Multiple Transposed Channel Matrix QR decompositions based Sphere Detection Algorithm for MLD-SDM/OFDM

概要

論文の詳細を見る
SDM/OFDM systems transmit a different data using the same frequency, so it is necessary to separate the simultaneously received signals in the receiver. Previous studies have shown that the maximum likelihood detection (MLD) has the best BER performance. However, the complexity in MLD is exponentially increased due to the modulation level and the number of transmitter antenna branches. Accordingly, it is not practical to use the conventional MLD without reducing the computational complexity because the computational complexity of MLD exceeds the level of an actual implementation. Recently, QRM-MLD has been proposed to reduce the system complexity with maintaining the system performance. However, the performance of QRM-MLD is largely depended on the number of surviving symbol replica candidates. When the QRM-MLD with small surviving symbol replica candidates, the performance is degraded. On the other hands, with large surviving symbol replica candidates, QRM-MLD requires large memory to keep a surviving symbol replica branch metrics. To reduce the above-mentioned QRM-MLD, in this paper, we propose multiple transposed channel matrix QR-decompositions based sphere detection algorithm for MLD-SDM/OFDM system.
社団法人電子情報通信学会の論文
2005-07-14