Method and apparatus providing transient control in optical add-drop nodes

[0001] The invention relates generally to the field of communications and, more specifically, to a method and apparatus for mitigating the effect of transients in optically amplified transmission systems.

[0002] Optical add-drop nodes such as optical add-drop multiplexers (OADMs) form a key functional element in dense wavelength-division multiplexed (DWDM) optical fiber networks. Optical amplifiers such as erbium doped fiber amplifiers, semi-conductor optical amplifiers, Raman amplifiers and the like are commonly deployed within such networks to overcome the attenuation of transmission fibers between nodes or the attenuation of components within network elements. The optical amplifiers typically operate in a saturated regime where, the total output power depends sub-linearly on the number of input channels or is essentially constant. If the number of input channels (i.e., wavelengths) passing through such an amplifier is suddenly reduced, the optical power of the remaining channels will be increased, potentially to the level that degrades optical quality (e.g., measured by a bit error rate) of these remaining channels. For example, in the case of an OADM node receiving a plurality of optical channels and adding a single channel, a fiber-cut upstream of the OADM node (or disconnected OADM input) will cause the sudden elimination of optical energy associated with the received through channels, while the remaining add channel will receive most of the total optical amplifier power that had previously been distributed among all the channels exiting the OADM. The sudden power change will have detrimental immediate effects not only on the added channel, but also on the stability of all the network elements downstream from the fiber cut. The optical power of the surviving channels can, in principle, be adjusted back to the desired value by re-adjusting the pump conditions of all optical amplifiers. However, in a large network with many WDM channels it is a very significant challenge to accomplish this in a time period sufficiently short to avoid noticeable effects on the network operation.

[0003] These and other deficiencies of the prior art are addressed by the present invention of a method and apparatus for determining if an optical input signal to an OADM has been interrupted and responsively replacing an interrupted optical input signal with a replacement optical signal having a similar optical profile, thereby suppressing transient power changes in uninterrupted add channels without requiring any changes to the optical amplifier operating parameters.

[0004] A method according to one embodiment of the invention comprises monitoring at least some of a plurality of optical signals to determine if the monitored optical signals have been interrupted, where the optical signals are adapted for use by an optical amplifier and, in response to the interruption of the monitored optical signals, replacing the monitored optical signals with other optical signals.

[0005] The teachings of the present invention can be readily understood by considering the following detailed description in conjunction with the accompanying drawings, in which:

[0006] FIG. 1 depicts a high level block diagram of an optical add-drop multiplexer (OADM) according to an embodiment of the present invention;

[0007] FIG. 2 depicts a high level block diagram of a controller suitable for use in the OADM of FIG. 1;

[0008] FIG. 3 depicts a flow diagram of a method according to an embodiment of the present invention;

[0009] FIGS. 4-5 depict a high level block diagram of alternate embodiments of a pair of OADMs within a bi-directional traffic environment; and

[0010] FIG. 6 depicts a synchronous optical network (SONET) useful in understanding an embodiment to the present invention.

[0011] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures.

[0012] The subject invention will be primarily described within the context of an optical add-drop multiplexer (OADM) which may be used in wavelength-division multiplexed (WDM) and dense WDM (DWDM) optical communications systems carrying various traffic types (e.g., SONET). However, it will be appreciated by those skilled in the art that the invention may be advantageously employed in any optical communications system in which it is desirable to avoid the need for fast optical amplifier response to transient conditions such as caused by, for example, a severed optical fiber.

[0013] FIG. 1 depicts a high level block diagram of an optical add-drop multiplexer (OADM) according to an embodiment of the present invention. Specifically, the OADM 100 of FIG. 1 comprises an input selector 105, a demultiplexer 110, a switching fabric 120, a multiplexer 130, an optical amplifier 140 and a controller 150. The OADM 100 of FIG. 1 receives an input signal IN, illustratively a WDM or DWDM optical signal comprising N optical signals having respective wavelengths ₁-_N and transported via an optical fiber and responsively produces a corresponding output signal OUT. The OADM 100 of FIG. 1 operates to controllably pass through, insert (add) and/or extract (drop) optical communications signals having specified wavelengths to implement thereby the known add-drop function. The OADM 100 is controlled by a network manager or element manager (not shown) via a control signal CONTROL.

