Optics: measuring and testing – By light interference – Spectroscopy
Reexamination Certificate
1999-02-23
2001-03-20
Font, Frank G. (Department: 2877)
Optics: measuring and testing
By light interference
Spectroscopy
C356S491000, C356S073100, C356S364000
Reexamination Certificate
active
06204924
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to a method and apparatus for measuring polarization mode dispersion of optical devices, especially, but not exclusively, components of optical telecommunications systems, including optical fibers, optical isolators, couplers, light amplifiers, and dispersion compensators.
BACKGROUND
Polarization mode dispersion (PMD) limits the performance of optical telecommunications systems by reducing bandwidth in digital telecommunications systems and contributing to distortion in analogue telecommunications systems. Techniques have been developed for measuring PMD so that it can be reduced or, if that is not possible, measures can be taken to mitigate its effects. The characterization or measurement of polarization mode dispersion in fiber optic telecommunications systems has become of great importance with the advent of high bit rate, long-haul telecommunications links. Optical components in such systems must be tested to verify that they do not add significant PMD. Measurements must be made in the laboratory environment during component and subsystem design, in the production environment when the components are being fabricated, and in the outside plant environment when the system has been installed.
The practical range over which PMD may need to be characterized now extends over five orders of magnitude from about 1 femtosecond to nearly 100 picoseconds. Moreover, in contemporary wavelength-division multiplex (WDM) systems, measurements must frequently be made through interposed, narrow path optical filters, and the PMD characterized for weak, intermediate and strong mode coupling. When the measurements must be made in the production environment or in the field, speed, robustness and portability are important factors.
Generally, to characterize the PMD, it is required to determine the state of polarization of the light at the output of the device under test. There are several standardized methods for measuring polarization mode dispersion, which can be classified into two categories according to the kind of data, i.e. time-domain or frequency-domain, that is collected. Thus, time-domain data is collected by the Interferometric method (INT). Frequency-domain data is collected by the fixed analyzer method (FA), also known as wavelength scanning, and by the polarimetric method, that is Jones Matrix Eigenanalysis (JME) method or the Poincaré Sphere (PS) method.
The following ANSI standards apply:
TIA/EIA-455-124 for the Interferometric method
TIA/EIA-455-122 for the Jones Matrix Eigenanalysis method
TIA/EIA-455-113 for the Fixed Analyzer method.
The interferometric measurement methods typically pass polarized light from a broadband light source through the device under test and to an interferometer, for example a Michelson interferometer or a Mach-Zehnder interferometer. For measurements in the field, the interferometric method usually is preferred because it provides measurements quickly and uses comparatively light apparatus. INT also gives the more reliable results in the important case of long installed fibres, owing to its insensitivity to fast fluctuations of the output state of polarization.
Disadvantages of previously-known interference methods include limitation of the measurement range to PMD values larger than about 150 femtoseconds, especially in the strong mode-coupling regime characteristic of most optical fibre measurements. Another disadvantage is that the measurement is sensitive to the shape of the spectrum transmitted through the device under test, so results are affected whenever some kind of filter is present in the path. Yet another disadvantage is that measurement is virtually impossible in the case of optical components which have limited spectral bandwidth.
Techniques have been proposed to improve the measurement of low PMD. For example, in U.S. Pat. No. 5,654,793 issued August 1997, A. J. Barlow et al. disclosed a “PMD-biasing” technique for measuring low PMD using a birefringent artefact with a stable PMD value in series with the fiber under test. The interferogram is biased away from the central autocorrelation peak and PMD is obtained by measuring the broadening of the peak. While lower PMD values may be measured in this way, ultra-low values, i.e. less than 10 femtoseconds, cannot.
What may be considered an improvement over Barlow's technique was disclosed in an article entitled “Interferometric Polarization Mode Dispersion Measurements with Femto Second Sensitivity” by T. Oberson et al., Journal of Light Wave Technology, 1997. Oberson et al. discussed standard techniques for PMD measurements using the Michelson interferometer and an envelope detector, the PMD delay being deduced from the width of the interferogram and concluded that such standard techniques were not suitable where the polarization mode delay is smaller than, or comparable to, the coherence time of the source. According to Oberson et al., a PMD of 104 femtoseconds represented the lower limit of PMD measurable using standard interferometric techniques. To extend the measurement range, Oberson et al. proposed modifying the standard set-up by inserting a high birefringent (HiBi) fiber with a PMD of about 0.5 picoseconds between the fibre under test and the analyzer. The HiBi fibre produced two side peaks, one each side of the central autocorrelation peak. The differential group delay (DGD) was determined from the extremes of separation between the side peaks.
While Oberson et al.'s technique may measure PMD values in the range of 10 femtoseconds, it does not address the problem of measurements being sensitive to the shape of the spectrum transferred to the device under test or the measurement of PMD of optical components with limited spectral bandwidth. Other limitations include a lengthy measurement time, application to weak mode-coupling only, and a requirement for a polarization controller in the light path which can introduce residual PMD which randomly adds to, or subtracts from, the measured PMD.
The Fixed Analyzer method makes a plurality of measurements at different wavelengths and analyzes the measured spectrum by counting extreme. Measurements at the different wavelengths may be obtained by varying the wavelength of the input light source, for example a tunable laser, or using a broadband source with a monochromator. The light from the source is polarized and passed through the device under test to a fixed analyzer and then to a photodetector. Alternatively, a broadband source could be used and the output from the polarizer/analyzer analyzed using an optical spectrum analyzer. The resulting spectrum exhibits a multiplicity of maxima and minima because the state of polarization at the output of the device under test, and hence at the input of the analyzer, changes with wavelength. The PMD is estimated by averaging the number of maxima and minima. While this might be satisfactory where the PMD is relatively large, it is not satisfactory for low PMD because the state of polarization does not change very much and there will be very few, perhaps only one or two, maxima and minima to average. Thus, its main disadvantage is that it can measure only PMD values which are much larger than the inverse of the source spectral width. For example, with one 1550 nm LED (20 THz width), the mean number of extrema is approximately 1 when PMD is equal to 25 fs.
The polarimetric methods use a laser or other narrowband light source that can be tuned across the range of wavelengths to be measured. The Poincaré sphere method, for example, then uses a polarimeter which measures state of polarization directly but at only one wavelength at a time. In addition to requiring an expensive tunable laser, this approach is time-consuming because the measurements must be repeated at the different wavelengths. Moreover, the scanning range of a tunable laser typically is about 100 nm, which is very restricted compared with that provided by a broadband source.
The Jones Matrix Eigenanalysis (JME) also uses a tunable laser, the output of which is passed through a pola
Adams Thomas
EXFO Electro-optical Engineering Inc.
Font Frank G.
Lee Andrew H.
LandOfFree
Method and apparatus for measuring polarization mode... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for measuring polarization mode..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for measuring polarization mode... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2444610