Title :
Optimal constellation labeling for iteratively decoded bit-interleaved space-time coded Modulation
Author :
Huang, Yuheng ; Ritcey, James A.
Author_Institution :
Dept. of Electr. Eng., Univ. of Washington, Seattle, WA, USA
fDate :
5/1/2005 12:00:00 AM
Abstract :
We consider the design of iteratively decoded bit-interleaved space-time coded modulation (BI-STCM) over fast Rayleigh-fading channels with Nt transmit and Nr receive antennas. We propose the design criterion to achieve the largest asymptotic coding gain inherited in the constellation labeling. In particular, for orthogonal space-time block codes, the labeling design criterion reduces to maximizing the (-NtNr)th power mean of the complete set of squared Euclidean distances associated with all "error-free feedback" events in the constellation. Based on this power mean criterion, we show that the labeling optimization problem falls into the category of quadratic assignment problems for constellations of any shape and with an arbitrary number of transmit and receive antennas. For a set of practical values of Nt and Nr, we present optimal labeling maps for 8-PSK, 16-QAM, and 64-QAM constellations.
Keywords :
Rayleigh channels; antenna arrays; iterative decoding; modulation coding; phase shift keying; quadrature amplitude modulation; receiving antennas; space-time codes; transmitting antennas; 16-QAM; 64-QAM; 8-PSK; BI-STCM; bit-interleaved space-time coded modulation; constellation labeling; error-free feedback events; fast Rayleigh-fading channels; iterative decoding; multiple antennas; optimization problem; phase shift keying; quadratic assignment problems; quadrature amplitude modulation; receive antennas; squared Euclidean distances; transmit antennas; Acoustic signal detection; Acoustic signal processing; Digital communication; Iterative decoding; Labeling; MIMO; Modulation coding; Phase modulation; Receiving antennas; Signal design; Bit-interleaved coded modulation (BICM); constellation labeling; multiple antennas; optimal code search; quadratic assignment problem (QAP); space–time coding;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2005.846409