On Throughput Optimization and Bound Analysis in Cache-Enabled Fiber-Wireless Access Networks
share
The backhaul of fiber-wireless (FiWi) access networks is facing the bandwidth crunch due to the increasing demands for bandwidth-hungry applications. In this paper, we consider utilizing the caches in FiWi access networks to mitigate backhaul bottleneck and improve network throughput. As both the power budget in wireless access networks and the bandwidth of backhaul are constrained, it is challenging to allocate power for caching and wireless transmission properly to achieve superior system performance. Our goal is to maximize the downlink throughput by considering power allocation and caching strategy jointly. We first propose a volume-adjustable backhaul-constrained water-filling method (VABWF) to derive the optimal power allocation for wireless transmission. A dynamic programming algorithm is then designed by reformulating the problem as a multiple-choice knapsack problem (MCKP). To explore the potential of throughput in cache-enabled FiWi access networks, we derive a tractable expression of the ergodic capacity (EC) of wireless links and provide a theoretical upper bound of the throughput by utilizing the analytical properties of EC. We show that an appropriate allocation of transmission power and caching power is essential to achieve higher throughput. Numerical and simulation results validate our theoretical analysis and demonstrate the effectiveness of our proposed optimization method.
READ FULL TEXT