Abstract
A method for determining the position of an irregularity in an optical transmission fiber using an optical time domain reflectometer, the method comprising the steps of emitting a succession of sampling light pulses into the optical transmission fiber, detecting reflected light pulses resulting from the reflection of the sampling light pulses at the irregularity in the optical transmission fiber and generating corresponding time-dependent detection signals, wherein different delays are associated with detection signals corresponding to different sampling light pulses, obtaining a combined signal from the detection signals, and analyzing the combined signal for determining the position of the irregularity in the optical transmission fiber with respect to the optical time domain reflectometer, wherein the combined signal corresponds to a superposition of the detection signals.
Claims
1. A method for determining a position of an irregularity in an optical transmission fiber using an optical time domain reflectometer, the method comprising the steps of: emitting a succession of sampling light pulses into the optical transmission fiber, wherein the sampling light pulses have a predefined pulse width; detecting each of a plurality of reflected light pulses that each result from a reflection of a respective one of the sampling light pulses that occurs at the irregularity in the optical transmission fiber, generating a plurality of detection signals that each correspond to a respective one of the plurality of detected reflected light pulses, wherein each detection signal is active for a time corresponding to the predefined pulse width; obtaining a combined signal from the plurality of detection signals; and analyzing the combined signal to determine the position of the irregularity in the optical transmission fiber with respect to the optical time domain reflectometer based on a time shift between the combined signal and a calibration signal; wherein the combined signal corresponds to a superposition of the plurality of detection signals; wherein the generation of each of the plurality of detection signals is time-dependent relative to a timescale; wherein a basis for the timescale comprises at least one of: a clock signal comprising a regular clock signal cycle, at least one trigger point; wherein at least one of the plurality of detection signals is, relative to the basis of the timescale, delayed by an amount of delay that differs from an amount by which at least one other of the plurality of detection signals is delayed relative to the basis of the timescale; wherein the calibration signal is obtained by the steps of: emitting into the optical transmission fiber a succession of calibration light pulses; detecting a plurality of reflected calibration light pulses that each result from a reflection of a respective one of the emitted calibration light pulses that occurs at a test irregularity located at a known distance from the optical time domain reflectometer; generating a plurality of calibration detection signals that each correspond to a respective one of the plurality of detected reflected calibration light pulses; generating the calibration signal using a superposition of the plurality of calibration detection signals.
2. The method of claim 1, wherein the test irregularity is a connection port of the optical time domain reflectometer configured for connecting the optical time domain reflectometer to an optical transmission fiber.
3. The method of claim 2, wherein the step of emitting the succession of calibration pulses is carried out when starting operation of the optical time reflectometer.
4. The method of claim 1, wherein the step of analyzing the combined signal further comprises normalizing an amplitude of the combined signal so that it has a same amplitude as the calibration signal.
5. The method of claim 4, wherein the time shift between the combined signal and the calibration signal is determined at a signal point corresponding to one of: the respective trailing edge of each of the combined signal and the calibration signal; or the respective leading edge of each of the combined signal and the calibration signal.
6. The method of claim 1, wherein the time shift between the combined signal and the calibration signal is determined at a signal point at which an amplitude of the combined signal and an amplitude of the calibration signal both have a predetermined value between 10% and 90% of a maximum signal amplitude.
7. The method of claim 1, wherein the step of analyzing the combined signal comprises fitting the combined signal and the calibration signal to an analytic function.
8. The method of claim 7, wherein the analytic function is a polynomial function.
9. The method of claim 1, wherein the step of analyzing the combined signal comprises calculating any of the following parameters: Surface:
Σ.sub.i=1.sup.Nc.sub.i or ∫.sub.−∞.sup.∞c(t)dt; Pulse center: Symmetry parameter: wherein c.sub.i stands for the combination signal at a time t.sub.i, wherein N stands for a number of sampling values.
10. The method of claim 1, wherein the sampling light pulses have a predefined pulse width.
11. The method of claim 10, wherein the predefined pulse width of the sampling light pulses is between 1 μs and 1 ms.
12. The method of claim 1, wherein each of the plurality of detection signals is, relative to the basis of the timescale, delayed by an amount of delay that differs from amounts by which remaining ones of the plurality of detection signals are delayed relative to the basis of the timescale, and wherein the amounts by which the plurality of detection signals are delayed relative to the basis of the timescale are each a different integer multiple of a predetermined time increment among a plurality of integer multiples.
