INTRODUCTION

An optical time-domain reflectometer (OTDR) is an optoelectronic instrument used to characterize an optical fiber. An OTDR injects a series of optical pulses into the fiber under test. It also extracts, from the same end of the fiber, light that is scattered (Rayleigh backscatter) or reflected back from points along the fiber. (This is equivalent to the way that an electronic time-domain reflectometer measures reflections caused by changes in the impedance of the cable under test.) The strength of the return pulses is measured and integrated as a function of time, and is plotted as a function of fiber length. An OTDR may be used for estimating the fiber's length and overall attenuation, including splice and mated-connector losses. It may also be used to locate faults, such as breaks, and to measure optical return loss. To measure the attenuation of multiple fibers, it is advisable to test from each end and then average the results, however this considerable extra work is contrary to the common claim that testing can be performed from only one end of the fiber. In addition to required specialized optics and electronics, OTDRs have significant computing ability and a graphical display, so they may provide significant test automation. However, proper instrument operation and interpretation of an OTDR trace still requires special technical training and experience. OTDRs are commonly used to characterize the loss and length of fibers as they go from initial manufacture, through to cabling, warehousing while wound on a drum, installation and then splicing. The last application of installation testing is more challenging, since this can be over extremely long distances, or multiple splices spaced at short distances, or fibers with different optical characteristics joined together. OTDR test results are often carefully stored in case of later fiber failure or warranty claims. Fiber failures can be very expensive, both in terms of the direct cost of repair, and consequential loss of service. OTDRs are also commonly used for fault finding on installed systems. In this case, reference to the installation OTDR trace is very useful, to determine where changes have occurred. Use of an OTDR for fault finding may require an experienced operator who is able to correctly judge the appropriate instrument settings to locate a problem accurately. This is particularly so in cases involving long distance, closely spaced splices or connectors, or PONs. OTDRs are available with a variety of fiber types and wavelengths, to match common applications. In general, OTDR testing at longer wavelengths, such as 1550 nm or 1625 nm, can be used to identify fiber attenuation caused by fiber problems, as opposed to the more common splice or connector losses. The optical dynamic range of an OTDR is limited by a combination of optical pulse output power, optical pulse width, input sensitivity, and signal integration time. Higher optical pulse output power, and better input sensitivity, combine directly to improve measuring range, and are usually fixed features of a particular instrument. However optical pulse width and signal integration time are user adjustable, and require trade-offs which make them application specific.

Basic principles of the method

The backscattering method was invented by M. Barnoskim and M. Jensen in 1976 , in time when technology of the optical fibre manufacturing was at early stages. The precise and reliable measurement of local losses on the fiber was very important for further improvement of quality of fibers. In the paper cited above the authors describe a new method for the loss distribution along the fiber. The basic idea of the proposed method consisted in launching a rather short and high power optical impulse into the tested fiber and a consequent detection of back scattered optical power as a response of the fiber to the test impulse. The detected signal provides the detail picture about the local loss distribution or reflections along the fiber caused by any of the attenuation mechanisms or some other nonhomogeneities on the fiber. An important feature of the method is non-destructivity and the fact that the access to only input end of the fiber is needed. The measurement of the time delay of the detected signal from the fiber end or from any perturbation on the fiber allows to derive the information about the perturbation localization provided that the index of refraction in the fiber core or group velocity of light propagation is known. In any point on the fiber the magnitude of the backscattered optical power is proportional to the local transmitted optical power. Due to the nonzero losses this power is gradually attenuated along the fiber and consequently also the backscattered power is also attenuated. The measurement of the backscattered power as a function of time or position on the fiber gives the information about the local distribution of the attenuation coefficient along the fiber. In this way one can evaluate the space distribution and magnitude of various nonhomogeneities along the fiber like optical connectors, splicings, micro- and macro-bend losses and others measurand-perturbances. The comparison of the losses closely before and after point of interest makes possible to evaluate insertion losses of the various optical components on the fiber link.

