Queue-Aware Distributive Resource Control for Delay-Sensitive Two-Hop MIMO Cooperative Systems

摘要

In this paper, we consider a queue-aware distributive resource controlalgorithm for two-hop MIMO cooperative systems. We shall illustrate that relaybuffering is an effective way to reduce the intrinsic half-duplex penalty incooperative systems. The complex interactions of the queues at the source nodeand the relays are modeled as an average-cost infinite horizon Markov DecisionProcess (MDP). The traditional approach solving this MDP problem involvescentralized control with huge complexity. To obtain a distributive and lowcomplexity solution, we introduce a linear structure which approximates thevalue function of the associated Bellman equation by the sum of per-node valuefunctions. We derive a distributive two-stage two-winner auction-based controlpolicy which is a function of the local CSI and local QSI only. Furthermore, toestimate the best fit approximation parameter, we propose a distributive onlinestochastic learning algorithm using stochastic approximation theory. Finally,we establish technical conditions for almost-sure convergence and show thatunder heavy traffic, the proposed low complexity distributive control is globaloptimal.