Method and apparatus for cancelling echo signals having different dynamic ranges

1. A physical layer transceiver for a wireline communications system, the physical layer transceiver transmitting data onto a wireline channel medium to a remote receiver and receiving data via the wireline channel medium from a remote transmitter, the physical layer transceiver comprising:

a receive path configured for coupling to the wireline channel medium and for communicating signals received from the wireline channel medium to a functional circuit;

a transmit path configured for coupling to the wireline channel medium and for communicating signals received from the functional circuit onto the wireline channel medium; and

echo cancellation circuitry coupled to the receive path and the transmit path and configured to cancel from the receive path echoes of signals transmitted on the transmit path, the echo cancellation circuitry including:

a first plurality of high-dynamic-range filter segments,

a second plurality of low-dynamic-range filter segments, and

control circuitry configured to:

detect on the receive path the echoes of the signals transmitted on the transmit path,

classify each respective echo signal as having a low dynamic range or a high dynamic range, and

deploy a respective filter segment in the first plurality of high-dynamic-range filter segments to a respective location of a respective echo signal classified as having a high dynamic range, and deploy a respective filter segment in the second plurality of low-dynamic-range filter segments to a respective location of a respective echo signal classified as having a low dynamic range.

2. The physical layer transceiver of claim 1 wherein the control circuitry is configured to classify each respective echo signal as having a low dynamic range or a high dynamic range by comparing a time-domain characteristic of the respective echo signal to a threshold.

3. The physical layer transceiver of claim 2 wherein the control circuitry configures the threshold.

4. The physical layer transceiver of claim 2 wherein:

the control circuitry is configured to detect on the receive path the echoes of the signals transmitted on the transmit path by performing cross-correlation of the receive path and the transmit path.

5. The physical layer transceiver of claim 4 wherein:

the control circuitry is configured to compare the time-domain characteristic of the respective echo signal to the threshold by comparing, to the threshold, a metric derived from the cross-correlation.

6. The physical layer transceiver of claim 5 wherein:

the control circuitry is configured to deploy the respective filter segment in the first plurality of high-dynamic-range filter segments to the respective location of the respective echo signal classified as having a high dynamic range by:

for each respective echo signal classified as having the high dynamic range, sorting the respective signal from a highest value of the metric in order of decreasing value of the metric, and

deploying filter segments in the first plurality of high-dynamic-range filter segments to locations starting with a location corresponding to the highest value of the metric and to additional locations corresponding to the values of the metric in the order of decreasing value of the metric.

7. The physical layer transceiver of claim 5 wherein:

the control circuitry is configured to deploy the respective filter segment in the second plurality of low-dynamic-range filter segments to the respective location of the respective echo signal classified as having a low dynamic range by:

for each respective echo signal classified as having the low dynamic range, sorting the respective signal from a highest value of the metric in order of decreasing value of the metric, and

deploying filter segments in the second plurality of low-dynamic-range filter segments to locations starting with a location corresponding to the highest value of the metric and to additional locations corresponding to the values of the metric in the order of decreasing value of the metric.

8. The physical layer transceiver of claim 1 wherein:

each filter segment in the first plurality of high-dynamic-range filter segments and the second plurality of low-dynamic-range filter segments is a rover filter segment configurable to correspond to a location along the wireline channel medium by adjustment of a respective delay line; and

the control circuitry is configured to deploy the respective filter segment in the first plurality of high-dynamic-range filter segments to the respective location of the respective echo signal classified as having the high dynamic range, and to deploy the respective filter segment in the second plurality of low-dynamic-range filter segments to the respective location of the respective echo signal classified as having the low dynamic range, by adjusting the respective delay line.

9. The physical layer transceiver of claim 1 wherein:

each filter segment in the first plurality of high-dynamic-range filter segments has a first resolution; and

each filter segment in the second plurality of low-dynamic-range filter segments has a second resolution smaller than the first resolution.

10. The physical layer transceiver of claim 9 wherein:

each filter segment in the first plurality of high-dynamic-range filter segments has tap coefficients having a first number of bits; and

each filter segment in the second plurality of low-dynamic-range filter segments has tap coefficients having a second number of bits smaller than the first number of bits.

