Method and apparatus for fast retraining of ethernet transceivers based on trickling error
A method of operation for an Ethernet transceiver is disclosed. The method includes operating the Ethernet transceiver in a data mode, and triggering a fast retrain sequence of steps based on trickling error information. The triggering includes detecting error information, averaging the detected error information over a time interval to generate the trickling error information, comparing the averaged detected error information to a selected threshold value, and initiating the fast retrain sequence of steps based on the comparing.
1. A method of operation for an Ethernet transceiver, the method comprising:
operating the Ethernet transceiver in a data mode;
selecting a time interval for monitoring error information; and
triggering a fast retrain sequence by
detecting error information during the time interval,
averaging at least two non-consecutive frame errors in the detected error information over the time interval to generate trickling error information,
comparing the trickling error information to a selected threshold value, and
performing the fast retrain sequence-based on the comparing.
2. The method according to claim 1 , wherein the error information comprises an error predictor.
3. The method according to claim 1 , wherein the error information comprises Low-Density Parity Check (LDPC) error code information.
4. The method according to claim 1 , wherein the error information comprises LDPC loop iteration information.
5. The method according to claim 1 , further comprising:
operating the Ethernet transceiver in accordance with an NBASE-T Ethernet protocol.
6. The method according to claim 1 , further comprising:
prior to operating the Ethernet transceiver in the data mode, performing a full training sequence.
7. An integrated circuit (IC) Ethernet transceiver chip comprising:
training logic to trigger a fast retrain sequence during a data mode, the training logic including
error detection circuitry to detect error information,
circuitry to average at least two non-consecutive frame errors in the detected error information over a user-selected time interval to generate trickling error information,
comparison circuitry to compare the trickling error information to a selected threshold value, and
circuitry to perform the fast retrain sequence based on the comparing.
8. The IC Ethernet transceiver chip according to claim 7 , wherein the error information comprises an error predictor.
9. The IC Ethernet transceiver chip according to claim 7 , wherein:
the error detection circuitry comprises low density parity check (LDPC) decoding circuitry.
10. The IC Ethernet transceiver chip according to claim 7 , wherein the error information comprises Low-Density Parity Check (LDPC) error code information.
11. The IC Ethernet transceiver chip according to claim 7 , wherein the error information comprises LDPC loop iteration information.
12. The IC Ethernet transceiver chip according to claim 7 , realized as an NBASE-T IC Ethernet transceiver chip.
13. A method of operation for an Ethernet transceiver, the method comprising:
operating the Ethernet transceiver in a data mode;
selecting a time interval for monitoring error information;
establishing a consecutive frame error threshold and a trickling error threshold;
detecting error information;
averaging at least two non-consecutive frame errors in the error information over the time interval to generate trickling error information, and
triggering a fast retrain sequence based on a comparison of the trickling error information during the time interval to the consecutive frame error threshold and the trickling error threshold.
14. The method according to claim 13 , wherein the detecting error information comprises:
detecting an error predictor.
15. The method according to claim 14 , wherein the error predictor is based on a detected signal-to-noise ratio (SNR) along a channel coupled to the Ethernet transceiver.