Redundancy minimizing techniques for robust transmission in wireless networks

In this paper, we consider a wireless multiple path network in which a transmitting node would like to send a message to the receiving node with a certain probability of success. These two nodes are separated by N erasure paths, and we devise two algorithms to determine minimum redundancy and optimal symbol allocation for this setup. We discuss the case with N =3 and then extend the case to an arbitrary number of paths. One of the algorithms minimum redundancy algorithm in exponential time is shown to be optimal in several cases, but has exponential running time. The other algorithm, minimum redundancy algorithm in polynomial time, is sub-optimal but has polynomial worst-case running time. These algorithms are based off the theory of maximum-distance separable codes. We apply the MRAET algorithm on maximum-distance separable, Luby transform, and Raptor codes and compare their performance.