Towards scalable and privacy preserving commercial content dissemination in social wireless networks

This paper proposes a Q-learning based Device-to-Device multicast routing framework for Social Wireless Networks. The goal of the proposed Scalable Q-learning based Gain-aware Routing (SQGR) content dissemination algorithm is to maximize a predefined economic gain for commercial content generators. This economic gain is defined as the revenue from delivery of a coupon minus the forwarding cost associated with that delivery. SQGR, with its embedding learning abilities, is expected to be robust in dynamic mobility environments. It also preserves scalability and privacy since it does not require storage of per-individual consuming interest and interaction profiles within the network. Using the DTN simulator software ONE, we evaluate functional validity and compare gain performance of SQGR with few existing protocols under various commercial, network and protocol parameters.