Method and apparatus for channel equalization

• US 7,400,677 B2
• Filed: 09/06/2005
• Issued: 07/15/2008
• Est. Priority Date: 07/01/2002
• Status: Expired due to Term

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: SolarFlare Communications Incorporated (Advanced Micro Devices, Inc.)
• Inventors: Jones, William
• Primary Examiner(s): VO, DON NGUYEN
• Application Number:
  US11/220,508
• Publication Number:
  US 20060007997A1
• Time in Patent Office: 1,043 Days
• Field of Search: 375/232, 375/231, 375/233, 375/350
• US Class Current: 375/232
• CPC Class Codes: H04L 25/03146 with a recursive structure H04L25/03127 takes precedence
• CPC Class Codes: H04L 25/03343 Arrangements at the transmitter end

Abstract

A method and apparatus is disclosed to overcome the effects of intersymbol interference during data transmission. Overcoming the effects of intersymbol interference makes possible higher data transmission rates for a given error rate. In one embodiment a receiver-transmitter pair is configured with a precode filter at the transmit side and a feed forward filter and a feedback filter at the receive side. Filter coefficients are calculated to reduce the undesirable effects of the channel, such as intersymbol interference. In one embodiment a training process occurs with the feedforward filter and a feedback filter, such that the first N coefficients of the feedback filter are set to zero. Thereafter, the coefficients of the feedforward filter are subject to spectral factorization and separated into minimum phase roots and maximum phase roots. The minimum phase roots comprise the precode filter coefficients and the maximum phase roots are established as feedforward filter coefficients.

Claims

1. A method for filter coefficient selection in a receiver-transmitter pair comprising:

training a first filter and a second filter to determine first filter coefficients and second filter coefficients such that the first filter and the second filter have a transfer function inverse to that of a channel, wherein during the training at least one second filter coefficient is forced to a predetermined value;

processing the first filter coefficients to isolate minimum phase coefficients and maximum phase coefficients;

transmitting the minimum phase coefficients to a third filter;

establishing coefficients of the third filter based on the minimum phase coefficients; and

setting the first filter coefficients based on the maximum phase coefficients.

2. The method of claim 1, wherein the first filter and the second filter are located in a receiver and the third filter is located in a transmitter.

3. The method of claim 1, wherein the first filter comprises a feedforward filter, the second filter comprises a feedback filter, and the third filter comprises a precode filter.

4. The method of claim 1, wherein transmitting comprises transmitting the minimum phase coefficients over one or more twisted pair conductors to a precode filter in a transmitter.

5. The method of claim 1, wherein processing comprises performing spectral factorization.

6. The method of claim 5, wherein processing further comprises forming a polynomial to obtain the minimum phase coefficients and the maximum phase coefficients.

7. The method of claim 1, wherein the at least one second filter coefficient that is forced to a predetermined value comprises four second filter coefficients that are forced to zero.

8. The method of claim 7, wherein the third filter comprises a five tap precode filter.

9. A method for equalizing a signal comprising:

receiving a signal at a precode equalizer, the signal to be transmitted through a channel;

processing the signal with the precode equalizer to pre-equalize the signal, wherein the precode equalizer utilizes minimum phase coefficients generated from the spectral factorization of the coefficients from a first equalizer, which is located at a receiver, during a training process;

receiving the signal, at the receiver, after transmission through the channel;

performing a first equalization process on the signal utilizing the first equalizer having maximum phase coefficients values, which are isolated from the spectral factorization of the coefficients from the first equalizer during the training process; and

performing a second equalization process on the signal utilizing a second equalizer, which is located at the receiver and configured with second equalizer coefficient values, wherein at least one of the second equalizer coefficient values are set to a predetermined value during the training process;

wherein the first equalization process and the second equalization process reduce the effects of transmission through the channel.

10. The method of claim 9, wherein the first equalizer comprises a feedforward equalizer.

11. The method of claim 9, wherein at least one of the second equalizer coefficient values are set to a fixed value during the second equalization process.

12. The method of claim 11, wherein the at least one of the second equalizer coefficient values comprises the first four second equalizer coefficient values and wherein a fixed value comprises a zero value.

