Joint Uplink-Downlink Resource Allocation for Multiuser IRS-Assisted Systems

We investigate the joint uplink-downlink configuration of an intelligent reflecting surface (IRS) for multi-user frequency-division-duplexing (FDD) and time-division-duplexing (TDD) systems. This is motivated in FDD since uplink downlink transmissions occur simultaneously hence IRS must be jointly configured both transmissions. In TDD, while a design not strictly necessary, it can significantly reduce feedback overhead, power consumption, periods associated with updating IRS. To compute trade-off between rates achieved by design, weighted-sum problem formulated optimized using developed block-coordinate descent algorithm. The resulting regions are investigated numerical simulation to gain insights into different scenarios. all scenarios some TDD scenarios, outperforms fixed-uplink (fixed-downlink) heuristic (downlink) assist (uplink) Moreover, substantially bridges gap individual upper bound allowing configurations downlink. Otherwise, remaining fixed-downlink designs nearly achieve performance overhead and/or complexity compared design.