Dynamic magnetic information storage or retrieval – Monitoring or testing the progress of recording
Reexamination Certificate
2002-04-04
2004-03-16
Hudspeth, David (Department: 2697)
Dynamic magnetic information storage or retrieval
Monitoring or testing the progress of recording
C360S046000
Reexamination Certificate
active
06707624
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to the field of perpendicular recording in data storage systems. More particularly, but not by limitation, the present invention relates to methods and apparatus for preprocessing perpendicular recording read-back signals in order to improve Bit-Error-Rate (BER) performance which is adversely affected by low frequency noise sources in the perpendicular recording read-back signals.
BACKGROUND OF THE INVENTION
Recording medium for recording data include magnetic discs, optical discs, magneto-optical discs, and magnetic tapes, for example. When data is digitally recorded on, or read-back from, a recording medium, it is preferable that the data is recorded at a high density Because of the enormous increase in demand for data storage capacity, research on general recording systems has resulted in the investigation of many potential methods and architectures for increasing the capacity of storage media. This is also true for magnetic recording, for example in magnetic disc drive data storage systems in which architectures which may result in a higher areal density are being explored.
One of the recent architectures in the field of magnetic recording, which is believed to have more potential than the already existing longitudinal recording architectures, is perpendicular recording. In longitudinal recording, the magnetic medium on the disc is magnetized parallel to the surface of the disc. In perpendicular recording, however, the medium is magnetized perpendicular to the surface of the disc.
In addition to its potential to achieve higher areal densities, the specific nature of perpendicular recording also brings its own difficulties. One of the main difficulties relates to low frequency components of the perpendicular read-back signals. A significant amount of the data in read-back signals is at low frequencies. In other words, there is a considerable amount of information about the written bits at the low frequency portion of the read-back signal. However, a significant amount of dominant noise sources is also located at low frequencies. Further, since the direct current (DC) component of a perpendicular read-back signal causes unwanted DC-coupling in the system, it is typically necessary to get rid of the DC content of the signal. Given that the significant amount of dominant noise sources is located at low frequencies, and that it is typically preferable to eliminate the DC content of the signal, it might seem to be beneficial to also get rid of the low frequency components of the signal. However while filtering out the low frequency components, a considerable amount of information about the data is also lost. Thus, there is a trade-off between removing the DC content and the dominant portions of the low frequency noise components, and losing some information about the data which might be used to increase system performance.
An ideal way to deal with this problem would be to use all kinds of information about the perpendicular channel, the data, and the noise to retrieve the data from the read-back signal. However, the dominant noise sources in the system are not Gaussian, and they are data dependent. For example, Base-Line Wander (BLW) and media noise are dependent on on-track data, while the neighboring track interference depends on data at adjacent tracks. The data dependency of noise and the non-Gaussian shape of its power spectrum makes it very difficult to come up with signal processing algorithms to effectively deal with the noise. Even if it can be done, this type of algorithm is generally the responsibility of the detection block of the electronics, which creates difficulty in the design of the detection block, likely resulting in a very complex algorithm and design.
Previously, the idea of transforming a perpendicular signal to a longitudinal one has been explored. Several factors served as motivation for this transformation. First, there generally is not a DC coupling problem or a problem with the low frequency components of read-back signals in longitudinal architectures. Second, transforming the perpendicular channel to a longitudinal one to get rid of DC coupling also makes use of existing longitudinal channel designs possible. Third, the noise at low frequencies will also be transformed, potentially allowing the existing longitudinal channel design to deal with the new longitudinal looking channel.
This idea of transforming a perpendicular signal to a longitudinal one does not consider any noise analysis in the system, and only expects the existing longitudinal channel designs to handle the new transformed noise effects in the system. However, while transforming perpendicular to longitudinal, some noise in the system is boosted, and/or some valuable portion of information about the data is lost. For this reason, transforming the perpendicular signals to longitudinal looking signals and applying the existing longitudinal architectures is likely to result in a worse Bit-Error-Rate (BER) performance than would result if the signal had not been transformed.
Consequently, an algorithm which both performs better than these different realizations of the idea of transforming perpendicular to longitudinal, and which also improves the overall system performance, would be a significant improvement in the art.
SUMMARY OF THE INVENTION
Embodiments of the present invention relate to methods and apparatus for preprocessing perpendicular read-back signals in order to improve the BER by removing low frequency noise while preserving low frequency signal content which contains information useful in determining values of bits read from a storage medium.
The present invention addresses the aforementioned problems by providing a method and apparatus for preprocessing low frequency components of a perpendicular read-back signal in a data storage system in order to reduce low frequency noise. First, a dominant known perturbation is introduced into the system. For this purpose, a dominant high-pass pole can be chosen. This high-pass pole masks the effects of other low frequency noise sources leaving the system with one dominant low frequency noise component which is artificially introduced into the system and is perfectly known. In presence of this high-pass pole, the system can be seen to have only the BLW effect produced by this pole. Then, the artificially introduced dominant perturbation is removed from the system.
In one embodiment, a method of the present invention includes introducing a dominant known perturbation, to a perpendicular read-back signal, which masks the effects of other low frequency noise sources and leaves the read-back signal with one dominant low frequency noise component. The method then further includes removing the dominant known perturbation from the read-back signal to recover low frequency portions of the read-back signal in order to determine values of bits read from a storage medium.
These and various other features as well as advantages which characterize embodiments of the present invention will be apparent upon reading of the following detailed description and review of the associated drawings.
REFERENCES:
patent: 4085427 (1978-04-01), Dunn
patent: 4736355 (1988-04-01), Kanamaru
patent: 4811262 (1989-03-01), White
patent: 5585986 (1996-12-01), Parkin
patent: 6046873 (2000-04-01), Hori et al.
Erden Mehmet Faith
Kurtas Erozan Mehmet
Hudspeth David
Rodriguez Glenda P.
Seagate Technology LLC
Westman Champlin & Kelly P.A.
LandOfFree
Method and apparatus for preprocessing low frequencies in... 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 and apparatus for preprocessing low frequencies in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for preprocessing low frequencies in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3254812