TY - JOUR
T1 - Enhancing Reflectarray Robustness Through Adjoint Optimization Enabled Recovery
AU - Chaky, Ryan J.
AU - Jenkins, Ronald P.
AU - Campbell, Sawyer D.
AU - MacKertich-Sengerdy, Galestan
AU - Werner, Douglas H.
N1 - Publisher Copyright:
© 1963-2012 IEEE.
PY - 2024/4/1
Y1 - 2024/4/1
N2 - Antenna array failures have been investigated since the early 1990s, often making use of array factor (AF) theory to accelerate the analysis. However, AF modeling is not always appropriate for reflectarray antennas (RAs) due to their operation in a scattering mode rather than being a driven antenna. Therefore, full-wave analysis is required to model RAs with failures. Unfortunately, this places a tremendous computational burden on any attempt at reoptimizing the RA to heal its performance. To overcome this issue, we introduce here an efficient method that exploits adjoint optimization (AO) to make RA healing tractable. To demonstrate the approach, a 25×25 element RA is analyzed under multiple failure scenarios to determine the expected gain recovery after optimization. The proposed optimization framework achieves a 550× speed improvement over conventional approaches, thus making optimization-based RA healing tractable.
AB - Antenna array failures have been investigated since the early 1990s, often making use of array factor (AF) theory to accelerate the analysis. However, AF modeling is not always appropriate for reflectarray antennas (RAs) due to their operation in a scattering mode rather than being a driven antenna. Therefore, full-wave analysis is required to model RAs with failures. Unfortunately, this places a tremendous computational burden on any attempt at reoptimizing the RA to heal its performance. To overcome this issue, we introduce here an efficient method that exploits adjoint optimization (AO) to make RA healing tractable. To demonstrate the approach, a 25×25 element RA is analyzed under multiple failure scenarios to determine the expected gain recovery after optimization. The proposed optimization framework achieves a 550× speed improvement over conventional approaches, thus making optimization-based RA healing tractable.
UR - http://www.scopus.com/inward/record.url?scp=85186076085&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85186076085&partnerID=8YFLogxK
U2 - 10.1109/TAP.2024.3365276
DO - 10.1109/TAP.2024.3365276
M3 - Article
AN - SCOPUS:85186076085
SN - 0018-926X
VL - 72
SP - 3362
EP - 3373
JO - IEEE Transactions on Antennas and Propagation
JF - IEEE Transactions on Antennas and Propagation
IS - 4
ER -