Link level performance prediction of MIMO PIC receivers through QR decomposition of channel matrix

Abstract

In this paper, an accurate QR decomposition based link level performance abstraction technique for a multi-input multi-output iterative receiver is proposed. The post-detection SNR values for an iterative receiver are estimated using QR decomposition of the channel matrix. The effect of channels on the received SNR is realized by scaling the signal in accordance with the leading diagonal elements of the R matrix. Both mutual and exponential mapping functions have been used for data compression. The proposed link level performance abstraction strategy employs a single lookup table without having to do the complex link level simulations. It is independent of the channel conditions and depends only on the type of the receiver used and the modulation and coding scheme adopted. The resulting link level performance prediction shows a good approximation of the simulated performance at different modulation and coding scheme.