[0014] The input selector 105, in response to a control signal SW produced by the controller 150, responsively couples one of the input signal IN and a replacement signal REP to an input of the demultiplexer 110. The replacement signal REP comprises a signal having optical characteristics similar to those of the input signal IN. Specifically, in the embodiment of FIG. 1 the replacement signal REP comprises a DWDM signal comprising the same number of multiplexed channels (i.e., ₁-_N) where each of the multiplexed channels has optical power characteristics similar to those of the channels they are replacing. In this manner, channel equalization and amplification characteristics used by the optical amplifier 140 are acceptable for processing optical signals derived from either the input signal N or replacement signal REP. In this manner, transitioning from the input signal N to replacement signals REP will not induce a large transient adjustment problem for the optical amplifier 140, and in particular will not cause a significant power excursion of any channels added via the switch fabric 120.

[0015] The demultiplexer 110 demultiplexes the selected DWDM input signal IN (or REP) to extract therefrom a plurality of optical signals having respective wavelengths denoted as ₁, ₂, and so on up to _N, which optical signals (channels) are coupled to respective inputs of the switch fabric 120. The switching fabric 120 also receives up to N add wavelengths or channels, denoted as ADD at respective inputs. The switch fabric 120, illustratively an MM switch fabric, provides at its output ports up to N respective signals or channels selected from the selected input channels IN/REP (pass through mode) and/or additional channels ADD. Additionally, the switch fabric 120 couples to a second group of output ports up to N channels to be dropped.

[0016] The optical signals selected for propagation as part of the multiplexed output signal OUT are coupled from the switch fabric 120 to the multiplexer 130, where they are multiplexed to form an optical signal corresponding in form to the input signal IN (i.e., a DWDM optical signal comprising up to N wavelengths). The multiplexed signal produced by the multiplexer 130 is then amplified by optical amplifier 140 to produce the output signal OUT.

[0017] The optical amplifier may comprise an erbium doped fiber amplifier, semi-conductor optical amplifier, Raman amplifier and the like. Such optical amplifiers typically include various modes of operation adapted to insure appropriate amplification for each of the wavelengths within a DWDM optical signal. The optical amplifier 140 of FIG. 1 may perform various equalization calculations, spectral analysis calculations, spectral analysis calculations, individual and aggregate power calculations and the like among the various wavelengths (channels) within the multiplexed DWDM optical signal. It is noted that such operations are typically not performed rapidly enough to avoid transient errors caused by, for example, an upstream cut fiber condition. However, the controller 150 operates to cause the input selector 105 to provide the replacement signal REP in place of the input signal IN should such a cut fiber condition (or OADM input open condition) be detected. Thus, the relatively slow adaptations made by the optical amplifier 140 to insure appropriate amplification are sufficient to keep the output signal OUT valid for at least those wavelengths including valid data.

[0018] The controller 150 controls various operations within the OADM 100 of FIG. 1. For example, in one embodiment the switch fabric 120 includes an input monitor 125 that monitors various parameters associated with the input signals (₁-_N) to the switch fabric 120. The input monitor 125 is capable of determining that the power level and/or data within the received wavelengths indicates that an upstream cut fiber condition or OADM open input condition exists. In response to the determination of such condition, the controller 150 causes the input selector switch 105 to route the replacement optical signal to the demultiplexer 110 instead of the input signal IN. Contemporaneously, the controller 150 causes the optical amplifier 140 to begin the equalization and power distribution calibration processes used to equalize optical energy between channels within the DWDM signal it is processing. An embodiment of the controller 150 will be discussed in more detail below with respect to FIG. 2.

[0019] In one embodiment of the invention, the input monitor 125 monitors a plurality of the individual wavelengths ₁-_N and determines thereby whether a condition associated with one of a fiber cut or disconnected input is present. In one embodiment of the invention, in response to this condition the controller 150 causes all of the wavelengths within the input signal IN to be replaced by selecting, using the input selector 105, a corresponding DWDM signal including replacement wavelength channels. In an alternate embodiment of the invention, the replacement signals REP are inserted via the add input signals. In this embodiment of the invention, individual wavelengths ₁-_N within the input signal IN may be selectively replaced by adding corresponding replacement signals and dropping those signals determined by the input monitor 125 to be defective or otherwise inappropriate.

[0020] In one embodiment of the invention, a portion (e.g., 1% power) of the input signal IN is sampled by a power detector (PD) 102 comprising an optical splitter and a photo detector to produce thereby a power detection signal PD. The power detection signal PD is coupled to the controller 150, which responsively determines if the power detection signal is above a threshold level indicative of a non-severed upstream fiber condition (or a non-open OADM input condition). If the power level is below the threshold level, then an error condition is assumed and the switch 105 adapted accordingly. It is noted that the power detector 102 may be used in any of the other embodiments of the invention discussed herein with respect to FIGS. 1-6.

