TY - JOUR
T1 - Robust Rank-Two Multicast Beamforming under a Unified CSI Uncertainty Model
AU - Liu, Binyue
AU - Shi, Long
AU - Xia, Xiang Gen
N1 - Publisher Copyright:
© 1994-2012 IEEE.
PY - 2016/10
Y1 - 2016/10
N2 - We study a beamformed Alamouti coding scheme for a multigroup multicast downlink system. In this letter, we take into account imperfect channel state information (CSI) at the base station, where the CSI uncertainty region is formed by the intersection of multiple ellipsoids. This unified CSI uncertainty model leads to a challenging robust rank-two multicast beamforming problem. We first develop a conservative approximation method to handle the intractable constraints in the formulated robust beamforming problem under the considered CSI uncertainty region. Then, we apply the semidefinite relaxation (SDR) technique to approximately solve this problem. Our rank-profile analysis shows that the SDR is tight when there are at most two users in each multicast group and the CSI uncertainty region is not too large. Finally, we carry out simulations to validate the effectiveness of the proposed conservative approximation method. The simulation results coincide with the rank-profile analysis.
AB - We study a beamformed Alamouti coding scheme for a multigroup multicast downlink system. In this letter, we take into account imperfect channel state information (CSI) at the base station, where the CSI uncertainty region is formed by the intersection of multiple ellipsoids. This unified CSI uncertainty model leads to a challenging robust rank-two multicast beamforming problem. We first develop a conservative approximation method to handle the intractable constraints in the formulated robust beamforming problem under the considered CSI uncertainty region. Then, we apply the semidefinite relaxation (SDR) technique to approximately solve this problem. Our rank-profile analysis shows that the SDR is tight when there are at most two users in each multicast group and the CSI uncertainty region is not too large. Finally, we carry out simulations to validate the effectiveness of the proposed conservative approximation method. The simulation results coincide with the rank-profile analysis.
KW - Conservative approximation
KW - multigroup multicast
KW - rank-profile analysis
KW - robust rank-two beamforming
UR - http://www.scopus.com/inward/record.url?scp=84990209088&partnerID=8YFLogxK
U2 - 10.1109/LSP.2016.2600618
DO - 10.1109/LSP.2016.2600618
M3 - Article
AN - SCOPUS:84990209088
SN - 1070-9908
VL - 23
SP - 1419
EP - 1423
JO - IEEE Signal Processing Letters
JF - IEEE Signal Processing Letters
IS - 10
M1 - 7544524
ER -