11. A method for cancelling, from signals received from a wireline channel medium on a receive path of a physical layer transceiver, echoes of signals transmitted from functional circuity onto the wireline channel medium via a transmit path of the physical layer transceiver, the method comprising:

detecting on the receive path the echoes of the signals transmitted on the transmit path;

classifying each respective echo signal as having a low dynamic range or a high dynamic range; and

deploying a respective filter segment in a first plurality of high-dynamic-range filter segments to a respective location of a respective echo signal classified as having a high dynamic range, and deploying a respective filter segment in a second plurality of low-dynamic-range filter segments to a respective location of a respective echo signal classified as having a low dynamic range.

12. The method for cancelling echoes according to claim 11 wherein classifying each respective echo signal as having the low dynamic range or the high dynamic range comprises comparing a time-domain characteristic of the respective echo signal to a threshold.

13. The method for cancelling echoes according to claim 12 further comprising configuring the threshold using control circuitry of the physical layer transceiver.

14. The method for cancelling echoes according to claim 12 wherein:

detecting on the receive path the echoes of the signals transmitted on the transmit path comprises performing cross-correlation of the receive path and the transmit path.

15. The method for cancelling echoes according to claim 14 wherein:

comparing the time-domain characteristic of the respective echo signal to the threshold comprises comparing, to the threshold, a metric derived from the cross-correlation.

16. The method for cancelling echoes according to claim 15 wherein:

comparing, to the threshold, the metric derived from the cross-correlation comprises comparing, to the threshold, a sum of squares of the cross-correlation at different time lags.

17. The method for cancelling echoes according to claim 15 comprising:

deploying the respective filter segment in the first plurality of high-dynamic-range filter segments to the respective location of the respective echo signal classified as having a high dynamic range by:

for each respective echo signal classified as having the high dynamic range, sorting the respective signal from a highest value of the metric in order of decreasing value of the metric, and

deploying filter segments in the first plurality of high-dynamic-range filter segments to locations starting with a location corresponding to the highest value of the metric and to additional locations corresponding to the values of the metric in the order of decreasing value of the metric.

18. The method for cancelling echoes according to claim 17 comprising, when undeployed filter segments in the first plurality of high-dynamic-range filter segments remain after deploying filter segments in the first plurality of high-dynamic-range filter segments to all locations of the respective echo signal classified as having a high dynamic range:

deploying the respective filter segment in the second plurality of low-dynamic-range filter segments to the respective location of the respective echo signal classified as having a low dynamic range by:

for each respective echo signal classified as having the low dynamic range, sorting the respective signal classified as having the low dynamic range from a highest value of the metric in order of decreasing value of the metric;

deploying the remaining undeployed filter segments in the first plurality of high-dynamic-range filter segments to locations starting with a location corresponding to the respective signal classified as having the low dynamic range with the highest value of the metric and to additional locations of the respective signal classified as having the low dynamic range corresponding to the values of the metric in the order of decreasing value of the metric; and

deploying filter segments in the second plurality of low-dynamic-range filter segments to additional locations, not filled by the filter segment s in the first plurality of high-dynamic-range filter segments, corresponding to the values of the metric in the order of decreasing value of the metric.

19. The method for cancelling echoes according to claim 15 comprising:

deploying the respective filter segment in the second plurality of low-dynamic-range filter segments to the respective location of the respective echo signal classified as having a low dynamic range by:

for each respective echo signal classified as having the low dynamic range, sorting the respective signal from a highest value of the metric in order of decreasing value of the metric, and

deploying filter segments in the second plurality of low-dynamic-range filter segments to locations starting with a location corresponding to the highest value of the metric and to additional locations corresponding to the values of the metric in the order of decreasing value of the metric.

20. The method for cancelling echoes according to claim 11 comprising, when each filter segment in the first plurality of high-dynamic-range filter segments and the second plurality of low-dynamic-range filter segments is a rover filter segment configurable to correspond to a location along the wireline channel medium by adjustment of a respective delay line:

deploying the respective filter segment in the first plurality of high-dynamic-range filter segments to the respective location of the respective echo signal classified as having the high dynamic range, and deploying the respective filter segment in the second plurality of low-dynamic-range filter segments to the respective location of the respective echo signal classified as having the low dynamic range, by:

adjusting the respective delay line.