Advanced MAB Schemes for WiGig-Aided Aerial Mounted RIS Wireless Networks.

This paper uses an aerial mounted RIS (ARIS) to assist WiGig base station (BS) in serving mobile equipments (MEs) located within hotspot zones. The aerial should cover numerous large-capacity hotspots in this context while anticipating its flying/hovering energy expenditures. Hence, two advanced multi armed bandit (MAB) approaches, i.e. perturbed history exploration (PHE) and mini-max optimal Thompson sampling (MOTS), are envisioned as applicable self-learning methodologies to deal with such a problem effectively. Simulation results ensure the excellent performance of the envisioned schemes over naive upper confidence bound (UCB) and Thompson sampling (TS) algorithms and traditional heuristic solutions.