Title :
APP processing for high performance MIMO systems [receiver symbol detector]
Author :
Garrett, David ; Davis, Linda ; Ten Brink, Stephan ; Hochwald, Bert
Author_Institution :
Lucent Technol. Bell Labs., Holmdel, NJ, USA
Abstract :
One of the most demanding components in a multiple-input multiple-output (MIMO) receiver is the symbol detector. While linear cancellation techniques are typically used, the optimum maximum likelihood (ML) search algorithm combined with soft output processing, e.g. a posteriori probability (APP), offers superior performance and is feasible in current technology. An ASIC can be built for a 4×4 QPSK detector for HSDPA in less than 1.68 mm2 in a 0.18 μm process. An MLAPP search grows exponentially as the constellation size is increased, but it is still feasible to achieve near ML performance for 4×4 16QAM using "spherical" techniques in a complexity that remains a minor portion of the overall MIMO receiver.
Keywords :
MIMO systems; application specific integrated circuits; maximum likelihood detection; quadrature amplitude modulation; quadrature phase shift keying; radio receivers; 0.18 micron; APP; APP processing; ASIC; HSDPA; MIMO receiver; ML performance; MLAPP detector; MLAPP search; QAM; QPSK detector; a posteriori probability; constellation size; high performance MIMO systems; multiple-input multiple-output receiver; optimum maximum likelihood search algorithm; soft output processing; spherical techniques; symbol detector; Application specific integrated circuits; Availability; Detectors; MIMO; Maximum likelihood detection; Maximum likelihood estimation; Multiaccess communication; Quadrature phase shift keying; Silicon; Wireless LAN;
Conference_Titel :
Custom Integrated Circuits Conference, 2003. Proceedings of the IEEE 2003
Print_ISBN :
0-7803-7842-3
DOI :
10.1109/CICC.2003.1249401