DocumentCode :
747299
Title :
A fast convergence algorithm for echo cancellers in full duplex transmission using back projection from slicer
Author :
Stopler, Danny ; Reuven, Ilan
Author_Institution :
Dept. of Electr. Eng.Syst., Tel Aviv Univ., Israel
Volume :
53
Issue :
4
fYear :
2005
fDate :
4/1/2005 12:00:00 AM
Firstpage :
1534
Lastpage :
1549
Abstract :
The problem of accelerating echo canceller tracking, during full duplex transmission, in single carrier (data) receivers with adaptive equalization is addressed. The scheme presented accelerates the convergence rate by orders of magnitude, with a minor increase in complexity, which can be traded for speed. The complexity may be reduced to a fraction of that of previously published acceleration schemes, while providing substantial speedup factors with respect to conventional echo cancellers. The new scheme is based on measuring the residual echo signal at the slicer (symbol-by-symbol detector) but applying the correction to the conventional echo canceller. Measuring changes in a low-noise (post-slicer) environment enables speedup of the adaptation. Updating the pre-equalizer echo canceller both reduces complexity and distills the echo from the noisy equalizer input.
Keywords :
adaptive equalisers; adaptive filters; computational complexity; convergence of numerical methods; digital filters; echo suppression; least mean squares methods; signal detection; tracking; adaptive digital filter; adaptive equalization; back projection; computational complexity; convergence acceleration; digital subscriber line; echo canceller; fast convergence algorithm; fast tracking; full duplex transmission; least mean square algorithm; sheer; symbol-by-symbol detector; Acceleration; Convergence; Crosstalk; DSL; Decision feedback equalizers; Echo cancellers; Frequency; Modems; Noise cancellation; Transversal filters; Adaptive digital filters; DSL echo; LMS algorithm; convergence acceleration; digital subscriber line; echo cancellation; fast convergence; fast echo canceller; fast tracking;
fLanguage :
English
Journal_Title :
Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1053-587X
Type :
jour
DOI :
10.1109/TSP.2005.843709
Filename :
1408202
Link To Document :
بازگشت