WO2006128254A1 - Surveillance de performance optique wdm - Google Patents

Surveillance de performance optique wdm Download PDF

Info

Publication number
WO2006128254A1
WO2006128254A1 PCT/AU2006/000762 AU2006000762W WO2006128254A1 WO 2006128254 A1 WO2006128254 A1 WO 2006128254A1 AU 2006000762 W AU2006000762 W AU 2006000762W WO 2006128254 A1 WO2006128254 A1 WO 2006128254A1
Authority
WO
WIPO (PCT)
Prior art keywords
filter unit
output
tunable
tunable filter
wdm
Prior art date
Application number
PCT/AU2006/000762
Other languages
English (en)
Inventor
Rodney Stuart Tucker
Robert Evans
Original Assignee
National Ict Australia Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from AU2005902897A external-priority patent/AU2005902897A0/en
Application filed by National Ict Australia Limited filed Critical National Ict Australia Limited
Publication of WO2006128254A1 publication Critical patent/WO2006128254A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
    • H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
    • H04B10/077—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using a supervisory or additional signal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
    • H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
    • H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
    • H04B10/0795—Performance monitoring; Measurement of transmission parameters
    • H04B10/07955—Monitoring or measuring power
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
    • H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
    • H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
    • H04B10/0795—Performance monitoring; Measurement of transmission parameters
    • H04B10/07957—Monitoring or measuring wavelength

