Sleeping Multi Armed learning For Fast Uplink Grant Allocation Using SWUCB In Machine Type Communications

Abstract

multi-armed bandits (MABs) along with Sliding window UCB  is used for fast uplink grant allocation in Machine type communication . The set of active MTDs changes over time due to the knowledge on the set of active MTDs is probabilistic, so that SWUCB is  maximize the defined metric parameters like bit rate, signaling over head. Analyse the regret as well as effect of the prediction error of the source traffic prediction algorithm on the performance of the proposed sleeping MAB algorithm is investigated. Moreover, to activate  fast uplink allocation for multiple MTDs at each time, so concept of best MTD  ordering in the MAB setting. Simulation results show that the proposed framework Sliding window UCB yields good results in terms of regret, maximum tolerable access delay, throughput compared to traditional methods.