Recursive QR packet combining for uplink single-carrier multi-user MIMO HARQ using near ML detection

Tetsuya Yamamoto, Koichi Adachi, Sumei Sun, Fumiyuki Adachi

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)


QR decomposition-based near maximum likelihood block detection significantly improves the transmission performance of uplink single-carrier (SC) multi-user multiple-input multiple-output (MU-MIMO). Hybrid automatic repeat request (HARQ) is an indispensable error control technique for high-quality packet data transmission. The achievable diversity gain of HARQ depends on the packet combining strategy. In uplink MU-MIMO HARQ, the received signal may consist of new packets and retransmitted packets as retransmission for each user acts independently. In this paper, a recursive QR packet combining scheme suitable for uplink SC MU-MIMO HARQ is proposed, which takes into account the number of retransmissions for each user in the detection order. The computational complexity and required storage size can be significantly reduced by the proposed scheme. Moreover, it improves the packet error rate performance and throughput performance significantly over the conventional bit-level log likelihood ratio packet combining, as shown by our computer simulation results.

Original languageEnglish
Pages (from-to)1652-1663
Number of pages12
JournalWireless Communications and Mobile Computing
Issue number18
Publication statusPublished - 2012 Dec 25


  • HARQ
  • M-algorithm
  • QR decomposition
  • packet combining
  • single-carrier


Dive into the research topics of 'Recursive QR packet combining for uplink single-carrier multi-user MIMO HARQ using near ML detection'. Together they form a unique fingerprint.

Cite this