Theoretical description of the conventional OTDR

The elementary experimental experience gives the relation describing the dependence of the optical power propagating along the optical fiber as a function of the distance x from the input end of the fiber (1)

whereP(x) is the total optical power at the distance x from the point of launching the input optical impulse, P0 is the value of the input optical power (x = 0), is the total attenuation coefficient in Np/km and 1(x) is the Heaviside step function. In practice the attenuation coefficient is usually expressed in dB/km. In this case the relation (1) can be rewritten into the form (2)

where' is the total attenuation coefficient given in units dB/km. The mutual relation between and ' is defined by (3)

Total losses in the fiber are caused by different mechanisms and the total attenuation coefficient can be different at any point on the fiber. As a result it is necessary to rewrite the relation (1) into more general form

(4)

where the local attenuation coefficient (x) is now a function of the distance x. It can be shown that the total attenuation coefficient can be roughly split into two components (5) wherea(x) represents the absorption losses and rs(x) represents the losses by Rayleigh scattering mechanism. The average value of the total attenuation coefficient (x) on the fiber section defined by distance (0, x) can be calculated according to the formula

(6)

and the relation(4) can be simplified as follows (7)

The elementary optical power dPrs scattered by the Rayleigh mechanism on each elementary fiber section dx (scattering centre) at the distance x from the input end of the fiber is given by (8)

where due to simplicity coefficient rs(x) was taken as constant along the fiber. Provided that rsdx << 1, (8) can be approximated by the relation (9)

In accordance with the relation (4) the propagating local optical power P(x) changes along the fiber. A part of the isotropically scattered optical power, described (9), is refracted at the boundary core/cladding and is totally lost and the other part is recaptured by the numerical aperture of the fiber and is directed in the forward and backward direction. The part directed backwards is called backscattered optical power. Its magnitude is directly proportional to the backscattering coefficient S what allows one to express the backscattered power from the elementary section dx on the fiber in the form

(10)

The backscattered power is, similarly as forward propagating total optical power, attenuated on the route to the input end of the fiber. The backward attenuation coefficient (let us denote it by ''(x)) is generally different from the forward attenuation coefficient (x). As a result one can write for the backscattered power from the elementary section dx in the point x, that can be detected at the input end of the fiber, the formula

(11)

If one takes A1 and A2 as the total average attenuation coefficients at the distance x in forward and backward direction respectively

(12)

andA will represent their arithmetical average A = 0.5(A1 + A2), then the relation (11) can be transformed into the form (13)

For the backscattering coefficient S one can derive the analytical relation describing its magnitude for the single-mode and multi-mode fibers with a given refraction index profile. Under some simplifications a rather simple relation for the backscattering coefficient for a single-mode optical fiber can be obtained in the form [2]

(14)

For the case of a multi-mode fiber with a step-index profile the backscattering coefficient can be described by (15)

(16)

whereNA = (n12 - n22)1/2 is the numerical aperture, n1, n2 are the refractive indexes of the core and cladding respectively, wo is the mode field diameter of the basic mode, a is the fiber core radius, V is so called normalized frequency V = (2a/)NA.

Let us now to analyze the time dependence of the backscattered power detected at the input end of the fiber as a response to the testing impulse of the optical power. For this purpose let us consider the optical fiber into which an optical impulse of the instantaneous power P0 and the width T0 was coupled in the time t = 0. The time dependence of this impulse is given by the relation P(t) = P0[1(t) - 1(t - T0)] (17)

where 1(t) is the Heaviside unit step function. One can imagine this impulse as a lit section of the fiber. The length of the region is given by x = vgT0, where vg is the group velocity of the impulse propagation in the fiber. The position of the trailing edge of the impulse at time (t T0) is given by x and the position of the leading edge is given by (x + x). The described situation is outlined in the Fig.1.

