Dynamic information storage or retrieval – Condition indicating – monitoring – or testing – Including radiation storage or retrieval
Reexamination Certificate
2000-09-27
2003-03-04
Psitos, Aristotelis M. (Department: 2653)
Dynamic information storage or retrieval
Condition indicating, monitoring, or testing
Including radiation storage or retrieval
C375S360000, C375S376000, C369S047350
Reexamination Certificate
active
06529459
ABSTRACT:
FIELD OF INVENTION
The invention relates to digital data storage and retrieval, and, more particularly, to retrieving binary data conveyed by an incident signal from a mobile carrier such as a digital disk.
BACKGROUND OF INVENTION
Digital disks, especially compact disks (CDs) (e.g., Read Only Memory compact disks (“CDROMs”)) and multifunction digital disks (e.g., Digital Video Disks (or “DVDs”)) are used for storing digital data in a compressed form. A digital disk includes a single spiral track whose relief is representative of the binary information stored on the track of the disk. The track of the disk is illuminated by an incident optical beam (e.g., a laser beam) and several photodetectors (e.g., four) detect the reflection of the light beam on the disk. The optical pickup formed by the photodetectors then delivers four elementary signals from the four photodetectors. The four elementary signals are also used for carrying out a slaving of the optical beam to the track of the disk. An overall or useful signal equal to the sum of the four elementary signals is also delivered by the optical pickup, from which the binary information read on the track may be extracted.
The coding of the binary information on the disk is standardized and well known to those of skill in the art (eg., RLL (2,10) coding). The length of the hollows and of the bumps present on the spiral track of the disk determines the number of logic 0 values flanked by two logic 1 values. Also, these lengths of hollows and bumps are all multiples of a base length commonly referred to by the designation “1T.” By way of example, the value of +he base length 1T is equal to 0.64 microns for a OVD disk and 1.6 microns for a CD-ROM.
When the digital disk is rotating, the useful signal containing the binary data, also referred to as the “incident signal”, herein includes a succession of transitions whose spacings are representative of the lengths of the pulses. Also, the higher the speed of rotation of the disk, the smaller the spacings between the transitions. Extracting the binary data conveyed by the incident signal thus includes detecting the transitions of the incident signal, calculating the distances separating the successive transitions, and determining the values of the data from the calculated distances.
At present, a digital phase-locked loop is used for extracting the binary data which, for each speed of rotation, uses a predefined value of a base distance corresponding to the base length (1T, also referred to as the “period 1T”). Furthermore, a phase-locked loop generally includes a digital filter, and the coefficients of the filter may depend upon on the speed of rotation. Of course, the linear rotation speed of a portion of a given track will vary depending upon on the distance of the portion from the center of the disk.
Additionally, when the laser beam is instructed (by the microprocessor of a computer, for example) to perform a displacement jump from one portion of a track to another portion of the track situated, for example, further out on the disk, the correct locking of the phase-locked loop requires the use of the predefined stored value of the base distance (1T), which corresponds to the new linear rotation value. Furthermore, the coefficients of the filter must likewise be modified by using a preprogrammed set of coefficients corresponding to the new rotation speed. Yet, if a positioning error is committed (i.e., if the information given by the photodetector leads to poor actual locating of the track portion and hence to a poor estimate of the new linear rotation speed), the phase-locked loop will use an inappropriate value for the new period 1T, as well as an inappropriate set of filter coefficients. Also, this may lead to a much longer locking of the phase-locked loop, and consequently to a much greater latency time before it is possible to extract correct data. In the worst case, locking is not performed and the disk is then ejected from the carrier.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a method for extracting binary data conveyed by an incident signal that obviates the above problems.
According to the invention, a method for extracting binary data conveyed by an incident signal is provided where the binary data is coded in the form of a pulsatile signal whose pulses have variable lengths which are multiples of a base pulse length (period 1T). The incident signal includes a succession of transitions whose spacings are representative of the lengths of the pulses. The process may include detecting the transitions of the incident signal, calculating the distances separating the successive transitions, and determining the values of the data from the calculated distances.
More particularly, the method may include an initialization phase wherein the value of a base distance corresponding to the base pulse length is determined from the contents of the incident signal. The method may also include an extraction phase in which a set of reference values corresponding respectively to various multiples of the determined base distance (1T, 2T, . . . , 14T in the case of a DVD disk) is formulated. In certain cases a zero reference value corresponding to 0T may be stored. For a calculated current distance, the values of the data corresponding to this current distance are determined from a comparison of the reference values and a current corrected distance. The current corrected distance is formulated from the current calculated distance, from a comparison error relating to the previous calculated distance, and from the filtered comparison error.
Stated otherwise, the invention involves estimating the value of the base distance from the actual content of the incident signal, thereby intrinsically taking into account the rotation speed. In contrast, prior art methods typically require predefined and preprogrammed values corresponding to this base distance as a function of the various possible rotation values of the disk.
Furthermore, in the extraction phase the phase error may be directly corrected on the detected transitions of the signal (i.e., in tempo with the detected transitions) rather than in tempo with the signal sampling frequency, as in the prior art. Therefore, the coefficients of the filter of the phase-locked loop according to the invention become independent of the rotation speed of the disk and depend only on the physical characteristics of the disk, such as the inaccuracies of etching the track, for example. Thus, correction of the phase error is much more effective and rapid. Consequently, the latency duration for obtaining correct extracted binary data is also reduced.
The estimation of the period 1T may advantageously be performed algebraically. More particularly, according to one embodiment of the invention, the initialization phase may include at least one first subphase including the formulation of at least one first threshold distance from the current maximum calculated distance and corresponding to a first threshold length (e.g., 3.5T) situated between first and second successive multiples of the base length (e.g., 3T and 4T). This first subphase may also include comparing each current calculated distance with the first threshold distance, adding a first predetermined number (e.g., 21) of values of current calculated distances less than the first threshold distance, and dividing the sum thus obtained by a first predetermined divisor (for example 64) to obtain an estimated value of the base distance.
Such an initialization phase allows very rapid estimation of the value of the period 1T, typically within a duration equal of about two frames, i.e., 120 microseconds for a rotation speed of 1× (where a rotation speed of 1× corresponds to 4 m/second). Indeed, it has been observed that 70% of the logic values of the data transmitted corresponded to multiples 3T and 4T. Also, the combination of the particular characteristics of this initialization phase, with the estimation of the value of the period 1T from the actual conten
Allen Dyer Doppelt Milbrath & Gilchrist, P.A.
Jorgenson Lisa K.
Psitos Aristotelis M.
STMicroelectronics S.A.
LandOfFree
Process and device for extracting binary data conveyed by an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process and device for extracting binary data conveyed by an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process and device for extracting binary data conveyed by an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3052185