Interference alignment with asymmetric complex signaling and multiuser diversity

In this paper, we consider a downlink transmission technique for multi-cell and multi-user interfering broadcast channels (IFBCs). In this IFBC model, the simplest way to manage inter-cell interference is an orthogonal access method, such as time-division multiple-access (TDMA). Recently, a novel idea of an interference management scheme named interference alignment (IA) has been introduced by Cadambe and Jafar, which requires symbol extension to guarantee a degree-of-freedom (DOF) gain. In this paper, we propose a new IA scheme without symbol extension which outperforms the TDMA and the conventional IA by exploiting multi-user diversity and asymmetric complex signaling. We first introduce an exhaustive scheduling algorithm which shows the optimum performance. Then we propose a simplified suboptimum method to reduce the computational complexity and the search size.