Fig.1. The position of the optical impulse in the fiber core at time t

Using the substitution form [3], [4]

, resp.

the relation (13) can be rewritten into the

(18)

The time dependence of the backscatter power generated by the whole testing impulse can be obtained by the integration

(19)

Provided that AvgT0 << 1, what is for high quality fibers always obeyed the relation (19) can be transformed into the form

(20)

Using the same substitution as it was done for the relation (18) the time dependence of the backscattered power can be described by (21)

In the case of high quality fiber the attenuation coefficient is given by = A. It makes possible to write for the backscattered power the well known formula (22)

If the fiber parameters (S, rs, , vg) are constant and the maximum launched power is P0, then the maximum detectable backscattered power can be enhanced by the increase of P0 and T0, which define the energy of the impulse. From the theoretical point of view one can consider a limiting situation when T0 approaches zero and P0 goes to infinity while the energy is constant equal to unit. It is truly the case of Dirack impulse shape given by . This gives the possibility to interpret the relation (20) as the impulse response of the optical fiber. It is assumed that the optical fiber is at the moderate optical powers a linear system. Of course it holds only for a relative low powers. At higher powers (cca 1 W) the Kerr nonlinear effect occurs. If the optical fiber can be considered as linear system then the response of the fiber (backscattered power Prbs(t)) to an arbitrary shape of testing impulse Pin(t)) will be given by the convolution of the impulse response h(t) and the input optical power [5] (23)

whereh(t) is defined by (20). In a real optical fiber the time dependence of the backscattered power is influenced except the attenuation also by all other changes or statistical deviations of the fiber parameters from an ideal fiber considered in the theory. It means that the detected power also in the case of ideal detection system is not represented by an ideal theoretical exponential function. In reality it is a complicated function shaped by the internal properties of the fiber reflecting the space distribution of various perturbation along the fiber mentioned in previous chapter. Of course the properties of the optical receiver mainly the noise significantly influences the shape of the detected signal. It follows from the fact that the power level of the backscattered signal in single-mode optical fiber is cca 60 dB under the level of the launched input impulse power. Provided the coupled optical power from a laser diode is usually at the level of a few mW then the detected optical backscattered power approaches the level of cca - 90 to 100 dBm what represents in absolute scale the power of tenth or units of pW. It is power deeply sunk in the noise of optical receiver.
7

Dynamical range and resolution OTDR

As it was briefly indicated above the signal S(t) at the output of the detection system of the conventional OTDR reflectometer is generally given by the convolution of the input testing signal Pin(t), impulse response of the tested optical fiberH(t) and the impulse response of the optical receiver R(t) [6] (24)

The convolution product for the case of an ideal fiber ( = const.) and the rectangular shape of the testing impulse of duration T0, as it follows from (20) and (23) is given by the relation (25)

Provided the width of the testing impulse is T0 the space localization of a point scattering center cannot be better than x = 0.5T0vg, neither in the case of the use of an ideal optical receiver. To get the best space resolution of the disturbance localization by the method one should use as narrow testing impulse as possible. However due to decrease of the impulse energy E = P0T0 such a choice deteriorate the conditions for the detection as it follows from the relation (3.3.2). In the frame of the conventional OTDR there are only two possibilities how to increase the energy of the testing impulse: a) through the increase of P0 of the laser diode or b) through the increase of the testing impulse duration T0. However the latter possibility indispensably implies the deterioration of the space resolution of the method. This mutual competitive relation between the testing impulse power and its duration or between the dynamical range and space resolution represents the fundamental limit of the conventional OTDR. Due to practical constraints (life time, heating, high excitation current, electromagnetic interference, ...) the output power of the laser diode can not be increased above the certain level. More, neglecting the above mentioned practical reasons the optical power can not be enhanced above the level corresponding to the occurrence of non-linear effects in the fiber. Optical power of the testing impulse is also limited by the saturation of the optical receiver caused by Fresnel reflections from various non-homogeneities like optical connectors, fiber breaks, low quality splicings and so on. The conditions for good detection are negatively affected also by a very short excitation impulse that requires a broader bandwidth of the receiver. Of course the enlargement of the receiver bandwidth implies automatically in the increase of the receiver equivalent noise power. The mutual relations between the testing impulse energy, the magnitude of the backscattered and (Fresnel) reflected optical powers as the functions of the testing impulse width are graphically outlined . The dependences were set for single-mode fiber with the attenuation coefficient = 0.35 dB/km and the wavelength = 1300 nm, the laser diode output power is taken 8 mW (9 dBm). Two testing impulse width are considered - 4 s and

