OPTICAL FIBRE AND NETWORK
This invention relates to optical fibre networks and to a method of monitoring them to determine fault location therein.
Optical fibre networks are becoming increasingly important in many area of telecommunication technology, and recently many forms of optical fibre networks have been developed for example for telephone systems, cable television, local area networks (LANs) and the like. A network typically may have a tree structure in which a single station end or head end, which may be a CATV receiver, an exchange or a LAN controller, is linked to a number of end users by means of a number of branches in the network. In the case of passive networks, that is to say, networks in which no components need be supplied with power between the station end and the end user, the nodes typically are formed as l.n splitters for example by fused or planar fibre couplers.
In all such optical networks it is important to be
able to monitor the network for breaks in the optical line or degradation of the line, for example due to the presence of moisture. A number of methods have been proposed for monitoring such optical networks based on optical time domain reflectometry (OTDR) and optical frequency domain reflectometry (OFDR) . In OTDR a pulse is sent from the station end of the network, and the reflected signals are analysed as a function of time in order to judge the distance between the station end and a break in the optical line. However, since the pulse will be reflected by the termination of the line at each end user in addition to any reflection that may be caused by a break in the line, a single pulse will cause a large number of reflections to be generated with different time delays, with the result that it may not possible to distinguish a fault in the network from the other reflections. Furthermore, if a fault occurs in one of a number of branches of a network, it is usually not possible to determine in which branch the fault is located. Alternatively, one could test the network by sending a signal from the end users' premises back to the station end, but it is not commercially viable to send a maintenance engineer to each end users' premises for the purpose of routine maintenance. U.K. patent specification No.2, 268, 652A discloses a method of monitoring a branched optical network by an OTDR or OFDR technique in which, once a fault is located in the network by monitoring at the station end, a signal is introduced into the network at successive nodes along the network in order to home in on the fault. However, such a method still involves the necessity for personnel
to travel to a number of locations before the position of the fault can be identified.
It has been proposed to incorporate a Bragg grating at each end user's premises which will reflect light of a wavelength that is characteristic to that subscriber, so that reflections observed when monitoring the network can be assigned to individual end users and faults in the network can therefore be assigned to particular branches. However, such a system has the problem that the number of different wavelengths that can be employed would severely restrict the possible number of end users, and that the number of end users will normally considerably exceed the possible number of different wavelengths that can be employed to monitor the system. According to one aspect, the present invention provides an optical network which includes a plurality of branches in an optical fibre line between a station- end and a plurality of end users, and in which traffic is modulated at one or more optical wavelengths wherein the network includes in the region of each end user a combination of reflectors which create a reflection pattern that is unique to that end user.
According to another aspect, the invention provides a method of monitoring a network that includes a plurality of branches in an optical fibre line between a station-end and a plurality of end users, and in which traffic is modulated at one or more optical wavelengths wherein the network includes in the region of each end user a combination of reflectors which create a reflection pattern that is unique to that end user, which method comprises sending light into the network in
the region of the station-end and observing the pattern of reflected light.
The wavelength(s) reflected by the reflectors will normally be different from the traffic wavelength in order to prevent any effect on the traffic.
The present invention has the advantage that it is possible to identify a large number of end users of the network by using only a small number of different wavelengths. Thus, for example, n different wavelengths could be used to identify 2n different end users if different combinations of the wavelengths were employed. However, it is possible to employ a spatial encoding of reflectors in addition to, or instead of, employing combinations of reflectors of different characteristic wavelength. Thus, for example, it is possible to employ a combination of reflectors that each reflect the same wavelength but which are separated from one another along the fibre line according to a pattern that is characteristic of that end user. One example of such a combination would be for reflectors to be separated from one another by multiples of a defined minimum separation so that the combination of reflectors becomes a binary bit pattern in which the presence of a reflector at a particular point can be regarded as a "1" and the absence of a reflector can be regarded as a "0". Since, in this case, the reflected signal is a binary bit pattern, it is possible to employ conventional data manipulation methods on them. For example, redundancy may be built in to the binary codes for the purpose of error checking and/or error correction, with the result that it may be possible to identify end users even with
relatively high levels of noise.
Even if only a single wavelength is employed for the testing, the reflection patterns are not limited to binary bit patterns. For example, higher module numbers may be achieved by using Bragg gratings of differing reflectivity which will generate reflected pulses of differing intensity. Also other forms of coding, for instance pulse width modulation or pulse position modulation may be used. The network can be monitored by a reflectometry method, either in the time domain (OTDR) or in the frequency domain (OFDR) . OFDR is preferred since it is possible to obtain higher spatial resolution without sacrificing dynamic range. Whichever method is employed the reflection pattern of the network can be recorded for example in a computer memory and reflection patterns subsequently obtained during the monitoring process can be compared with the original reflection pattern stored in the memory and any changes in the reflection patterns identified as possible network faults. The network and method according to the invention has the significant advantage over conventional networks that are monitored by OTDR or OFDR methods in that it is not necessary to ensure that branches of the network were of different lengths (which was necessary in order to be able to resolve the reflections due to different branches in the reflection pattern) . According to the present invention, different branches of the network may have substantially the same length so that reflection patterns due to different combinations of reflectors overlie one another in the reflection pattern of the
network. In this case, if one branch is subjected to a fault so that the reflection pattern of one of the reflector combinations is attenuated, the relevant reflection pattern will clearly be identified simply by subtracting the reflection pattern of the network from the network reflection pattern stored in memory. In addition, since the majority of reflection patterns will be accounted for, there will only be a small number of possible combinations of bit patterns that overlie each other at any position.