Definitions

  • the present invention relates broadly to a monitoring system for WDM signals or channels, and to a method of monitoring WDM signals.
  • WDM and DWDM optical communications systems multiple optical signals at different optical wavelengths share the same optical fibre transmission medium.
  • WDM and DWDM systems it is important that all signals are at their assigned channel wavelength and that the power level of all signals is controlled to within certain allowable levels. If a system fault occurs, such as a cable cut or a device failure, it is necessary to detect a lost or degraded signal and to isolate the failed part of the system.
  • many characteristics of the system including non-uniform amplifier gain spectra and nonlinear effects such as cross-phase or self-phase modulation can cause differential changes in the power level across optical signals in the system and degradation of system performance. It is therefore desirable to ensure that a monitor simultaneously measures the wavelengths of all channels and their optical power levels.
  • a number of techniques have been used in the past for optical monitoring in WDM and DWDM systems. These include conventional grating-based optical spectrum analysers, fixed-grating devices with detector arrays, interferometer-based wavelength meters, conventional multi-channel WDM and DWDM demultiplexers with multiple detectors, and high-finesse scanning Fabry-Perot filters.
  • the previous techniques for optical monitoring are often unsuitable because of: low operating speed, complicated calibration schemes, and poor performance.
  • techniques based on scanning Fabry-Perot filters are difficult to calibrate in the wavelength domain, and require complicated calibration schemes, using input reference circuits. Other techniques do not provide high wavelength resolution.
  • a method of monitoring individual signals or channels contained in a WDM optical spectrum comprising filtering the WDM optical spectrum utilising a tunable filter unit, filtering the output of the tunable filter unit utilising a fixed filter unit having a known transfer function as a function of wavelength, measuring the optical power at the output of the tunable filter unit for different tuning conditions of the tunable filter unit, measuring the corresponding optical power of the output of the fixed filter unit for said different tuning conditions, and determining the optical power level of each WDM signal and the error-corrected wavelength of each WDM signal from an analysis of the measured optical power at the output of the tunable filter unit and the corresponding measured optical power at the output of the fixed filter unit for said different tuning conditions.
  • the determining of the optical power level of each WDM signal and the error-corrected wavelength of each signal preferably comprises a deconvolution process to extract power levels and un- corrected values of the wavelength of each WDM signal from the power measurements at the output of the tunable filter unit for said different tuning conditions, and a deconvolution process to extract a measure of the fixed filter transfer function at each WDM signal wavelength from the corresponding optical power measurements at the output of the fixed filter unit for said different tuning conditions, and determining the error-corrected wavelength of each WDM signal from the results of the two deconvolutions.
  • the two deconvolution processes utilise the same algorithm.
  • the determining of the error-corrected wavelengths advantageously comprises determining a measure of the fixed filter transfer function at each individual WDM signal wavelength from an optical power measurement at the output of the fixed filter unit and a corresponding optical power measurement at the output of the tunable filter unit for the tuning condition for which a maximum intensity for said individual WDM signal is detected at the output of the tunable filter unit, and determining the corrected wavelengths from said measure of the fixed filter transfer function at said tuning condition and the known transfer function of the fixed filter as a function of wavelength.
  • the method may further comprise measuring reflected power from the input of the fixed filter unit at said different tuning positions of the tunable filter, whereby determining of the error corrected wavelengths of the WDM signals may further comprise an analysis of the measured reflected power for said different tuning conditions.
  • the analysis of the measured reflected power preferably comprises forming the difference of the measured reflected power and the corresponding measured power at the output of the fixed filter unit for each of said different tuning conditions of the tunable filter.
  • the method may further comprise inhibiting reflected signals from the input of the fixed filter unit from reaching the output of the tunable filter unit
  • the method further may comprise selectively manipulating the polarisation of the input signal to the tunable filter unit, whereby the method is adapted to conduct polarisation independent measurements.
  • the method preferably further comprises polarising the input to the tunable filter unit.
  • a monitoring system for monitoring individual signals or channels contained in a WDM optical spectrum, the system comprising: a tunable filter unit for filtering the WDM optical spectrum, a fixed filter unit for filtering the output from the tunable filter unit, the fixed filter unit having a known transfer function as a function of wavelength, a first power detection unit for measuring the optical power at the output of the tunable filter unit for different tuning conditions of the tunable filter unit, a second power detection unit for measuring the corresponding power at the output of the fixed filter unit for said different tuning conditions, and a processing unit for determining the optical power levels of each WDM signal and the error-corrected wavelengths of each individual WDM channel from an analysis of the measured optical power at the output of the tunable filter unit and the corresponding optical power at the output of the fixed filter unit for said different tuning conditions.
  • the processing unit is preferably arranged, to perform a deconvolution process to extract power levels and un-corrected values of the wavelength of each WDM signal from the optical power measurements at the output of the tunable filter unit for said different tuning positions, and to perform a deconvolution process to extract a measure of the fixed filter transfer function at each WDM signal wavelength from the corresponding optical power measurements at the output of the fixed filter unit for said different tuning conditions, and to determine the error corrected wavelengths of each WDM signals from the results of the two deconvolution processes.
  • the processing unit is arranged, to utilise the same algorithm in the two deconvolution processes.
  • the processing unit is preferably arranged, to determine a measure of the fixed filter transfer function at each individual WDM signal wavelength from the optical power measurement at the output of the fixed filter unit and the corresponding optical power measurement at the output of the tunable filter unit for the tuning condition for which a maximum intensity for said individual WDM signal is detected at the output of the tunable filter unit, and to determine the error-corrected wavelengths from said measure of the fixed filter transfer function at said tuning condition and the known transfer function of the fixed filter unit as a function of wavelength.
  • the system may further comprise a third power detection unit for measuring reflected power from the input of the fixed filter unit for said different tuning conditions of the tunable filter, and wherein the processing unit may be further arranged, to perform an analysis of the measured reflected power at said different tuning conditions for the determining of the corrected wavelengths of the WDM signals.
  • the processing unit is arranged, to form a difference of the measured reflected power and the corresponding measured power at the output of the fixed filter unit for each of said different tuning conditions of the tunable filter unit.