100 ns. Due to the use of the 3 dB optical power splitter at the input end of the fiber the absolute value of the launched optical power in the fiber is equal to 4 mW (6 dBm).

Fig. 2. Optical power relations in the conventional OTDR

The particular lines represent in dB scale a) the dependence of the coupled optical power in the fiber, b) the dependence of the Fresnel reflected power (for the case of reflection at the boundary glass-air the reflection coefficient is taken equal to 0.04, in dB it is approximately 14 dB under the local power level), c) the dependence of backscattered power for T0 = 4 s and 100 ns respectively as the functions of distance on the fiber measured from the fiber input end. The maximum level of the backscattered power for 4 s impulse is roughly - 40 dBm and for 100 ns impulse it is nearly - 56 dBm. For the reliable testing of a 100 km long section of fiber with the attenuation coefficient = 0.35 dB/km, what represents the one way loss of 35 dB, one needs an optical receiver with the minimum dynamical range of 70 dB. It is interesting that for the 4 s excitation impulse the detected backscattered power changes in the range from 100 nW to 10 fW, for the 100 ns impulse it does from 2.5 nW to 0.25 fW. Moreover if Fresnel reflections are taken into account one can see that the optical receiver can be saturated by the reflected power at least on the first tenths or hundreds of meters of the fiber. The saturation of the receiver implies the loss of the information carried by the backscattered power coming from the certain section of fiber closely located to the point of reflection and called the dead zone of the reflectometer. The dead zone of the reflectometer is one of the most important quality indicators of the optical reflectometer. Dynamical range of the OTDR reflectometer is determined by the signal to noise ratio (SNR). SNR characterizes
9

the ability of the device to detect minimum and maximum optical powers. High quality reflectometers should be capable to process high power level signal without remarkable distorting caused by saturation of the receiver and simultaneously to process as low signals as possible without the deterioration by the noise. The noise of the optical receiver is of the most significance for the SNR and therefore it should be as small as possible. The total noise power is proportional to the bandwidth of the receiver that is determined by the testing impulse width. The shorter impulse implies the better space resolution but also the greater bandwidth of the receiver and consequently greater noise power and lower dynamical range. Therefore in this conventional OTDR there is a need of some compromise between these performance parameters of the OTDR reflectometer. One of the possibilities how to reduce the influence of the noise is the numerical filtration of the signal from the noise. It is frequently called the method of signal averaging. The basic idea of this approach consists in the fact that the average value of the noise component contained in the signal is zero. As a result after many times repetitions of the measurement process and consequent addition of the particular measured data and calculating the average the noise is automatically eliminated. It can be shown that after the N times repeated measurements the SNR obtained by this signal processing is times greater than the SNR obtained in one measurement. This effect is results from the fact that in repeated measurements the signal increases proportionally with the number of repetitions Nwhile the noise only proportionally to the [5]. Provide the optical receiver is characterized by a given noise equivalent power PNEP (NEP - Noise Equivalent Power) it is possible to express the resulting SNR obtained after N times repeated measurements by the formula

(26)

If Pinit represents the maximum value of the backscattered optical power from the input end section of the fiber, that is for x = 0, then (27)

Taking the logarithm of the (26) one can obtain the general relation for the SNR, frequently called in anglo-saxon special literature as "OTDR maker's formula" [7] (28)

It describes the SNR as a function of distance x measured from the input end of the fiber and the basic parameters of the optical source and receiver - Pinit and PNEP respectively. Here Noct = log2N.

