Granted Patent B2

US 8,315,528 · App. 12/644,409 · Granted Nov 20, 2012

Abstract

Claims (40)

1. In a coherent receiver of an optical communications system, a method of data symbol recovery comprising:

a frequency error estimator computing an estimate of a frequency offset Δf between a transmit laser and a Local Oscillator (LO) of the receiver;

a phase rotator applying a phase rotation κ(n) to a plurality of data symbol estimates to generate corresponding rotated symbol estimates, the phase rotation κ(n) being derived from the estimated frequency offset Δf; and

a filter block processing the rotated symbol estimates using a bounded filter function to generate corresponding decision values of each data symbol.

2. The method of claim 1 , wherein each data symbol is encoded in accordance with a predetermined encoding scheme, and wherein the predetermined encoding scheme is not a differential encoding scheme.

3. The method of claim 1 , further comprising:

processing two or more SYNC bursts having a known symbol sequence and periodicity to derive an estimate of the frequency offset Δf between a transmit laser and a Local Oscillator (LO) of the receiver.

4. The method of claim 3 , wherein processing two or more SYNC bursts comprises:

calculating a phase slope ηφ SYNC (i) between a pair of successive SYNC bursts; and

deriving a phase offset parameter ψ based on the calculated phase slope.

5. The method of claim 4 , wherein calculating the phase slope between a pair of successive SYNC bursts comprises:

calculating a respective phase error Δφ SYNC (i) of each SYNC burst, relative an ideal phase of corresponding symbol values within a phase space of an encoding scheme used to encode the data symbols;

calculating the phase slope ηφ SYNC (i) using the respective phase errors and the known periodicity of the SYNC bursts.

6. The method of claim 4 , wherein the phase offset parameter ψ is equal to the calculated phase slope.

7. The method of claim 4 , wherein the phase offset parameter ψ is equal to an average of respective phase slopes of two or more successive pairs of SYNC bursts.

8. The method of claim 4 , wherein computing the phase rotation κ(n) comprises accumulating, for each successive data symbol estimate, a corresponding incremental phase rotation corresponding to μ 1 ψ, where μ 1 is a scaling factor defining a phase adjustment step size for each successive symbol estimate.

9. The method of claim 1 , wherein processing the rotated symbol estimates comprises, for each rotated symbol estimate:

calculating a respective symbol phase error Δφ(n) of the rotated symbol estimate;

filtering the rotated symbol estimate based on the calculated symbol phase error Δφ(n); and

selecting a decision value of a corresponding data symbol based on the filtered symbol estimate.

10. The method of claim 9 , wherein filtering the rotated symbol estimate comprises:

comparing the respective symbol phase error Δφ(n) to a predetermined threshold; and

when the respective symbol phase error Δφ(n) exceeds the predetermined threshold, clipping a phase of the rotated symbol estimate.

11. The method of claim 10 , wherein clipping a phase of the rotated symbol estimate comprises:

calculating a second phase rotation τ(n) based on a difference between the respective symbol phase error Δφ(n) and the predetermined threshold; and

applying the second phase rotation τ(n) to the rotated symbol estimate.

12. The method of claim 9 , wherein the predetermined threshold value is based on a width of decision regions of a phase space of an encoding scheme used to encode the data symbols.

13. The method of claim 12 , wherein the predetermined threshold value corresponds with a phase offset of boundaries of each decision region, relative an ideal phase of symbols of the encoding scheme.

14. The method of claim 12 , wherein the predetermined threshold value differs from a phase offset of boundaries of each decision region, relative an ideal phase of symbols of the encoding scheme.

15. The method of claim 9 , wherein the predetermined threshold value is based on a bandwidth of a filter block of the receiver for filtering the rotated symbol estimates.

16. The method of claim 9 , wherein filtering the rotated symbol estimate comprises, for each symbol estimate:

computing a respective second order phase rotation term μ 2 Δφ(n+1) based on the symbol phase error Δφ(n), where μ 2 is a scaling factor defining a phase adjustment step size for each successive symbol estimate; and

adding the second order phase rotation term μ 2 Δφ(n+1) to the phase rotation κ(n) applied to the symbol estimate.

17. The method of claim 9 , wherein selecting the decision value comprises:

analysing a phase of the filtered symbol estimate to identify a decision region, of a phase space of an encoding scheme used to encode the data symbols, in which the filtered symbol estimate is located; and

selecting a symbol value of the encoding scheme associated with the identified decision region as the decision value.

18. A coherent receiver of an optical communications system, the coherent receiver comprising:

a frequency error estimator for computing a frequency offset Δf between a transmit laser and a Local Oscillator (LO) of the receiver;

a phase rotator for applying a phase rotation κ(n) to a plurality of data symbol estimates to generate corresponding rotated symbol estimates, the phase rotation κ(n) being derived from the estimated frequency offset Δf; and

a filter block for processing the rotated symbol estimates using a bounded filter function to generate corresponding decision values of each data symbol.

Assignments (8)