  • the system may comprise more than one fixed filter unit, whereby the processing unit may be arranged, to base the determination of each of the corrected wavelength positions on the measured power levels at the output of at least one of the fixed filter units for said different tuning conditions.
  • the fixed filter units have the same free spectral range, but positions of the peaks in the filter responses are offset.
  • the spectral peaks in the filter response of one of the fixed filter units preferably fall substantially halfway between the spectral peaks of one of the other fixed filter units.
  • the system may comprise more than one tunable filter unit, and the processing unit in such embodiments may be arranged, to base the determination of each corrected wavelength position on the measured power levels at the output of at least one of the tunable filter units for different tuning conditions.
  • the system is preferably arranged in a manner such that, reflected signals from the input of the fixed filter unit(s) are inhibited from reaching the output of the tunable filter unit(s).
  • the system may further comprise a polarisation manipulation unit for selectively manipulating the polarisation of an input signal to the tunable filter unit(s), whereby the system is adapted, to conduct polarisation independent measurements.
  • the system advantageously further comprises a polariser unit disposed, between the polarisation manipulation unit and the input of the tunable filter unit(s).
  • the polarisation manipulation unit may comprise a controllable waveplate or retarder.
  • the tunable filter unit(s) may comprise, but is not limited to, one or more of the group of a scanning fibre Fabry-Perot filter, a Micro electromechanical Systems (MEMS) Fabry-Perot tunable filter, a tunable optical planar waveguide filter, a tunable optical fibre filter such as a tunable Fibre Bragg Grating (FBG) filter, a tunable interferometer- based filter, and a tunable filter using an anisotropic tuning medium such as a liquid crystal Fabry-Perot tunable filter or other tunable filter based on liquid crystal materials.
  • MEMS Micro electromechanical Systems
  • FBG Fibre Bragg Grating
  • the fixed filter unit(s) may comprise, but is not limited to, one or more of the group of a Fabry Perot filter, an optical planar waveguide filter, and an optical fibre filter such as a Fibre Bragg Grating (FBG) filter, an interferometer-based filter, or a filter based on wavelength-dependent absorption or reflection.
  • a Fabry Perot filter such as a Fibre Bragg Grating (FBG) filter, an interferometer-based filter, or a filter based on wavelength-dependent absorption or reflection.
  • FBG Fibre Bragg Grating
  • the first power detection unit may comprise an optical splitting element and a photodiode.
  • the system may be implemented utilising waveguide optics or free-space optics.
  • the system may be implemented for WDM signals or DWDM signals.
  • FIG. 1 is a schematic diagram illustrating a WDM monitoring system embodying the present invention.
  • Figure 2 is a schematic diagram illustrating a preferred embodiment of a. WDM monitoring system.
  • Figure 3 shows spectra and transfer functions characterising one embodiment of the WDM monitoring system of Figure 2.
  • Figure 4 shows a detail of Figure 3(c).
  • FIG. 5 shows spectra and transfer functions characterising another WDM monitoring system embodying the present invention.
  • Figure 6 shows a detail of Figure 5(c).
  • Figure 7 shows spectra and transfer functions characterising another WDM monitoring system embodying the present invention.
  • Figure 8 shows a detail of Figure 7(c).
  • Figure 9 is a schematic signal flow graph illustrating a process of monitoring WDM signals embodying the present invention.
  • FIG. 10 is a flowchart illustrating a method of monitoring WDM signals embodying the present invention.
  • FIG 11 shows spectra and transfer functions characterizing another WDM monitoring system embodying the present invention.
  • FIG 12 is a schematic diagram of another WDM signal monitoring system embodying the present invention.
  • Figure 13 shows spectra and transfer functions characterising the WDM signal ' monitoring system of Figure 12.
  • Figure 14 is a schematic diagram illustrating another WDM signal monitoring system embodying the present invention.
  • Figure 15 shows spectra and transfer functions characterising the WDM signal monitoring system of Figure 14.
  • FIG 16 is a flowchart illustrating a method of monitoring WDM signals embodying the present invention.
  • FIG 17 is a schematic diagram of another WDM signal monitoring system embodying the present invention.
  • FIG. 18 is a schematic diagram of another WDM signal monitoring system embodying the present invention.
  • FIG 19 is a schematic diagram of another WDM signal monitoring system embodying the present invention.
  • FIG 20 is a schematic diagram of another WDM signal monitoring system embodying the present invention.
  • Figure 1 shows how an optical monitor 10 embodying the present invention can be placed in a WDM or DWDM system.
  • An optical splitter 12 taps off a small part (typically 10% or less) of the WDM or DWDM spectrum to be monitored at the WDM input 11.
  • the monitor 10 measures the optical wavelengths ⁇ i to X n , and the optical power levels P 1 to P n , of the n optical channels.
  • the optical wavelengths will normally be close to one of the standard ITU-designated optical wavelengths, but if one or more of the transmitters is faulty, these wavelengths may deviate from the ITU standard values.
  • FIG. 2 shows a preferred embodiment of the invention.
  • the optical input to the monitor 20 passes through a tunable optical filtering element 22.
  • the tunable optical filtering element 22 is tuned under the control of a filter controller 24 that is, in turn, controlled by the monitor controller and digital signal processing circuit 26.
  • the control signal to the tunable optical filtering element is typically a voltage, labeled in Fig. 2 as V.
  • the tunable optical filtering clement 22 provides a wavelength-dependent filtering response that attenuates different optical wavelengths by different amounts. When the controller input to the filtering element 22 changes, the attenuation characteristic as a function of optical wavelength changes.
  • the tunable optical filtering element 22 is a tunable band-pass filter such as a scanning Fabry Perot etalon or a tunable fibre Bragg grating or a bulk Bragg grating.
  • a tunable band-pass filter such as a scanning Fabry Perot etalon or a tunable fibre Bragg grating or a bulk Bragg grating.
  • other filtering elements and/or responses are also possible. It will be appreciated by a person skilled in the art that any tunable optical filtering element, such as a tunable optical filtering element for which its response as a function of wavelength is changed when the control signal V from the controller is changed, may be used in other embodiments.
  • An optical power detection unit 23 determines the optical power at the output of the tunable filtering element 22.
  • the power detection unit 23 in this embodiment comprises an optical splitter 21 and a photodiode 1.
  • the splitter 21 splits the output from the tunable optical filtering element 22 into two paths. One path is directed to photodiode
  • FIG. 3(a) shows an input WDM or DWDM spectrum
  • 3(b) shows the tunable filter transfer function for an embodiment in which the tunable filter 22 is a tunable band-pass filter and its passband response is sufficiently selective to discriminate one WDM or DWDM signal from all the other WDM or DWTM signals in Fig 3 (a).
  • the signal from photodiode 1 is sent to the controller and signal and processing circuit 26.
  • the dotted curve represents the "skirt" around the tuned wavelength position of the tunable filter element 22 (see Figure 2), i.e. the finite bandwidth of the tunable filter element 22 (see Figure 2) at each tuning position.
  • a complete measurement of all channels of the WDM spectrum requires the centre wavelength of the passband of the tunable filter element 22 to be scanned across all wavelengths of the input WDM spectrum.
