EESM-based link adaptation in OFDM: Modeling and analysis

In orthogonal frequency division multiplexing systems, such as Long Term Evolution (LTE) and WiMAX, the different subcarriers over which a codeword is transmitted may see different signal-to-noise-ratios (SNRs). Thus, adaptive modulation and coding (AMC) in these systems must be based on a vector of subcarrier SNRs seen by the codeword, and is considerably more involved. Exponential effective SNR mapping (EESM) simplifies the problem by mapping the vector of SNRs into a single equivalent flat-fading SNR. However, the analysis of AMC using EESM is challenging owing to its non-linear nature and because it uses an SNR scaling parameter that depends on the modulation and coding scheme. We first propose a novel statistical model for EESM based on the Beta distribution, which is motivated by the central limit approximation for the sum of random variables with finite support. Unlike several ad hoc statistical models, which require three or more parameters to be computed numerically, the proposed model requires only two parameters, for which closed-form expressions are derived for both correlated and uncorrelated subcarrier SNRs. Despite its simplicity, it is as accurate as the ad hoc models. We then present a novel, tight upper bound and an accurate approximation in closed-form for the throughput of a frequency-selective system that uses EESM for AMC.