Multi-User Small Base Station Association via Contextual Combinatorial Volatile Bandits

We propose an efficient mobility management solution to the problem of assigning small base stations (SBSs) multiple mobile data users in a heterogeneous setting. formalize using novel sequential decision-making model named contextual combinatorial volatile multi-armed bandits (MABs), which each association is considered as arm, volatility arm imposed by dynamic arrivals users, and context additional information linked with user SBS such user/SBS distance transmission frequency. As next-generation communications are envisioned take place over highly links millimeter wave (mmWave) frequency band, we consider unknown channel distribution limited feedback form acknowledgments under absence state (CSI). dynamically varying, assignment cannot be solved offline. Thus, online algorithm able solve user-SBS multi-user time-varying environment, where number varies time. Our strikes balance between exploration exploitation achieves sublinear time regret optimal dependence on structure dynamics departures. In addition, demonstrate via numerical experiments that our significant performance gains compared several benchmark algorithms.