10

The main aspects of the signal processing in conventional OTDR

A simplified block diagram of the OTDR-based optical reflectometer is given in the Fig.3 The main blocks of the reflectometer are the generator of the testing impulse and the detection system of the backscattered light. The remaining blocks provide the suitable timing of signals (clock generator) and the interpretation of the measured data (display). A 3-dB fiber power splitter makes possible to couple the optical excitation power impulse into the tested fiber and simultaneously to deviate the backscattered power to the optical receiver.

Fig. 3. The simplified block diagram of the conventional OTDR based reflectometer. The optical path is depicted by the blue color.

The crucial element of the device is the block for the processing of the signal from the optical receiver. For the signal recovery a technique of signal sampling using the A/D converter simultaneously with the signal averaging method is used. For the elimination of the dead zone a blind subsidiary fiber put between the optical source and the input end of the tested fiber is used. In this way the Fresnel reflection from the input end of the tested fiber and subsequent dead zone occurs in the time corresponding to the section of subsidiary fiber and no information coming from the tested fiber is lost due to dead zone.

11

How Does an OTDR work?

Unlike sources and power meters, which measure the loss of the fibre optic cable plant directly, the OTDR works indirectly. The source and meter duplicate the transmitter and receiver of the fibre optic transmission link, so the measurement correlates well with actual system loss. The OTDR, however, uses unique phenomena of fibre to imply loss. The biggest factor in optical fibre loss is scattering. It is like billiard balls bouncing off each other, but occurs on an atomic level between photons (particles of light) and atoms or molecules. If you have ever noticed the beam of a flashlight shining through foggy or smokey air, you have seen scattering. Scattering is very sensitive to the colour of the light, so as the wavelength of the light gets longer, toward the red end of the spectrum, the scattering gets less. Very much less in fact, by a factor of the wavelength to the fourth power - that's squared-squared. Double the wavelength and you cut the scattering by sixteen times! You can see this wavelength sensitivity by going outside on a sunny day and looking up. The sky is blue because the sunlight filtering through the atmosphere scatters like light in a fibre. Since the blue light scatters more, the sky takes on a hazy blue cast.

Figure 4 Scattering in an Optical Fibre In the fibre, light is scattered in all directions, including back toward the source as shown in Figure 4. The OTDR uses this "backscattered light" to make its measurements. It sends out a very high power pulse and measures the light coming back. At any point in time, the light the OTDR sees is the light scattered from the pulse passing through a region of the fibre. Think of the OTDR pulse as being a "virtual source" that is testing all the fibre between itself and the OTDR as it moves down the fibre. Since it is possible to calibrate the speed of the pulse as it passes down the fibre, the OTDR can correlate what it sees in backscattered light with an actual location in the fibre. Thus it can create a display of the amount of backscattered light at any point in the fibre, Figure 2.

12

Figure 5. OTDR Display There are some calculations involved. Remember the light has to go out and come back, so you have to factor that into the time calculations, cutting the time in half and the loss calculations, since the light sees loss both ways. The power loss is a logarithmic function, so the power is measured in dB. The amount of light scattered back to the OTDR is proportional to the backscatter of the fibre, peak power of the OTDR test pulse and the length of the pulse sent out. If you need more backscattered light to get good measurements, you can increase the pulse peak power or pulse width as shown in Figure 3.