13. The method of claim 12, wherein the respective delay of each of the plurality of detection signals is not, relative to delays of previously generated ones of the plurality of detection signals, monotonously increasing or decreasing.
14. The method of claim 12, wherein the predetermined time increment is between 100 ns and 75 μs.
15. The method of claim 1, wherein each of the plurality of reflected light pulses is detected using a predefined sampling period.
16. The method of claim 15, wherein the predefined sampling period is between 100 ns and 75 μs.
17. The method of claim 1, wherein each of the plurality of detection signals is, relative to the basis of the timescale, delayed by a respective time increment that has no integer multiple relationship to respective time increments by which remaining ones of the plurality of detection signals are delayed relative to the basis of the timescale.
18. The method of claim 1, wherein the step of emitting the succession of sampling light pulses comprises emitting a predetermined first number of sampling light pulses each with a first same delay and emitting a predetermined second number of sampling light pulses each with a second same delay that is different than the first same delay, wherein the step of generating the plurality of detection signals comprises generating a first plurality of detection signals that each correspond to a respective one of a first plurality of reflected light pulses that is detected as a result of a reflection of one of the predetermined first number of sampling light pulses, wherein the step of generating the plurality of detection signals further comprises generating a second plurality of detection signals that each correspond to a respective one of a second plurality of reflected light pulses that is detected as a result of a reflection of one of the predetermined second number of sampling light pulses, and further comprising the step of generating a first average detection signal by averaging over the first plurality of detection signals and generating a second average detection signal by averaging over the second plurality of detection signals, and wherein the combined signal corresponds to a superposition of the first and second average detection signals.
19. The method of claim 1, wherein the basis for the timescale is the at least one trigger point.
20. The method of claim 1, wherein the basis for the timescale is the clock signal.
21. An optical time domain reflectometer for determining a position of an irregularity in an optical transmission fiber, the optical time domain reflectometer comprising: a light source configured for emitting a succession of sampling light pulses into the optical transmission fiber, wherein the sampling light pulses have a predefined pulse width; a light receiver configured for detecting each of a plurality of reflected light pulses that each result from a reflection of a respective one of the sampling light pulses that occurs at the irregularity in the optical transmission fiber and for generating a plurality of detection signals that each correspond to a respective one of the detected plurality of reflected light pulses, wherein each detection signal is active for a time corresponding to the predefined pulse width; a processing unit operatively connected to the light receiver and configured for obtaining a combined signal from the plurality of detection signals; an analyzing unit operatively connected to the processing unit and configured for analyzing the combined signal to determine the position of the irregularity in the optical transmission fiber with respect to the optical time domain reflectometer based on a time shift between the combined signal and a calibration signal; wherein the processing unit is further configured for obtaining the combined signal from a superposition of the plurality of detection signals; wherein the generation by the light receiver of each of the plurality of detection signals is time-dependent relative to a timescale; wherein a basis for the timescale comprises at least one of: a clock signal comprising a regular clock signal cycle, at least one trigger point; wherein at least one of the plurality of detection signals is delayed, relative to the basis of the timescale, by an amount of delay that differs from an amount by which at least one other of the plurality of detection signals is delayed relative to the basis of the timescale; wherein the calibration signal is obtained by the steps of: emitting into the optical transmission fiber a succession of calibration light pulses; detecting a plurality of reflected calibration light pulses that each result from a reflection of a respective one of the emitted calibration light pulses that occurs at a test irregularity located at a known distance from the optical time domain reflectometer; generating a plurality of calibration detection signals that each correspond to a respective one of the plurality of detected reflected calibration light pulses; generating the calibration signal using a superposition of the plurality of calibration detection signals.
22. The optical time domain reflectometer of claim 21, further comprising a control unit operatively connected to the light source and configured for controlling the light source.
23. The optical time domain reflectometer of claim 22, wherein at least one of the processing unit and the control unit comprise at least one of an analog/digital converter and a digital/analog converter.
24. The optical time domain reflectometer of claim 22, wherein an integrated device comprises the processing unit, the analyzing unit, and the control unit.
25. The optical time domain reflectometer of claim 21, further comprising an optical component assembly and a connector, wherein the connector is configured for connecting the optical time domain reflectometer to an optical transmission fiber and wherein the optical component assembly is configured for: directing, towards the connector, light pulses coming from the light source; and directing, towards the light receiver, light pulses coming from the connector.
