The approximate optimality of simple schedules for half-duplex multi-relay networks

In ISIT2012 Brahma, Özgür and Fragouli conjectured that in a half-duplex diamond relay network (a Gaussian noise network without a direct source-destination link and with N non-interfering relays) an approximately optimal relay scheduling (achieving the cut-set upper bound to within a constant gap uniformly over all channel gains) exists with at most N + 1 active states (only N + 1 out of the 2N possible relay listentransmit configurations have a strictly positive probability). Such relay scheduling policies are said to be simple. In ITW2013 we conjectured that simple relay policies are optimal for any halfduplex Gaussian multi-relay network, that is, simple schedules are not a consequence of the diamond network's sparse topology. In this paper we formally prove the conjecture beyond Gaussian networks. In particular, for any memoryless half-duplex N-relay network for which the cut-set bound is approximately optimal to within a constant gap under some conditions (satisfied for example by Gaussian networks), an optimal schedule exists with at most N + 1 active states. The key step of our proof is to write the minimum of a submodular function by means of its Lovasz extension and use the greedy algorithm for submodular polyhedra to highlight structural properties of the optimal solution. This, together with the saddle-point property of minmax problems and the existence of optimal basic feasible solutions in linear programs, proves the claim.