Cost efficient media streaming algorithms for rate-dependent pricing strategies in heterogeneous wireless networks

The future of wireless networking is envisioned as an integrated system of wireless radio access technologies with heterogeneous features. This heterogeneity combined with the characteristic diversity of provided services create promising opportunities for improving the utility of both operators and users. In this paper, we investigate the opportunity to reduce the average cost of streaming sessions by benefiting from the system embedded heterogeneity and streaming application buffering capability. First, we analyze the optimal streaming strategy for a theoretical infinite session. Based on this analysis, we propose pseudo-optimal and greedy-optimal adaptive media streaming algorithms for heterogeneous wireless networks. The performance of these algorithms is compared to a naive greedy streaming approach using NS2 simulations. The results show that the greedy-optimal algorithm reduces the average session cost down to 73.9% of the average cost incurred on using greedy algorithm. This cost saving is realized at an insignificant increase in signaling load and session blocking probability. Hence, we strongly recommend the developed greedyoptimal algorithm for media streaming in next-generation heterogeneous wireless networks.