AU2008314503A1 - Optical performance monitoring - Google Patents

Description

WO 2009/049368 PCT/AU2008/001532 1 "Optical performance monitoring" Cross-Reference to Related Applications The present application claims priority from Australian Provisional Patent Application No 2007905657 filed on 16 October 2007, the content of which is incorporated herein 5 by reference. Technical Field This invention concerns optical performance monitoring, and in particular relates to the use of low speed multi-tap sampling and calculation of a higher order correlation 10 coefficient to identify signal bit rate. Background Art Optical performance monitoring (OPM) is important for monitoring highly transparent and reconfigurable optical networks and switching systems. OPM allows measurement 15 of channel quality and impairments without prior knowledge of the origin, transport history and content of a signal. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a 20 context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each claim of this application. 25 Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps. 30 Summary of the Invention According to a first aspect the present invention provides a method for monitoring an optical signal, the method comprising: sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples 35 from the optical signal which are separated in time by a tap delay; retrieving a plurality of sample sets over time; WO 2009/049368 PCT/AU2008/001532 2 determining from the plurality of sample sets a higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a second sample; altering the tap delay and repeating the sampling, retrieving and determining, to 5 obtain a higher order correlation coefficient for the altered tap delay; and analysing a higher order correlation coefficient function to identify at least one characteristic of the optical signal. According to a second aspect, the present invention provides a system for monitoring 10 an optical signal, the system comprising: a multi-tap sampling device to obtain from the optical signal a plurality of sample sets each comprising at least two samples which are separated in time by a tap delay, and to obtain a plurality of such sample sets for differing values of tap delay; a processor to compute for each tap delay a higher order correlation coefficient 15 comprising an expected value of a higher order combination of a first sample with a second sample; and to compute at least one characteristic of the optical signal by analysing a higher order correlation coefficient function. A computer program product comprising computer program code means to make a 20 computer execute a procedure for monitoring an optical signal, the computer program element comprising: computer program code means for sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples from the optical signal which are separated in time 25 by a tap delay; computer program code means for retrieving a plurality of sample sets over time; computer program code means for determining from the plurality of sample sets a higher order correlation coefficient comprising an expected value of a higher order 30 combination of a first sample with a second sample; computer program code means for altering the tap delay and repeating the sampling, retrieving and determining, to obtain a higher order correlation coefficient for the altered tap delay; and computer program code means for analysing a higher order correlation 35 coefficient function to identify at least one characteristic of the optical signal.

WO 2009/049368 PCT/AU2008/001532 3 The expected value of higher order combination of the first sample with the second sample may comprise the expected value of: the square of the first sample multiplied by the square of the second sample. More generally the expected value of higher order combination of the first sample with the second sample may comprise: 5

