Performance Analysis of KMV-Cast with Imperfect Prior Knowledge

Abstract

It is predicted that the global mobile data traffic will increase nearly eightfold in the next five years, and the cloud applications will consist of 90 percent of total mobile data traffic by the end of 2019. There is no doubt that the increasing high quality video requirements and huge similar information stored in cloud servers enforce to setup a new communication paradigm. In this paper, we will take a brief review over the knowledge-enhanced mobile video broadcasting (KMV-Cast) framework, and analyze its performance with imperfect prior knowledge, which means that the prior knowledge will be transmitted through Gaussian channel. The simulation results have shown that KMV-Cast with imperfect prior knowledge performs the worst compared with the other two schemes in terms of PSNR. It indicates that the prior knowledge is very important for the video reconstruction in KMV-Cast. Meanwhile, with the SNR increasing for prior knowledge transmission, the performance still performs poor and has an optimum point.