A Sphere Decoder with Approximate QR Decomposition for Frequency-Selective Channels

This paper presents a method to significantly reduce the preprocessing complexity of the sphere decoder (SD) in frequency-selective channels. The method consists of calculating an approximate QR decomposition (AQRD) of the channel matrix, making use of its special Toeptliz and block-Topelitz structure in single and multiple-antenna frequency-selective channels, respectively. The AQRD obtains the QR decomposition of a small submatrix of the channel matrix and extends that result to the rest of the matrix, resulting in a considerable complexity reduction compared to the original full QR decomposition (FQRD). Simulation results show that, despite the lower complexity of the AQRD, it causes only a small bit error rate (BER) performance degradation in the SD.

[1]  Jorma Rissanen,et al.  Properties of infinite covariance matrices and stability of optimum predictors , 1969, Inf. Sci..

[2]  J. Rissanen.  Algorithms for triangular decomposition of block Hankel and Toeplitz matrices with application to factoring positive matrix polynomials , 1973 .

[3]  Babak Hassibi,et al.  Maximum-Likelihood Sequence Detection of Multiple Antenna Systems over Dispersive Channels via Sphere Decoding , 2002, EURASIP J. Adv. Signal Process..

[4]  Giuseppe Caire,et al.  On maximum-likelihood detection and the search for the closest lattice point , 2003, IEEE Trans. Inf. Theory.

[5]  M. O. Damen,et al.  A unified framework for tree search decoding: rediscovering the sequential decoder , 2005, SPAWC 2005.

[6]  Tharmalingam Ratnarajah,et al.  On the Complexity of the Sphere Decoder for Frequency-Selective MIMO Channels , 2008, IEEE Transactions on Signal Processing.

[7]  Martin Haardt,et al.  Comparative study of joint-detection techniques for TD-CDMA based mobile radio systems , 2001, IEEE J. Sel. Areas Commun..

[8]  Martin Haardt,et al.  Schur algorithms for joint-detection in TD-CDMA based mobile radio systems , 1999, Ann. des Télécommunications.

[9]  Urbashi Mitra,et al.  Sphere-constrained ML detection for frequency-selective channels , 2006, IEEE Transactions on Communications.

[10]  Martin J. Gander,et al.  Asymptotic properties of the QR factorization of banded Hessenberg–Toeplitz matrices , 2005, Numer. Linear Algebra Appl..

[11]  Hamid Reza Karimi,et al.  A novel and efficient solution to block-based joint-detection using approximate Cholesky factorization , 1998, Ninth IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (Cat. No.98TH8361).

[12]  Gene H. Golub,et al.  Matrix computations , 1983 .

[13]  T. Kailath,et al.  Fast Parallel Algorithms for QR and Triangular Factorization , 1987 .