E[X

1

"X

2 m ] where X 1 denotes the first sample of each of the sample sets and X 2 denotes the respective second sample of each of the sample sets, and at least one of n and m is greater than one. Additionally, the higher order combination of the first sample with the second sample may comprise other mathematical operations such as the sine, 10 cosine, logarithm or other operations upon either or both of X 1 and X 2 , provided that the higher order correlation coefficient function, such as the function of higher order correlation coefficient vs. tap delay, yields at least one desired characteristic of the optical signal. Correlation coefficients derived by raising one or both of X 1 and X 2 to an order higher than 2 may also be used to derive more information from the original 15 signal. Preferably, the higher order correlation coefficient is obtained for each of a large number of different tap delay values within a tap delay range of interest, to improve resolution of the higher order correlation coefficient function within that range. The 20 tap delay range of interest may be positioned many bit periods away from the origin of the correlation coefficient function, in order to capture a substantially periodic portion of the function not influenced by effects close to the origin. The or each tap delay between the at least two sample points may be applied in the 25 electrical domain, for example by buffering, or in the optical domain, for example by splitting the optical signal into paths of different lengths. The present invention may be of utility in determining a bit rate of an optical data signal, such as a non-return-to-zero (NRZ) data signal or return-to-zero (RZ) data 30 signal. In a third aspect, the invention is a method for automatically evaluating the bit rate of a non-return-to-zero (NRZ) data stream, comprising the steps of: splitting an original signal into two identical signals; 35 delaying one of the two identical signals using an asynchronous sampling technique; WO 2009/049368 PCT/AU2008/001532 4 computing the correlation coefficient of the power of each of the original and delayed signals over a range of relative delays; and determining the repetition rate of the original signal from the computed cross correlation coefficients. 5 Embodiments of the invention may provide for automated identification of bit rate in a manner that is robust against multiple optical impairments such as dispersion and noise. Some embodiments of the invention may be used to enhance the performance of a multi-impairment optical performance monitoring system and the management of 10 intelligent reconfigurable optical add drop multiplexers (ROADM) in future optical networks. Asynchronous sampling may be implemented using an asynchronous sampling device with a two-tap delay line. Using asynchronous sampling, the invention only requires 15 low-frequency sampling such that a time between the gathering of a first sample set and a second sample set is substantially greater than a bit period of the optical signal. Samples may be collected at two time points economically for the calculation of the correlation coefficients. 20 In a fourth aspect, the invention is a system for automatically evaluating the bit rate of a non-return-to-zero (NRZ) data stream, comprising: an asynchronous sampling device to split an original signal into two identical signals and to delay one of the two identical signals; and a processor to compute the correlation coefficient of the power of each of the 25 two signals over a range of relative delays and to determine the repetition rate of the original signal from the computed correlation coefficients. In a fifth aspect, the invention is software to perform the method. 