A Buffer Aware Resource Allocation Framework for Video Streaming over LTE

Abstract

Latency sensitive high bandwidth multimedia data is expected to capture more than $70\%$ of the available bandwidths in LTE and future wireless networks. This is expected to pose enormous challenges on the radio resource and management mechanisms in these networks. One of the most important factors that diminish the quality of viewing experience of delivered videos is frequent client side rebuffering events. This paper proposes a buffer-aware downlink scheduling framework (called BA-TLS) for streaming video applications. The problem of scheduling a set of video streams over LTE has first been formulated as an Integer Linear Programming (ILP) problem. It has been shown that a conventional Dynamic Programming (DP) solution for the ILP imposes prohibitive online overheads. Then, by utilizing analytical properties of the problem, a genetic algorithm based stochastic solution has been implemented. Further, a fast and efficient deterministic heuristic known as Proportionally Balanced Robustness-level Allocator (PBRA) has been implemented over the proposed framework. Experimental results show that while the optimal DP based solution performs significantly better at reducing rebuffering events with respect to the stochastic as well as the deterministic solutions, both of them incur much lower temporal overheads compared to DP. Further, it may be observed that although the stochastic and deterministic solutions are comparable in performance, the deterministic heuristic solution PBRA is about 300 to 600 times faster on average compared to its stochastic counterpart.