13. The method of claim 9, wherein setting at least one of the second equalizer coefficient values to a predetermined value reduces error propagation associated with the second equalizer.

14. The method of claim 9, wherein the first equalizer comprises a feedforward equalizer, the second equalizer comprises a feedback equalizer, and at least one of the first five second equalizer coefficient values are set to zero.

15. The method of claim 9, wherein performing a second equalization process further includes providing a feedback signal to a decision device.

16. A system for channel equalization comprising:

a first filter configured to utilize one or more first filter coefficients, the first filter comprising maximum phase filter;

a second filter having two or more second filter coefficients with at least one of the two or more second filter coefficients forced to a predetermined value; and

a precode filter having one or more precode filter coefficients, the precode filter comprising a minimum phase filter;

wherein the first filter and the second filter are located in a receiver and the precode filter is located in a transmitter and wherein the first filter, second filter, and precode filter are trained with coefficient values to operate as a system to equalize a signal and during training the second filter has one or more coefficient forced to a predetermined value.

17. The system of claim 16, wherein the predetermined value comprises zero.

18. The system of claim 16, further including a processor configured to calculate the first filter coefficients, second filter coefficients and the precode filter coefficients.

19. The system of claim 16, wherein the first filter comprises a feed forward filter and the second filter comprises a feedback filter.

20. The system of claim 16, wherein the first filter, second filter and precode

filter coefficient values are determined during a training process and one aspect of the training process comprises spectral factorization of the first filter coefficients.

21. A system for exchanging data between a first location and a second location, the system comprising:

a first transceiver comprising;

a data source interface configured to receive data from a data source;

a precode filter configured to equalize the data to at least partially account for transmission through a channel;

a digital to analog converter configured to convert the data to a first analog signal;

a line driver configured to adjust the power of the first analog signal for transmission over a channel;

a receiver configured to receive data from a second transceiver;

a second transceiver comprising;

an amplifier configured to receive a second analog signal from the channel, the second analog signal comprising a distorted version of the first analog signal;

an analog to digital converter configured to convert the second analog signal to data;

a first filter configured to process the data to at least partially reverse the effects of transmission through the channel;

a second filter configured to process the data to at least partially reverse the effects of transmission through the channel;

a decision device configured to quantize portions of the data into two or more distinct values;

a transmitter configured to transmit precode filter coefficient data to the precode filter in the first transceiver;

wherein the precode filter coefficient data is generated by processing the first filter coefficients, which are generated during training while at least one coefficient of the second filter is set to zero, to isolate minimum phase coefficients of the first filter coefficients.

22. The system of claim 21, wherein the precode filter is configured as a minimum phase filter and the first filter is configured as a maximum phase filter.

23. The system of claim 21, wherein second filter comprises a feedback filter having greater than N coefficients and the first N coefficients are set to zero, wherein N comprises an integer value having magnitude of less than 20.

24. The system of claim 21, wherein the second transceiver further comprises a processor configured to calculate coefficients for the precode filter, the first filter, and the second filter.

25. A computer program product comprising a computer useable medium having computer program logic recorded thereon for calculating equalizer coefficients, comprising:

computer program code logic configured to initiate a training sequence;

computer program code logic configured to train coefficients of a feedforward filter as a mixed phase filter;

computer program code logic configured to train coefficients of a feedback filter, wherein at least one of the coefficients of the feedback filter is set to zero;

computer program code logic configured to process the coefficients of the feedforward filter to calculate minimum phase coefficients and maximum phase coefficients;

computer program code logic configured to establish the feedforward coefficients based on the maximum phase coefficients; and

computer program code logic configured to establish precoder coefficients based on the minimum phase coefficients.

26. The computer program product of claim 25, wherein computer program code logic configured to process the coefficients comprises computer program code logic configured to perform spectral factorization.

27. The computer program product of claim 25, wherein establishing precoder coefficients based on the minimum phase coefficients comprises forming a polynomial from the maximum phase feedforward coefficients.

28. The computer program product of claim 25, wherein the computer program product is located in a communication receiver.