TY - GEN
T1 - Tuning Algorithm to Compensate for Hardware Errors of IBFD Analog Device
AU - Morgenstern, Carl W.
AU - Rong, Yu
AU - Herschfelt, Andrew
AU - Molnar, Alyosha C.
AU - Apsel, Alyssa B.
AU - Landon, David G.
AU - Bliss, Daniel W.
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - This study investigates processing techniques to alleviate the impact of hardware imperfections for in-band full-duplex (IBFD) systems. The studied IBFD model considers a multi-tap digitally controlled analog filter to provide RF cancellation. With precise optimization, non-negative tap weights reconstruct the multi-path channel to mitigate the self-interference (SI). In our simulation study, the SI cancellation significantly degrades when the initial tap weights are optimized without accounting for the timing errors induced by hardware imperfections. Therefore, we develop a new adaptive algorithm to tune the tap weights and recover the performance degradation caused by the timing errors. Our results indicate that the proposed data-driven method compensates for the timing offsets. Additionally, by considering the finite bit depth of the non-negative tap weights the performance improves compared to when there are no timing errors present.
AB - This study investigates processing techniques to alleviate the impact of hardware imperfections for in-band full-duplex (IBFD) systems. The studied IBFD model considers a multi-tap digitally controlled analog filter to provide RF cancellation. With precise optimization, non-negative tap weights reconstruct the multi-path channel to mitigate the self-interference (SI). In our simulation study, the SI cancellation significantly degrades when the initial tap weights are optimized without accounting for the timing errors induced by hardware imperfections. Therefore, we develop a new adaptive algorithm to tune the tap weights and recover the performance degradation caused by the timing errors. Our results indicate that the proposed data-driven method compensates for the timing offsets. Additionally, by considering the finite bit depth of the non-negative tap weights the performance improves compared to when there are no timing errors present.
UR - http://www.scopus.com/inward/record.url?scp=85150198046&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85150198046&partnerID=8YFLogxK
U2 - 10.1109/IEEECONF56349.2022.10051934
DO - 10.1109/IEEECONF56349.2022.10051934
M3 - Conference contribution
AN - SCOPUS:85150198046
T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers
SP - 1228
EP - 1232
BT - 56th Asilomar Conference on Signals, Systems and Computers, ACSSC 2022
A2 - Matthews, Michael B.
PB - IEEE Computer Society
T2 - 56th Asilomar Conference on Signals, Systems and Computers, ACSSC 2022
Y2 - 31 October 2022 through 2 November 2022
ER -