Ultra-dense heterogeneous networks (Ud-HetNets) have been put forward toimprove the network capacity for next-generation wireless networks. However,counter to the 5G vision, ultra-dense deployment of networks wouldsignificantly increase energy consumption and thus decrease network energyefficiency suffering from the conventional worst-case network designphilosophy. This problem becomes particularly severe when Ud-HetNets meet bigdata because of the traditional reactive request-transmit service mode. In viewof these, this article first develops a big-data-aware artificial intelligentbased framework for energy-efficient operations of Ud-HetNets. Based on theframework, we then identify four promising techniques, namely big dataanalysis, adaptive base station operation, proactive caching, andinterference-aware resource allocation, to reduce energy cost on both large andsmall scales. We further develop a load-aware stochastic optimization approachto show the potential of our proposed framework and techniques in energyconservation. In a nutshell, we devote to constructing green Ud-HetNets of bigdata with the abilities of learning and inferring by improving the flexibilityof control from worst-case to adaptive design and shifting the manner ofservices from reactive to proactive modes.
展开▼