30 According to a sixth aspect the present invention provides a method for monitoring impairments in an optical signal, the method comprising: sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples from the optical signal which are separated in time by a tap delay; 35 retrieving a plurality of sample sets over time; WO 2009/049368 PCT/AU2008/001532 5 determining from the plurality of sample sets a first higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a second sample; determining from the plurality of sample sets a second higher order correlation 5 coefficient comprising an expected value of a higher order combination of a first sample with a second sample and being of a different order to the first higher order correlation coefficient; and analysing the first and second higher order correlation coefficients to identify at least one characteristic of the optical signal. 10 Even with a single value of tap delay, obtaining multiple coefficients of differing order may provide valuable information as to impairments and/or signal format, even without retrieving the function of coefficient correlation vs. tap delay. For example first, second and third order correlation coefficients may be obtained, as may higher order 15 correlation coefficients. Brief Description of Drawings An example of the invention will now be described with reference to the accompanying drawings, in which: 20 Fig. 1 is a diagram of the optical performance monitoring system exemplifying the invention; Fig. 2 is a flowchart of the steps performed by the asynchronous sampling module and the bit rate identification module in Fig. 1; Fig. 3a is a plot of a first order correlation coefficient of a non return to zero 25 (NRZ) optical signal with sampling tap delays ranging from zero to a few bit periods; Fig. 3b is a plot of a second order correlation coefficient of a NRZ optical signal with sampling tap delays ranging from zero to a few bit periods; and Fig. 3c is a plot of a second order correlation coefficient of a NRZ optical signal with large sampling tap delays ranging over a few bit periods; 30 Fig. 4 is a schematic of an experimental setup exemplifying the invention; Fig. 5a is a plot of experimentally obtained second order correlation coefficients of NRZ input signals over a range of delays, each signal having a different bit rate; and Fig. 5b is a plot of experimentally obtained second order correlation coefficients of input signals over a range of delays, each signal having a different level of dispersion; 35 Figure 6a shows plots of the first order correlation function vs. tap delay for a 1OGbit/s NRZ signal when subjected to: 0 ps/nm dispersion and 32dB OSNR; 0 ps/nm WO 2009/049368 PCT/AU2008/001532 6 dispersion and 17dB OSNR; and 850 ps/nm dispersion and 17dB OSNR, respectively; Figure 6b is a chart of plots of the second order correlation function vs. tap delay for a lOGbit/s NRZ signal when subjected to: 0 ps/nm dispersion and 32dB OSNR; 0 ps/nm dispersion and 17dB OSNR; and 850 ps/nm dispersion and 17dB OSNR, respectively; 5 and Figure 6c is a chart of plots of the second order correlation function vs. tap delay, in a range of large tap delays, for a 1OGbit/s NRZ signal when subjected to: 0 ps/nm dispersion and 32dB OSNR; 0 ps/nm dispersion and 17dB OSNR; and 850 ps/nm dispersion and 17dB OSNR, respectively; Figure 7a is a chart of plots of the robustness to OSNR of the second order 10 correlation and the first order correlation, respectively for a 10 Gbit/s signal; Figure 7b is a chart of plots of the robustness to chromatic dispersion of the second order correlation and the first order correlation, respectively for a 10 Gbit/s signal; and Figure 7c is a chart of the accuracy of NRZ bit rate estimation of the second order correlation function. 15 Figures 8a, 8b and 8c illustrate the effects of chromatic dispersion upon the second order (E[X 2

