TY - JOUR
T1 - Intelligent Reflecting Surface Backscatter Enabled Multi-Tier Computing for 6G Internet of Things
AU - Xu, Sai
AU - Liu, Jiajia
AU - Kato, Nei
AU - Du, Yanan
N1 - Publisher Copyright:
© 1983-2012 IEEE.
PY - 2023/2/1
Y1 - 2023/2/1
N2 - This paper investigates a novel framework of intelligent reflecting surface (IRS) backscatter enabled multi-tier computing system. In such a hierarchical network, the data bits of computational task requested by each user equipment (UE) are broken up into three parts, which are respectively computed at tier-1 UEs, tier-2 access points (APs) and a tier-3 central server. Distinguished from conventional active antennas, IRS backscatter at the UEs is leveraged to offload data bits to the APs. Based on the established network framework, an optimization problem is formulated, which aims at maximizing the sum computational bits of system during the considered time block by jointly optimizing the active beamforming at the power beacon, the passive beamforming at the UEs, the active beamforming at the APs, the bandwidth and power allocation among all the UEs, as well as the time of local computing. To seek the optimal solution, the optimization problem is decomposed into two, namely the maximization of stage-1 sum computational bits and the minimization of stage-2 delay. By the objective function conversion and alternative optimization methods, the two problems are addressed. Extensive simulations are performed to confirm the feasibility of the proposed system and show the achievable performance in processing computational bits.
AB - This paper investigates a novel framework of intelligent reflecting surface (IRS) backscatter enabled multi-tier computing system. In such a hierarchical network, the data bits of computational task requested by each user equipment (UE) are broken up into three parts, which are respectively computed at tier-1 UEs, tier-2 access points (APs) and a tier-3 central server. Distinguished from conventional active antennas, IRS backscatter at the UEs is leveraged to offload data bits to the APs. Based on the established network framework, an optimization problem is formulated, which aims at maximizing the sum computational bits of system during the considered time block by jointly optimizing the active beamforming at the power beacon, the passive beamforming at the UEs, the active beamforming at the APs, the bandwidth and power allocation among all the UEs, as well as the time of local computing. To seek the optimal solution, the optimization problem is decomposed into two, namely the maximization of stage-1 sum computational bits and the minimization of stage-2 delay. By the objective function conversion and alternative optimization methods, the two problems are addressed. Extensive simulations are performed to confirm the feasibility of the proposed system and show the achievable performance in processing computational bits.
KW - backscatter
KW - intelligent reflecting surface (IRS)
KW - Multi-tier computing system
KW - optimization
KW - sum computational bits
UR - http://www.scopus.com/inward/record.url?scp=85146240308&partnerID=8YFLogxK
U2 - 10.1109/JSAC.2022.3231861
DO - 10.1109/JSAC.2022.3231861
M3 - 文章
AN - SCOPUS:85146240308
SN - 0733-8716
VL - 41
SP - 320
EP - 333
JO - IEEE Journal on Selected Areas in Communications
JF - IEEE Journal on Selected Areas in Communications
IS - 2
ER -