Performance Analysis of Push-Based Converged Networks With Limited Storage

Abstract

By pushing popular contents directly to users through broadcast networks while serving individual requests using cellular networks, a wireless converged network provides a highly effective means to cope with the wireless traffic bottleneck. In this paper, we analyze the network capacity and the pushing and caching schemes for the push-based converged network with limited user storage. Both uniform and non-uniform storage cases are investigated. Specifically, we first derive the converged network capacity under the traditional popularity-based pushing and caching scheme. Furthermore, we establish push-limited and cache-limited regions for the uniform case and an additional transitional region for the non-uniform case, to reveal the constraining factor on the network capacity growth. Moreover, the optimal pushing and caching schemes are investigated from both the network and user perspectives. It is shown that both network and user criterion reduce to the same offloading optimization problem, which can be modeled as the 0-1 knapsack problem and effectively solved by the greedy algorithm. We point out that the optimal scheme depends on not only the traffic load, but also the region where the system operates. Finally, the numerical results are provided to confirm the accuracy of the developed analytical results.