Adaptive Power and Rate Control for Mixed Proactive Pushing and On-demand Traffic: A CMDP Approach

Changkun Li,Wei Chen
DOI: https://doi.org/10.1109/globecom46510.2021.9685341
2021-01-01
Abstract:Proactive pushing has been recognized as a promising solution to support the dramatically increasing global data traffic, thereby gaining much attention recently. By proactively pushing popular files to users prior to their requests, the data traffic load over peak hours can be relieved, leading to substantially reduced content access latency. However, extra power consumption is incurred due to pushing. How to efficiently schedule proactive pushing and on-demand transmission becomes an important problem. In this paper, we present a unified framework for joint proactive pushing and on-demand transmission. Based on the joint proactive pushing and on-demand transmission scheme, we minimize the average content queueing latency through a Constrained Markov Decision Process (CMDP) approach while satisfying a power constraint. Through linear programming (LP) formulation, we obtain the optimal delay-power tradeoff of the traditional on-demand transmission only scheme and the joint pushing and on-demand transmission scheme, respectively. Finally, simulation results present that joint scheduling of pushing and on-demand transmission significantly reduces the content queueing delay.
What problem does this paper attempt to address?