Energy-Efficient Content Pushing based on Rate and Power Adaptation with Delay Constraints

Proactive caching is capable of offloading data traffic during on-peak hours and improving system capacity, thereby attracting much attention recently. However, the energy efficiency (EE) gain brought by proactive caching has not been fully exploited yet in fading channels. In this paper, we present energy-efficient content pushing methods in fading channels, while guaranteeing the average delay or delay-outage constraint. In particular, a general rate and power adaptation scheme is adopted to support content pushing and on-demand delivery, helping to improve the EE. To maximize the EE, we derive the average delay, delay-outage probability, and EE for any given adaptive policy. Based on these results, EE maximization problems are formulated for a specific adaptive scheme, namely, the water-filling scheme. Formulated problems are then converted into one-dimensional optimization problems to reduce computational complexity. Simulations indicate a request probability threshold for content pushing, above which the content pushing performs better than the traditional on-demand transmission.