  • the second output from the optical splitter 21 is fed into a fixed optical filter 28.
  • an optical isolator 30 is placed between the tunable filtering element 22 and the fixed filter 28, which reduces undesirable cavity effects between these two devices.
  • the fixed filter element 28 can take many forms.
  • the fixed filter is a Fabry-Perot etalon made using a stable material such as quartz.
  • the fixed optical filter 28 can be placed in a stabilised environment 32 to minimise temperature changes and other environmental changes that would change its characteristics.
  • the output power from the fixed optical filler 28 is detected using photodiode 2 and the output from this photodiode is fed to the controller and digital signal processing circuit 26.
  • Figures 3 (a) and 3(b) show the input WDM or DWDM optical spectrum, and the transfer function of the tunable filter element 22 for a preferred embodiment (see Figure 2).
  • Fig. 3(c) shows the fixed filter 28 transfer function (see Figure 2) for the same preferred embodiment.
  • the digital signal processing circuit 26 determines the average signal power level in the filtered WDM or DWDM spectrum. In determining this power level, the signal processing circuit calibrates the measurements to power levels at the WDM input 11 in Fig. 1. This calibration procedure accounts for the responsivity of photodiode 1, splitting losses in the optical splitter 12 in Figure 1, splitting losses in the optical splitter 21 and losses in other components in the monitor 20 ( Figure T), including the optical isolator 30. Using standard optical and electrical calibration techniques, it is possible to determine optical losses of these components. From this calibration, it is possible to determine the power level P k in the signal at wavelength ⁇ k from the measured output of photodiode 1.
  • Dl is the calibrated signal detected by photodiode I 9 P A is the optical power in signal k and RT ⁇ is the tunable filter transfer function at ⁇ k.
  • RT ⁇ is the tunable filter transfer function at ⁇ k.
  • the fixed filter 28 has a periodic characteristic as a function of wavelength, with a period (or free spectral range) equal to or approximately equal to twice the ITU standard wavelength (or frequency) spacing used in DWDM systems (typically 100, GHz or 50 GHz, but other spacings are possible).
  • Figure 4 shows a detail of the fixed filter transfer function (of Figure 3(c)) in the region of the signal in the input WDM or DWDM spectrum at wavelength ⁇ k .
  • the fixed filter transfer function at wavelength ⁇ k has a value RF k .
  • Also shown in Figure 4 are the locations of neighbouring WDM signals (at wavelengths ⁇ k-7, ⁇ k -i and X ⁇ 1 ), and their alignment with respect to the fixed filter transfer function.
  • the transfer function of the fixed filter element 28 (see Figure 2) (i.e. the attenuation of the filter as a function of wavelength) is known or has previously been measured. This measurement could be carried out when the monitor instrument is first constructed, and possibly repeated from time to time to ensure that calibration has not drifted.
  • An accurate (error-corrected) measure of the wavelength ⁇ k is determined from the measured input power level to the fixed filter 28 (e.g. from the measurement of Dl provided by photodiode 1 (see Figure 2) and a measurement of the output at photodiode 2 (see Figure 2) D2. The ratio of these two quantities provides a measure of the transfer function RF k of the fixed filter 28 (see Figure 2) at wavelength ⁇ *, and by comparing this measure of the transfer function with the known transfer function versus wavelength characteristic, the wavelength can be determined accurately.
  • the approximate value of ⁇ k is determined using the tunable filter element 22 (see Figure 2).
  • the fixed filter 28 has a periodic filter characteristic, such as in the example embodiment of Figures 3 and 4, it is preferable that the tunable filter 22 (see Figure 2) can locate ⁇ k approximately to within one half of a free spectral range of the fixed filter element 28 (see Figure 2).
  • the tunable filter 22 transfer function in Fig. 3(b) is a snapshot showing the tunable filter 22 tuned to one of many possible centre wavelengths (or frequencies).
  • the filter 22 is tuned so that the output of the photodetector 1 (see Figure 2) in response to the signal at wavelength ⁇ k is maximized.
  • the centre wavelength of the tunable filter 22 is scanned across the full DWDM spectrum to be measured.
  • scanning of the tunable filter 22 would be achieved by using a triangular or sawtooth waveform for the filter control signal V, but other waveforms are possible.
  • the outputs from photodiode 1 and photodiode 2 are sampled repeatedly in the digital signal processing circuit 26 (see Figure 2) and the waveforms of the optical signal into photodiode 1 and photodiode 2 (see Figure 2) are reconstructed. From the measured waveform of the signal at photodiode 1 (see Figure 2), and by identifying the peaks of the waveform, the approximate wavelength (or frequency) of the DWDM signals can be estimated.
  • the power of each of the n DWDM signals is obtained from the amplitude of the peaks of the photodiode 1 waveform, and, as explained above, the approximate wavelengths of the n DWDM signals are given by the location of the peaks on the photodiode 1 waveform and the known sweep waveform of the tunable filter 22 control signal (see
  • the ratio of the signals from the two photodiodes 1 and 2 provides a measure of the value of the transfer function RF k of the fixed filter 28 (see Figure 2) at wavelength ⁇ k , and from the known transfer function versus wavelength characteristic, the wavelength ⁇ k can be determined accurately. It will be appreciated that this procedure involves determining optical losses in the system of Fig. 2, including optical coupling losses and the responsivity or quantum efficiency of the two photodiodes 1 and 2, which is a straightforward procedure for a person skilled in the art.
  • the individual channels in the input WDM spectrum will have spectral widths that depend on a number of factors, including the modulation data rate. It will be appreciated that the wavelength ⁇ k obtained using the method described in the previous paragraphs is related to the mean or average wavelength of the signal in each channel.
  • the number of samples required to reconstruct the waveform from photodiode 1 and photodiode 2 depends on a number of factors, including the spectral width of each of the WDM optical signals, the passband bandwidth of the tunable filter 22, and the desired accuracy of the measurements. For example, in a system with DWDM channel spacings of 100 GHz, at least 10 samples are preferably required as the centre frequency of the tunable filter 22 sweeps between two channels. If the measurement range of the monitor covers 100 DWDM channels, at least 1,000 samples are preferably required across the entire measurement range.
  • Figure 5(c) shows the fixed filter 28 transfer function for another embodiment, in which the free spectral range of the fixed filter element 28 is substantially equal to the ITU standard wavelength spacing.
  • the characteristics in Figure 5(a) and Figure (b) are identical to the corresponding characteristics in Figure 3 (a) and Figure 3(b).
  • Figure 6 shows a detail of the fixed filter 28 transfer function for this embodiment, and the value RFk of the fixed filter 28 transfer function at wavelength ⁇ k.
  • the fixed filter 28 in preferred embodiments is its free spectral range, its finesse, and the wavelengths of the peaks of its transfer function.
  • a number of factors affect the choice of the free spectral range of the fixed filter 28. These include the need to ensure that the tunable filter 22 can approximately locate ⁇ k , to within one half of one free spectral range of the fixed filter 28, and the resolution of the fixed filter.
  • the fixed filter 28 is, however, not limited to fixed filters having a periodic filter characteristic. Rather, the present invention may be implemented using fixed filter elements having non-periodic characteristics, including e.g. fixed filters having a monotonic filter characteristic. It will be appreciated by a person skilled in the art that in such embodiments, the filter transfer function shape and, in the case of periodic filter functions, the free spectral range has to be balanced against variations in the accuracy of determining the signal wavelengths and power levels.
