Robust Iterative Interference Alignment with Limited Feedback

In this paper we propose a robust iterative interference alignment (IA) technique for the downlink of OFDM-based (Orthogonal Frequency Division Multiplexing) broadband wireless systems with limited feedback. The channel frequency responses (CFR) associated to each link between base station (BS) and user terminal (UT) are quantized and feedback from the UT to the BS, which sends it to the other BSs through a limited-capacity backhaul network. This information is employed by each BS to perform a robust overall IA design, which takes into account the statistical characterization of the CFR quantization errors. An analytical method for obtaining this statistical characterization is derived taking advantage of the Gaussian characteristics of the CFR. Simulations show the relevance of the proposed technique, almost perfectly canceling multi-user interference, and the high accuracy of the theoretical characterization of the CFR quantization error. Moreover, the robust IA technique with the proposed analytical characterization of the CFR error is as good as genie-aided techniques that know the CFR error characteristics, outperforming non-robust IA designs, as well as techniques that employ an approximate characterization based on the uniform distribution of the quantization error.