Skip to Main Content
We formulate the problem of network-coding (NC)-based scheduling for media transmission to multiple users over a wireless-local-area-network-like or WiMAX-like network as a Markov decision process (MDP). NC is used to minimize the packet losses that resulted from unreliable wireless channel conditions, whereas the MDP is employed to find the optimal policy for transmissions of unequally important media packets. Based on this, a dynamic programming technique is used to give an optimal transmission policy. However, this dynamic programming technique quickly leads to computational intractability, even for scenarios with a moderate number of receivers. To address this problem, we further propose a simulation-based dynamic programming algorithm that has a much lower run time yet empirically converges quickly to the optimal solution.
Date of Publication: March 2011