  • the finesse of the fixed filter element 28 in different embodiments of the invention can take a wide range of values. Typically, the finesse is of the order of 10, but higher and lower values are possible. A finesse of around 10 will generally provide a reasonable compromise between accuracy and the wavelength range over which the fixed filter 28 provides calibrated outputs.
  • the absolute value of the gradient of the fixed filter 28 transfer function as a function of wavelength in the wavelength region of the WDM channel to be measured is large. Therefore it is desirable that the wavelength of the channels in the WDM or DWDM spectrum to be measured are not aligned close to the peaks of the transfer function of the fixed filter 28, where the gradient is small, hi other words, it is preferable to ensure that no peaks of the fixed filter 28 transfer function are aligned close to ITU standard wavelengths to be measured. Note that in Figure 4 and Figure 6, the WDM channel wavelengths to be measured are aligned away from the peaks of the fixed filter 28 transfer fixation.
  • FIG. 7(c) shows the fixed filter 28 transfer function and the location of the main signal and two unwanted signals from adjacent channels, at the input to the fixed filter 28.
  • the free spectral range of the fixed filter 28 in the embodiment shown in Figure 7(c) is approximately twice the TTU standard channel spacing, but other values of free spectral range and other fixed filter transfer functions can be used.
  • a process of deconvolution could remove the above-mentioned errors in channel power and wavelength, using the known filter response of the tunable filtering element 22.
  • a deconvolution of the tunable filter 22 output is not capable of removing wavelength errors caused by non-linearities and instabilities in the filter tuning mechanism.
  • the present invention overcomes these limitations by performing a deconvolution that removes the unwanted crosstalk signals from the main signal at wavelength ⁇ k at the output of the tunable filter 22 and also at the output of the fixed filter 28. From the results of these two deconvolutions, and the known transfer function of the fixed filter 28, the wavelengths of all channels can be determined accurately. The method is described below in the case where two adjacent channels pass through the tunable filter in addition to the main signal.
  • Equation (2) shows only three wavelengths at a time passing through the tunable filter 22.
  • the deconvolution technique also works for more than three wavelengths. It also works for tunable filters 22 with a wide range of different transfer functions. Note again that the accuracy of the channel wavelengths resulting from this deconvolution will be limited by uncertainties associated with uncertainties in the V versus wavelength tuning characteristic of the tunable filter.
  • deconvolved data from the fixed filter 28 is used. This deconvolved data is obtained from the signal from photodiode 2, as described below.
  • Figure 8 shows a detail of the fixed filter 28 transfer function in Figure 7(c), together with the location of the main signal at wavelength ⁇ and two unwanted signals from adjacent channels at wavelengths ⁇ k -i and ⁇ k+ i.
  • the values of the fixed filter transfer function at wavelengths ⁇ k-i, ⁇ k, and ⁇ jt + ; are RF k -L RFk and RFk+i, respectively.
  • the total power D2 detected by photodiode 2 (sec Figure 2), and processed by the signal processing circuit 26 to calibrate out the responsivity of photodiode 2, losses in the optical splitter 11 in Figure 1, and losses in the monitor 20 in Figure 2, is
  • A P k . 1 RF k . 1 RT k . 1 (4)
  • the same deconvolution algorithm as is applied to the measured data Dl is also applied to measured data D2 (see Figure 2) for a range of centre wavelengths of the tunable filter 22.
  • a measure of RFk is determined for all values of k. Once RF k is known, ⁇ k can be determined for Jc-I, ...n, based on the known transfer function of the fixed filter 28.
  • Figure 9 is a signal flow graph showing the process of deconvolution and accurate measurement of the wavelengths ⁇ k and channel power levels P k .
  • the boxes labelled Dl and D2 represent the two photodiodes (see Figure 2) and the calibration process to account for optical losses.
  • the boxes labelled R 1 and R ⁇ represent deconvolution procedures or algorithms for the data from photodiode 1 and photodiode 2, respectively. In general, those two algorithms can be different, but in a preferred embodiment, they are identical.
  • the tuning voltage that controls the centre wavelength of the tunable filter RT( ⁇ , Fi) is V.
  • the output from the deconvolved Dl data is the channel powers, P ⁇ and the approximate channel wavelengths, ⁇ & where k ⁇ l,..n and the hat indicates approximate values of ⁇ *.
  • Figure 10 is a flowchart of the steps needed to make a complete measurement in a preferred embodiment.
  • Figure 10 shows that under control of the control signal V, the tunable filter 22 is scanned across all wavelengths to be measured. As the tunable filter 22 is scanned, the output of photodiode 1 and photodiode 2 are recorded at m different values of the filter control signal V.
  • the value of the number m will depend on the number of WDM channels to be measured and the required accuracy. It also depends on the finesse of the tunable filter 22. Choice of the value of m is also influenced by how much deconvolution is needed. If only two weak interfering signals are present (as shown in Fig.
  • the deconvolution process is quite "robust" and the number of samples necessary would generally be similar to the number of samples needed in the system characterised in Fig. 3. If ten interfering signals were present, for example, it would generally be preferable to increase the number of samples (i.e. reduce the wavelength of frequency spacing between channels) by a factor ranging roughly between 2 and 10. It will be appreciated by the person skilled in the art how to address these trade-offs. As explained earlier the value of m is typically at least ten times the number of channels n to be measured.
  • the recorded data from photodiode 1 and photodiode 2 are corrected (i.e. calibrated) to account for optical losses, to give the data sets Dl and D2.
  • These data i.e. Vk , Xk ,and PkRFb
  • Vk , Xk ,and PkRFb are then combined with the known transfer function of the fixed filter 28 to obtain error-corrected estimates of the channel power levels and wavelengths, as described above.
  • deconvolution techniques that can be used to perform the deconvolution functions R 1 and R 2 in Figure 9 and Figure 10.
  • These techniques include statistical methods such as least-squares and maximum likelihood estimation and can be implemented in a variety of forms such as finite impulse response and infinite impulse response filters, block deconvolution using matrix inversion, and techniques based on Fourier transforms.
  • Block methods collect all the data from photodiode 1 and photodiode 2 from a full sweep of the tunable filter across all WDM or DWDM channels and store these data in the memory of the digital signal processing system 26, where the data is subsequently processed as a block.
  • the filtering based methods the data from photodiode 1 and photodiode 2 is processed serially as the data is collected.
  • the use of these techniques will be familiar to a person who is skilled in the art.
  • Figure 11 shows the input WDM channels and the tunable filter 22 transfer function for an embodiment of the invention that uses a tunable filter 22 with a broader and more complex transfer function than the embodiments of Figure 3 and Figure 7.
  • Figure 1 l(a) shows the input WDM spectrum
  • Figure ll(b) shows the tunable filter 22 transfer function for two different values (VI and V2) of the tunable filter control signal V.
