Noise-tolerant optical modulation

1. A method of transmitting data, the method comprising:

an encoder encoding a data signal to be transmitted as symbols of an N-symbol constellation defined in an M-dimensional space, wherein N<2 M , and a composite distance between a given pair of symbols within the constellation increases with increasing Hamming distance between data words encoded in each of those symbols; and

a modulator modulating at least M dimensions of a carrier light in accordance with the symbols.

2. The method of claim 1 wherein the composite distance includes any one or more of an angular component and a Euclidean component.

3. The method of claim 1 wherein the composite distance includes a radial component.

4. The method of claim 1 wherein the composite distance includes an anti-correlation component.

5. The method of claim 4 wherein the anti-correlation component is with respect to phase angle.

6. The method of claim 4 wherein the anti-correlation component is with respect to a polarization angle.

7. The method of claim 1 wherein modulating at least M dimensions of the carrier light comprises modulating each polarization of the carrier light in accordance with a respective one of at least two 2-dimensional projections of the constellation.

8. The method as claimed in claim 7 wherein the at least two 2-dimensional projections of the constellation are distributed across any one or more of time, polarization, and I-Q axes of the carrier light.

9. The method as claimed in claim 7 , wherein a first one of the at least two 2-dimensional projections is related to a second one of the at least two 2-dimensional projections by a rotation of the constellation about at least one axis of the M-dimensional space.

10. The method as claimed in claim 9 , wherein the at least one axis of rotation is selected such that any two constellation points that have a small Euclidian separation in one 2-dimensional projection will have a large Euclidian separation in at least one other 2-dimensional projection.

11. The method as claimed in claim 9 , wherein the at least one axis of rotation is selected such that a phase angle between a pair of constellation points in one 2-dimensional projection is different from a corresponding phase angle between the pair of constellation points in another 2-dimensional projection.

12. The method as claimed in claim 9 , wherein at least two 2-dimensional projections are associated with respective time-frames, and wherein the at least one axis of rotation is selected such that a given pair of constellation points that have a polarization orientation vector with respect to each other in one time-frame, have a different polarization orientation vector with respect to each other in the other time-frame.

13. The method as claimed in claim 9 , wherein the at least one axis of rotation is selected to minimize signal power variations between different 2-dimensional projections.

14. The method as claimed in claim 9 , wherein the at least one axis of rotation is selected to increase radial distance values between constellation points within the at least one 2-dimensional projection.

15. The method as claimed in claim 7 , wherein the at least two 2-dimensional projections comprise an even time-frame associated with a corresponding even time slot, and an odd time-frame associated with a corresponding odd time slot, and wherein the respective projections of each of the even and odd time-frames are selected such that a first symbol has a different polarization orientation vector with respect to a second symbol in each of the even and odd time-frames.

16. The method as claimed in claim 7 , wherein the at least two 2-dimensional projections are selected such that a phase angle between a pair of constellation points in one projection is different from the phase angle between that same pair of constellation points in another projection.

17. The method as claimed in claim 1 , wherein a phase angle between co-planar symbols of the constellation is selected to vary with Hamming distance between b=log 2 (N)-bit data words encoded in each symbol.

18. A transmitter for transmitting data in an optical communications system, the transmitter comprising:

an encoder configured to encode a data signal to be transmitted as symbols of an N-symbol constellation defined in an M-dimensional space, wherein N<2 M , and a composite distance between a given pair of symbols within the constellation increases with increasing Hamming distance between data words encoded in each of those symbols; and

a modulator configured to modulate the symbols onto at least M dimensions of a carrier light.

19. The transmitter as claimed in claim 18 , wherein the composite distance includes any one or more of:

an angular component;

a Euclidean component;

a radial component; and

an anti-correlation component.

20. The transmitter as claimed in claim 18 , wherein the modulator is configured to modulate each polarization of the carrier light in accordance with a respective one of at least two 2-dimensional projections of the constellation.

21. The transmitter as claimed in claim 20 , wherein the at least two 2-dimensional projections of the constellation are distributed across any one or more of time, polarization, and I-Q axes of the carrier light.

22. The transmitter as claimed in claim 20 , wherein a first one of the at least two 2-dimensional projections is related to a second one of the at least two 2-dimensional projections by a rotation of the constellation about at least one axis of the M-dimensional space.

23. The transmitter as claimed in claim 20 , wherein the at least two 2-dimensional projections comprise an even time-frame associated with a corresponding even time slot, and an odd time-frame associated with a corresponding odd time slot, and wherein the respective projections of each of the even and odd time-frames are selected such that a first symbol has a different polarization orientation vector with respect to a second symbol in each of the even and odd time-frames.