UP Paper 1292 US-M-JDOWN
Reduced Complexity Stack-Based Limited Tree Searching Algorithm for V-BLAST Systems
Kang,JoonhyukInformation and Communications University
Lee,NamjeongInformation and Communications University
Lee,KeonkookInformation and Communications University
Gil,GyetaeProfessional Research Group, Advanced Technology Laboratory, KT
Cha,JongsubInformation and Communications University
The Vertical Bell Labs Layered Space-Time (V-BLAST) architecture is considered as an efficient technology for achieving high spectral efficiency over a rich-scattering environment. In this paper, we propose a computationally efficient near-optimal detection algorithm for V-BLAST systems. The proposed scheme employs minimum mean-square error sorted QR decomposition (MMSE-SQRD) of the channel matrix and stack-based limited tree searching method. For reducing computational complexity, the number of branches required for the extension of the tree structure is conditionally chosen according to the channel condition, unlike the conventional stack algorithm. The receiver checks the super-diagonal term in R matrix to examine the channel condition. Some small super-diagonal elements of R indicate that the channel condition is bad. In this case, the larger number of branches is assigned for tree searching. To provide an efficient way of saving complexity at the receiver, it is important to find the suitable number of branches in accordance with the channel condition. Simulation results show that the computational effort of the proposed method dramatically decreased compared with the limited tree searching with constant value. It requires a fraction of an effort to achieve near-SD performance in comparison with sphere detection (SD). The complexity analysis of the proposed algorithm together with other conventional schemes will be derived numerically as well.

Namjeong Lee was born in Seoul, Korea. She received the B.S. degree in electrical and computer engineering from the Information and Communications University in 2006. Since 2006, she has been a M.S. student at the school of engineering, Information and Communications University, Daejeon, Korea. Her research interest covers MIMO and multiuser detection.