  • the tunable filter 22 response in Figure 11 is asymmetrical. It is important to note that the principles of the present invention, including the deconvolution process and the use of the fixed filter 28 to obtain error-corrected values of the channel wavelengths also apply to asymmetric alters of this type.
  • Figure 12 shows another embodiment of the invention, in which an additional photodiode, 3, is added.
  • An optical coupler 27 couples some of the optical output from the tunable filter 22 into photodiode 1.
  • the coupler 27 couples some of the optical power reflected from the input of the fixed filter 28 into photodiode 3.
  • the reflected power filter transfer function is plotted alongside the filter transmission transfer function in Figure 13(c). Note that when the transmission is at a peak, the reflection is at a minimum, and vice- versa
  • the graphs shown in Figure 13 (a) and 13(b) are identical to the graphs shown in Figure 7(a) and Figure 7(b) respectively.
  • the signal from photodiode 3 is subtracted from the signal from photodiode 2 before being sent to a controller and digital signal processing circuit 23.
  • FIG. 14 Another embodiment of the invention is shown in Figure 14.
  • the output from a tunable filtering element 40 is split into three separate paths using optical splitter 41 and optical splitter 42.
  • Optical splitter 41 directs part of the output signal from the tunable filtering element 40 to photodiode 1 and the other part of this signal to optical splitter 42, which splits the signal into two paths.
  • the two signals output by splitter 42 are then fed into two fixed optical filters 44, 46.
  • a variety of different means can be used to split the signal from the tunable filtering element 40 into three paths, including three-way optical splitters and
  • the output powers from the two fixed optical filters 44, 46 are detected by two photodiodes: photodiode 2A and photodiode 2B.
  • the two fixed optical filters 44, 46 have the same free spectral range, but the position of the peaks of the filter responses are offset so that the wavelengths of the peaks of the filter response of one of the filters fall approximately half way between the wavelengths of the peaks and the wavelengths of the minima of the filter response of the other of the filters. This ensures that the absolute value of the gradient of at least one of the fixed filters 44, 46 transfer functions as a function of wavelength is non-zero at all wavelengths.
  • the embodiment of Figure 14 therefore provides improved dynamic range and improves the wavelength range over which the fixed filters 44, 46 provide accurate wavelength calibration.
  • Figure 15 shows the filter transfer functions for the embodiment of Figure 14.
  • Figure 16 shows the signal flowgraph for data processing in the embodiment shown in
  • RFA k and RFB k are the transfer functions of fixed optical filter 44 and fixed optical filter 46, respectively, at wavelength ⁇ k .
  • the embodiment in Figure 14 requires three separate deconvolutions, one for data from each of the three photodiodes 1, 2A, and 2B.
  • the three deconvolution algorithms, R 1 , R 2 , and R 3 can be different, but in a preferred embodiment, the three deconvolution algorithms are all identical. It is noted that for the determination of individual WDM signal wavelengths, output signals from at least one of the two fixed optical filters 44, 46 is used, i.e. from at least one of the two photodiodes 2A and 2B.
  • Figure 17 shows an alternative embodiment of Figure 2, using free-space optics rather than fibre optics to interconnect the two filters.
  • an optical beam splitter 50 splits off some of the free space output from a tunable optical element 52 and directs it to photodiode 1.
  • the fixed optical filter 54 is tilted by a small amount away from the optical axis, and the reflected beam from the fixed filter element 54 does not enter the tunable optical filter 52, thereby reducing undesirable cavity effects between the two filter elements.
  • Figure 18 is an alternative embodiment of Figure 12, using free-space optical techniques.
  • reflected light from the input of the fixed filter 54 is partially reflected by a beam splitter 60 and is incident on photodiode 3.
  • the fixed optical filter 54 is tilted by a small amount away from the optical axis, and the reflected beam from the fixed filter element 54 does not enter the tunable optical filter 52.
  • Figure 19 is an alternative embodiment of Figure 14, using free-space optical techniques.
  • an additional optical splitter 70 partially reflects the free- space output from the tunable optical element 52 (after optical splitter 50).
  • the reflected portion is incident on a fixed optical filter B, where as the transmitted portion is incident on a fixed optical filter A.
  • Both the fixed optical A and the fixed optical filter B are tilted by a small amount away from the relevant optical axes, so that reflected signals do not enter the tunable optical filter 52.
  • the state of polarisation of the input light to the optical monitor 10 in Figure 1 will, in general, be random, in order to accurately measure the channel wavelengths and power levels, it is preferable that the optical monitor 10 is able to provide accurate measurements independent of the state of polarisation of the input signal.
  • Polarisation-independence can be achieved in a number of ways.
  • One method for obtaining polarisation independence is to use polarisation independent components in the monitor.
  • the monitor 10 will provide accurate polarisation independent data.
  • Low PDL is generally not difficult to achieve in fixed Fabry-Perot filters, and can be low in tunable filters such as fibre Fabry-Perot tunable filters and Microelectromechanical Systems (MEMS) tunable filters.
  • MEMS Microelectromechanical Systems
  • some tunable filters such as some optical planar waveguide devices and tunable filters using, linear anisotropic tuning media, such as liquid crystal Fabry-Perot tunable filters exhibit significant birefringence or polarisation dependence.
  • One technique for obtaining polarisation independence is to use a polarisation diversity technique.
  • the input signal is passed through a polarising beamsplitter and the two polarised outputs of the beamsplitter are fed into two nominally identical polarisation sensitive monitors.
  • the total monitor output is obtained by adding the outputs of both monitors.
  • a disadvantage of the polarisation diversity technique is that it requires two optical monitors.
  • Figure 20 shows an embodiment of the invention for use when a tunable filter is 80 is polarisation-sensitive.
  • This embodiment does not require two monitors.
  • a linear polariser 82 is placed at the input of the tunable filter.
  • the axis of the polariser is aligned with the preferred input polarisation of the tunable filter 80. It is noted that if the tunable filter 80 provides a polarising function, the polariser is not necessary.
  • the input to the monitor passes through a controllable waveplate or retarder 84 and is incident on the polariser 82 and' tunable filter 80.
  • the optic axis of the controllable waveplate 84 is set at 45 degrees to the orientation of the polariser 82.
  • a control signal from a controller 86 of the monitor is used to adjust the retardance of the retarder 84.
  • the controller 86 switches the retarder 84 repetitively between zero retardance and half-wave retardance.
  • the retardance is zero
  • the input polarisation is unchanged.
  • the retardance is half-wave
  • the vertical component of the input polarisation is switched to horizontal and vice versa.
  • the optical powers PZj to PZ n are determined for the wave plate 84 set to zero retardance and the optical powers PH 1 to PH 2 are determined for the wave plate 84 set to half-wave.
  • the total (polarisation independent) channel powers Pi to P n are obtained by adding the two sets of measurements,
  • the error-corrected channel wavelengths are obtained using the techniques described earlier for other embodiments of the invention.
  • the calibrated data from the photodiodes arc obtained with the wave plate 84 set to zero retardance and then set to half-wave retardance.
  • the two sets of calibrated data for the two retardances are then added for each photodiode before application of the deconvolution algorithm.

