Intermediate Performance of Rateless Codes over Dying Erasure Channel

In this paper, we propose a new rateless coding scheme for dying erasure channels with random channel lengths and time-varying packet error rates. Firstly, we propose a heuristic approach for suboptimal degree distributions based on AND-OR tree technique to achieve higher intermediate performance. Secondly, the optimal code length for the maximum average date delivery ratio is derived and analyzed. Simulation results demonstrate that our coding scheme outperforms existing conventional rateless codes with significantly better performance, which can be employed as a solution to the efficient large bulk transmission on dynamical and unreliable channels.