26. The optical time domain reflectometer of claim 21, wherein the basis for the timescale is the clock signal.
27. The optical time domain reflectometer of claim 21, wherein the basis for the timescale is the at least one trigger point.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) FIG. 1 shows a schematic representation of a conventional method for determining the position of an irregularity in an optical transmission fiber using an OTDR.
(2) FIG. 2 shows a schematic representation of a succession of sampling light pulses emitted into the optical transmission fiber with different delays relative to a clock signal: a. shows a succession of sampling light pulses having monotonously increasing delays; b. shows a succession of sampling light pulses having delays which are not monotonously increasing or decreasing.
(3) FIG. 3 shows a schematic representation of time-dependent detection signals resulting from a succession of sampling light pulses and the effects of jitter: a. shows detection signals corresponding to a succession of ideal sampling light pulses having monotonously increasing delays in the absence of jitter; b. shows detection signals corresponding to a succession of sampling light pulses having monotonously increasing delays and the effect of jitter; c. shows detection signals corresponding to a succession of ideal sampling light pulses having disorderedly increasing delays and the cancellation of the effect of jitter.
(4) FIG. 4 shows a schematic representation of time-dependent detection signals resulting from the succession of sampling light pulses of FIG. 2 and the obtained combined signal, according to an embodiment of the invention. The effects of fiber attenuation and of different reflectivities of irregularities are not taken into account: a. refers to a first (test) irregularity; b. refers to a second irregularity.
(5) FIG. 5 shows a schematic representation of the calibration signal and the combined signal of FIG. 4.
(6) FIG. 6 compares the impact of distortions of the pulse shape on the precision of the determined position of the irregularity for the conventional method and for the method of the invention.
(7) FIG. 7 shows the effects of fiber attenuation on the combined signal: a. shows the combined signals as resulting from the superposition; b. shows the combined signals normalized to have a maximum amplitude of 1.
(8) FIG. 8 shows normalized combined signals obtained for irregularities located at different positions of the optical transmission fiber.
(9) FIG. 9 shows plots of different optical parameters of the combined signals of FIG. 5 a. shows the pulse center of the combined signals of FIG. 5 as a function of the distance between the irregularity and the OTDR. b. shows the symmetry parameter of the combined signals of FIG. 5 as a function of the distance between the irregularity and the OTDR. c. shows a plot of pulse center against symmetry parameter for the combined signals of FIG. 5.
(10) FIG. 10 shows the effects of different shapes of the sampling light pulses on the resulting combined signal.
(11) FIG. 11 shows a comparison of the accuracy requirements of a conventional OTDR to those of the invention.
(12) FIG. 12 shows different techniques for associating delays to the detection signals.
(13) FIG. 13 shows a schematic representation of an OTDR according to an embodiment of the invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
(14) For the purposes of promoting an understanding of the principles of the invention, reference will now be made to a preferred embodiment illustrated in the drawings, and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated apparatus and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur now or in the future to one skilled in the art to which the invention relates.
(15) FIG. 1 shows a schematic representation of a conventional method for determining the position of an irregularity in an optical transmission fiber using an OTDR commonly known in the art. The figure shows in the upper row the time evolution of a light pulse emitted by the OTDR and in the lower row the reflected light pulse resulting from the reflection at a distance x from the OTDR of the light pulse of the upper row detected by the OTDR with a predefined sampling period ΔT. Also shown are the corresponding sampling points T1 to T3. As shown in the figure, where both the upper row and the lower row share a common time scale on the horizontal axis, the reflected light pulse is detected in the OTDR after a time 2x/c measured from the time of emission, where c stands for the speed of light within the fiber. If the sampling period is denoted by ΔT the receiver is able to take samples every ΔT, which corresponds to the distance between T1 and T2, and between T2 and T3. The position of the irregularity can be determined by measuring the time elapsed between the emission of the sampling light pulse and the detection of the reflected light pulse. Assuming that the reflected light pulse of the figure results from a sampling light pulse which has been reflected almost completely by the corresponding irregularity, this interval of time might, for instance, be measured between the emission of the leading edge of the sampling pulse and the detection of the leading edge of the reflected pulse. However, it can only be concluded that the leading edge of the pulse is detected at some time between the first sampling point T1 and the second sampling point T2. This means that the position x of the irregularity can be determined with a precision Δx=c.Math.ΔT/2 only. For example, for a sampling period ΔT=1 μs and a speed of light in the fiber c=2.Math.10.sup.8 m/s, the position of the irregularity can be determined with a precision Δx=100 m only.
