Dynamic information storage or retrieval – Binary pulse train information signal – Including sampling or a/d converting
Reexamination Certificate
1999-01-29
2003-03-04
Young, W. R. (Department: 2653)
Dynamic information storage or retrieval
Binary pulse train information signal
Including sampling or a/d converting
C369S044340, C369S124050
Reexamination Certificate
active
06529460
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to an improvement in a method and system for estimating pulse peak amplitudes and pulse peak instances (the times or instants at which the amplitude of pulse peak occurs) of serial analog pulse sequences in which the pulses have varying amplitude, pulse widths and base levels in the presence of noise. Specifically, this is a simple but elegant digital processing approach applied to determining pulse peak amplitudes and peak instances of data pulses recovered from stored data retrieved from magneto-optical (MO) drives.
BACKGROUND
With respect to parent case Ser. No. 09/132,962 referenced above and incorporated herein by reference, there is disclosed a typical disk with servo and data sectors. The servo sector is described to have a particular architecture, i.e., a physical arrangement or pattern, of data pits, grouped into patterns. The patterns define radial servo-timing-marks (STM) marks, formed of contiguous pits of constant size between circular inner diameter ID and outer diameter OD boundaries; the patterns form data track and servo sector address marks and Position-Error-Sensing (PES) marks.
In the system of Ser. No. 09/132,962, data may be modulated by pit position modulation. Therefore, the data pit and laser spot size must be about the same size to keep pulse amplitude about the same. Returned signal pulses have a narrow full width half height maximum FWHM pulse width at the OD and a wide pulse width at the ID. The circumferential spacing between radii in the servo sector is wider at the OD than at the ID. The pits toward the OD are spaced farther apart laterally than the pits toward the ID, even though they have the same width.
With regard to
FIGS. 6
a
and
6
b
in the present application, typical signal pulses that may be recovered from a disk in drive system are depicted. In a typical servo sector, the first digital data to be recovered are the STM pulses. Pulses in such systems exhibit variation in pulse width, peak amplitude, base line displacement from reference zero level, and noise on the waveform.
With regard to
FIG. 4
a
in the present application, some characteristics of disk drive data pulses not treated by the disclosure Ser. No. 09/132,962 are shown. A signal pulse
452
recovered from an inside data (ID) track has a wider pulse width ID PW
460
than a pulse width OD PW
462
from a signal pulse recovered from an outside (OD) track OD pulse
454
. Even though the peak amplitudes Apj(ID)
464
and Apj(OD)
466
of the ID and OD pulses may be the same, OD PW
462
is narrower than ID PW
460
due to the higher relative speed of the data bit and the read head toward the perimeter of the disk. Variation is also typical in the level of background reflectance which corresponds to the base signal level Base Line
1
470
. This is indicated by a lower base signal level Base Line
2
472
. The base line signal level Base Line
1
470
appears as an offset &bgr; in the signal level, from a zero level reference line
474
. The maximum pulse deflections from the Base Line
470
represent the peak amplitude of the pulses, i.e., Apj(ID)
464
for the ID pulse and Apj(OD)
466
for the OD pulse. With component aging and disk contamination build up over time, Apj(ID)
464
and Apj(OD)
466
also decrease. This is indicated by the dashed lines of
FIG. 4
a
in the present application. Digital processing of the signals ID pulse
452
and OD pulse
454
require sampling of the signals with an analog to digital converter at clock tick, k, and with a sufficiently small sample period
480
, with a analog-to-digital conversion device having a sampling range
484
sufficiently large to cover the Base Line
470
expected. Random system noise
486
on the pulse signals ID pulse
452
and OD pulse
454
adds quantizing noise to any digital signal processing performed by the system
100
.
With regard to Ser. No. 09/132,962, a pulse data recovery system is used in a magneto-optic disk drive. A laser spot is directed to a disk surface from a read/write head. Light reflected from the disk surface is received by the head and processed by a signal channel. The magnitude of the reflected signal from the disk surface is a constant (a base line level from a zero level reference) where the surface is flat. Pits formed in the disk surface during a mastering process cause the reflected signal near a pit to decrease as the laser spot passes over the pit edge because of destructive interference.
The resulting magnitude variation from a constant base value to a minimum peak and back to the base value is detected as a pulse signal. Achieving maximum peak pulse amplitude depends on having pit width and spot width of similar size. The pit width can be optimized for maximum signal robustness to width variations by diffraction modeling, One typical case has a pit size of about 350 nm and a 550 nm lambda wavelength laser full width half maximum (FWHM) spot size of 660 nm. The detected pulse width from the reflection of, an illuminating laser beam returned from a pit on the disk surface is related to the size of the laser spot and the size of the pit. Since pits near the OD are traveling at a higher linear speed (constant angular velocity at a greater radius) than those at the ID, the data pulse widths at the OD are correspondingly narrower. As the disc size and rotation speed are increased to achieve greater data capacity and data transfer rates, the difference between the detected pulse widths near the ID and near the OD becomes greater. In a typical application this could lead to a ratio of pulse widths of 2:1 or greater.
In Ser. No. 09/132,962, a digital signal processing channel (PDC) for recovering peak pulse amplitude and peak pulse instance is disclosed. The PDC is an invention of a digital circuit implementation of a pulse detection method and system using quadratic interpolation, peak amplitude estimation. The mariner of how the previous PDC works in combination with a Pulse Peak Synchronizer (PPS), a Servo Timing Mark Detector (STMD), a servo sector architecture, and cooperating system electronics (DDCS) is briefly summarized here.
The PDC invention of Ser. No. 09/132,962 provides an estimate, Ep(j), for the peak amplitude Ap(j) and an estimate, Toffset, for the offset of the peak instance tpj of a detected pulse from a sampling instance, e.g., a sampling clock SYSCLK. The estimates Ep(j) and Toffset, relative to the center sample of a multi-sample frame, are derived from an equation for a curve fitting parabola when the curve fitting parabola is fit to three adjacent samples of a respective data pulse. When the amplitude of a center sample is greater than or equal to the amplitude of one of the adjacent samples and is greater than the amplitude of the other adjacent sample comparator, logic sub-circuits in the PDC give indication to the system that the peak instance has occurred next to the center sample.
Pj is sampled asynchronously with SYSCLK having a sampling period Tclk more than about ⅕ and less than about ⅓ a nominal minimum pulse period Tpsmin. Tclk is provided in the system
100
at such a rate that the each pulse is consecutively sampled above a threshold value.
In a particular embodiment of the invention disclosed in Ser. No. 09/132,962, the pulse waveforms amplitudes are sampled at about 50 MHZ. The pulse amplitudes are sampled with a high-speed A to D Converter (ADC). The sampled amplitude values and identifying sample clock ticks are processed by subsystems of the invention to determine the accurate time estimates of the instance of a data pulse peak relative to the timing of a system logic bit frame. Further processing of sampled data pulse amplitudes and identifying sample clock ticks by embodiments of this invention provide accurate estimates of the instances of the pulse peaks and estimates of the pulse peak amplitudes. These estimates are provided for use by the detection and control electronics (DDCS) of the disk drive system to enable system performance enhancements, e.g. PES processing and the like.
In Ser. No.
Berger Derek J.
Cesari Kirk A.
Chu Kim-Kwok
Seagate Technology LLC
Young W. R.
LandOfFree
Detection of pulse peak instance and amplitude in a storage... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Detection of pulse peak instance and amplitude in a storage..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Detection of pulse peak instance and amplitude in a storage... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3031113