[0021] In one embodiment of the invention, the optical amplifier A is not used to amplify the DWDM signal provided by the multiplexer 130. Rather, each of the individual wavelengths ₁-_N provided by the switch fabric 120 are individually amplified by respective amplifiers (not shown). In this embodiment, the controller 150 communicates with each of the individual wavelength amplifiers to effect thereby an initiation of a recalibration process. It is also noted that where individual switch fabric outputs are amplified, the controller 150 may selectively cause only those optical amplifiers operative to process replacement signals to enter a calibration process.

[0022] In one embodiment of the invention, the power detector 102, input monitor 125 and related logic within the controller 150 may be implemented as a stand-alone detector/control function in which detection of a power level below an appropriate threshold level causes a change in state of a control signal (e.g., SW), thereby cause a replacement of an input WDM or output WDM signal with a replacement signal. In any event, the signal provided at the output will comprise a replacement signal in the case of an inappropriate input signal power level. In embodiments discussed below (e.g., FIGS. 2-6) the replacement signal may comprise a looped back signal provided by a physically proximate OADM servicing a second traffic path (e.g., in and East/West traffic path topology).

[0023] FIG. 2 depicts a high level block diagram of a controller suitable for use in the OADM of FIG. 1. Specifically, the controller 200 of FIG. 2 is suitable for implementing the controller 150 within the OADM 100 of FIG. 1.

[0024] The controller 200 of FIG. 2 comprises a processor 230 as well as memory 240 for storing various control programs and other programs 242. The processor 230 cooperates with conventional support circuitry 220 such as power supplies, clock circuits, cache memory and the like as well as circuits that assist in executing the software routines stored in the memory 340. As such, it is contemplated that some of the steps discussed herein as software processes may be implemented within hardware, for example as circuitry that cooperates with the processor 230 to perform various steps. The controller 200 also contains input/output (I/O) circuitry 210 that forms an interface between the various functional elements communicating with the controller 200.

[0025] Although the controller 200 of FIG. 2 is depicted as a general purpose computer that is programmed to perform various control functions in accordance with the present invention, the invention can be implemented in hardware as, for example, an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). As such, the process steps described herein are intended to be broadly interpreted as being equivalently performed by hardware, software, or a combination thereof.

[0026] The memory 240 is used to store various software instructions including those useful in implementing different embodiments of the present invention, such as the steps discussed below with respect to FIG. 3. For example, in one embodiment of the invention the switch fabric 120 includes an input monitor 125 that monitors some or all of the signals (₁ through _N) received by the switch fabric 120. In response to a loss of signal indicative of, for example, a severed upstream fiber optic cable or a disconnected OADM input, the input monitor 125 provides indicia of the options of one or more input signals ₁ through _N to the controller 150. The controller 150 responsively causes one or more replacement signals REPLACE to be inserted in lieu of the missing input signals such that the optical amplifiers 130 will continue to correctly process the output signal OUT.

[0027] FIG. 3 depicts a flow diagram of a method according to an embodiment of the present invention. Specifically, FIG. 3 depicts a flow diagram of a method 300 suitable for use in the OADM 100 of FIG. 1 for identifying a loss of input signal condition and responsively providing a replacement signal such that optical amplification of one or more added signals is performed in an appropriate manner.

[0028] The method 300 of FIG. 3 is entered at step 310, where the input signals (i.e., ₁ through _N) are monitored. Referring to box 315, the group power of the input signals (i.e., the aggregate powers of signals ₁ through _N) may be monitored, individual input signals may be monitored or other monitoring techniques may be employed. At step 320, a query is made as to whether the monitored input signals are in a good status (i.e., no apparent fiber cut or OADM input disconnect condition). If the monitored input signals appear satisfactory, the method continues to monitor and query at steps 310 and 320. In the event of a non-satisfactory condition, the method 300 proceeds to step 330.

[0029] At step 330, a determination is made as to whether all or some of the input signals are unsatisfactory. Referring to box 335, this determination may be made with respect to the entirety of the input signals, individual groups of input signals, or each input signal individually, as well as other techniques.