X

2 2 ]), third order E[X 1 3

X

2 3 ]and sixth order E[X 1 6

X

2 6 ] correlation functions, respectively; Figure 9a plots the first order correlation coefficient function obtained from an amplitude differential phase shift keying (A-DPSK) signal and the first order 20 correlation coefficient function obtained from a NRZ signal, respectively; and Figure 9b plots the second order correlation coefficient function obtained from an A-DPSK signal and the second order correlation coefficient function obtained from a NRZ signal, respectively; and Figure 10 is a plot of the first and second order correlation coefficient functions 25 of a 40 Gbit/s RZ signal, illustrating the effect of dispersion. Disclosure of the Invention Some portions of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operations on data bits within a computer 30 memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, though not 35 necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has WO 2009/049368 PCT/AU2008/001532 7 proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like. It should be borne in mind, however, that all of these and similar terms are to be 5 associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussion, it is appreciated that throughout the description, discussions utilizing terms such as "processing" or "computing" or "calculating" or "determining" or "displaying" or the like, refer to the action and processes of a computer system, or 10 similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices. 15 The present invention also relates to apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may comprise a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in 20 a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus. 25 The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general purpose systems may be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The 30 required structure for a variety of these systems will appear from the description below. In addition, the present invention is not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the invention as described herein. 35 A machine-readable medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, a WO 2009/049368 PCT/AU2008/001532 8 machine-readable medium includes read only memory ("ROM"); random access memory ("RAM"); magnetic disk storage media; optical storage media; flash memory devices; electrical, optical, acoustical or other form of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.); etc. 5 Referring first to Fig. 1, the optical performance monitoring system 100 comprises: an asynchronous sampling module 110, a bit rate identification module 120, a format identification module 130 and a multi-impairment monitoring module 140. 10 A NRZ data stream 105 is first tapped from an optical network being monitored and passed through the asynchronous sampling module 110. Asynchronous delay tap sampling combines asynchronous sampling with a two-tap delay line to sample the original signal. Referring now to Fig. 2, the input signal is first split into two identical signals; step 112. One of the two signals is then delayed by At and both signals are 15 then sampled; step 114. Notably, in alternative embodiments the delay could be applied in the electrical domain, after optical-to-electrical detection. Each sample point includes two measurements separated by a fixed time corresponding to the delay length At. The outputs of the asynchronous sampling module 110 are two arrays of data denoting the original signal at two instants: 20 X(t) and X(t +At). The first order correlation coefficient Rxx of the input signal can be computed as the expected value of a product of the original signal and its delayed version: Rxx(t, t + At) = E[X(t) X(t +At)]. 25 Fig. 3(a) shows a plot of the first order correlation coefficient function Rxx(t, t+ At) over a range of delays At for a NRZ signal. While the width of the first order correlation function could be used to estimate the bit rate, this width is sensitive to impairments which thus cause errors in bit rate estimation. 30 Data Bit Rate Identification At the bit identification module 120, the sensitivity to impairments is substantially overcome by using the second order correlation coefficient, namely the correlation coefficient of the power of the signals, R 2 xx, to extract periodic characteristics of the 35 NRZ data stream. The power of each of the two signals X(t) and X(t+At) is first WO 2009/049368 PCT/AU2008/001532 9 computed to obtain the following second order correlation coefficient; steps 122 and 124 in Fig. 2: R xx(t, t + At) = E[X 2 (t) X 2 (t +At)]. 5 It will be appreciated that although a power of two has been used here, a higher order correlation coefficient may also be used. For example, R'xx(t, t + At), where n > 3 may also be used. Further, although it is currently preferred that X(t) and X(t+At) be raised to the same order, this might not be the case in alternative embodiments. It is noted that R xx is obtained directly from the sampled data and thus does not entail any pre 10 processing. Fig. 3(b) shows a plot of the second order power correlation coefficient function R 2 xx(t, t + At) over a range of relative delays. In contrast to the first order correlation coefficient function, R 2 xx exhibits a periodic structure. The repetition rate, i.e. bit rate, 15 of the sampled NRZ data stream can then be estimated using the time difference (T) between the multiple peaks on the correlation curve; step 128 in Fig. 2. The amplitude and time of the first peak and notch of the second order (power) correlation curve in Fig. 3(b) shows sensitivity to impairments and introduces 20 systematic error to the repetition period estimation. Therefore, in order to minimise error and further improve the accuracy of the bit rate estimation, the time delay At may be set to vary over a range which covers a few bit periods and which has a lower limit of at least several bit periods. The second order correlation for such large tap delay times is shown in Figure 3(c). This can be simply obtained by using sample pairs with 25 a small tap delay and then offsetting sample pairs such that the sampling rate contributes to tap delay, for example by choosing X 1 = X(ti) and X 2 = X(ti+ 1 ±At). In turn, the power correlation coefficient becomes:

