Patent Application
20060291859
This invention relates to a computer architecture that provides for timely automated end-to-end wavelength provisioning and turn-up of an optical transport system and more particularly to the measuring and automatically correcting for the dispersion, attenuation, and other characteristics of optical fiber, and for mapping and adjusting optical amplifiers, optical spectrum analyzers, regenerators, and other components in the optical transport system.
A goal of many modern long haul optical transport systems is to provide for the efficient transmission of large volumes of voice and data traffic over trans-continental distances at low costs. Various methods of achieving these goals include time division multiplexing (TDM) and wavelength division multiplexing (WDM). In time division multiplexed systems, data streams comprised of short pulses of light are interleaved in the time domain to achieve high spectral efficiency, high data rate transport. In wavelength division multiplexed systems, data streams comprised of short pulses of light of different carrier frequencies, or equivalently wavelength, are co-propagated in the same fiber to achieve high spectral efficiency, high data rate transport.
The transmission medium of these systems is typically optical fiber. In addition there is a transmitter and a receiver. The transmitter typically includes a semiconductor diode laser, and supporting electronics. The laser may be directly modulated with a data train with an advantage of low cost, and a disadvantage of low reach and capacity performance. After binary modulation, a high bit may be transmitted as an optical signal level with more power than the optical signal level in a low bit. Often, the optical signal level in a low bit is engineered to be equal to, or approximately equal to zero. In addition to binary modulation, the data can be transmitted with multiple levels, although in current optical transport systems, a two level binary modulation scheme is predominantly employed.
Typical long haul optical transport dense wavelength division multiplexed (DWDM) systems transmit 40 to 80 10 Gbps (gigabit per second) channels across distances of 3000 to 6000 km in a single 30 nm spectral band. A duplex optical transport system is one in which traffic is both transmitted and received between parties at opposite ends of the link. In current DWDM long haul transport systems transmitters different channels operating at distinct carrier frequencies are multiplexed using a multiplexer. Such multiplexers may be implemented using array waveguide (AWG) technology or thin film technology, or a variety of other technologies. After multiplexing, the optical signals are coupled into the transport fiber for transmission to the receiving end of the link.
At the receiving end of the link, the optical channels are de-multiplexed using a de-multiplexer. Such de-multiplexers may be implemented using array waveguide (AWG) technology or thin film technology, or a variety of other technologies. Each channel is then optically coupled to separate optical receivers. The optical receiver is typically comprised of a semiconductor photodetector and accompanying electronics.
The total link distance may in today's optical transport systems be two different cities separated by continental distances, from 1000 km to 6000 km, for example. To successfully bridge these distances with sufficient optical signal power relative to noise, the total fiber distance is separated into fiber spans, and the optical signal is periodically amplified using an in-line optical amplifier after each fiber span. Typical fiber span distances between optical amplifiers are 50-100 km. Thus, for example, 30 100 km spans would be used to transmit optical signals between points 3000 km apart. Examples of in-line optical amplifiers include erbium doped fiber amplifiers (EDFAs) and semiconductor optical amplifiers (SOAs).
A duplex optical transport system is one in which voice and data traffic is both transmitted and received between parties at opposite end of the link. There are several architectures that support duplex operation in fiber optical transport systems. Each suffers from some limitation.
Long-distance transmission of data at high speed is increasingly dependent on optical fibers and efficient state-of-the-art light wave transmission systems. These transmission systems typically include, the optical fiber itself and among other things, end terminals, programmable optical amplifiers, optical spectrum analyzers, regenerators, and to be generically referred to as components in the transmission system. Besides transmitting customer data at high data rates, fiberoptic transmission systems that span large geographic distances must be able to send maintenance information or telemetry from one part of the system to another. In traditional regenerated systems, maintenance information is added as overhead to customer data which is transmitted on the fiber. Access to the overhead data is allowed at each regenerator site. The data can be provided to a local user or passed through the data collection station unchanged. In systems employing optical amplifiers, maintenance information is transmitted on a separate wavelength on the same optical fiber. Access to customer data is usually not possible because it is transmitted on a different wavelength.
