Queue-aware learning for scheduling in healthcare clouds

Abstract

This paper presents an adaptive algorithm for the scheduling of randomly deployed 60 GHz IEEE 802.11ad access points (APs) with the concept of stochastic message-passing in in-hospital medical healthcare cloud platforms. To formulate this scheduling problem, this paper uses max-weight independent set (MWIS) formulation where the weight is defined as the queue-backlog size; and then it approximately solves the problem with the theory of stochastic learning, i.e., stochastic message-passing.