The thesis proposes a new multicast scheme which combines scalable video with FMS (flexible multicast service) for handling user heterogeneity problem in video multicast. In scalable video, content may be provided in a variable format depending on the service type. FMS is a service that the source separates the multicast streams and transmits respective streams at specific time, when the source has multiple multicast streams to transmit. In video multicast, it is difficult to satisfy the requirements of all users, because they differ in channel conditions, available energy capacity, preference between video quality and energy consumption, etc. In the proposed scheme, the BL (base layer) and the EL (enhancement layer) of the scalable video are separated into different streams and then transmitted. The BL and the EL streams have different transmission times and users can selectively receive those packets. As a consequence, users can receive services that are better tailored for individual users. Noting that multicast is unreliable in general because retransmission is not used, we transmit the coded packets by using network coding. We define the utility function to reflect the video quality, energy consumption and the preference of users. At the user side, users determine their actions to maximize their own utility considering their situation when EL packets are transmitted. At the source side, the source determines the PHY rate and the number of EL packets to maximize the sum of all user’s utilities after determining user’s behavior. Through simulations, we have confirmed that the proposed scheme has about 15 % to 34 % gain in utility over the legacy IEEE 802.11 multicast scheme and alleviates the user heterogeneity problem.
PDF
download
878987797
028-85220240
