Browse Prior Art Database

Adaptive Digital Readback Equalization for Recording Devices

IP.com Disclosure Number: IPCOM000110505D
Original Publication Date: 1992-Dec-01
Included in the Prior Art Database: 2005-Mar-25
Document File: 2 page(s) / 90K

Publishing Venue

IBM

Related People

Baumann, GW: AUTHOR

Abstract

This invention is an adaptive digital filter for the read channel of a magnetic or optical recording device. When the readback signal is attenuated by a defect (e.g., debris between magnetic tape surface and read-back head gap), the readback amplitude versus frequency is monitored through the defect region. If a clocking error is then defected, and the normal error recovery procedure fails, the digitized signal is equalized by a slewing digital filter which is tied to the record of frequency-dependent attenuation versus time (or bit). The resultant equalized signal is then passed again through the clocking and detection algorithm to regenerate the bit stream.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Adaptive Digital Readback Equalization for Recording Devices

       This invention is an adaptive digital filter for the read
channel of a magnetic or optical recording device.  When the readback
signal is attenuated by a defect (e.g., debris between magnetic tape
surface and read-back head gap), the readback amplitude versus
frequency is monitored through the defect region.  If a clocking
error is then defected, and the normal error recovery procedure
fails, the digitized signal is equalized by a slewing digital filter
which is tied to the record of frequency-dependent attenuation versus
time (or bit).  The resultant equalized signal is then passed again
through the clocking and detection algorithm to regenerate the bit
stream.

      Digital recording devices, including magnetic tape drives,
magnetic disc drives, and optical disc drives, are engineered to have
very robust hard error rates and to achieve very high areal density.
These objectives, while both desirable, compete with one another.
High areal density results in low signal levels and low signal to
noise ratios.  These weak signals then result in high soft error
rates (i.e., high levels of correctable read errors).  To achieve
acceptable hard error rates, high level (or layered) error correction
codes (ECCs) are used.  In recent products, such as the Exabyte 8mm
EXB-8200 CTS drive, as many as 400 bytes of ECC are used to protect
1024 bytes of user data.  As signal to noise ratios decrease with
higher density, this ECC overhead grows, penalizing both user
capacity and user data rate.  Moreover, as the level or layering of
ECC increases, the power relationship between hard and soft errors
increases, so that slight increases in soft errors cause disastrous
increases in hard errors.  Hence, ECC alone will not indefinitely
push back the limitations to areal density and error rates.

      This new equalization artificially boosts the apparent signal
to noise ratio of the device, enabling improvements to either areal
density or da...