R

2 xx(t, t + At') = E[X 2 (t) X 2 (t+At')] where At'= (ti.

1 - ti)+ At. 30 We note that larger delays and improved accuracy may be effected by offsetting by more than one sampling period, and optionally building the function from many different such sampling period offsets. This in turn may allow a smaller number of optical delay settings to be provided. Fig. 3(c) shows the power correlation function R~xx over a range of delays At'. Compared to Fig. 3(c), the periodic trend of the curve 35 is enhanced and its repetition period (T) becomes more distinguishable and less WO 2009/049368 PCT/AU2008/001532 10 affected by the offsets caused by correlation when At < -T. The bit rate of the input NRZ data stream is 1/T; step 128 in Fig. 2. The second order correlation coefficient function may also be usefully applied to 5 measure the bit period of a RZ signal. For a non-distorted RZ signal the first order correlation function has a periodic tail from which the bit rate may be identified. However, the amplitude of this tone may be small and the period difficult to measure. Figure 10 is a plot of the first and second order correlation coefficient functions of a 40 Gbit/s RZ signal, illustrating the effect of dispersion at a level of 160 ps/nm on each 10 function. In contrast to the first order function, the second order correlation function shows a more clearly defined periodic structure, particularly for At above about 100 ps, from which the bit rate can be more effectively determined. Data Format Identification and Multi-Impairment Monitoring 15 Correlation coefficients R 2 xx(t, t + At) may also be used to determine data format of the input signal 105 and to enhance network impairment monitoring. Experimental Setup - Bit Rate Identification An exemplary experimental setup to demonstrate the principle of the invention will 20 now be explained with reference to Fig. 4. Firstly, a 1552.5 nm tunable laser 205 is externally modulated with a Mach-Zehnder modulator (MZM) 210 and a 2 2-1 pseudo random bit sequence (PBRS) generator 215 to produce a NRZ signal at bit rates ranging from 5 to 40 Gb/s. The signal is then passed through an erbium-doped fiber amplifier (EDFA) 225 to compensate the loss of the modulator 210. 25 Optical impairments such as dispersion and amplified spontaneous emission (ASE) noise are then added to the signal. OSNR is varied by combining the signal with an ASE noise source 220 attenuated using a variable optical attenuator (VOA) 225. The signal is then launched into a piece of standard single mode fiber (SMF) 230 for the 30 introduction of dispersion. Variable dispersion is achieved by switching the signal into different lengths of SMF with a dispersion coefficient of 16 ps/nm/km. To compensate the difference in the link loss, a variable optical attenuator (VOA) 235 may be added right after the SMF. 35 The signal with dispersion and noise may be monitored using an optical spectrum analyzer (OSA) 240. The signal is then re-amplified by an EDFA 245 to overcome any WO 2009/049368 PCT/AU2008/001532 11 system losses, and optically de-multiplexed at 250 to remove ASE noise contributed by other frequency channels using an arrayed waveguide grating. The resulting signal is then passed through an asynchronous delay tap sampling module 5 260. The signal is first split optically into two signals using a 3dB coupler 265. One arm 270 of the 3 dB coupler is connected to a variable optical delay (VOD) device 275 variable between 0 and 300ps in lps increments, while the other arm 280 is connected by an optical patchcord directly to the DCA. 10 Both arms are then directed to the optical inputs of a "free-running" two-channel digital communications analyzer (DCA) 285, where asynchronous sampling of the signal occurs. Delay-tap samples X(t) and X(t+At) are then captured and transferred to bit rate identification device 290 for bit rate identification, as described with reference to Figs. 1 and 2. For each tap delay setting of VOD 275, 4000 sample sets were obtained. 