IRS-Based MEC for Delay-Constrained QoS Over RF-Powered 6G Mobile Wireless Networks

Mobile-edge computing (MEC) and intelligent reflecting surface (IRS) have been recognized as two key technologies for 6G mobile networks. Consequently, we propose the IRS-based MEC schemes for the delay-constrained quality-of-service (QoS) provisioning over radio frequency (RF) powered 6G mobile networks with
non-linear
energy harvesting (EH) model. Using the IRS, multiple mobile users (MUs) first harvest energy from a multi-antenna base station (BS) equipped with an MEC server and then transmit their data to the BS for data-processing. First, we formulate a delay minimization problem for MUs under their QoS requirements, by jointly optimizing the IRS's phase-shift matrices, the MEC server's finite computation resource allocation, the MIMO based BS's multi-MU detection (MUD) coefficients, and the energy/data transmission time and task allocation coefficient of each MU. Second, we define the total delay of each MU as the sum of its EH time, data-transmission time, and data-processing time. Since our formulated joint-optimization problem is non-convex with multiple coupled variables, we apply the
block coordinate descending
(BCD) method to decompose it into several subproblems which can thus be iteratively solved by a low complexity algorithm. Third, we extend our proposed scheme to IRS-based MEC over Terahertz (THz) mobile networks. Finally, we validate and evaluate our developed delay-minimization schemes through numerical analyses, showing that the total delay of the RF-powered MUs can be significantly reduced by using our proposed schemes.