Figure 6. Increasing the pulse width increases the backscatter level Note on the display shown in Figure 2, some events like connectors show a big pulse above the backscatter trace. That is a reflection from a connector, splice or the end of the fibre. They can be used to mark distances or even calculate the "back reflection" of connectors or splices, another parameter we want to test in singlemode systems

13

Figure 7. OTDR Trace Information The slope of the fibre trace shows the attenuation coefficient of the fibre and is calibrated in dB/km by the OTDR. In order to measure fibre attenuation, you need a fairly long length of fibre with no distortions, or non linearities, caused by the OTDR initial pulse, along its length or at the launch cable connection at the near end or at the far end caused by "ghosts" or highly reflective events. If the fibre looks nonlinear at either end, especially near a reflective event like a connector, avoid that section when measuring loss. Connectors and splices are called "events" in OTDR jargon. Both should show a loss, but connectors and mechanical splices will also show a reflective peak. The height of that peak will indicate the amount of reflection at the event, unless it is so large that it saturates the OTDR receiver. Then the peak will have a flat top and tail on the far end, indicating the receiver was overloaded. Sometimes, the loss of a good fusion splice will be too small to be seen by the OTDR. That's good for the system but can be confusing to the operator. It is very important to know the lengths of all fibres in the network, so you know where to look for events and won't get confused when unusual events show up (like ghosts, which we'll describe below.) Reflective pulses can show you the resolution of the OTDR. You cannot see two events closer than is allowed by the pulse width. Generally longer pulse widths are used to be able to see farther along the cable plant and narrower pulses are used when high resolution is needed, although it limits the distance the OTDR can see.

14

Micro OTDR Operating via a USB Computer Port

The uOR-100 and uOR-200 Micro PC-Based OTDR measures the attenuation in optical fibres and splices, as well as the length and the distance to any event, such as a break in the fibre link. The Micro OTDR is the perfect tool for the construction, maintenance and restoration of cable plants and also as an educational tool. The Micro OTDR is accurate enough for laboratory use but portable and rugged enough for field applications. The Micro OTDR consists of the single, small rugged case; it is lightweight and uses minimal power. The Micro OTDR is used in conjunction with a PC or Laptop with a USB interface, eliminating the need for batteries completely. The USB connection allows the Micro OTDR to be controlled from a laptop or PC and runs off of Windows operating system. The Micro OTDR software allows the user to determine all the necessary characteristics of the Optical fibre and is capable of displaying, storing, reading, printing and analysing several traces at the same time.

Optical Fibre Type Wavelength, nm Dynamic range

Attenuation Dead Zone Event Dead Zone Pulsewidth Distance Range Loss Resolution Distance accuracy Refractive Index (n) Range 1.0000 to 2.0000 Setting Optical Connector Type Various options available Supply Current From USB Port <200mA Size 6.25inches x 3inches x 1.25inches; 15.9cm x 7.6cm x 3.18cm Weight 1.5 lbs, 0.7kg Operation Temperature 0 to +40C Relative Humidity 95% without condensation ** Dual WavelenghSinglemode Version Available Versions (Dynamic Range) 29/29 dB 32/31 dB 40/38 dB 43/42 dB *** Single WavelenghSinglemode Version Available Versions (Dynamic Range) 30dB (1310nm) 30dB (1550nm) Order Number 101 Wavelength 850/1300nm Decription Multimode W/ST*
15

TECHNICAL SPECIFICATIONS Multimode Singlemode 850nm/1300nm 20nm 1310nm/1550nm 30nm ** 4 Dual Versions available 26/25 dB *** 2 Single Versions available 12 metres 6 metres 3 metres 1.5 metres 10 to 1000 ns 10 to 20000 ns 2, 5, 10, 20, 40, 80, 120, 160, 240 km 0.001 dB ((0.5+(5 x 10 to power-5 x L)+(dn/n) x L)) where L=length

Dynamic Range 26/25 dB

201 203 204 301 200A 200T 200F

Singlemode W/FC Singlemode W/FC Singlemode W/FC Multimode and 850/1300/1310/1550nm Singlemode W/ST+FC 1310/1550nm Singlemode W/FC 1310nm Singlemode W/FC 1550nm Singlemode W/FC * = With ST connector interface.

1310/1550nm 1310/1550nm 1310/1550nm

32/31dB 40/38dB 43/42dB 26/26(m/mode)dB 28/29(s/mode)dB 29/29dB 30dB 30dB

16