Adaptive FIR filter and method

1. A finite impulse response filter comprising:

a block configured to generate a coefficient adjustment signal based, at least in part, on a voltage level of a received symbol in a pulse amplitude-modulated signal;

an integrator configured to generate an undated filter coefficient based, at least in part, on a tap coefficient associated with the finite impulse response filter and the coefficient adjustment signal; and

a multiplier coupled to the integrator and configured to multiply the received symbol with the updated filter coefficient to generate a correction signal;

wherein the block is further configured to generate the coefficient adjustment signal with a value such that the updated filter coefficient is equal to the tap coefficient in response to receipt of a symbol having a first specific voltage level and to generate the coefficient adjustment signal with a value such that the updated filter coefficient is different from the tap coefficient in response to receipt of a symbol having a second specific voltage level.

2. The finite impulse response filter of claim 1 , wherein the block comprises a lookup table configured to store an association between the received symbol and a value used to generate the updated filter coefficient.

3. The finite impulse response filter of claim 1 , wherein the pulse amplitude-modulated signal comprises an eight-level pulse amplitude-modulated signal, and wherein the block is further configured to generate the coefficient adjustment signal with a value such that the updated filter coefficient is equal to the tap coefficient in response to receipt of a symbol having an amplitude of +7 or −7 and to generate the coefficient adjustment signal with a value such that the updated filter coefficient is different from the tap coefficient in response to receipt of a symbol having an amplitude other than 7 and −7.

4. The finite impulse response filter of claim 1 , wherein the block is further configured to generate the coefficient adjustment signal based on a signal-to-noise ratio of the received symbol.

5. The finite impulse response filter of claim 1 , wherein the block is further configured to receive a first signal corresponding to a sign of the received symbol and a second signal corresponding to a sign of an error signal corresponding to a difference between the received symbol and a corresponding decoded data signal, and wherein the block is further configured to generate the coefficient adjustment signal based, at least in part, on a multiplication of the first and second signals in response to receipt of a symbol having the second specific voltage level and to generate the coefficient update signal equal to a constant tap coefficient in response to receipt of a symbol having the first specific voltage level.

6. The finite impulse response filter of claim 5 , wherein the block comprises a lookup table configured to store an association between the voltage level of the received symbol, the sign of the received symbol, and the coefficient adjustment signal.

7. The finite impulse response filter of claim 5 , wherein the block comprises a logic circuit.

8. A method of adjusting a tap coefficient for a finite impulse response filter, the method comprising:

in response to receipt of a pulse amplitude-modulated signal including a first data symbol having a first specific voltage level, adjusting the filter coefficient based on a sign of the received first data symbol and a sign of an error signal corresponding to a difference between the received first data symbol and a decoded data signal; and

maintaining a constant tap coefficient in response to receipt of a second data symbol having a second specific voltage level.

9. The method of claim 8 , wherein the second specific voltage level is less than the first specific voltage level.

10. The method of claim 8 , wherein the first data symbol has a higher signal-to-noise ratio than the second data symbol.

11. The method of claim 8 , further comprising accessing a lookup table storing an association between the voltage level of the received data symbol, the sign of the received data symbol, and the coefficient adjustment signal.

12. The method of claim 8 , wherein said adjusting the filter coefficient comprises generating a filter coefficient adjustment signal based, at least in part, on multiplying the sign of the received data symbol and the sign of the error signal.

13. The method of claim 8 , wherein said maintaining a constant tap coefficient comprises generating a filter coefficient adjustment signal corresponding to the constant tap coefficient.

14. A decision feedback equalizer comprising:

a decision circuit configured to receive a data signal and output decoded data;

an error calculator configured to generate an error signal corresponding to a difference between the data signal and the decoded data;

a plurality of taps coupled in series, wherein the plurality of taps are configured to receive a data signal, delay the data signal, and couple the delayed data signal to a next tap, each tap comprising a finite impulse response filter having a corresponding filter coefficient, wherein each finite impulse response filter is configured to generate a correction signal based on the data signal and the filter coefficient, and wherein each finite impulse response filter includes a filter coefficient updater configured to update the filter coefficient based, at least in part, on a specific voltage level of a received symbol; and

a summer configured to receive the respective correction signals from the plurality of taps, wherein the summer is coupled to the decision circuit and configured to receive an input signal and to generate the data signal based, at least in part, on the correction signals.

15. A finite impulse response filter having an associated filter coefficient, the finite impulse response filter comprising:

means for delaying an input signal;

means for updating the filter coefficient based on the filter coefficient and a symbol of the delayed input signal, wherein the filter coefficient is updated to a different filter coefficient in response to receipt of a first symbol having a first specific voltage level, and wherein the filter coefficient is maintained constant in response to receipt of a second symbol having a second specific voltage level; and

means for generating a correction signal based on the delayed input signal and the updated filter coefficient.