Error detection/correction and fault detection/recovery – Pulse or data error handling – Digital data error correction
Reexamination Certificate
1999-09-30
2003-12-09
Moise, Emmanuel L. (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Digital data error correction
C714S796000, C713S400000, C375S262000, C375S265000, C375S341000
Reexamination Certificate
active
06662338
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
The invention is related generally to electronic circuits, and more particularly to a parity-sensitive Viterbi detector and technique for recovering information from a read signal. In one embodiment, the Viterbi detector is parity sensitive, and recovers only data sequences having the correct parity. Such parity checking allows the Viterbi detector to more accurately recover information from a read signal having a reduced signal-to-noise ratio (SNR). By allowing the read signal to have a reduced SNR, the Viterbi detector allows one to increase the area density (number of storage locations per square inch) of a storage disk.
BACKGROUND OF THE INVENTION
The storage capacity of a magnet disk is typically limited by the disk surface area and the minimum read-signal SNR specified for the data recovery circuit . Specifically, the diameter of the disk, and thus the disk surface area, are typically constrained to industry-standard sizes. Therefore, the option of increasing the surface area of the disk to increase the disk's storage capacity is usually unavailable to disk-drive manufacturers. Furthermore, the SNR of the read signal is a function of the data-storage density on the surface or surfaces of the disk; the higher the storage density, the lower the SNR of the read signal, and vice-versa. Typically, as the SNR of the read signal decreases, the number of read errors that the recovery circuit introduces into the recovered data increases. Unfortunately, an increase in the number of read errors may degrade the effective data-recovery speed of a disk drive to unacceptable levels.
FIG. 1
is a circuit block diagram of part of a conventional disk drive
10
, which includes a magnetic storage disk
12
and a read channel
14
for reading data from the disk
12
. The read channel
14
includes a read head
16
for sensing the data stored on the disk
12
and for generating a corresponding read signal. A read circuit
18
amplifies and samples the read signal and digitizes the samples, and a digital Viterbi detector
20
recovers the stored data from the digitized samples.
Typically, the greater the data-storage density of the disk
12
, the greater the noise the read head
16
picks up while reading the stored data, and thus the lower the SNR of the read signal. The disk
12
typically has a number of concentric data tracks (not shown in
FIG. 1
) that each have a respective number of data-storage locations. The storage density of the disk
12
is a function of the distances between storage locations along the circumferences of the respective tracks and the distances between respective tracks. The smaller these distances, the higher the storage density, and thus the closer the surrounding storage locations to the read head
16
when it is reading the surrounded location. The closer the surrounding locations to the read head
16
, the greater the magnitudes of the magnetic fields that these locations respectively generate at the head
16
, and thus the greater the Inter Symbol Interference (ISI). The greater the ISI, the smaller the root-mean-square (rms) amplitude of the read signal. In addition, as the storage density increases, the media noise increases. Generally, the media noise results from the uncertainty in the shapes of the read pulses that constitute the read signal. This uncertainty is caused by unpredictable variations in the positions of the data storage locations from one data-write cycle to the next. Moreover, for a given disk spin rate, as the linear storage density along the tracks increases, the bandwidth of the read head
16
must also increase. This increase in bandwidth causes an increase in the white noise generated by the read head
16
. The SNR of the read signal for a particular storage location is the ratio of the rms amplitude of the corresponding read pulse to the sum of the amplitudes of the corresponding media and white noise. Thus, the lower the rms amplitudes of the read pulses and the greater the amplitudes of the media and/or white noise, the lower the SNR of the read signal.
Unfortunately, the Viterbi detector
20
often requires the read signal from the head
16
to have a minimum SNR, and thus often limits the data-storage density of the disk
12
. Typically, the accuracy of the detector
20
decreases as the SNR of the read signal decreases. As the accuracy of the detector
20
decreases, the number and severity of read errors, and thus the time needed to correct these errors, increases. Specifically, during operation of the read channel
14
, if the error processing circuit (not shown) initially detects a read error, then it tries to correct the error using conventional error-correction techniques. If the processing circuit cannot correct the error using these techniques, then it instructs the read channel
14
to re-read the data from the disk
12
. The time needed by the processing circuit for error detection and error correction and the time needed by the read channel
14
for data re-read increase as the number and severity of the read errors increase. As the error-processing and data re-read times increase, the effective data-read speed of the channel
14
, and thus of the disk drive
10
, decreases. Therefore, to maintain an acceptable effective data-read speed, the read channel
14
is rated for a minimum read-signal SNR. Unfortunately, if one decreases the SNR of the read signal below this minimum, then the accuracy of the read channel
14
degrades such that at best, the effective data-read speed of the disk drive
10
falls below its maximum rated speed, and at worst, the disk drive
10
cannot accurately read the stored data.
Overview of Conventional Viterbi Detectors, Read Channels, and Data Recovery Techniques
To help the reader more easily understand the concepts discussed above and the concepts discussed below in the description of the invention, a basic overview of conventional read channels, digital Viterbi detectors, and data recovery techniques follows.
Referring again to
FIG. 1
, the digital Viterbi detector
20
“recovers” the data stored on the disk
12
from the digitalized samples of the read signal generated by the read circuit
18
. Specifically, the read head
16
reads data from the disk
12
in a serial manner. That is, assuming the stored data is binary data, the read head
16
senses one or more bits at a time as the surface of the disk
12
spins it, and generates a series of sense voltages that respectively correspond to the sensed bits. This series of sense voltages composes the read signal, which consequently represents these sensed data bits in the order in which the head
16
sensed them. Unfortunately, because the disk
12
spins relatively fast with respect to the read head
16
, the read signal is not a clean logic signal having two distinct levels that respectively represent logic 1 and logic 0. Instead, the read signal is laden with noise and inter-symbol interference (ISI), and thus more closely resembles a continuous analog signal than a digital signal. Using the sample clock, which is generated with circuitry that is omitted from
FIG. 1
, the read circuit
18
samples the read signal at points that correspond to the read head
16
being aligned with respective bit storage locations on the surface of the disk
12
. The read circuit
18
digitizes these samples, and from these digitized samples, the Viterbi detector
20
ideally generates a sequence of bit values that is the same as the sequence of bit values stored on the disk
12
as described below.
FIG. 2
is a block diagram of the Viterbi detector
20
of FIG.
1
. The detector
20
receives the digitized read-signal samples from the read circuit
18
(
FIG. 1
) on an input terminal
22
. A data-sequence-recovery circuit
24
processes these samples to identify the bits represented by the read signal and then provides these identified bits to shift registers
26
, which reproduce the stored data sequence from these bits. The detector
20
then provides this reproduced data sequence on an output terminal
28
as
Marrow Marcus
Rezzi Francesco
Jorgenson Lisa K.
Moise Emmanuel L.
Santarelli Bryan A.
STMicroelectronics Inc.
LandOfFree
Parity- sensitive Viterbi detector and method for recovering... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Parity- sensitive Viterbi detector and method for recovering..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Parity- sensitive Viterbi detector and method for recovering... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3145960