Adaptive equalizer with gain controlled initialization

Electrical computers: arithmetic processing and calculating – Electrical digital calculating computer – Particular function performed

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C375S231000

Reexamination Certificate

active

06804694

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to signal processing generally, and more specifically to adaptive equalization devices.
BACKGROUND OF THE INVENTION
AMI (Alternate Mark Inversion) transmission line code is extensively used for signal transmission over metallic interfaces, such us T1/E1 or SONET/SDH. The characteristics and length of the cable used as transmission channel may be unknown, so blind adaptive equalization techniques plays a very important role in order to restore the original transmitted signal at the receiver. However, most adaptive equalization techniques fail to equalize AMI signals given the source correlation (memory effect) of this transmission code.
The Decision-Directed Least Mean Squares algorithm (DD-LMS) has shown effectiveness when equalizing AMI signals despite the source correlation, but this technique may be affected by convergence problems if the equalizer coefficients are not initialized properly. The adaptation of the filter coefficients can easily lead to a solution where all the filter coefficients grow to an infinite value or are reduced to zero. In both cases the transmitted signal is not recovered.
An equalizer for AMI signals has been implemented, in which the DD-LMS algorithm is used to equalize the AMI signal. In order to avoid wrong convergence, an analog automatic gain controlled amplifier based on the input signal envelope and a set of fixed IIR (Infinite Impulse Response) filters were provided.
The filter that showed the least error, depending on a certain criterion, was chosen to pre-equalize the signal. The pre-equalization performed by the fixed IIR filter reduces the risk of wrong convergence. However, due to the absence of accurate control of the equalizer initialization, convergence to the zero solution (all coefficients equal to zero) can be easily achieved for long cables. A wide range of cable types and lengths, are appropriate for T1/E1 applications, but correct performance (i.e., avoiding the zero solution) on different conditions cannot be assured.
An improved method and apparatus for ensuring correct convergence of an adaptive equalizer is desired.
SUMMARY OF THE INVENTION
The present invention is a method and apparatus for processing an input signal. A gain is applied to an input signal to form an amplified signal. An autocorrelation value of the amplified signal is calculated. The autocorrelation value is compared to a predetermined constant. The gain is adjusted, based on which of the group consisting of the autocorrelation value and the predetermined constant is greater. The gain application, autocorrelation calculation, comparison and gain adjustment are repeated, until the autocorrelation value is sufficiently close to the predetermined constant to satisfy a convergence criterion. The amplified signal is then filtered.


REFERENCES:
patent: 3659229 (1972-04-01), Milton
patent: 5267266 (1993-11-01), Chen
patent: 5537437 (1996-07-01), Kaku et al.
patent: 6674795 (2004-01-01), Liu et al.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Adaptive equalizer with gain controlled initialization does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Adaptive equalizer with gain controlled initialization, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive equalizer with gain controlled initialization will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-3326813

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.