Service providers who assemble and maintain light wave transmission systems, and network providers who manage networks of such optical transport systems are under increasing pressure to reduce costs. Equipment costs, maintenance costs and the cost of personnel to operate the transmission system all must be kept at a minimum in order for service providers to remain as competitive as possible. In current optical transport systems it is necessary to manually find and track the very large number of components, subsystems and modules that comprise the optical transport system. In order to manage, maintain and grow the optical transport system it is necessary to have an up-to-date, correct inventory of the various components, subsystems and modules that comprise the optical transport system. This inventory comprises a complete and correct listing of the type and quantity of the various components, subsystems and modules that comprise the optical transport systems. Human error combined with changing configurations currently lead to inefficient incorrect inventory process. Consequently there is a need for an optical transport system that automatically discovers the various components, subsystems and modules that comprise the optical transport systems. Such a capability is referred to as autodiscovery. Similarly, there is a need for an optical transport system that automatically inventories the various components, subsystems and modules that comprise the optical transport systems. Such a capability is referred to as autoinventory. In order to manage, maintain and grow the optical transport system it is additionally necessary to have an up-to-date, correct topology of the various components, subsystems and modules that comprise the optical transport system. This comprises a complete and correct arrangement of the connections and relative locations of the various components, subsystems and modules that comprise the optical transport systems. Human error combined with changing configurations currently lead to inefficient incorrect topology process. Consequently there is a need for an optical transport system that automatically determines the topology of the various components, subsystems and modules that comprise the optical transport systems. Such a capability is referred to as autotopology.
Provisioning a new channel to provide duplex operation between terminals that may be located in different cities is currently a tedious, labor intensive process. Consequently there is a need for an optical transport system that automatically provisions a new channel to provide duplex operation between distant terminals. Such a capability is referred to as auto-provisioning.
Certain physical characteristics of each optical fiber and the components used in the transportation system impose limitations on the data rate and transmission distance of any light wave transmission system. The performance of the components used, depends on, among other factors, the settings of the components, for example the gain setting on an in line amplifier. The performance of the fiber is based mainly on attenuation and chromatic dispersion properties of the fiber. The physical characteristics of each optical fiber cannot be readily adjusted for optimal performance. However, the components used in the transport system can be adjusted for optimal performance.
Attenuation is a primary limitation in the design of a light wave system. Attenuation is the loss of optical energy which is caused by impurities in the fiber, scattering and other phenomena. Loss of optical energy negatively affects information transmission by reducing the amplitude of the transmitted signal. Attenuation can be corrected, within limits, by regeneration or the addition of optical amplifiers along the path of the optical fiber to amplify the signal to replace lost optical energy. Further, the relative attenuation of a channel operating at wavelength as compared to a channel operating at a second wavelength is also important in an optical transport system. Different channels may see different relative attenuations because of the different spectral filtering of the different components, subsystems and modules that comprise the optical transport system.
Another important limitation is chromatic dispersion. Chromatic dispersion, or more precisely, group velocity dispersion, is caused by a variation in the group velocity in a fiber with changes in optical frequency. Chromatic dispersion causes the spreading of pulses in a light wave signal. Pulse spreading leads to timing problems and increased error rates.
In order to maximize the data transmission distance and speed, it is necessary to measure or characterize attenuation and chromatic dispersion for each fiber. After characterization, it is also necessary to correct for these limitations as far as is possible to maximize data transmission speed and distance. In the prior art, characterization takes place manually requiring a labor-intensive process of measuring each parameter at the time of installation. Consequently, there is a need for an optical transport system that automatically characterizes the physical properties of the fiber plant and other components, subsystems and modules that comprise the optical transport system. Such a capability is referred to as autocharacterization. After characterization, the system operator must currently manually correct for attenuation and chromatic dispersion by setting certain parameters on each amplifier specification. Consequently, there is a need for an optical transport system that automatically corrects the physical properties of the fiber plant and other components, subsystems and modules that comprise the optical transport system. Such a capability is referred to as autotuning. Autotuning will automatically adjust the control parameters of the optical transport system to account for changes in the number and character of the channels that are operational at a given time.
