Asymptotic performance of BI-GDFE for large isometric and random precoded systems

Block-iterative generalized decision feedback equalizer (BI-GDFE), has been proposed to achieve single user matched filter bound (MFB) for linearly precoded systems, such as CP-CDMA and MC-CDMA, which usually suffer from strong interferences. This paper analyzes the asymptotic performance of BI-GDFE for large isometric and random precoded systems. It is shown that the output signal-to-interference-plus-noise ratios (SINRs) for each of the iterations of the BI-GDFE receiver converge in probability to deterministic values depending on the probability distribution of the channel coefficients when N → ∞, P/N → constant α (α ≤ 1 for isometric case). The asymptotic convergence of BI-GDFE has been evaluated through numerical simulations.