Read channel apparatus for asynchronous sampling and synchronous equalization

1. A method to read information encoded in a data storage medium, comprising the steps of:

supplying a read channel comprising an analog to digital converter, an upsampler interconnected with said analog to digital converter, an interpolator interconnected with said upsampler, a fractionally-spaced equalizer interconnected with said interpolator, a gain control module interconnected with said fractionally-spaced equalizer, a sequence detector interconnected with said gain control module, a timing control circuit interconnected with said sequence detector and with said interpolator;

reading a data track written to a data storage medium to form an analog signal, wherein said data track comprises information encoded at a first rate;

asynchronously sampling at a fixed sampling rate said analog signal to form sampled signals;

upsampling in the asynchronous domain at an upsampling rate, wherein said upsampling rate is greater than said first rate;

forming interpolated samples;

providing said interpolated samples to said fractionally-spaced equalizer at an interpolation rate, wherein said interpolation rate is greater than said first rate;

forming synchronous, equalized samples at an equalization rate, wherein said equalization rate is less than said interpolation rate;

adjusting the gain of said equalized samples;

generating a timing signal;

providing said timing signal to said interpolator;

generating an equalizer error signal;

providing said equalizer error signal to said fractionally-spaced equalizer;

detecting data symbols representing a maximum likelihood sequence estimate using said gain-adjusted, synchronous, equalized samples,

generating signal estimates formed by decisions from said sequence detector;

wherein said generating an equalizer error signal step further comprises generating an equalizer error signal comprising the difference between said signal estimates and a delayed equalized signal, before gain control.

2. The method of claim 1 , further comprising the steps of:

generating a gain control error signal comprising the difference between a delayed gain-adjusted signal and said signal estimates;

providing said gain control error signal to said gain control module;

wherein said generating an equalizer error signal step and said generating a gain control error signal steps are decoupled.

3. The method of claim 2 , wherein said supplying a sequence detector step further comprises supplying a sequence detector comprising a target, said method further comprising the steps of:

generating an adjusted target;

providing said adjusted target to said sequence detector;

setting said target to said adjusted target.

4. The method of claim 2 , wherein said supplying a sequence detector step further comprises supplying a Partial Response Maximum Likelihood sequence detector, and wherein said detecting step further comprises detecting data symbols representing a maximum likelihood sequence estimate using said Partial Response Maximum Likelihood sequence detector.

5. The method of claim 2 , wherein said supplying a sequence detector step further comprises supplying a Noise Predictive Maximum Likelihood sequence detector, and wherein said detecting step further comprises detecting data symbols representing a maximum likelihood sequence estimate using said Noise Predictive Maximum Likelihood sequence detector.

6. The method of claim 5 , wherein said Noise Predictive Maximum Likelihood sequence detector comprises a branch metric computation algorithm and a plurality of predictor filters, wherein said predictor filters are embedded into said branch metric computation algorithm, wherein each transition of said algorithm is associated with a different one of said plurality of predictor filters.