A comparative study of interference alignment schemes with LTE-compliant turbo coding

Abstract

It was recently theoretically shown that the sum capacity of the K-user interference channel scales linearly with a factor of K/2 in the high signal-to-noise ratio (SNR) regime. This scaling can be accomplished with interference alignment (IA), which is a transmission technique that requires perfect global channel state information to be available at all transceivers. For the general case of K > 3 users, a number of different iterative algorithms for achieving the IA sum-rate scaling at high SNRs has been lately proposed. These algorithms compute transmit and receive filters aiming at optimizing a certain performance objective, thus ranging in sum-rate performance vs. complexity tradeoffs. However, most of the performance evaluation studies to date do not include error correction coding, thus taking away from our ability to quantify the true sum-rate potential of the various aforementioned algorithms in practice. In this paper, we perform a comparative study of several IA schemes in the spatial dimension when combined with an error correction coding scheme. Our results shed light into how IA would perform in LTE-compliant systems.