Equalizer for heavily clipped or compressed communications signals
View Patent ↗Apparatus and methods mitigate a problem of equalizing communications signals that have been distorted by severe non-linearities such as clipping or harsh compression. For example, severe non-linearity occurs when signal compression or signal clipping occurs at rates above 20% of the data transmission interval. Severe non-linearities may significantly reduce system performance. Disclosed techniques selectively apply DSP equalization based on the detection of non-linearity for a present sample or one or more samples prior to the present sample. These techniques can be implemented in relatively low-cost high-speed SerDes designs to improve eye openings and reduce sensitivity to InterSymbol Interference (ISI) and to improve bit error rate (BER).
1. A method of equalization in a receiver, the method comprising:
receiving a communications signal having symbols embedded therein;
converting the received communications signal from analog form to digital form to generate a digital received signal comprising soft symbols, wherein the digital received signal is quantized to at least 3 bits; and
selectively equalizing a current soft symbol sample based at least partly on a determination of whether or not at least one of the current soft symbol sample or an immediately prior soft symbol sample to the current soft symbol sample has been non-linearly distorted beyond a threshold, wherein the threshold indicates clipping and/or harsh compression, wherein selectively equalizing further comprises:
applying no equalization to the current soft symbol sample when the current soft symbol sample is above the threshold; and
applying linear equalization to the current soft symbol sample when the current soft symbol sample is below the threshold.
2. The method of claim 1 , wherein selectively equalizing comprises:
selecting a weight w from a plurality of weights, wherein the weight w varies depending upon whether or not non-linear distortion is indicated;
multiplying the selected weight w to an immediately prior soft symbol sample to generate a weighted prior soft symbol sample; and
summing the current soft symbol sample with the weighted prior soft symbol sample to generate an equalized current soft symbol sample.
3. The method of claim 2 , further comprising slicing the equalized current soft symbol sample to generate a hard symbol sample.
4. The method of claim 2 , wherein at least a portion of values for the plurality of weights is determined based on eye diagnostics.
5. The method of claim 1 , wherein the current soft symbol sample is equalized using a plurality of prior soft symbol samples.
6. The method of claim 1 , wherein equalizer coefficients are selected based on a non-linear distortion status of the current soft symbol sample and a plurality of prior soft symbol samples.
7. The method of claim 1 , wherein selectively equalizing is performed by firmware or software instructions executed by a processor, wherein the firmware or software instructions are stored on a non-transitory computer-readable recording medium.
8. A method of equalization in a receiver, the method comprising:
receiving a communications signal having symbols embedded therein;
converting the received communications signal from analog form to digital form to generate a digital received signal comprising soft symbols, wherein the digital received signal is quantized to at least 3 bits;
selectively equalizing a current soft symbol sample based at least partly on a determination of whether or not at least one of the current soft symbol sample or an immediately prior soft symbol sample to the current soft symbol sample has been non-linearly distorted beyond a threshold, wherein the threshold indicates clipping and/or harsh compression, wherein selective equalizing further comprises:
applying no equalization to the current soft symbol sample when the current soft symbol sample is above the threshold;
applying linear equalization to the current soft symbol sample when both the current soft symbol sample and the symbol sample immediately prior to the current soft symbol sample are below the threshold; and
applying modified equalization to the current soft symbol sample when the current soft symbol sample is below the threshold and the symbol sample immediately prior to the current soft symbol sample is above the threshold, wherein the modified equalization comprises modifying an effective value of an equalizer coefficient by a modified weight w mod , wherein w mod is greater than or equal to 1 and less than or equal to w_max, wherein
w_max
=
1
-
2
-
num
_
A
D
C
_
bits
clip_level
wherein num_ADC_bits represents a number of bits quantized by an analog-to-digital converter for the soft symbol samples and wherein clip_level represents a magnitude level wherein a soft symbol is clipped.
9. An apparatus for equalization in a receiver, the apparatus comprising:
a detection circuit configured to receive a digital received signal comprising soft symbols, wherein the digital received signal is quantized to at least 3 bits, wherein the detection circuit is configured to determine whether a sample has been non-linearly distorted beyond a threshold, wherein the threshold indicates clipping and/or harsh compression; and
an equalizer configured to receive the digital received signal, the equalizer configured to selectively equalize a current soft symbol sample based at least partly on a determination of whether or not at least one of the current soft symbol sample or an immediately prior soft symbol sample to the current soft symbol sample has been non-linearly distorted beyond the threshold, wherein the equalizer is further configured to:
bypass equalization to the current soft symbol sample when the current soft symbol sample is above the threshold; and
apply linear equalization to the current soft symbol sample when the current soft symbol sample is below the threshold.
10. The apparatus of claim 9 , wherein the equalizer comprises:
a multiplexer circuit coupled to a plurality of weights w, wherein the plurality of weights w vary depending upon whether or not non-linear distortion is indicated;
a pattern analyzer configured to recognize whether or not non-linear distortion is indicated in the current soft symbol sample and one or more prior soft symbol samples prior to the current soft symbol sample, wherein the pattern analyzer is coupled to the multiplexer circuit to select a weight w based on a recognized pattern;
a multiplier circuit configured to apply the selected weight w to a prior soft symbol sample of the one or more prior soft symbol samples to generate a weighted prior soft symbol sample; and
a summing circuit configured to sum the current soft symbol sample with at least the weighted prior soft symbol sample to generate an equalized current soft symbol sample.
11. The apparatus of claim 10 , further comprising a slicer configured to slice the equalized current soft symbol sample to generate a hard symbol sample.
12. The apparatus of claim 10 , wherein at least a portion of values for the plurality of weights is determined based on eye diagnostics.
13. The apparatus of claim 9 , wherein the equalizer is configured to equalize the current soft symbol sample with a plurality of prior soft symbol samples.
14. The apparatus of claim 9 , wherein the equalizer is configured to apply equalizer coefficients based on a non-linear distortion status of the current soft symbol sample and a plurality of prior soft symbol samples.
15. An apparatus for equalization in a receiver, the apparatus comprising:
a detection circuit configured to receive a digital received signal comprising soft symbols, wherein the digital received signal is quantized to at least 3 bits, wherein the detection circuit is configured to determine whether a sample has been non-linearly distorted beyond a threshold, wherein the threshold indicates clipping and/or harsh compression; and
an equalizer configured to receive the digital received signal, the equalizer configured to selectively equalize a current soft symbol sample based at least partly on a determination of whether or not at least one of the current soft symbol sample or an immediately prior soft symbol sample to the current soft symbol sample has been non-linearly distorted beyond the threshold, wherein the equalizer is further configured to:
bypass equalization to the current soft symbol sample when the current soft symbol sample is above the threshold;
apply linear equalization to the current soft symbol sample when both the current soft symbol sample and the soft symbol sample immediately prior to the current soft symbol sample are below the threshold; and
apply modified equalization to the current soft symbol sample when the current soft symbol sample is below the threshold and the soft symbol sample immediately prior to the current soft symbol sample is above the threshold, wherein the modified equalization comprises modifying an effective value of an equalizer coefficient by a modified weight w mod , wherein w mod is greater than or equal to 1 and less than or equal to w_max, wherein
w_max
=
1
-
2
-
num
_
A
D
C_
bits
clip_level
wherein num_ADC_bits represents a number of bits quantized by an analog-to-digital converter for the soft symbol samples and wherein clip_level represents a magnitude level wherein a soft symbol is clipped.