(16) FIG. 2 shows a schematic representation of a succession of sampling light pulses s.sub.1 to s.sub.5 emitted into the optical transmission fiber with different delays relative to a clock signal, such that the corresponding detection signals are associated with such different delays relative to the clock signal. The clock signal provides a regular and stable timescale which allows referencing the delays to evenly spaced clock signal points φ.sub.1 to φ.sub.5, separated by regular clock signal cycles Δφ. The sampling light pulses s.sub.1 to s.sub.5 are timely ordered, such that one pulse is emitted at a time. All pulses have a predefined pulse width W. The different delays of the sampling light pulses with respect to the clock signal differ by multiples of a predetermined time increment Δt. FIG. 2a shows a succession of sampling light pulses having monotonously increasing delays with respect to the clock signal. A first sampling light pulse s.sub.1 is emitted with a delay Δt with respect to the clock signal point φ.sub.1, that is at a time T=φ.sub.1+Δt. A second sampling light pulse s.sub.2 is emitted with a delay 2Δt with respect to the clock signal point φ.sub.2, that is at a time T=φ.sub.2+2.Math.Δt, namely a time Δφ+Δt later, and so on. In this exemplary embodiment, the pulse width W is chosen to be smaller than the corresponding clock signal cycle Δφ such that the individual sampling light pulses do not overlap in time. Hence, each of the sampling light pulses s.sub.1 to s.sub.5 is characterised by the corresponding delay with respect to the clock signal.
(17) FIG. 2b shows a succession of sampling light pulses emitted into the optical transmission fiber with disordered delays relative to the clock signal corresponding to multiple integers of the predetermined time increment Δt. Hence, the corresponding detection signals will have delays associated therewith that correspond to integer multiples of the predetermined time increment Δt which are not monotonously increasing or decreasing. By associating delays to detection signals corresponding to a succession of sampling light pulses in such a disordered way, the effects of jitter and wander are circumvented.
(18) FIG. 3 illustrates the effects of jitter and wander on the combined signal obtained detection signals resulting, for example, from a succession of sampling light pulses emitted in the way shown in FIGS. 2a and 2b. FIG. 3a shows a succession of detection signals associated with continuously increasing delays corresponding to continuously increasing integer multiples of the predetermined time increment Δt with respect to an ideal clock signal that is not affected by jitter or wander. FIG. 3b shows the effects of jitter or wander on the succession of detection signals of FIG. 3a. Small black rectangles indicate the timely variation in the start time of the detection signals introduced by jitter or wander. In addition to affecting the aforementioned start time, jitter and wander also effect the width of the pulses. As seen in the figure, the jitter period is in this case such that the three first detection signals of the succession are retarded with respect to the ideal foreseen delay due to jitter, whereas the three last detection signals of the succession are advanced with respect to the ideal foreseen delay due to jitter. As a consequence, the maximum of the triangular function fitted to the combined signal obtained from the detection signals of FIG. 3b is shifted with respect to the maximum of the triangular function fitted to the combined signal obtained from the detection signals of FIG. 3a, which corresponds to the ideal case in the absence of jitter or wander. As a result, a smaller delay is measured for the combined signal and hence the irregularity is assumed to be closer to the OTDR than it actually is.
(19) FIG. 3c shows the combined signal obtained if the same detection signals of FIGS. 3a and 3b are associated with delays corresponding to integer multiples of the predetermined time increment in a disordered way, according to the pattern shown in FIG. 2b. As seen in FIG. 3c, the shift in the triangular function fitted to the combined signal due to jitter is significantly reduced in this case, such that the final shape is very similar to that of the triangular function fitted to the combined signal of FIG. 3a. The use of a larger number of pulses may even result to the shift due to jitter and wander being averaged out.
(20) A large number of techniques are known to disorderly associate delays corresponding to integer multiples of a predetermined time increment to a succession of signals. In the following, a solution that can be implemented very easily is explained. Assuming that N pulses have to be generated, their delay D can be determined using the equation D=[mod(i.Math.M, N)+1].Math.Δt, where i is an index identifying the pulse and ranging from 1 to N and Δt is the predetermined time increment. The operation mod denotes the modulo operator as used in modular arithmetic. In detail, the modulo operation r=mod(a,b) is defined in such a way, that the remainder r and the input variables a and b satisfy the condition α=n.Math.b+r with n representing a non-negative integer and o≤r<b. Furthermore, M stands for an integer smaller than N, where M and N do not have any common divisor other than 1. Without affecting the functionality, the term +1 can be omitted in this equation. The table below shows the distribution of the delays for an example with N=8 and M=5.
