Title :
Application of joint channel and data estimation algorithms for short block detection
Author :
Nefedov, Nikolai
Author_Institution :
Nokia Res. Center, Espoo, Finland
Abstract :
This paper presents studies of the expectation-maximization (EM) algorithm and its modifications for joint channel estimation and short block detection in frequency selective channels. To facilitate the EM convergence in case of short blocks, we introduce a penalized update for channel re-estimates in semi-blind detection. For channels with intersymbol interference, we suggest a simple method to improve accuracy of channel re-estimates (channel model inverse) that may be embedded into EM-based detectors. Application of the EM-based detectors for schemes with transmitting diversity in EDGE (enhanced data rates for GSM evolution) provides more than 1dB gain
Keywords :
cellular radio; convergence of numerical methods; diversity reception; intersymbol interference; optimisation; parameter estimation; signal detection; telecommunication channels; EDGE; ISI; channel estimation; data estimation; enhanced data rates for GSM evolution; enhanced data rates for global evolution; expectation-maximization algorithm; frequency selective channels; intersymbol interference; short block detection; transmitting diversity; Channel estimation; Delay; Detectors; Frequency estimation; Ice; Intersymbol interference; Inverse problems; Parameter estimation; Transmitting antennas; Wireless communication;
Conference_Titel :
Personal, Indoor and Mobile Radio Communications, 2001 12th IEEE International Symposium on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-7244-1
DOI :
10.1109/PIMRC.2001.965507