Efficient Square-Root and Division Free Algorithms for Inverse LDLT Factorization and the Wide-Sense Givens Rotation with Application to V-BLAST

We propose efficient square-root and division free algorithms for inverse LDL' factorization and the wide-sense Givens rotation, mainly for conventional multiply-add digital computers. The algorithms are then employed to propose a fast algorithm for vertical Bell Laboratories Layered Space-Time architecture (V-BLAST). The proposed inverse LDL' factorization avoids the conventional back substitution of the Cholesky factor, and then has the speedups of 1.19 over the square root and division free alternative Cholesky factorization plus the back substitution. The proposed wide-sense Givens rotation requires the same computational complexity as the efficient Givens rotation. The speedups of the proposed V-BLAST algorithm over the recursive V-BLAST algorithm are 1.05~1.4. The difference between the complexity of the proposed V-BLAST algorithm and that of the square-root V-BLAST algorithm is trivial and even negligible. However, the square-root V-BLAST algorithm requires many square-root and division operations, while the proposed V-BLAST algorithm requires no square-root operations, and requires only one division for each transmit signal, which can be postponed until the signal estimation is required.

[1]  Wen Chen,et al.  Improved Fast Recursive Algorithms for V-BLAST and G-STBC with Novel Efficient Matrix Inversion , 2009, 2009 IEEE International Conference on Communications.

[2]  E. J. Baranoski.  Triangular factorization of inverse data covariance matrices , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[3]  M. Ylinen,et al.  A fixed-point implementation of matrix inversion using Cholesky decomposition , 2003, 2003 46th Midwest Symposium on Circuits and Systems.

[4]  C. L. Nikias,et al.  Algorithms for Statistical Signal Processing , 2002 .

[5]  Babak Hassibi,et al.  An efficient square-root algorithm for BLAST , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[6]  James Demmel,et al.  On computing givens rotations reliably and efficiently , 2002, TOMS.

[7]  Wen Chen,et al.  Reducing the Computational Complexity for BLAST by Using a Novel Fast Algorithm to Compute an Initial Square-Root Matrix , 2008, 2008 IEEE 68th Vehicular Technology Conference.

[8]  Robert H. Halstead,et al.  Matrix Computations , 2011, Encyclopedia of Parallel Computing.

[9]  Linda M. Davis.  Scaled and decoupled Cholesky and QR decompositions with application to spherical MIMO detection , 2003, 2003 IEEE Wireless Communications and Networking, 2003. WCNC 2003..

[10]  K. J. Ray Liu,et al.  A class of square root and division free algorithms and architectures for QRD-based adaptive signal processing , 1994, IEEE Trans. Signal Process..

[11]  Reinaldo A. Valenzuela,et al.  V-BLAST: an architecture for realizing very high data rates over the rich-scattering wireless channel , 1998, 1998 URSI International Symposium on Signals, Systems, and Electronics. Conference Proceedings (Cat. No.98EX167).

[12]  Jacob Benesty,et al.  A fast recursive algorithm for optimum sequential signal detection in a BLAST system , 2003, IEEE Trans. Signal Process..