Joint Chance-Constrained Predictive Resource Allocation for Energy-Efficient Video Streaming

摘要: Predictive resource allocation (PRA) techniques that exploit knowledge of the future signal strength along roads have recently been recognized as promising approaches to save base station (BS) energy and improve user quality service (QoS). Recent studies on human mobility patterns wireless measurements buses trains indeed supported practical potential PRA. An unresolved challenge, however, is modeling uncertainty in predictions, developing real-time robust solutions incorporate probabilistic QoS guarantees. This paramount importance PRA due prediction time horizon adds considerable complexity increases rate problem. With these developments mind, this paper addresses energy-efficient applied stored video streaming using chance constrained programming. The proposed solution incorporates: 1) uncertainty predicted rates; 2) a joint level constraint satisfaction over a horizon; 3) both optimal gradient-based guided heuristic solutions. Our framework fundamentally differs from previous work literature where nonstochastic with assumptions perfect were primarily used demonstrate energy savings gains. Numerical simulations based standard compliant long term evolution (LTE) system are provided examine compare developed solution. Unlike existing PRA, achieves desired under imperfect channel predictions. robustness attained without compromising energy-efficiency compared opportunistic schedulers, thus supports implementation practice.