Preferably the network includes further combinations of reflectors between the end users and the station end, normally being located at one or more nodes in the network, and especially the network includes a combination of reflectors at each node in the network. In such a network, receipt of a reflection pattern corresponding to that of a given node will indicate continuity of the network up to that node. However, the provision of reflectors at the nodes has an additional advantage in that it is possible to compare the intensity of light reflected at one point in the network (be it an end user or a node) with the intensity of light reflected at a second point connected thereto by a fibre link and located nearer to the station end of the network. Such a comparison, together with a knowledge of the length of the fibre link and the expected attenuation of signals at the relevant wavelength, can provide an indication of the quality of the optical fibre link. Thus, for example, if the attenuation of the monitoring signals in the fibre link begins to rise, this can indicate deterioration of that link, for
example due to hydrogen or water permeation or due to stress on the optical fibre, and can allow remedial action to be undertaken before the fibre link fails.
The reflectors may be based on any of a number of devices. For example, they may comprise optical filters that may reflect a single wavelength or range of wavelengths e.g. a monochromatic reflecting optical filter, while allowing transmission of all other wavelengths. Alternatively and preferably the reflectors may comprise a Bragg grating which will transmit signals modulated at all wavelengths but which will reflect part of a signal back along the fibre if the signal has the appropriate wavelength. Such a Bragg grating may be formed in an optical fibre by shining ultraviolet light onto parts of the fibre in order to alter the refractive index of the fibre core. If positions of the fibre that have been irradiated with ultraviolet light are separated by a distance d the grating so formed will reflect light of wavelength 2d back along the fibre.
Yet a further advantage of the network according to the invention is that if a break occurs at any point in the network where a number of fibres need to be re- spliced, the engineer can conduct an OTDR or OFDR measurement into each end of the broken fibres in order to ascertain which line any given fibre end corresponds to. Such measurement may be conducted non-intrusively, for example, by means of equipment for forming microbending and/or macrobending taps. Preferably the optical network is a passive optical network (PON). However, the invention is not, in fact,
limited to networks and indeed the coding of the reflection pattern can be used to identify fibres. Thus, according to another aspect, the invention provides an optical fibre which incorporates a combination of reflectors (e.g. Bragg gratings) at one or more locations along its length (perhaps every few metres) which create a reflection pattern that is unique to that fibre. The reflection pattern can then be employed to identify the fibre and acts, in essence, as a bar code.
The invention will now be described by way of example with reference to the following:
Figure 1 is a schematic layout of a simplified network; and Figure 2 shows a combination of Bragg gratings employed in the network. Referring to the accompanying drawing figure 1 shows schematically a simplified passive optical network in which a station end or head end 1 is connected to a number of end users or subscribers 2 to 5 by optical fibres in a tree configuration. As shown, an optical fibre leg 8 extends from the head end 1 to a first node 10 comprising a 1:4 splitter, and one arm 12 of the node 10 extends to a second node 14 comprising a further 1:4 splitter. Each of the arms 16, 18, 20 and 22 of the second node extend to its respective end user 2 to 5.
Each node 10 and 14 has associated with it a combination 24 and 26 of Bragg gratings and each of the end users 2 to 5 has an associated combination 32 to 35 of Bragg gratings. Each combination of Bragg gratings begins with a start bit 41 and ends with a stop bit 42
formed from short grating sections which may, if desired, be of greater reflectivity than the remaining bits in order for ease of identification. Between the start and stop bits additional spatially separated grating sections 44 provide bits that provide a reflection pattern that is unique to each grating combination.
The Bragg gratings in the optical fibre may be formed in a number of ways. For example, the grating may be recorded as a hologram, and an image of the grating subsequently may be formed in the optical fibre by shining ultraviolet light through the hologram. Alternatively, a beam of ultraviolet light may be shone through a mesh having the appropriate grating printed thereon and focused onto the fibre. In yet another method, a line of ultraviolet light may be passed along a portion of the fibre at a defined speed and its intensity may be modulated at a frequency that will give the desired line spacing along the fibre. Whichever method is employed the refractive index of the optical fibre core will vary periodically in accordance with the intended line spacing. Such a fibre will transmit light over a range of wavelengths, but if the light wavelength
(in vacuum) λ is related to the line spacing d of the grating by the Bragg formula λ =-2 n d where n is the mean refractive index of the fibre, then part of the signal is reflected back by the grating. Such Bragg gratings may be formed for example as described in European patent applications Nos. 438759A
and 507882A.
One typical combination of Bragg gratings that may be located at an end user or at one of the nodes is shown schematically in Figure 2. A length of optical fibre 41 has been irradiated in order to provide a Bragg grating, but, during irradiation, a mask had been interposed so that only certain parts of the fibre were irradiated to form a unique pattern. In this case the irradiated fibre has been provided with a pattern comprising a start bit 42, a stop bit 44 and, between the start and stop bit an eight bit code reading 01100101 binary corresponding to 101 denary. If desired the start and stop bits may be made distinctive from the other bits, for example by irradiating the fibre to a different extent thereby increasing their intensity.
The network can be monitored by an optical reflectometry method in which light is sent into the network at point 36 and reflected light is stored in computer 38. The reflection pattern can be compared with a standard reflection pattern of the system that has been stored in the computer 38 and if any part of the network is damaged, leading to attenuation of the signal, the difference between the signals will show the unique reflection of the nodes and end users affected, thereby indicating where the damage occurs.
Even if more than one combination of gratings is located at exactly the same distance from the station- end 1 and are affected by the damage, the computer 38 can be programmed to identify which combinations of gratings the reflected signal corresponds to be comparing it with the signal that would be reflected
from different permutations of Bragg grating combinations.