Therefore, this paper suggests an optimization method to find the bit patterns causing the lowest received high symbol, and the highest received low symbol, at the sampling time point. Difference of these values can be used to estimate the worst-case eye height. The proposed approach is based on a mapping method and Bayesian optimization, which provides a significant speedup compared to the traditional transient eye. This optimization technique is capable of solving both non-linear and non-convex problems. A numerical example is provided to show performance of the proposed approach.
UR - http://www.scopus.com/inward/record.url?scp=85084289367&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85084289367&partnerID=8YFLogxK
U2 - 10.1109/LASCAS45839.2020.9069049
DO - 10.1109/LASCAS45839.2020.9069049
M3 - Conference contribution
AN - SCOPUS:85084289367
T3 - 2020 IEEE 11th Latin American Symposium on Circuits and Systems, LASCAS 2020
BT - 2020 IEEE 11th Latin American Symposium on Circuits and Systems, LASCAS 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th IEEE Latin American Symposium on Circuits and Systems, LASCAS 2020
Y2 - 25 February 2020 through 28 February 2020
ER -