Optimal Bandwidth Allocation with Edge Computing for Wireless VR Delivery

摘要

In this paper, we propose a optimal bandwidth allocation scheme which takes advantage of the user computation capability in wireless VR delivery system. To avoid the redundant transmission, according to the similarity, the VR videos are divided into the shared part transmitted by multicast and the separate part transmitted by unicast. In particular, to minimize the maximum user transmission and computation delay, we formulate the bandwidth allocation problem into a convex problem which can be solved by KKT conditions. Moreover, we analyze the special case where the shared part equal to zero, it means that there is no similarity between the videos requested by each user. Compared with other bandwidth allocation schemes, numerical results show the advantage of the proposed scheme.