15 Results Analysis - Bit Rate Identification Results of the experimental setup in Fig. 4 will now be explained with reference to Figs. 5(a) and 5(b). In the experiment, signals of different bit rates from 5 Gb/s to 12.5 Gb/s, and 40 Gb/s are measured. Fig. 5(a) shows eight correlation curves with bit rates 20 ranging from 9 Gb/s to 11.5 Gb/s. These curves can be approximately fitted with sinusoidal functions, which are then used to calculate the repetition period of the sampled data stream. Alternatively the period can be obtained from other means such as the time between zero crossing points or alternatively from the frequency content as obtained from the Fourier transform (or FFT) of the correlation function. The same 25 method applies to a 40 Gb/s signal. The estimated bit rates, together with the percentage error and standard deviation of the estimation, are tabulated in Table 1 and plotted in Figure 7c. The percentage error is the highest at 40 Gb/s because the variable optical delay line is adjusted in a 1 ps step. 30 Due to the short period of 40 Gb/s signal, a 1 ps delay step causes about 2% error to the final estimation. However, a lower estimation error may be achieved by improving the delay step resolution to better than 0.1 ps per capture at DCA, and/or by capturing further data to extending the function over many more periods. 35 Table 1(a) Results of bit rate identification over a range of signal bit rates WO 2009/049368 PCT/AU2008/001532 12 Actual Bit Rate Estimated Bit Rate Error (Gb/s) (Gb/s) (%) 5.00 5.01 0.28 6.00 6.01 0.09 7.00 7.01 0.13 8.00 8.01 0.08 8.50 8.53 0.33 9.00 9.01 0.15 9.50 9.53 0.28 10.00 10.01 0.13 10.31 10.33 0.21 10.51 10.54 0.25 10.66 10.67 0.12 10.71 10.72 0.10 11.00 11.01 0.10 11.50 11.52 0.17 12.00 12.01 0.07 12.50 12.53 0.24 40.00 39.83 0.42 Systems with different levels of OSNR and fibre length are also analysed. The estimated average bit rate of five input signals having 32dB OSNR and different chromatic dispersion levels are shown in Table 1(b) and Fig. 5(b). The dispersion level 5 of the signal is varied from 0 to 850 ps/nm by ranging the SMF length from 0 to 500 km. As shown, the estimation error is less than 0.4% in all cases. Table 1(b): Estimation results of a 10 Gb/s signal with different dispersion levels Dispersion Average Bit Rate Error (ps/nm) (Gb/s) % 0 10.03 0.3% 170 10.03 0.3% 425 10.03 0.3% 595 10.00 0.4% 850 10.03 0.3% WO 2009/049368 PCT/AU2008/001532 13 The effect of OSNR on the estimation results is also analysed using a 10 Gb/s NRZ data stream with OSNR ranging from 32 dB to 12 dB. As shown in Table 1(c), the one bit period of the 10 Gb/s signal, which is 100 ps, can be detected with a percentage error of less than 0.3%, except when the OSNR is degraded to 12 dB. 5 Table I(c): Estimation results of a 10 Gb/s signal with different OSNRs OSNR Average Bit Rate Error (dB) (Gb/s) (%) 32dB 10.03 0.3% 28dB 10.03 0.3% 24dB 10.03 0.3% 20dB 10.00 0.0% 16dB 10.03 0.3% 12dB 10.16 1.6% The results demonstrate that the invention is capable of detecting the period of a signal with multiple optical impairments such as noise and dispersion effectively. It is also 10 shown that these impairments only generally affect the height of the peaks and notches of the correlation coefficient function, and have little effects on the period of the correlation coefficient; see Fig. 5(b) for example. The described embodiments of the invention thus use the asynchronous delayed 15 samples to determine bit rate from the period of the second order auto correlation function of the signal intensity. In contrast to methods based on the first order auto correlation functions, it is demonstrated that such embodiments are robust to impairments. 20 Fig 8 demonstrates the sensitivity of the 2 d, 3rd and 6 th order correlation coefficient functions to chromatic dispersion. It is proposed that the analysis of simultaneously measured correlation functions can distinguish and measure multiple simultaneous impairments. 25 The present invention further provides for determination of signal format, even in the presence of signal impairments. For example, in Figures 9a and 9b is plotted the 1st and 2nd order auto correlation functions for 10 Gb/s NRZ and A- DPSK signals for comparison. Differences are apparent in both orders. Although the first order WO 2009/049368 PCT/AU2008/001532 14 correlation function is sufficient to distinguish format in this ideal case the higher order correlation function will enable format identification in more realistic cases of signal impairment, for which the 1st order may not be sufficient. 5 The present invention may be applied in conjunction with the techniques set forth in International Patent Publication Numbers WO 2007/041808 and WO 2007/041807, the contents of each application being incorporated herein by reference. It will be appreciated by persons skilled in the art that numerous variations and/or 10 modifications may be made to the invention as shown in the specific embodiments without departing from the scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims (18)

