SMDP-based resource allocation for video streaming in cognitive vehicular networks

Video streaming over vehicular network is playing a more and more important role in our daily life. In this paper, considering a cognitive radio-enabled roadside infrastructure network, which is potentially of flexible spectrum, a joint call admission control and channel allocation scheme is proposed to facilitate the transmission of scalable video coding (SVC)-based video service for vehicle users. To capture the dynamic of vehicle users´ mobility and the change of cognitive bands´ availability, we formulate the joint resource allocation problem as a semi-Markov decision process (SMDP). Specifically, taking into account the system state consisting of the spatial distribution of vehicle users with different numbers of video layers and the availability of the cognitive bands, the optimal resource allocation is addressed to maximize a long-term reward, which aims at striking a good balance between the two key aspects of video quality, namely, high peak signal-to-noise ratio (PSNR) and smooth playback. Simulation results show the effectiveness of the proposed scheme in terms of improving both aforesaid aspects of video quality for vehicle users in a vehicular network.