Granted Patent B2

US 7,623,790 · App. 11/233,587 · Granted Nov 24, 2009

Abstract

Claims (20)

1. A method for identifying characteristics of an optical signal, comprising:

receiving a data sample representing a first optical signal; said receiving step comprising the steps of:

providing an optical fiber having an amount of optical power therein; tapping off a fraction of the optical power present in the optical fiber; obtaining an optical power spectrum; and parsing the optical power spectrum into a plurality of data samples associated with particular channels of a wavelength division multiplexing communication system; wherein each of the data samples comprises a plurality of data points; and wherein obtaining the optical power spectrum is limited to a first resolution by the characteristics of an optical spectrum analyzer, and wherein at least one signal characteristic attributed to the first optical signal has a resolution finer than the first resolution;

comparing the data sample to a set of pre-stored model curves, each model curve representing at least one signal characteristic using a curve-fit analysis;

determining, for each comparison, an error value representing a quantity of difference between the data sample and a particular one of the pre-stored model curves;

selecting one of the pre-stored model curves with an error value having a least quantity of difference;

attributing to the first optical signal the at least one signal characteristic of the selected one of the pre-stored model curves; and

determining center frequency detuning from an ITU grid for the optical signal by performing a least-squares estimate for each characteristic functional form several times with small discrete frequency shifts to the functional form applied each time.

2. The method of claim 1 , wherein comparing the data sample to a set of pre-stored model curves comprises comparing the data sample to each of the pre-stored model curves in parallel.

3. The method of claim 1 , wherein comparing the data sample to a set of pre-stored model curves comprises comparing the data sample to at least a subset of the pre-stored model curves in parallel.

4. The method of claim 1 , further comprising performing a best-fit to a line.

5. The method of claim 1 , wherein the data sample comprises a plurality of power spectrum values, and the plurality of power spectrum values fall within the bandwidth of a DWDM channel.

6. The method of claim 1 , wherein the curve-fit analysis provides the amplitude multiplier and offset value of the pre-stored model curve.

7. The method of claim 6 , wherein the step of determining an error value prior to selecting one of the pre-stored model curves, includes determining an error value that considers the amplitude multiplier and the offset value.

8. The method of claim 6 , further comprising obtaining an OSNR value based, at least in part, on the ratio of the amplitude multiplier and the offset value.

9. The method of claim 1 , wherein the curve-fit analysis includes frequency shifting of the model curves to measure frequency detuning of the first optical signal.

10. The method of claim 1 , wherein the act of comparing the data sample to a set of pre-stored model curves comprises performing a least squares curve-fit between the data sample and a particular one of the pre-stored model curves.

11. The method of claim 1 , wherein the pre-stored model curves reside in memory of a computer.

12. The method of claim 1 , wherein the method is performed by an optical monitoring device in conjunction with a computer.

13. The method of claim 1 , wherein the method is performed by a computer.

Assignments (8)