(21) TABLE-US-00001 index delay i = 1 6 .Math. Δt i = 2 3 .Math. Δt i = 3 8 .Math. Δt i = 4 5 .Math. Δt i = 5 2 .Math. Δt i = 6 7 .Math. Δt i = 7 4 .Math. Δt i = 8 Δt
(22) The numbers above have been exemplary chosen for illustrating the technique. In a preferred embodiment of the invention, N equals 200 and M is equal to 164.
(23) FIG. 4a shows a schematic representation of time-dependent detection signals d.sub.1 to d.sub.5, resulting from the succession of sampling light pulses of FIG. 2. The obtained combined signal C is also shown. Again, for simplicity, ideal pulses are considered, assuming instantaneous pulse increments and complete reflection, and neglecting the effects of attenuation. As in FIG. 3, the horizontal axis corresponds to a clock signal time scale t with respect to which the delays are determined, not to be confused with the absolute time scale T of FIGS. 1 to 2. The detection signal di corresponds to the sampling light pulse s.sub.i emitted with a delay relative to the clock signal t.sub.i=(i−1).Math.Δt, with i=1, . . . , 5. The sampling light pulse s.sub.i is reflected back at an irregularity resulting in a reflected light pulse r.sub.i. The reflected light pulse r.sub.i starts being detected at a time t.sub.i+τ, where τ is the time elapsed between the emission of the sampling light pulse s.sub.i and the detection, for example, of the leading edge of the reflected light pulse r.sub.i measured in the scale of clock signal time t. τ is proportional to the distance between the irregularity and the OTDR. The detection signal d.sub.i resulting from the reflected light pulse r.sub.i is active for a time W, that is between times t=t.sub.i+τ and t=t.sub.i+τ+W.
(24) The resulting combined signal C corresponds to a superposition of the detection signals d.sub.1 to d.sub.5 and is shown in the bottom diagram. Also shown in the figure is a triangular function to which the combined signal has been fitted.
(25) The time dependent detection signals d.sub.1 to d.sub.5 of FIG. 4a could result from the reflection of sampling light pulses at a test irregularity located at a known distance from the OTDR, corresponding for example to a connection port of the OTDR configured for connecting the OTDR to an optical transmission fiber. In that case, τ would be proportional to the distance between the test irregularity and the OTDR, and the combined signal of FIG. 4a could be used as a calibration signal, to which a subsequent combined signal might be compared. Such a subsequent combined signal is shown in FIG. 4b.
(26) FIG. 4b displays the same elements as FIG. 4a but since the time-dependent detection signals therein result from the reflection of sampling light pulses at an irregularity located further away from the OTDR than the test irregularity of FIG. 4a, all time values including τ in FIG. 4a will include instead a different parameter τ′, with τ′>τ, and τ′ being proportional to the distance between the irregularity and the OTDR. Hence the calibration signal (i.e. the combined signal of FIG. 3a) and the combined signal of FIG. 4b differ by a time shift Δτ=τ′−τ, due to the different distances between the OTDR and the test irregularity and the OTDR and the irregularity. Since the distance between the OTDR and the test irregularity, x.sub.test, is known, the distance between the OTDR and the irregularity can be obtained from Δτ: it corresponds to x.sub.test+c.Math.Δτ/2.
(27) FIG. 5 shows the calibration signal of FIG. 4a and the combined signal of FIG. 4b on the same clock signal time scale. The signals are shifted with respect to one another by a time shift Δτ=τ′−τ, due to the different distances between the OTDR and the test irregularity of FIG. 4a and between the OTDR and the irregularity of FIG. 4b. According to this embodiment of the invention, the position of the irregularity in the optical transmission fiber is determined by analyzing the time shift Δτ.
