Coded data generation or conversion – Analog to or from digital conversion – Analog to digital conversion
Reexamination Certificate
2000-07-07
2002-04-09
Tokar, Michael (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Analog to digital conversion
C341S157000, C341S162000, C341S144000, C341S151000, C341S152000, C341S126000, C341S139000
Reexamination Certificate
active
06369741
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates in general to analog signal processing circuits, and more particularly, to processing systems of analog signals originating from electromagnetic transducers (heads) for reading data stored on magnetic media, for example, in read channels of hard disk drives (HDD).
BACKGROUND OF THE INVENTION
Systems for reading data from magnetic media of mass memory devices such as hard disks, magnetic tape, floppy disks and the like, used in personal computers, should have the highest possible signal
oise ratio (SNR) in order to comply with the reliability and speed requisites of data transfer. Depending on the type of read channel (PR4, EPR4, EEPR4 . . . s), the requirements to fulfill may be more or less rigorous. However, along the analog path of the signal originating from a read head (pick-up), it may be necessary to operate a pre-equalization to accentuate the gain of the read channel in the frequency range in which the signal presents maximum energy, thus making the equalization process carried out by an adaptive filter (FIR) more efficient. The adaptive filter may be either analog or digital.
Typical functional schemes of a read channel for HDD are depicted in
FIGS. 1 and 2
. In the examples of these figures, the read channels, which always include, besides a pre-amplification stage, an automatic gain control circuit (AGC) commonly realized by a variable gain amplifier (VGA) controlled via the digital/analog converter DAC_VGA, that keeps the amplitude of the signal input to the analog/digital converter ATOD constant, irrespective of the amplitude of the input signal or of the filter gain in the signal band.
The Magnetic Resistive Asymmetry block (MRA) eliminates or in any case decisively attenuates the second harmonic (that is, the contribution of the term ax
2
) of the analog input signal originating from the read pick-up (MR). This operation also corrects the signal spectrum and is dynamically controlled through the digital/analog converter DAC_MRA. Depending on the required equalization and on the characteristics of the transducers, the boost of the characteristic (transfer function) of the low pass filter LPS is programmed through the DAC_BOOST converter, whereas the cut-off function of the pre-equalization filter is programmed through the DAC_FC converter. The programming digital commands (Word FC and Word Boost) read for example from a read only buffer, are converted through these converters into analog signals that control the transfer function of the low pass filter LPF.
An OFFSET STAGE, whose function is to eliminate the eventual offset existing on the output of such a low pass filter of pre-equalization LPF, is coupled in cascade of the low pass filter LPF. Even in this case, the DIGITAL POST PROCESSING block controls the offset counterbalancing stage through the digital/analog converter DAC_OFF. The control block referred to as DIGITAL POST PROCESSING operates, through the control loop realized by the DAC converter and by the VCO (Voltage Control Oscillator), the preservation of a correct sampling synchronism by the ATOD converter. The equalization is thereafter completed in an adaptive mode or in analog mode through an analog FIR (
FIG. 1
) or else in a digital mode through a digital FIR (FIG.
2
).
Normally, the parameters (fc and boost) of the low pass filter of pre-equalization are preset during the fabrication process of the drive. Usually, the programming values of the parameters are stored for every track of the disk, and are then read while the drive is operating. These pre-programmed values are generally optimized on each single disk in case of high quality disks, whereas for less expensive disks the same values are used for all the disks (statistical optimization).
In both cases the FIR allows for the completion of the equalization even if the parameters of the pre-equalization filter are not perfectly adjusted to the characteristics of the particular disk or equivalent magnetic support. In this case however, the FIR operates under non-optimal conditions, as it must recover a certain precision of equalization on signals that are exceedingly off equalization.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method of defining, in a relatively precise manner, the programming values of the parameters, overshoot (boost) and cut-off frequency (fc), of the low pass filter of pre-equalization, thus ensuring that the adaptive filter of equalization operates under optimal conditions.
The method of the invention includes establishing the digital values for the programming of the parameters, boost and cut-off frequency, of the low pass filter of pre-equalization for each magnetic-medium, as a function of the instantaneous operating parameters, purposely sensed, of the adaptive filter that carries out the definitive equalization of the signal. The method of the invention allows for an easy and effective programming of the parameters of the filter of pre-equalization that can be economically carried out for each drive, that is, for each disk or similar magnetic mass storage medium. Such an individual programming of the parameters of each drive may be carried out by using a dedicated program that executes the algorithm of the invention.
The procedure of the invention includes carrying out a read scanning of the magnetic medium while detecting the working coefficients of the adaptive filter at the scanning phase or instant i. Then, the procedure includes generating instantaneous programming digital values of the boost and cut-off frequency of the low pass filter of pre-equalization for a successive scanning phase or instant i+1, as values rounded to the next integer value of the sum of the relative digital value at the instant i and of the offsets of the values of the relative work coefficients of the adaptive filter at the instant i in respect to their mean value. These values are weighed by the gain of the respective regulation loop of the parameter of the pre-equalization filter.
The pre-programming algorithm of the operating parameters of the low pass filter of pre-equalization may be implemented inside the chip of the read channel, that is according to a hardware solution. Alternatively, the algorithm of the invention may also be implemented in the form of an optimizing software to trim devices lacking such a hardware implementation.
REFERENCES:
patent: 5541600 (1996-07-01), Blumenkrantz et al.
patent: 5598160 (1997-01-01), Matsushima
Bollati Giacomino
Demicheli Davide
Demicheli Marco
Marchese Stefano
Allen, Dyer, Doppelt, Milbrath & Gilchrist, P.A
Jorgenson Lisa K.
Mai Lam T
STMicroelectronics S.r.L.
Tokar Michael
LandOfFree
Optimization of the parameters of a pre-equalization low... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Optimization of the parameters of a pre-equalization low..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimization of the parameters of a pre-equalization low... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2877867