Performance Assessment of Multi User Multi Cell Interference Alignment with Limited Feedback

Abstract

The optimum performance of a closed loop transmission scheme depends upon perfect feedback information at the transmitter. The perfect feedback requires unlimited resources in the reverse communication link. However, the feedback resources are limited in a practical system. Therefore, there is a trade off between the feedback load and the performance of transmission scheme. In this work, we evaluate the performance of multi user inter cell interference alignment (MUICIA) based transmit precoding scheme in a downlink MIMO system. It has been shown that in a system with unlimited feedback resources, MUICIA outperforms state of the art transmit precoding schemes. Here we assess the gains of MUICIA in a limited feedback uplink which is an important step in realization of this transmission scheme in practical systems. We consider two vector quantization based feedback designs. One is based on random generation of codebooks and the other is based on hierarchical codebooks.

Interference Alignment - based Precoding and User Selection with Limited Feedback in Two - cell Downlink Multi - user MIMO Systems

“…The gain of system throughput brought by IA for limited feedback systems that adopt the IA scheme depends on the quantization quality of CSI. System performance with imperfect CSI was investigated in an interfering broadcast channel [15]. An optimal IA scheme [16] was obtained for the two-cell multiple access channel by quantizing the beamforming vector with the criterion of minimizing total residual inter-cell interference.…”