1. A method for monitoring an optical signal, the method comprising: sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples 5 from the optical signal which are separated in time by a tap delay; retrieving a plurality of sample sets over time; determining from the plurality of sample sets a higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a second sample; 10 altering the tap delay and repeating the sampling, retrieving and determining, to obtain a higher order correlation coefficient for the altered tap delay; and analysing a higher order correlation coefficient function to identify at least one characteristic of the optical signal.

2. The method of claim 1 wherein the expected value of higher order combination 15 of the first sample with the second sample comprises: E[X 1 "X 2 "] where X 1 denotes the first sample of each of the sample sets and X 2 denotes the respective second sample of each of the sample sets, and at least one of n and m is greater than one. 20

3. The method of claim 1 or claim 2 wherein the higher order combination of the first sample with the second sample comprises at least one mathematical operation selected from the sine, cosine and logarithm, carried out upon at least one of X1 and X 2 .

4. The method of any one of claims 1 to 3 wherein correlation coefficients derived by raising one or both of Xi and X 2 to an order higher than 2 are analysed to identify at 25 least one characteristic of the optical signal.

5. The method of claim 4, wherein the expected value of higher order combination of the first sample with the second sample comprises the expected value of: the square of the first sample multiplied by the square of the second sample.

6. The method of any one of claims 1 to 5 wherein the higher order correlation 30 coefficient is obtained for each of a large number of different tap delay values within a tap delay range of interest, to provide improved resolution of the higher order correlation coefficient function within that range. Substitute Sheet m 1 xn nIATT WO 2009/049368 PCT/AU2008/001532 16

7. The method of claim 6 wherein the tap delay range of interest is positioned many bit periods away from the origin of the correlation coefficient function, in order to capture a substantially periodic portion of the function not influenced by effects close to the origin. 5

8. The method of any one of claims 1 to 7 wherein the or each tap delay between the at least two sample points is applied in the electrical domain.

9. The method of any one of claims 1 to 7 wherein the or each tap delay between the at least two sample points is applied in the optical domain.

10. A system for monitoring an optical signal, the system comprising: 10 a multi-tap sampling device to obtain from the optical signal a plurality of sample sets each comprising at least two samples which are separated in time by a tap delay, and to obtain a plurality of such sample sets for differing values of tap delay; a processor to compute for each tap delay a higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a 15 second sample; and to compute at least one characteristic of the optical signal by analysing a higher order correlation coefficient function.

11. The system of claim 10 wherein the processor is configured to compute the expected value of higher order combination of the first sample with the second sample as: 20 E[X 1 "X 2 m ] where X 1 denotes the first sample of each of the sample sets and X 2 denotes the respective second sample of each of the sample sets, and at least one of n and m is greater than one.

12. The system of claim 10 or claim 11 wherein the processor is further configured 25 to analyse correlation coefficients derived by raising one or both of Xi and X 2 to an order higher than 2, to identify at least one characteristic of the optical signal.

13. The system of any one of claims 10 to 12 further comprising electrical delay means to effect the or each tap delay between the at least two sample points in the electrical domain. Substitute Sheet (Rule 26) RO/AU WO 2009/049368 PCT/AU2008/001532 17

14. The system of any one of claims 10 to 12 further comprising optical delay means to effect the or each tap delay between the at least two sample points in the optical domain.

15. A computer program product comprising computer program code means to 5 make a computer execute a procedure for monitoring an optical signal, the computer program element comprising: computer program code means for sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples from the optical signal which are separated in time 10 by a tap delay; computer program code means for retrieving a plurality of sample sets over time; computer program code means for determining from the plurality of sample sets a higher order correlation coefficient comprising an expected value of a higher order 15 combination of a first sample with a second sample; computer program code means for altering the tap delay and repeating the sampling, retrieving and determining, to obtain a higher order correlation coefficient for the altered tap delay; and computer program code means for analysing a higher order correlation 20 coefficient function to identify at least one characteristic of the optical signal.

16. A method for automatically evaluating the bit rate of a non-return-to-zero (NRZ) data stream, comprising the steps of: splitting an original signal into two identical signals; delaying one of the two identical signals using an asynchronous sampling 25 technique; computing the correlation coefficient of the power of each of the original and delayed signals over a range of relative delays; and determining the repetition rate of the original signal from the computed cross correlation coefficients. 30

17. A system for automatically evaluating the bit rate of a non-return-to-zero (NRZ) data stream, comprising: Substitute Sheet (Rule 26) RO/AU WO 2009/049368 PCT/AU2008/001532 18 an asynchronous sampling device to split an original signal into two identical signals and to delay one of the two identical signals; and a processor to compute the correlation coefficient of the power of each of the two signals over a range of relative delays and to determine the repetition rate of the 5 original signal from the computed correlation coefficients.

18. A method for monitoring impairments in an optical signal, the method comprising: sampling the optical signal from at least two tap points to retrieve a sample set comprising at least two samples, the at least two tap points adapted to retrieve samples 10 from the optical signal which are separated in time by a tap delay; retrieving a plurality of sample sets over time; determining from the plurality of sample sets a first higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a second sample; 15 determining from the plurality of sample sets a second higher order correlation coefficient comprising an expected value of a higher order combination of a first sample with a second sample and being of a different order to the first higher order correlation coefficient; and analysing the first and second higher order correlation coefficients to identify at 20 least one characteristic of the optical signal. Substitute Sheet (Rule 26) RO/AU