Distributed rotation recovers spatial diversity

In relay networks, a conventional way to exploit spatial diversity is to introduce distributed space-time processing at the relays. In our work, we show that even simple time-varying distributed rotation can recover spatial diversity. The main idea is to convert the inherent spatial diversity to time diversity by creating an artificial fast fading channel. It turns out that the proposed framework is both tractable from the theoretical point of view and simple from the practical point of view. Furthermore, the framework is quite general and can be applied to a wide range of linear/nonlinear relaying strategies. As applications, we first propose a linear relaying scheme called rotate-and-forward for multiple-antenna two-hop layered networks. It is shown that, in some non-trivial setting, this scheme outperforms existing schemes and achieves the optimal diversity-multiplexing tradeoff. The second application is a decode-and-forward scheme based on the same idea in the single-antenna multiple-relay channel. It is shown to achieve the maximum diversity with low signaling complexity.