Coded data generation or conversion – Digital code to digital code converters – To or from run length limited codes
Reexamination Certificate
2003-09-22
2004-09-28
Tokar, Michael (Department: 2819)
Coded data generation or conversion
Digital code to digital code converters
To or from run length limited codes
C341S058000, C360S044000, C360S046000, C360S053000, C369S044270, C369S044250
Reexamination Certificate
active
06798363
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method for compensating asymmetry in a reproduction signal from an optical recording medium. The invention further relates to an apparatus for reading from and/or writing to optical recording media using such method.
BACKGROUND OF THE INVENTION
For high data storage densities on optical recording media, the modulation transfer function drops very steeply. The high frequency components of the analog reproduction signal are, therefore, attenuated considerably compared to the low frequency components. In case of a blu-ray disk (BD), which is currently under development, with a storage capacity of about 25 gigabytes, the shortest run-length components (
2
T) are attenuated by a factor of more than 20 dB compared to the longest run-length components (
8
T). This results in a large amount of inter-symbol interference. The eye-pattern, i.e. the high frequency signal obtained by summing the output signals of a photodetector array used in an apparatus for reading from and/or writing to optical recording media (“reproduction signal”) is even without noise nearly closed. Furthermore, the reproduction signal is also non-linear, which results in a strong asymmetry of the eye-pattern. This can, inter alia, be caused by non-optimum recording conditions like an excessive write power, leading to different lengths of marks and spaces on the optical recording medium. The amount of this asymmetry can be larger than the amplitude of the shortest run length signal.
For a reliable data detection, the mid-level signal, which is necessary for further processing of the reproduction signal, has to be placed exactly in the middle of the shortest run-length component of the reproduction signal. This can be achieved by subtracting an offset compensation signal, which is generated by an offset compensator, from the reproduction signal.
Such a solution is, for example, disclosed in the U.S. Pat. 6,324,144. The document shows an apparatus for correcting asymmetry existing in a reproduction signal by processing the reproduction signal in digital form. An analog-to-digital converter converts the analog reproduction signal to a digital reproduction signal. A predetermined asymmetry compensation signal is then added to the digital signal to obtain an asymmetry compensated signal. Finally, a binary non-return-to-zero-inverted (NRZI) data signal is detected from the asymmetry compensated signal. This binary data signal is output by the apparatus. For determining the asymmetry compensation signal, the zero crossing points of the digital reproduction signal are detected. The zero crossing points are needed for controlling a sign bit counting operation, which is used for controlling an asymmetry corrector.
The solution disclosed in the document, which has been developed for DVD-RAM where the shortest run-length is
3
T, has the disadvantage that the detection of zero crossing points cannot be reliably performed if the amplitude of the shortest run-length components is smaller than the asymmetry of the reproduction signal. In this case the shortest run-length components are nearly vanishing in the noise floor and the zero crossing points cannot easily be detected.
It is, therefore, an object of the invention to propose a method for compensating an offset in an asymmetric reproduction signal capable of compensating the offset even if the amplitude of the shortest run-length components is smaller than the asymmetry of the reproduction signal, i.e. if the detection of zero crossing points is not possible for the shortest run-length components.
SUMMARY OF THE INVENTION
According to the invention, this object is achieved by a method for compensating an offset in an asymmetric reproduction signal, whereby an offset compensation signal is subtracted from the reproduction signal, the offset compensation signal being generated by an offset compensator, comprising the steps of:
detecting a binary data signal from the asymmetric reproduction signal; and
using the binary data signal for obtaining the offset compensation signal.
Using the binary data signal, which only assumes two discrete values, for obtaining the offset compensation signal has the advantage that the offset compensation signal can be obtained with a much higher reliability compared with the digital reproduction signal, which assumes a plurality of discrete values. Even when a detection of zero crossing points is not possible, the offset compensation signal can still be obtained.
Favorably, the method further comprises the step of detecting the shortest run-length components of the binary data signal for obtaining the offset compensation signal. Since the shortest run-length components are most affected by the asymmetry of the reproduction signal, it is sufficient to use only these components for obtaining the offset compensation signal. In this case, every time a shortest run-length component is detected, an enable signal is generated for enabling the offset compensation. Of course, it is also possible to detect signal components with another run-length and to generate the respective enable signal. A secure run-length detection based on the digital reproduction signal as as known from prior art, i.e. before detecting the binary data signal, is not possible if the amplitude of the shortest run-length components is smaller than the asymmetry of the reproduction signal.
Advantageously, the method further comprises the step of delaying the asymmetric reproduction signal before obtaining the offset compensation signal and/or before subtracting the offset compensation signal from the reproduction signal. This allows to compensate for the processing delay caused by the detection of the binary data signal from the asymmetric reproduction signal and by the run-length detection, so that the enable signal, and correspondingly the asymmetry compensation signal, coincides exactly with the shortest run-length samples of the reproduction signal. Delaying the asymmetric reproduction signal can, for example, be performed by a register chain.
Favorably, the method further comprises the step of centering the asymmetric reproduction signal with regard to a digital zero before detecting the binary data signal. This centering is, for example, performed by passing the reproduction signal through a slicer. Centering the asymmetric reproduction signal without compensating the offset is sufficient for a reliable run-length detection until the offset compensation has settled to a final offset compensation signal.
Advantageously, a partial response maximum likelihood detector or a bit-by-bit detector is used for detecting the binary data signal. Both detectors deliver a non-return-to-zero (NRZ) data stream at their output, which can be used for obtaining the asymmetry compensation signal. While the partial response maximum likelihood detector, e.g. a partial response equalizer in combination with a Viterbi detector, delivers a lower bit error rate and has a higher performance, the bit-by-bit detector is less expensive and simplifies the necessary delay of the reproduction signal samples.
According to a further refinement of the invention a plurality of run-lengths of the binary data signal are detected for obtaining run-length dependent offset compensation signals and enabling the offset compensation accordingly. For each signal sample the offset compensation signal corresponding to the run-length of the signal sample is used for offset compensation. In this way, the offset is not only compensated for the shortest run length, but selectively also for other run-lengths, which are allowed by the channel modulation, leading to an even more reliable data detection and hence to a lower bit error rate. In this case, it is perfectly possible to use a partial response maximum likelihood detector for detecting the binary data signal used for obtaining the different offset compensation signals, and to use a simple bit-by-bit detector for controlling a multiplexer used for selecting the appropriate offset compensation signal.
According to anoth
Mai Lam T.
Reitseng Lin
Thomson Licensing S.A.
Tokar Michael
Tripoli Joseph S.
LandOfFree
Method for compensating an offset in an asymmetric... 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 compensating an offset in an asymmetric..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for compensating an offset in an asymmetric... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3257231