TY - GEN
T1 - A dimension-reduced artificial neural network for the compact modeling of semiconductor devices
AU - Huang, Andong
AU - Zhong, Zheng
AU - Guo, Yong-Xin
AU - Wu, Wen
N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].
PY - 2018/6/29
Y1 - 2018/6/29
N2 - A dimension-reduced artificial neural network (DRANN) is proposed for the compact modeling of semiconductor devices. The fully connected feedforward artificial neural network (FFANN) is known for its universal fitting ability, however, measurement data are usually not sufficient to train the FFANN, for example, semiconductor devices are commonly measured under limited (2 or 3) ambient temperatures, which will cause overfitting problem if temperature is directly taken as the input of FFANN. In this paper, DRANN is proposed to address the overfitting problem for multi-dimensional data mapping. The dimensions (such as port voltages) with enough datasets are modeled by FFANN, whilst other dimensions (such as thermal and traps) with few datasets are represented by Taylor expansion. DRANN is finally a combination of multiple low-dimensional FFANNs. The DRANN is verified by the accurate prediction of pulsed I-Vs (PIVs) of GaN HEMT with various thermal and trap states. © 2018 IEEE.
AB - A dimension-reduced artificial neural network (DRANN) is proposed for the compact modeling of semiconductor devices. The fully connected feedforward artificial neural network (FFANN) is known for its universal fitting ability, however, measurement data are usually not sufficient to train the FFANN, for example, semiconductor devices are commonly measured under limited (2 or 3) ambient temperatures, which will cause overfitting problem if temperature is directly taken as the input of FFANN. In this paper, DRANN is proposed to address the overfitting problem for multi-dimensional data mapping. The dimensions (such as port voltages) with enough datasets are modeled by FFANN, whilst other dimensions (such as thermal and traps) with few datasets are represented by Taylor expansion. DRANN is finally a combination of multiple low-dimensional FFANNs. The DRANN is verified by the accurate prediction of pulsed I-Vs (PIVs) of GaN HEMT with various thermal and trap states. © 2018 IEEE.
KW - artificial neural network
KW - dimension reduction
KW - GaN HEMT
KW - Taylor expansion
KW - thermal
KW - trap
UR - http://www.scopus.com/inward/record.url?scp=85050383481&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85050383481&origin=recordpage
U2 - 10.1109/IEEE-IWS.2018.8400840
DO - 10.1109/IEEE-IWS.2018.8400840
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 9781538663462
T3 - 2018 IEEE MTT-S International Wireless Symposium, IWS 2018 - Proceedings
SP - 1
EP - 4
BT - 2018 IEEE MTT-S International Wireless Symposium, IWS 2018 - Proceedings
PB - IEEE
T2 - 2018 IEEE MTT-S International Wireless Symposium, IWS 2018
Y2 - 6 May 2018 through 9 May 2018
ER -