[0030] At step 340, at least the unsatisfactory input signal or signals are replaced by alternative or replacement optical signals having at last a similar optical power and spectral characteristics. In the embodiment 100 of FIG. 1, the input monitor 125 and controller 150 operate to replace all of the wavelengths within the input signal IN with corresponding wavelengths within a DWDM replacement signal REP using the input selector switch 105. Referring to box 345, the replacement signals may comprise one or more dummy signals, loop back signals from another OADM (or portions of the OADM 100 of FIG. 1 that are not shown) or other signal sources. In any case, the replacement optical signals comprise optical signals having characteristics similar to the missing optical signals such that the optical amplifiers used within the OADM are not subjected to a fast transient change condition. In this manner, those optical signals added by the OADM 100 of FIG. 1 are appropriately amplified and tend to avoid any increase in bit error rate (BER) or other quality of service (QoS) degradation due to the sudden loss of received input signal IN.

[0031] At step 350, software adjustment procedures for the optical amplification stage are initiated. That is, in one embodiment the controller 128 of the switch fabric 120 indicates to the optical amplifier stage that certain input signals are not present and that the optical amplifier should perform various readjustment procedures such as channel equalization and the like as performed within standard optical amplifier adjustment processes.

[0032] The method 300 of FIG. 3 is primarily described within the context of a particular OADM topology (i.e., that described above with respect to FIG. 1). However, modifications to the method 300 may be made by those skilled in the art and informed by the teachings of the present invention to enable similar methodology within the context of the OADMs of FIGS. 4-6.

[0033] The present invention is especially well suited to optical backbone networks (as opposed to access networks), which are typically symmetrical and bi-directional networks providing an East connection to complement a West connection having a substantially equal bandwidth. Backbone networks are typically implemented as two fiber links between nodes using one fiber-optic cable (i.e., a cable including at least two fibers). The architecture and configuration of an OADM node in this environment may be symmetric or asymmetric with respect to add-channels, drop-channels and through-channels, though a symmetric configuration is more common. Therefore, through-traffic in the west direction will have the same channel count and spectral distribution as through-traffic in the east direction. The inventors have determined that the through channels in either direction tend to be equalized to a desired power level at the OADM, and that such redirected or replacement channels provide an excellent substitute (with respect to optical amplifier performance considerations) for lost channels, such that there is no immediate need to reconfigure any optical amplifiers within the network.

[0034] FIG. 4 depicts a high level block diagram of a pair of OADMs within a bi-directional traffic environment. In the embodiment of FIG. 4, to suppress optical amplifier transients and stabilize the optical layer downstream of a cable cut, the upstream traffic is looped back and inserted into the downstream direction via a fast optical selector switch. That is, the replacement channels REP discussed above with respect to FIG. 1 are derived from the opposite direction traffic channels.

[0035] Specifically, referring to FIG. 4, a pair of OADMs 100_W and 100_E are shown. Each of the East 100_E and West 100_W OADMs includes a respective input switch 105, a respective add-drop multiplexer core (110-150) and a respective splitter 107. Each of the OADMs receives a respective input DWDM signal IN at a first input of its respective selector switch 105, and a split portion of the opposed OADM output at a second input of its respective selector switch 105. The output of each selector switch is coupled to the input of its respective OADM core.

[0036] Each ADM core comprises the functionality of the demultiplexer 110, switch fabric 120, multiplexer 130, optical amplifier 140 and controller 150 described above with respect to FIG. 1. The output signal produced by each ADM core is coupled to its respective splitter 107, which splits the output signal into two reduced power (e.g., 3 dB) signals. The first split signal is provided as a respective output signal OUT for the next node in the network, while the second split signal REP is provided as the second input signal (i.e., the replacement signal) to the switch 105 of the opposite direction OADM.

[0037] In operation, referring to the East signal path, an input signal IN_E comprises a DWDM signal including up to N individual wavelengths. One of the East input signal IN_E and a portion of the West output signal OUT_W is selectively coupled to the East ADM core via the East OADM input switch 105_E in response to a switch control signal SW_E provided by the controller 150 within the East ADM core. The East ADM core responsively adds up to N signals ADD_E and drops up to N signals DROP_E as previously discussed and produces a DWDM output signal which is coupled to the East splitter 107_E.

[0038] Similarly, referring to the West signal path, an input signal IN_W comprises a DWDM signal including up to N individual wavelengths. One of the West input signal IN_W and the portion of the East output signal OUT_E is selectively coupled to the West ADM core via the West OADM input switch 105_W in response to a switch control signal SW_W provided by the controller 150 within the West ADM core. The West ADM core responsively adds up to N signals ADD_W and drops up to N signals DROP_W as previously discussed and produces a DWDM output signal which is coupled to the West splitter 107_W.

