Per-survivor based adaptive equalizer
A method used in an adaptive survivor based channel equalizer, the method comprises selecting at a decision time a survivor in a Viterbi trellis and a corresponding equalizer, adaptively updating at the decision time the corresponding equalizer to define a new corresponding equalizer for use at a next decision time, retrieving the new corresponding equalizer as defined at an earlier decision time, and using the new corresponding equalizer as defined at an earlier decision time as an equalizer for other survivors in the Viterbi trellis at the next decision time. A corresponding adaptive survivor based channel equalizer includes a fixed pre-filter configured to provide a pre-filtered signal to a reduced state sequence estimator (RSSE) which is configured for providing recovered symbols. A coefficient adaptor is coupled to the RSSE and configured to essentially perform the method.
1. A method used in an adaptive survivor based channel equalizer, the method comprising:
selecting at a decision time a survivor in a Viterbi trellis, the survivor having a corresponding equalizer defined at an immediately earlier decision time;
adaptively updating at the decision time the corresponding equalizer to define the corresponding equalizer for use at a next decision time;
retrieving, from memory, the corresponding equalizer as defined at an earlier decision time; and
using the corresponding equalizer including associated coefficients as defined at said earlier decision time as an equalizer including the associated coefficients for other survivors in the Viterbi trellis at the next decision time.
2. The method of claim 1 further comprising sorting error metrics, each error metric associated with one survivor in the Viterbi trellis, to provide a sorted error metrics list and wherein the selecting the survivor comprises selecting the survivor in accordance with the sorted error metrics list.
3. The method of claim 2 wherein the selecting the survivor further comprises selecting a strongest survivor with a best error metric based on the sorted error metrics list.
4. The method of claim 2 further comprising choosing one or more additional survivors in accordance with the sorted error metrics list and adaptively updating at the decision time one or more additional equalizers, the one or more additional equalizers corresponding one to one to the one or more additional survivors.
5. The method of claim 4 wherein the choosing the one or more additional survivors in accordance with the sorted error metrics list further comprises selecting one or more additional survivors that have associated error metrics, which are next best error metrics after a strongest survivor with a best error metric has been selected.
6. The method of claim 1 wherein the retrieving the corresponding equalizer as defined at the earlier decision time further comprises retrieving the corresponding equalizer as defined at the immediately earlier decision time.
7. The method of claim 1 wherein the Viterbi trellis is a 64 state trellis and the other survivors comprise 63 survivors.
8. The method of claim 1 wherein the Viterbi trellis is a 32 state trellis and the other survivors comprise 31 survivors.
9. The method of claim 1 wherein the Viterbi trellis is a 16 state trellis and the other survivors comprise 15 survivors.
10. The method of claim 1 wherein the adaptively updating at the decision time the corresponding equalizer further comprises updating feed forward filter coefficients and feedback filter coefficients based on the symbols corresponding to the survivor.
11. An adaptive survivor based channel equalizer comprising:
a fixed pre-filter configured to be coupled to a receive signal and provide a pre-filter signal;
a reduced state sequence estimator (RSSE) coupled to the pre-filter signal and configured for providing recovered symbols, the RSSE including an adaptive feedback filter configured to extend each survivor in a Viterbi trellis at a first time to a second time; and
a coefficient adaptor coupled to the RSSE and configured:
to adaptively update at the first time first coefficients for the adaptive feedback filter, the adaptive feedback filter defined by the first coefficients included with and used for first one or more equalizers associated with a first portion of survivors, and
to assign at the first time second coefficients for the adaptive feedback filter, the adaptive feedback filter defined by the second coefficients included with and used for second equalizers associated with a second portion of the survivors, the second coefficients being first coefficients for the adaptive feedback filter used for one or more of the first one or more equalizers at an earlier time.
12. The adaptive survivor based channel equalizer of claim 11 wherein the coefficient adaptor is configured to sort error metrics to provide a sorted error metrics list, each error metric associated with one survivor in the Viterbi trellis, the first one or more equalizers determined in accordance with the sorted error metrics list.
13. The adaptive survivor based channel equalizer of claim 12 wherein the coefficient adaptor is configured to select a strongest survivor based on a best error metric from the sorted error metrics list and adaptively update first coefficients for the adaptive feedback filter, the adaptive filter defined by the first coefficients included with and used for an equalizer associated with the strongest survivor.
14. The adaptive survivor based channel equalizer of claim 13 wherein the coefficient adaptor is configured to select additional survivors based on the sorted error metrics list and adaptively update first coefficients for the adaptive feedback filter, the adaptive filter defined by the first coefficients included with and used for additional equalizers associated one for one with the additional survivors.
15. The adaptive survivor based channel equalizer of claim 14 wherein the coefficient adaptor is configured to choose the additional survivors based on the sorted error metrics list where the additional survivors have associated error metrics, which are next best error metrics after a strongest survivor with a best error metric has been chosen.
16. The adaptive survivor based channel equalizer of claim 11 wherein the coefficient adaptor is configured to assign the second coefficients for the adaptive feedback filter used for the second equalizers wherein the second coefficients are first coefficients at one decision time earlier.
17. The adaptive survivor based channel equalizer of claim 11 further comprising an adaptive feed forward filter configured to compensate the pre-filter signal to provide a compensated pre-filter signal and couple the compensated pre-filter signal to the RSSE and wherein the coefficient adaptor provides filter coefficients to the adaptive feed forward filter.
18. The adaptive survivor based channel equalizer of claim 17 wherein the coefficient adaptor is further configured:
to adaptively update at the first time first filter coefficients for the adaptive feed forward filter used for the first one or more equalizers associated with the first portion of survivors; and
to assign at the first time second filter coefficients for the adaptive feed forward filter used for the second equalizers associated with a second portion of the survivors, the second filter coefficients being first filter coefficients for the adaptive feed forward filter used for one or more of the first one or more equalizers at the earlier time.
19. The adaptive survivor based channel equalizer of claim 18 wherein the coefficient adaptor is configured to sort error metrics to provide a sorted error metrics list, each error metric associated with one survivor in the Viterbi trellis, the first one or more equalizers determined in accordance with the sorted error metrics list.
20. The adaptive survivor based channel equalizer of claim 19 wherein the coefficient adaptor is configured to select a strongest survivor based on a best error metric from the sorted error metrics list and adaptively update first coefficients for the adaptive feedback filter and adaptively update first filter coefficients for the adaptive feed forward filter, both filters used for an equalizer associated with the strongest survivor.
21. The adaptive survivor based channel equalizer of claim 19 wherein the coefficient adaptor is configured to select additional survivors based on the sorted error metrics list and adaptively update first coefficients for the adaptive feedback filter and adaptively update first filter coefficients for the adaptive feed forward filter, both filters used for additional equalizers associated one for one with the additional survivors.
22. The adaptive survivor based channel equalizer of claim 11 wherein the coefficient adaptor configured to adaptively update at the first time the first coefficients is further configured to adaptively update the first coefficients by calculating new first coefficients based on previous coefficients and an error.