Multi-data Delivery Based on Network Coding in On-demand Broadcast

On-demand data broadcast is widely deployed to achieve high scalability in a mobile computing environment. However, traditional on-demand data broadcasting assumes that each time slot includes only one data item. Therefore, the above constraint requires the mobile users to wait until the next broadcast cycle to retrieve a data item if they miss the item in this cycle. The above constraint also limits the number of users that can be served in each time slot. In this paper, we propose a new on-demand data broadcast model with modified network coding. Our approach enables a server to encode multiple data items in each time slot, while each mobile user retrieves the data items by decoding the encoded data items with the locally stored data items. Our approach is different from the traditional network coding because each time slot encodes only a subset of data items, which are decided according to identities of the requested and stored data items of users. The simulation results show that our algorithm can reduce the access time by 66% as compared to the traditional schemes.