(28) The impact of bandwidth limitation when processing the reflected light pulses and generating the corresponding detection signals and of the sampling rate on measurement precision is illustrated in FIG. 6. As already explained with respect to FIG. 1, the location of the irregularity can be determined only up to a deviation Δx=c.Math.ΔT/2 when using the described conventional technique. Thus, acceptable precision conventionally requires the use of fast and hence costly hardware components able to offer short sampling periods ΔT. This requires large electrical bandwidths for processing the reflected light pulses and generating the corresponding detection signals. Hence, limited electrical bandwidth and effects such as attenuation and dispersion typically lead to distortions of the shape of the sampling and reflected light pulses, as illustrated in FIG. 6. Measurements from such distorted light pulses can be carried out with reduced accuracy only, in particular when the shape of the pulses is no longer symmetrical to its center point. When using the method of the invention, the pulses also suffer from signal distortions, such that typically, the form of the pulses is smoothed. In spite of these distortions of the shape of the individual pulses and of a significantly longer sampling period, the combined signal can nevertheless be determined with high accuracy. In the shown example, it is assumed that an analytic function is fitted to the combined signal by means of a least mean square fitting. A remarkable advantage stems from the fact that the combined signal, and hence the corresponding fitting function as well, correspond to a superposition of the detection signals corresponding to many sampling light pulses, whence signal distortions that might be noticeable in the individual pulses or detection signals are rendered irrelevant for the purposes of determining the position of an irregularity in an optical transmission fiber.
(29) FIG. 7 shows the effects of attenuation upon combined signals obtained for a given irregularity and for different assumed levels of uniform fiber attenuation. FIG. 7a shows the combined signals as resulting from the superposition of the pulses without normalization, whereas a the combined signals presented in FIG. 7b have been normalized to have a maximum amplitude of 1. All combined signals have been fitted to analytic functions. As can be seen in the figure, fiber attenuation results in a signal shift which is smaller the larger the fiber attenuation is. The imbalance between the leading edge and the trailing edge is also inversely proportional to fiber attenuation. The data shown in FIG. 7b may be comprised in a set of stored data to which a calibration signal obtained by the OTDR can be compared. As a result of this comparison, the level of fiber attenuation can be determined and taken into account for subsequent measurements.
(30) FIG. 8 shows combined signals obtained for irregularities located at different positions of the optical transmission fiber. All combined signals have been normalized to have a maximum amplitude of 1. Alternatively, the zero of the spatial scale could be set to coincide with the location of a test irregularity. The combined signals shown are obtained by emitting a succession of 10 sampling light pulses with a pulse width of 100 μs and with different delays relative to a clock signal, differing by a time increment of 10 μs. In this case, a sampling period of 1.2 μs could be used.
(31) Further, the effects of attenuation can be appreciated in FIG. 8. Normalization and attenuation cause the trailing edge and the leading edge of the signals to increasingly separate for increasing distance to the irregularity. Thereby, the combined signals become broader and more imbalanced, since the shifts of the trailing edges are larger than the shifts of the leading edges. This is caused by the exponential increase of Rayleigh backscattering with propagation distance induced by fiber attenuation.
(32) A determination of the time shift between the combined signals provides a measure of the distance between the OTDR and the corresponding irregularity which is not limited by the sampling period, the pulse width, or by any other parameter related to the individual sampling light pulses imposed by the speed of the OTDR hardware components. For example, a resolution of 1 m has been achieved with a sampling rate of approximately 600 kHz by using pulses with a width of approximately 330 μs and by superimposing 196 pulses. Furthermore, by conducting several measurements it has been shown that the position of an irregularity located 1 m apart from the device could be determined with high repeatability. In contrast, the resolution of a conventional technique using the same sampling rate would be 167 m only.
(33) FIG. 9 shows plots corresponding to different optical parameters of the combined signals of FIG. 7. In FIG. 9a, the pulse center τ.sub.center of each of the combined signals is displayed as a function of the distance between the irregularity and the OTDR. FIG. 9b shows the symmetry parameter of each of the combined signals plotted against the distance between the irregularity and the OTDR. FIG. 9c is a parametric representation of the symmetry parameter of the signals against the pulse center. By obtaining, for example, the pulse center and the symmetry parameter of a given combined signal, the corresponding fiber attenuation can be inferred from FIG. 9c. Once the level of fiber attenuation is known, FIG. 9a or FIG. 9b can be consulted to determine the position of the irregularity in the optical transmission fiber. This way of determining the position of the irregularity in the optical transmission fiber takes the effects of fiber attenuation into account and does not require resorting to a calibration signal. In particular, information like that contained in FIG. 9a-c can be comprised as stored data in the OTDR.
