Pulse or digital communications – Equalizers – Automatic
Reexamination Certificate
1998-11-02
2003-02-11
Chin, Stephen (Department: 2634)
Pulse or digital communications
Equalizers
Automatic
C375S348000, C375S349000, C375S350000
Reexamination Certificate
active
06519282
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The invention relates to a method for digital transmission of data over copper lines, wherein digital signals are precoded with a fixed precoder on the transmitter side and wherein the signals are recovered on the receiver side through blind equalization (see “IEEE Journal on Selected Areas in Communication”, Vol. SAC-13, December 1995, pp. 1622 to 1633).
This method can be used for transmission of information over copper lines as well as for wireless transmission. While subsequent discussion relates mainly to applications for transmission over copper lines, it is in no way intended to be the limit of the applicability of the invention.
2. Description of the Prior Art
Fast digital transmission methods over copper cable, that include several or a large number of so-called two-wire lines placed next to each other, have become increasingly important because of their potential to enable a variety of new digital services in the near future. They also provide a gradual transition to glass fiber networks and have to be able to operate reliably over long cable lengths, because the number of repeaters between the telephone exchange of a telephone network and the subscriber has to be kept small for economic reasons.
Attenuation and distortion during the land-line transmission of information signals produce channel-related inter-symbol interference (ISI). The channel-related ISI in digital information transmission can be effectively eliminated through equalization with decision feedback equalization (DFE). This method is highly efficient and not very complex. However, DFE cannot be combined directly with coded modulation because the feedback filters in DFE require an immediate decision about the transmitted signals.
Tomlinson (see Electronics Letters, Vol. 7, March 1971, pp. 138 and 139) and Harashima/Miyakawa (see IEEE Transactions on Communications, Vol. COM-20, August 1972, pp. 774 to 780) propose a solution for this problem. They propose to shift the feedback filter of the DFE to the transmitter. In addition, they introduce a modulo operation to limit the amplitude. This method is referred to as Tomlinson-Harashima precoding (THP). However, a transmission with optimal THP requires information about the channel state on the transmitter side which has to be transmitted via a return channel. This requires complex protocols for setting up the connection to prevent mutual blocking of the two transmission directions. Also, not every application may have a return channel.
This problem can be sidestepped, as described in the above referenced article, by employing a fixed precoder which simplifies the set-up of the connection. However, preceding that is not optimized produces residual interference which has to be equalized linearly on the receiver side. The linear equalizer for the residual interference is adapted blindly, i.e. without using a training sequence. However, in a system using THP, it is essentially impossible to provide blind equalization at the symbol clock rate because the signal that has to be reconstructed has an approximately discrete Gaussian distribution. The precoding is therefore modified to limit the dynamic range of the effective transmitted signal and to produce signals with better statistics. An example for such a method is “Dynamics Shaping” (DS). The DS method permits blind equalization while maintaining or even increasing the excellent efficiency of THP. For uncoded transmission with the DS method, a simple standard method for blind equalization can be employed, using the so-called Sato algorithm (see IEEE Transactions on Communications, Vol. COM-23, June 1975, pp. 679 to 682).
SUMMARY OF THE INVENTION
It is the object of the invention to improve the method described above and also the blind equalization process.
The invention solves the object by subdividing the blind equalization into an equalization of the magnitude followed by an equalization of the phase.
This method for digital data transmission is easy to implement. Most importantly, the signal which is to be recovered in the equalizer may be correlated, whereas the methods known in the art can only recover white signals. The method is very efficient due to the fixed precoding and two-stage blind equalization. The preceding is adapted to a fixed reference application, so that channel information does not have to be retransmitted. Blind equalization removes the resulting residual interference. Equalization of the magnitude takes into consideration any correlation present in the transmitted sequence. The blind equalization can then be robust even if the signal-to-interference ratio at the equalizer input is very small. In particular, after the magnitude has been equalized, a simple blind algorithm can be used to equalize the phase. The convergence is very fast, in spite of strong residual interference at the input of the magnitude equalizer, and in spite of a low signal-to-noise ratio (SNR) and the presence of correlated symbols that have to be reconstructed.
The cable path that can be spanned can be significantly lengthened by further coding of the signals, for example by using trellis-coding. In this case, equalizing the magnitude before separately equalizing the phase has also proven to be particularly advantageous due to the small signal-to-noise ratio at the equalizer input.
The invention will be fully understood when reference is made to the following detailed description taken in conjunction with the accompanying drawing.
REFERENCES:
patent: 5572262 (1996-11-01), Ghosh
patent: 5909466 (1999-06-01), Labat et al.
patent: 6026121 (2000-02-01), Sadjadpour
patent: 6134265 (2000-10-01), Long
patent: 0314360 (1989-05-01), None
da Rocha et al., “Self-Learning Deconvolution Using a Cascade of Magnitude and Phase Equalizers”, 1996, Circuits and Systems, 1995, Proceedings of the 38th Midwest Symposium, vol. 1, 255-258.*
“Dynamics Limited Precoding, Shaping, and Blind Equalization for Fast Digital Transmission over Twisted Pair Lines” by R.F.H. Fischer et al, IEEE Journal on Selected Areas in Communications, vol. 13, No. 9, Dec. 1995, pp. 1622-1633.
Robert F. H. Fischer, Wolfgang H. Gerstacker and Johannes B. Huber, all IEEE Members, entitled “Dynamics Limited Precoding, Shaping, and Blind Equalization for Fast Digital Transmission over Twisted Pair Lines”, IEEE Journal On Selected Areas In Communications, pp. 1622-1633, dated Dec. 1995 , vol. 13, No. 9.
M. Tomlinson, entitled “New Automatic Equaliser Employing Modulo Arithmetic”, Electronics Letters, dated Mar. 25, 1971, pp. 138-139, vol. 7, Nos. 5/6.
Hiroshi Harashima and Hiroshi Miyakawa, IEEE Member, entitled “Matched-Transmission Technique for Channels With Intersymbol Interference”, IEEE Transactions On Communications, dated Aug. 1972, pp. 774-780, vol. Com-20, No. 4.
Yoichi Sato, entitled “A Method Of Self-Recovering Equalization for Multilevel Amplitude-Moducation Systems” IEEE Transations On Communications, dated Jun. 1975, pp. 679-682.
Wolfgang H. Gerstacker, Robert F. H. Fischer, and Johannes B. Huber, entitled “A Transmission Scheme For Twisted Pair Lines With Coding, Precoding, and Blind Equalization”.
Fischer Robert
Gerstacker Wolfgang
Huber Johannes
Alcatel
Chin Stephen
Lugo David B.
LandOfFree
Method for digital transmission of information does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for digital transmission of information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for digital transmission of information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3147211