[0039] In one embodiment of the invention, the portion of the output signals diverted to the input of the opposing ADM cores is scrambled by either an East 410_E or West 410_W scrambler. In this manner, the physical layer error of a fiber cut or open input is correctly interpreted as an error by a higher logical layer of, for example, a SONET system. That is, the rerouted data will likely not cause a loss of signal (LOS) error within the system, even thought the data is invalid. To induce an error (since the rerouted or replacement data is not valid to the link(s) into which it is routed), the scrambling of such data will likely cause at least loss of frame (LOF) error.

[0040] FIG. 5 depicts a high-level block diagram of a pair of OADMs within a bi-directional traffic environment. Specifically, the OADM pair 500 of FIG. 5 comprises a modified version of the OADM pair 400 of FIG. 4. That is, each of an East 100_E and West 100_W OADMs of FIG. 5 includes a respective output switch 105 (rather than an input switch), a respective add-drop multiplexer core (110-150), a respective input splitter 107 (rather than an output splitter) and an optional scrambler 510. In FIG. 5, the position of the switches 105 and splitters 107 is reversed from that of FIG. 4 such that the looped back replacement signal(s) comprises and input signal IN rather than an output signal OUT.

[0041] In operation, referring to the East signal path, an input signal IN_E comprises a DWDM signal including up to N individual wavelengths. The East splitter 107_E provides one portion (e.g., half the power) of the input signal IN_E to the East ADM core and the other portion to an input of the West OADM output switch 105_W. The East ADM core responsively adds up to N signals ADD_E and drops up to N signals DROP_E as previously discussed and produces a DWDM output signal which is coupled to the East output switch 105_E. In response to a switch control signal SW_E provided by the controller 150 within the East ADM core, the East output switch 105_E couples one or the output of the East ADM core and a portion of the West input signal IN_W to its output as the east output signal OUT_E. The West signal path works in a similar manner.

[0042] In an alternate embodiment of the OADM pairs of FIGS. 4 and 5, the 12 switches and splitter used for each of the East and West paths may be replaced with a 22 switch. Specifically, rather than splitting the input (or output) optical signals, a 22 switch is used to nominally route each respective input signal IN through its respective ADM core and to its respective output signal path out. In case of a fiber cut condition, the 22 switch operates to route active data signals as output signals OUT.

[0043] Within the context of a synchronous optical network (SONET) system, such as a ring network, a loss of signal (LOS) on the line side of an OADM does not automatically result in a LOS for SONET client equipment (i.e., the client side laser and modulator are not switched off immediately). Where an automatic protection switch (APS) is used, it is especially important to give error signaling to the network manager and such that the network traffic may be routed via an alternate path.

[0044] FIG. 6 depicts a synchronous optical network (SONET) useful in understanding an embodiment to the present invention. Specifically, FIG. 6A depicts a SONET at work comprising a first SONET network element 610 communicating with a second SONET network element 670 via East and West communications paths including, in the order named, a first optical translator (OT) that creates a suitable DWDM wavelenth signal, a plurality of optical ADMs 630-650 and a second OT. FIG. 6B depicts a fiber cut occurring between OADMs 640 and 650. In this instance, it is noted that the East bound signal provided by SONET element 610 is guided back towards SONET element 610 via a loop back element 645 within OADM 640. Similarly, a West bound signal provided by SONET element 670 is routed back to SONET element 670 via a loop back element 655 within OADM 650.

[0045] The system 600 of FIG. 6 uses a two step process to handle the fiber cut. As a first step, the optical amplifier transient avoidance scheme discussed above with respect to FIGS. 1-5 is employed, thereby ensuring appropriate optical amplification within the various OADMs and OTs as the various optical amplifiers avoid transient-induced error and refine their operation under software control. In this manner, signals added and dropped between OADMs having paths that avoid the fiber cut are still valid and usable. Optionally, by scrambling loop back channels, loss of frame (LOF) errors may be generated to inform the SONET management system that a major signal degradation has occurred, and thereby to induce automatic protection switching at the SONET layer.

[0046] The scramblers discussed above with respect to the various figures may be implemented by using a Lyot depolarizer which comprises, in part, a high second order PMD element. This may be implemented using approximately 80 meters of PM fiber in two sections spliced under 45 degree angles. An alternate embodiment comprises an all pass filter loop, in which a 3DBM coupler has one output spliced back into its inputs.

[0047] Although various embodiments that incorporate the teachings of the present invention have been shown and described in detail herein, those skilled in the art can readily devise many other varied embodiments that still incorporate these teachings.