(34) Further advantageous aspects of the present invention are illustratively displayed in FIG. 10. The shaded elements shown therein illustrate three different possible shapes of single sampling light pulses. In contrast to the rectangular shape shown on the leftmost side, the pulse in the middle has inclined flanks, whereas the pulse shown on the right has edges that have been smoothed due to filtering effects or linear and nonlinear fiber effects. The combined signal generated from the shown sampling light pulses is represented for each of the pulse shapes by a solid curve and the corresponding analytical function fitted thereto is represented by a dashed line.
(35) Due to different shapes of the individual sampling light pulses, the corresponding combined signals display different shapes. The combined signal in the middle of FIG. 10 displays a perfect triangular shape, the combined signal on the left corresponds to a stair function, whereas the combined signal on the right has irregular variations due to the smoothed edges of the individual sampling light pulses. However, the resulting analytical function fitted to the respective combined signals is the same for all considered pulse shapes. As a result, the particular shape of the individual pulses has negligible effects on the resulting combined signal and hence on the determination of the position of the irregularity in the optical transmission system.
(36) FIG. 11 highlights the advantages of the invention over the prior art solutions with respect to the possibility of achieving an acceptable accuracy for determining the position of an irregularity in an optical transmission fiber without having to resort to short sampling period saw pulse widths, namely, without having to resort to costly high-speed hardware components. This also applies to the possibility of reconfiguring existing optical equipment, designed for purposes other than those of an OTDR, i.e. other than for determining the position of an irregularity in an optical transmission fiber according to the method of the invention, like an OTDR. For the cases shown on the left and on the right hand side, parameter values allowing to detected irregularities that are 1 m away from the OTDR are provided as examples.
(37) The two drawings placed one above the other on the left side illustrate the role of the sampling rate when using the conventional OTDR technique elucidated above with respect to FIG. 1. The sampling points are signaled as dark dots. In the lower illustration, the sampling period has been halved as compared to the upper one. In both the upper and the lower left illustrations, the shaded area provides an estimate of the inaccuracy of the conventional OTDR technique of FIG. 1. The only statement that can be made from the detected samples is that the leading edge must lay within this area. As seen in the figure, decreasing the sampling period results in the width of the shaded area decreasing and thus leads to a more precise determination of the location of the irregularity.
(38) The drawings on the right side of FIG. 11 illustrate the situation when using the invention with respect to combined signals. As seen in the figure, the time axis of the right hand side figure, corresponding to an embodiment of the invention, is significantly compressed as compared to the conventional OTDR technique of the left hand side, and hence corresponds to a larger timescale. The time divisions along the time axis of the left hand side correspond to 10 ns, whereas on the right hand side they correspond to 1.6 μs. This illustrates the lower requirements on time resolution posed by the invention for a given spatial resolution in the detection of the irregularity.
(39) The same sampling period is assumed for both drawings on the right side of FIG. 11. However, a different resolution of the involved analog/digital converters is assumed, that results in different resolutions in the detection of the amplitude of the reflected light signals and in the generation of the corresponding detection signals. For illustration purposes, three different amplitude sensitivity levels are considered in the upper drawing, denoted by l.sub.0, l.sub.1/2 and l.sub.1. Due to the limited resolution, the analytic function fitted to the combined signal (solid curve) deviates significantly from the leading edge of the ideal signal to be detected (dash-dotted curve). In the lower drawing, the resolution in the detection of the amplitude of the reflected light signals and in the generation of the corresponding detection signals is increased and five amplitude sensitivity levels l.sub.0, l.sub.1/4, l.sub.1/2, l.sub.3/4, and l.sub.1 are possible. With this modification, the fitting curve provides a significantly improved estimate of the ideal curve.
(40) Thus FIG. 11 illustrates that, whereas conventional OTDR techniques rely on a better time resolution, e.g. shorter sampling periods, for improving the accuracy with which the position of an irregularity in an optical transmission fiber can be determined, according to the method of the invention this accuracy can be improved by simply resorting to sufficiently sensitive analogue/digital converters, i.e. providing sufficient resolution. Hence, for the purpose of achieving a desired level of accuracy for determining the position of an irregularity in an optical transmission fiber, the method of the invention allows for the use of longer sampling light pulses. Further, while conventional methods call for costly high-speed hardware components, the aforementioned accuracy can be improved by means of the method of the invention via widely available non-costly analogue/digital converters. In particular, the resolution of standard analog/digital converters proves fully sufficient in terms of carrying out the invention.
