A Markovian model for satellite integrated cognitive and D2D HetNets

Next-generation wireless systems are expected to provide bandwidth-hungry services in a cost-efficient and ubiquitous manner. D2D communications, spectrum sharing and heterogeneous network architectures (HetNets) are touted as crucial enablers to attain these goals. Moreover, the shifting characteristics of network traffic towards content consumption necessitate content-centric architectures and protocols. In this work, we propose a comprehensive analytical model for a content-oriented heterogeneous wireless network with cognitive capability. We model our HetNet architecture with a Continuous Time Markov Chain (CTMC) and characterize the trade-off between energy efficiency and system goodput. We elaborate on novel elements in our model, namely the integration of universal source concept (modeling the content retrieval operation from external networks), caching and overlaying in D2D mode. Besides, our investigation on network mode selection provides further insight on how resource allocation and performance are intertwined.