An Iterative SNR Estimator for Link Adaptation in IEEE 802.11n System

A dynamic and reliable link adaptation algorithm is essential to achieving higher throughput limits in a distributed and diverse wireless network. However, SNR based Link Adaptations reported in literature thus far assume that SNR estimation is always ideal. In this paper, ideal is defined as the absolute performance of the SNR estimator which approaches the Cramer-Rao Bound (CRB) with almost no bias or variance. In real wireless communication systems, this assumption doesn’t hold true. We extend the SNR estimator for Single Input Single Output (SISO) system to a Multiple Input Multiple Output (MIMO) Physical Layer (PHY) model, and show the actual performance of a cross layer SNR based link adaptation. Simulation results prove that the SNR estimates are unstable and fluctuate in a multi-path Rayleigh fading channel, degrading the effectiveness of link adaptation algorithm. This motivates us to propose an iterative SNR estimator for link adaptation algorithm that allows the efficient utilization of channel bandwidth to achieve a throughput gain for IEEE 802.11n based system.