(41) FIG. 12 shows two exemplary methods to generate the delays associated with the detection signals. On behalf of clarity, only two pulses are shown for each method. While sampling light pulses emitted by the OTDR and launched into the fiber to be measured are shown in the upper row, the resulting reflected light pulses received by the OTDR are shown in the middle row. The lower row shows the detection signals that are superimposed in order to generate the corresponding combined signal.
(42) FIG. 12a shows sampling light pulses emitted with a predetermined delay t.sub.i relative to trigger points ψ.sub.1 and ψ.sub.2. These trigger points might be virtually generated in a random manner, but they might also be derived from a clock signal. In the latter case, equidistantly distributed trigger points are obtained that correspond to the clock signal points φ.sub.1 and φ.sub.2 shown in FIG. 2. Each reflected light pulse received at the OTDR is delayed by 2x/c with respect to the time at which the respective sampling pulse has been emitted and is delayed by 2x/c+t.sub.i with respect to the respective trigger point, where t.sub.i stands for the delay introduced in FIG. 4. The corresponding detection pulses might be further delayed by a processing time Δt.sub.p and thus are delayed in total by 2x/c+t.sub.i+Δt.sub.p relative to the respective trigger point. The same delay appears on a time scale t=T−ψ.sub.i representing the time that has lapsed since the respective trigger point. Thus, two reflected light pulses labelled i and j are delayed by a time t.sub.i-t.sub.j with respect to each other. Superposition of the corresponding detection signals results in the combined signals as described above.
(43) An alternative method is shown in FIG. 12b. Sampling light pulses are emitted by the OTDR without delay (or with the same delay) relative to trigger points and the reflected light pulses are received with a delay of 2x/c relative to the trigger points (or further delayed by the same delay). When processing the received reflected light pulses, the delays t.sub.i associated with the resulting detection signals are introduced preferably in the digital domain, i.e. after an analog to digital conversion. Finally, the same detection signals as shown in FIG. 12a are obtained.
(44) Both methods can be also be combined. In particular, some additional delays can be introduced in the digital domain in order to compensate for instabilities (jitter and wander) of a clock signal.
(45) FIG. 13 shows a schematic representation of an OTDR 10 for detecting an irregularity in an optical transmission fiber according to an embodiment of the invention. The OTDR 10 comprises a light source 12, a light receiver 16, a processing unit 18, an analogue/digital converter 20 and a digital/analog converter 22, an analyzing unit 24, a circulator 26, a connector 28, and a control unit 30. The light source 12 is configured for emitting a succession of sampling light pulses into the optical transmission fiber 14 via a second connector 29 connected to the connector 28 of the OTDR. The sampling light pulses pass through the circulator 26 and are reflected at the connector 28 or at an irregularity in the optical transmission fiber 14, resulting in reflected light pulses. The reflected light pulses circulate back to the circulator 26 and are directed to the light receiver 16. The light receiver 16 is configured for detecting the reflected light pulses and for generating corresponding digital time-dependent detection signals by means of the analogue/digital converter 20. A processing unit 18 is operatively connected to the light receiver 16 via the analogue/digital converter 20 and is configured for generating a clock signal, for associating detection signals corresponding to different sampling light pulses with different delays with respect to the clock signal, and for obtaining a combined signal from the time-dependent detection signals generated by the light receiver 16. The analyzing unit 24 is operatively connected to the processing unit 18 and is configured for analyzing the combined signal for determining the position of the irregularity in the optical transmission fiber with respect to the OTDR. The control unit 30 is operatively connected to the light source 12 via the analogue/digital converter 22 and is configured for controlling the light source 12. In the embodiment shown, the processing unit 18, the analyzing unit 24, and the control unit 30 are comprised in an integrated device 32.
(46) Although preferred exemplary embodiments are shown and specified in detail in the drawings and the preceding specification, these should be viewed as purely exemplary and not as limiting the invention. It is noted in this regard that only the preferred exemplary embodiments are shown and specified, and all variations and modifications should be protected that presently or in the future lie within the scope of protection of the invention as defined in the claims.
REFERENCE SIGN LIST
(47) 10 optical time domain reflectometer 12 light source 14 optical transmission fiber 16 light receiver 18 processing unit 20 analog/digital converter 22 digital/analog converter 24 analyzing unit 26 component assembly 28 connector 29 connector 30 control unit 32 integrated device
