Pulse or digital communications – Equalizers – Automatic
Reexamination Certificate
1998-08-31
2001-04-24
Le, Amanda T. (Department: 2634)
Pulse or digital communications
Equalizers
Automatic
C375S350000, C708S323000
Reexamination Certificate
active
06222879
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention.
This invention relates in general to a data channel, and more particularly to a method for automating the convergence of tap weights in an equalizer for a data channel.
2. Description of Related Art.
Computers often include auxiliary memory storage having media on which data can be written and from which data can be read. Disk drive units incorporating stacked, commonly rotated rigid magnetic disks are one example of storage media. Tape drives and optical storage units are other examples of storage media.
In a disk drive, the data are stored in magnetic form on the disk surfaces. Data are recorded in concentric, radially spaced data information tracks arrayed on the surfaces of the disks. Transducer heads driven in a path toward and away from the drive axis write data to the disks and read data from the disks.
To achieve high data density in writing and reading data on storage media a partial response maximum likelihood (PRML) channel is often used. To obtain full advantage of the PRML channel, the received signal or the read signal must be filtered through a specially designed equalizing filter. For example, a common problem encountered when electronically reading or transmitting data is that it becomes corrupted by such things as background noise, impulse noise, fades, etc. Usually this data corruption is statistical phenomenon which causes additive and/or multiplicative transformations to the originally transmitted data. Thus, the original data undergoes changes such as frequency translation, non-linear or harmonic distortion, and time dispersion. In addition, high speed data transmission over channels of limited bandwidth results in a type of distortion commonly referred to as intersymbol interference.
One technique for reducing intersymbol interference includes equalizing the data using an equalizer that compensates for the average of the range of expected channel amplitude and delay characteristics. However, adaptive equalizers suffer from a relatively long convergence time for a least means square (LMS) algorithm. Another limitation of equalizers is that since they are implemented in digital circuitry, the data must be quantized prior to being processed by, for example, a finite impulse response (FIR) filter.
Current method of equalization for PRML channels involve tester software that executes a sequence of steps for optimizing the taps of the filter. For example, a set of tap weights are loaded into a finite impulse response filter (FIR) of the channel. A measurement window is set with a start and stop byte count which defines the data zone that will be used to make a measurement. A read command is then issued to the file. During the read command, the channel measures the mean squared error of the data that is inside the measurement window.
After the read completes, the value of the error is read into the tester software. The tester software then modifies the tap weights using one of several possible convergence methods and re-loads new tap weights into the channel. A new measurement is then made. If the error is smaller, then the new taps are stored and a new trial is attempted. Eventually, the optimum taps are obtained and the algorithm stops.
Nevertheless, this process is complicated and requires a tester to operate. Further the process is time consuming and the tester software must be rewritten for each new product.
It can be seen that there is a need for a simple, automated equalization method.
It can also be seen that there is a need for an automated equalization method that is contained within the channel itself.
SUMMARY OF THE INVENTION
To overcome the limitations in the prior art described above, and to overcome other limitations that will become apparent upon reading and understanding the present specification, the present invention discloses a method for automating the convergence of tap weights in an equalizer for a data channel.
The present invention solves the above-described problems by providing a simple, automated equalization method that is contained within the channel itself.
A method in accordance with the principles of the present invention includes obtaining a temporary copy of each of a plurality of current tap weights from a current tap weight memory, determining a direction for modifying the temporary copy of each of the plurality of current tap weights, constraining the modifications to the temporary copy of each of the plurality of current tap weights to produce adjusted tap weights and at the end of a read operation, storing the adjusted tap weights to the current tap weight memory for use during a next read operation.
Other embodiments of a system in accordance with the principles of the invention may include alternative or optional additional aspects. One such aspect of the present invention is that the determining further includes calculating the sign-sign LMS products for each of the plurality of current tap weights, summing the sign-sign LMS products for each of the plurality of current tap weights to produce a count and comparing the count for each of the plurality of current tap weights to a threshold produce a desired tap direction indicator for each of the plurality of current tap weights.
Another aspect of the present invention is that the desired tap direction includes a positive value if the count is more positive than the threshold and a negative value is if the count is more negative than the threshold.
Another aspect of the present invention is that the summing further includes incrementing the count by +1 if the sign product is positive and by −1 if the sign product is negative.
Another aspect of the present invention is that the constraining further includes moving odd taps in pairs having opposite signs so that the sum of the odd taps is unchanged.
Another aspect of the present invention is that the constraining further includes moving even taps in pairs having opposite signs so that the sum of the even taps is unchanged.
Another aspect of the present invention is that the constraining further includes moving even taps according to −T
0
+T
2
−T
4
+T
6
=−71 lsbs, wherein Ti represents the ith tap, and wherein the taps are moved in pairs having opposite signs so that the sum of the even taps remains −71 lsbs.
Another aspect of the present invention is that the constraining further includes moving odd taps according to −T
1
+T
3
−T
5
+T
7
=−71 lsbs, wherein Ti represents the ith tap, and wherein the taps are moved in pairs having opposite signs so that the sum of the odd taps remains −71 lsbs.
Another aspect of the present invention is that a finite impulse response filter is provided that includes a delay line having N−1 delay latches for receiving a data signal and processing the data signal through the N−1 delay latches to produce N sampled signals at N taps, N multipliers, coupled to the N taps of the delay line, for multiplying each of the N sampled signals by a corresponding Nth tap weight to produce a product signal, an adder, coupled to the multiplier, for summing the product signals of the N multipliers to produce an equalized sample and a subtractor, coupled to the adder, for subtracting an ideal equalized sample from the equalized sample to produce an error signal, the error signal being used to set the Nth tap weight for the N multipliers.
These and various other advantages and features of novelty which characterize the invention are pointed out with particularity in the claims annexed hereto and form a part hereof. However, for a better understanding of the invention, its advantages, and the objects obtained by its use, reference should be made to the drawings which form a further part hereof, and to accompanying descriptive matter, in which there are illustrated and described specific examples of an apparatus in accordance with the invention.
REFERENCES:
patent: 5268848 (1993-12-01), Coker et al.
patent: 5426541 (1995-06-01), Coker et al.
patent: 54915
Cideciyan Roy D.
Coker Jonathan D.
Eleftheriou Evangelos S.
Galbraith Richard L.
Stephenson John J.
Hollingsworth Mark A.
International Business Machines - Corporation
Le Amanda T.
Lynch David W.
LandOfFree
Self contained equalization method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Self contained equalization method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Self contained equalization method and apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2445726