Landscapes

  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Optical Communication System (AREA)
  • Spectrometry And Color Measurement (AREA)

Abstract

L'invention concerne un procédé de surveillance de canaux ou de signaux individuels contenus dans un spectre optique WDM. Ledit procédé consiste à filtrer le spectre optique WDM au moyen d'un ensemble filtre accordable, à filtrer la sortie de l'ensemble filtre accordable au moyen d'un ensemble filtre fixe présentant une fonction de transfert connue en fonction de la longueur d'onde, à mesurer la puissance optique à la sortie de l'ensemble filtre accordable dans différentes conditions d'accord de l'ensemble filtre accordable, à mesurer la puissance optique correspondante à la sortie de l'ensemble filtre fixe dans lesdites différentes conditions d'accord et à déterminer le niveau de puissance optique pour chaque signal WDM et la longueur d'onde à erreurs corrigées de chaque signal WDM à partir d'une analyse de la puissance optique mesurée à la sortie de l'ensemble filtre accordable et de la puissance optique correspondante à la sortie de l'ensemble filtre fixe dans lesdites différentes conditions d'accord.
PCT/AU2006/000762 2005-06-03 2006-06-02 Surveillance de performance optique wdm WO2006128254A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
AU2005902897 2005-06-03
AU2005902897A AU2005902897A0 (en) 2005-06-03 WDM optical performance monitoring

Publications (1)

Publication Number Publication Date
WO2006128254A1 true WO2006128254A1 (fr) 2006-12-07

Family

ID=37481159

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/AU2006/000762 WO2006128254A1 (fr) 2005-06-03 2006-06-02 Surveillance de performance optique wdm

Country Status (1)

Country Link
WO (1) WO2006128254A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010022327A2 (fr) * 2008-08-21 2010-02-25 Nistica, Inc. Moniteur de canal optique
CN109687901A (zh) * 2017-10-19 2019-04-26 福州高意通讯有限公司 一种光性能监测器
WO2020011175A1 (fr) * 2018-07-09 2020-01-16 中兴通讯股份有限公司 Procédé et dispositif de détermination d'informations de longueur d'onde d'un signal optique, et support de stockage
IT201900020554A1 (it) * 2019-11-07 2021-05-07 Milano Politecnico Sistema ottico comprendente un dispositivo riconfigurabile e metodo di controllo del sistema ottico

