Granted Patent B2

US 7,701,842 · App. 12/030,242 · Granted Apr 20, 2010

Abstract

Claims (77)

1. A method of suppressing effects of aliasing in a system for digitally processing a high speed signal having a symbol rate of 1/T, the method comprising:

filtering the high speed signal using a low-pass filter characteristic having a cut-off frequency corresponding to approximately ½T, and a roll-off greater than 20 dB per decade;

sampling the high speed signal at a fractional multiple (N) of the symbol rate, wherein N<2, to generate a corresponding sample stream; and

digitally processing the sample stream to compensate phase distortions due to the filtering.

2. The method as claimed in claim 1 , wherein N<3/2.

3. The method as claimed in claim 1 , wherein N>1.

4. The method as claimed in claim 1 , wherein the high speed signal is an input data stream x(t) to be transmitted through a link of an optical communications system, and wherein sampling the high speed signal comprises generating a stream of multi-bit samples of a desired complex optical E-field envelope of an optical signal for conveying the high speed signal through the link, based on the input data stream x(t).

5. The method as claimed in claim 4 , wherein the fractional multiple N=n/d, and wherein generating the stream of multi-bit samples comprises, during each clock cycle of period dxT:

converting the high speed signal into a set of n parallel input vectors, each input vector having a respective predetermined delay;

for each input vector, computing a respective multi-bit sample value of the desired complex optical E-field envelope at a predetermined phase shift relative the input vector; and

interleaving the computed multi-bit sample values.

6. The method as claimed in claim 5 , wherein computing the respective multi-bit sample value of the desired complex optical E-field envelope comprises applying a compensation operator C[] to the input vector, the compensation operator C[] being computed to compensate at least chromatic dispersion of the link.

7. The method as claimed in claim 6 , wherein digitally processing the sample stream to compensate phase distortions due to the filtering comprises:

characterizing the phase distortions as a function of frequency; and

computing the compensation operator C[] using the phase distortion characteristic so as to compensate the phase distortions.

8. The method as claimed in claim 5 , wherein the predetermined delay of each input vector and the predetermined phase shift used for computation of the respective multi-bit sample value for each input vector, are cooperatively selected such that the n multi-bit sample values computed during each clock cycle represent successive instantaneous values of the desired complex optical E-field envelope that are evenly spaced in time.

9. The method as claimed in claim 8 , wherein at least the step of computing, for each input vector, the respective multi-bit sample value is performed by a set of n parallel filter blocks, and wherein, for values of n=9 and d=8, the predetermined delay of each input vector and the predetermined phase shift used for computation of the respective multi-bit sample value for each input vector are:

Filter Block

Delay

Phase Shift

0

0

+4T/9

1

 T

+3T/9

2

2T

+2T/9

3

3T

 +T/9

4

4T

0

5

5T

 −T/9

6

6T

−2T/9

7

7T

−3T/9

8

7T

+5T/9

10. The method as claimed in claim 4 , wherein generating the stream of multi-bit samples comprises:

computing a Fast Fourier Transform (FFT) of the input data stream x(t) to generate an array {R} representing a frequency domain spectrum of the input data stream x(t);

processing the array {R} to generate a retimed array {R′} having a width corresponding to N times the width of the array {R}; and

computing an Inverse Fast Fourier Transform (IFFT) of the modified array {M}.

11. The method as claimed in claim 10 , wherein computing the Fast Fourier Transform (FFT) of the input data stream x(t) comprises steps of:

encoding the input data stream x(t) using a selected encoding scheme to generate a corresponding multi-bit complex valued symbol stream; and

computing a Fast Fourier Transform (FFT) of the multi-bit complex valued symbol stream.

12. The method as claimed in claim 11 , wherein the encoding scheme is selected from the set consisting of: Phase Shift Keying (PSK); Quadrature Phase Shift Keying (QPSK); and Optical Orthogonal Frequency Domain Multiplexing (O-OFDM).

13. The method as claimed in claim 10 , wherein processing the array {R} to generate a retimed array {R′} comprises increasing the width of the array {R} by a factor of N.

14. The method as claimed in claim 13 , wherein increasing the width of the array {R} comprises:

supplying a lowermost set of taps of the array {R} to a corresponding set of lowermost taps of the retimed array {R′}; and

supplying an uppermost set of taps of the array {R} to a corresponding set of uppermost taps of the retimed array {R′};

wherein the lowermost set of taps of the array {R} at least partially overlaps the uppermost set of taps of the array {R}, and wherein the set of lowermost taps of the retimed array {R′} is contiguous with the set of uppermost taps of the retimed array {R′}.

15. The method as claimed in claim 10 , wherein digitally processing the sample stream to compensate phase distortions due to the filtering comprises:

characterizing the phase distortions as a function of frequency; and

computing the compensation vector using the phase distortion characteristic so as to compensate the phase distortions.

16. The method as claimed in claim 1 , wherein the high speed signal is modulated on at least one transmitted polarization of an optical signal received through an optical link of an optical communications system, and wherein sampling the high speed signal comprises:

generating respective in-phase and quadrature multi-bit raw digital sample streams for each received polarisation of the optical signal, a sample rate of the multi-bit raw digital sample streams being N times the symbol rate of the high speed signal;

processing the multi-bit raw sample streams to compensate at least chromatic dispersion and polarization effects of the optical link; and

retiming the processing result to the symbol rate of the high speed signal.

17. The method as claimed in claim 16 , wherein processing the multi-bit raw sample streams and retiming the processed sample stream comprises:

computing a Fast Fourier Transform (FET) of the at least two multi-bit raw digital sample streams to generate at least two arrays {R}, each array {R} representing a frequency domain spectrum of a respective received polarization of the received optical signal, and

digitally processing the arrays {R} to generate at least one modified array {V}, each modified array {V} representing a frequency-domain spectrum of a corresponding transmitted polarization of the optical signal.

18. The method as claimed in claim 17 , wherein retiming the processing result comprises changing a width of each modified array {V} by a factor of 1/N.

19. The method as claimed in claim 18 , wherein changing a width of each modified array {V} comprises:

defining contiguous upper and lower sets of taps of the modified array {V}, each set of taps encompassing half of the taps of the modified array {V};

overlapping the upper and lower sets of taps by a predetermined number of taps in a center portion of the modified array {V};

adding the overlapped taps together, and supplying the addition result to corresponding central taps of a retimed array {V}′;

supplying uppermost taps lying above the center region of the modified array {V} to corresponding uppermost taps of the retimed array {V}′; and

supplying lowermost taps lying below the center region of the modified array {V} to corresponding lowermost taps of the retimed array {V}′.

Assignments (8)