Claims
- 1. A sequential decoder for decoding convolutional code, comprising:
a computing device comprising a Fano technique, the Fano technique including a plurality of variables being normalized to change a point of reference of the technique, one of the variables being a current node metric, the variables being normalized such that the current node metric is set to approximately zero.
- 2. The sequential decoder of claim 1 wherein the current node metric before being normalized is subtracted from each of the variables.
- 3. The sequential decoder of claim 1 wherein the Fano technique is embodied in a register transfer level (RTL) architecture and a finite state machine.
- 4. The sequential decoder of claim 3 wherein the RTL architectur e includes a branch metric unit to compute the current branch metric.
- 5. The sequential decoder of claim 4 wherein the RTL architecture further includes a sequence memory to store sequence data; and
the branch metric unit computes the current branch metric based upon the sequence data.
- 6. The sequential decoder of claim 5 wherein the finite state machine includes a look/move forward and tighten if needed state, a tighten or look/move forward state, and a look/move back state.
- 7. The sequential decoder of claim 5 wherein the Fano technique includes speculative computation of at least one variable.
- 8. The sequential decoder of claim 7 wherein the at least one variable comprises a threshold minus a selected branch metric and the threshold plus a threshold adjustment level.
- 9. A sequential decoder for decoding convolutional code, comprising:
a Fano technique embodied in a register transfer level architecture and a finite state machine, the Fano technique including speculative data execution of at least two variables of a plurality of variables.
- 10. The sequential decoder of claim 9 wherein the plurality of variables are normalized to change a point of reference of the technique, the variables including a current node metric, the variables being normalized such that the current node metric is set to zero.
- 11. The sequential decoder of claim 9 wherein the at least two variables include at least two branch metrics.
- 12. The sequential decoder of claim 9 wherein the at least two variables include a threshold minus a selected branch metric.
- 13. The sequential decoder of claim 9 wherein the at least two variables include a threshold plus a threshold adjustment level.
- 14. The sequential decoder of claim 9 wherein the at least two variables include a threshold plus a threshold adjustment level.
- 15. The sequential decoder of claim 9 wherein the register transfer level architecture and the finite state machine are implemented in a computing device.
- 16. The sequential decoder of claim 15 wherein the computing device is selected from the group consisting of processors and gate arrays.
- 17. A sequential decoder for decoding convolutional code, comprising:
a decoder unit having an operating parameter; and a controller, responsive to a trigger event, to dynamically control the operating parameter such that a performance characteristic is attained.
- 18. The sequential decoder of claim 17 wherein the trigger event is selected from the group consisting of a signal-to-noise level, a packet error rate, and a buffer utilization level.
- 19. The sequential decoder of claim 17 wherein the operating parameter is selected from the group consisting of a supply voltage level, a clock frequency, a threshold adjustment level, and a traceback limit.
- 20. The sequential decoder of claim 17 wherein the performance characteristic is selected from the group consisting of average power consumption of the decoder and execution speed of the decoder.
- 21. The sequential decoder of claim 19 wherein the performance characteristic is selected from the group consisting of average power consumption of the decoder and execution speed of the decoder.
- 22. A method of decoding convolutional code, comprising:
using a Fano technique to decode the convolutional code, the Fano technique comprising a plurality of variables including a current node metric; and normalizing the variables such that the current node metric is set to zero.
- 23. The method of claim 22 wherein using the Fano technique further includes:
executing a finite state machine comprising a look/move forward and tighten if needed state, a tighten or look/move forward state, and a look/move back state.
- 24. The method of claim 23 further including speculatively executing values for at least two of the variables.
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Application Serial No. 60/242,190, filed on Oct. 20, 2000, which is hereby incorporated by reference in its entirety.
Provisional Applications (1)
|
Number |
Date |
Country |
|
60242190 |
Oct 2000 |
US |