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001015349A1 (fr) * 1999-08-20 2001-03-01 Anritsu Company Systeme de surveillance pour multiplexeur en longueur d'onde dense (dwdm)
US6262822B1 (en) * 1998-05-13 2001-07-17 Jds Fitel Inc. Circuit for monitoring optical signals
WO2003017533A2 (fr) * 2001-08-15 2003-02-27 Optinel Systems, Inc. Dispositif de surveillance de canal optique utilisant de multiples bandes passantes a filtre fabry-perot
US20030112836A1 (en) * 2001-12-18 2003-06-19 Kim Bong Kyu Multiwavelength locking method and apparatus using acousto-optic tunable filter
JP2004132704A (ja) * 2002-10-08 2004-04-30 Sun Tec Kk 波長モニタ及びその基準値設定方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6262822B1 (en) * 1998-05-13 2001-07-17 Jds Fitel Inc. Circuit for monitoring optical signals
WO2001015349A1 (fr) * 1999-08-20 2001-03-01 Anritsu Company Systeme de surveillance pour multiplexeur en longueur d'onde dense (dwdm)
WO2003017533A2 (fr) * 2001-08-15 2003-02-27 Optinel Systems, Inc. Dispositif de surveillance de canal optique utilisant de multiples bandes passantes a filtre fabry-perot
US20030112836A1 (en) * 2001-12-18 2003-06-19 Kim Bong Kyu Multiwavelength locking method and apparatus using acousto-optic tunable filter
JP2004132704A (ja) * 2002-10-08 2004-04-30 Sun Tec Kk 波長モニタ及びその基準値設定方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
PATENT ABSTRACTS OF JAPAN *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010022327A2 (fr) * 2008-08-21 2010-02-25 Nistica, Inc. Moniteur de canal optique
WO2010022327A3 (fr) * 2008-08-21 2010-04-15 Nistica, Inc. Moniteur de canal optique
CN102150385A (zh) * 2008-08-21 2011-08-10 尼斯迪卡有限公司 光信道监控器
US9031404B2 (en) 2008-08-21 2015-05-12 Nistica, Inc. Optical channel monitor
CN109687901A (zh) * 2017-10-19 2019-04-26 福州高意通讯有限公司 一种光性能监测器
WO2020011175A1 (fr) * 2018-07-09 2020-01-16 中兴通讯股份有限公司 Procédé et dispositif de détermination d'informations de longueur d'onde d'un signal optique, et support de stockage
IT201900020554A1 (it) * 2019-11-07 2021-05-07 Milano Politecnico Sistema ottico comprendente un dispositivo riconfigurabile e metodo di controllo del sistema ottico
WO2021090205A1 (fr) * 2019-11-07 2021-05-14 Politecnico Di Milano Système optique comprenant un dispositif reconfigurable et procédé de commande du système optique
US11646790B2 (en) 2019-11-07 2023-05-09 Politecnico Di Milano Optical system comprising a reconfigurable device and optical system control method

Similar Documents

Publication Publication Date Title
US7756369B2 (en) OSNR monitoring apparatus and method using polarization splitting
US8345238B2 (en) Measuring optical spectral property of light based on polarization analysis
US5995255A (en) Concatenated fiber grating optical monitor
US5969834A (en) Method and apparatus for monitoring wavelength multiplexed optical signals
EP1432150A2 (fr) Procédé de surveillance du rapport signal sur bruit optique
US20070189776A1 (en) Polarization mitigation technique
US6104492A (en) Optical signal monitor for multiwave optical signals
US7589840B2 (en) Broad- and inter-band multi-wavelength-reference method and apparatus for wavelength measurement or monitoring systems
CN110383714B (zh) 使用频谱相关性对已调制信号的频谱形状进行无噪声测量
EP1972083A2 (fr) Système de mesure des signaux optiques
US20030161631A1 (en) Optical channel monitor device and method
US8160443B2 (en) Calibration factor for interferometric optical signal-to-noise ratio measurement
EP2486677B1 (fr) Système et procédé de mesure du rapport signal optique-bruit
WO2006128254A1 (fr) Surveillance de performance optique wdm
US8731400B2 (en) Characterizing a frequency channel in an optical network
US8160442B2 (en) Interferometric optical signal-to-noise ratio measurement using a calibration factor
US6624889B1 (en) Cascaded filter employing an AOTF and narrowband birefringent filters
US6614527B2 (en) Spectral bandwidth calibration of an optical spectrum analyzer
US10422721B2 (en) Measurement system and method to interrogate birefringent optical sensors with a frequency swept source based interrogator
Roudas et al. Coherent frequency-selective polarimeter for polarization-mode dispersion monitoring
EP1936841A2 (fr) Appareil de surveillance RSB optique et procédé d'utilisation de séparation de polarisation
Boroditsky et al. Technique for in-situ measurements of polarization mode dispersion
US9013694B2 (en) System and method for measuring a wavelength-resolved state of polarization of an optical signal
Zalevsky et al. Communication Networks: Performance Monitoring
Audet et al. OSNR in Next-Gen Networks: Polarization-Resolved Optical Spectrum Analysis Allows Fast and Accurate In-Band OSNR Measurement

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

WWW Wipo information: withdrawn in national office

Country of ref document: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06741179

Country of ref document: EP

Kind code of ref document: A1