Attenuation is corrected for in the prior art by adjusting the gain and tilt of each amplifier in the system. The gain of each amplifier causes a linear difference in amplification over the range of wavelengths amplified. The linear difference is known as tilt. Additionally, stimulated Raman scatterings can cause significant power tilt in the fiber which is further amplified. Skill and experience of the operator are required to account for and correctly adjust each component of the transmission system for maximum efficiency at each wavelength and to compensate for tilt as much as possible. However, this prior art procedure is prone to human error and inconsistency.
Chromatic dispersion is corrected for in the prior art by manually measuring the dispersion which has taken place in the signal at any of the amplifier locations. “Compensators” are then fabricated which compensate for the dispersion measured by the operator and installed to correct the dispersion. Fabrication of compensators is time consuming and can take up to several weeks to complete. The process is also so specific that minor changes in fiber characteristics due to repairs and fiber aging can require a repeat of the process. Also, because each fiber is typically measured only once, changes in the system are largely unaccounted for during system operation, reducing the efficiency of the system and the maximum data rate over time.
Certain prior art systems have attempted to address these problems with varying success.
U.S. Pat. No. 5,914,794 to Fee, et al., entitled “Method of And Apparatus For Detecting and Reporting Faults In An All-Optical Communications System”, discloses a method and system for reporting and detecting faults in an optical communications systems. A system monitors each optical fiber and uses the optical supervisory channel to detect and report faults. However, Fee does not disclose or suggest a way to eliminate the time consuming necessity for manual adjustment of the transmission system.
U.S. Pat. No. 5,225,922 to Chraplyvy, et al., entitled “Optical Transmission System Equalizer”, discloses an invention which selectively equalizes the optical gain or optical signal noise ratios of channels of a wavelength multiplexed optical transmission system. The optical output powers and the signal to noise ratios are selectively equalized by adjusting the optical input powers through a controller connected to an end terminal of the transmission system. However, the Chraplyvy system has drawbacks. It is difficult to apply on long-haul systems and does not account for dispersion characterization or measurement. Additionally, the Chraplyvy system does not provide for adjustment of amplifier ripple as opposed to amplifier gain.
U.S. Pat. No. 5,940,209 to Nguyen, entitled “Interactive Optical Fiber Amplifier, System And Method”, discloses a system and method for selectively amplifying an optical signal, depending upon the amplification needs of the system. This invention adjusts amplifier gains controlled by remote systems so that an optical path in the system can be changed and automatic amplification provided. Nguyen has drawbacks, however, in that it requires a reflected power detector which adds to equipment costs and complexity. Additionally, the Nguyen invention does not address automatically adjusting amplifiers for ripple or remotely sensing chromatic dispersion.
U.S. Pat. No. 5,737,118 to Sugaya, et al., entitled “Optical Amplifying Apparatus”, discloses a computer controlled optical amplifying apparatus which includes an optical amplifying unit including an optical amplifying controller which controls the amplifier. The controller disclosed by Sugaya monitors the status of the optical amplifier to report abnormal occurrences in relation to the relaxation time of the amplifier. The Sugaya invention, as with the other prior art inventions, does not address the need for the automatic characterization and correction for chromatic dispersion or continued operation of the amplifiers during a fault sequence.
U.S. Pat. No. 6,317,231 to Al-Salameh, et al., entitled “Optical Monitoring Apparatus And Method For Network Provisioning And Maintenance”, provides for elimination of optical to electrical and electrical to optical signal conversion for monitoring certain maintenance functions of an optical network. The network controller provided analyzes the values of the optical signals to determine certain fault conditions and monitors certain channel power requirements. Additionally, the invention provides for monitoring signal noise ratio, channel continuity and network provisioning. Al-Salameh does not provide for automatic detection of chromatic dispersion or automatic adjustment of amplifier gain or tuning of components in the transmission system to eliminate the need for human intervention to correct system faults.
U.S. Pat. No. 6,163,392 to Condict, entitled “Distributed Intelligence Wavelength Division Multiplexed Network”, provides for a fiber-optic communication network in which processors associated with each network element periodically transmit identification and status information to other processors in the network. In addition a service channel carries diagnostic and span topology information that can be transmitted through each span. While Condict does provide for information to be transmitted on a reserved wavelength, the information is limited to identification and status information and routing data. Condict does not disclose or suggest a system that provides for automatic detection of chromatic dispersion, automatic adjustment of amplifier gain or tuning of components in the transmission system to eliminate the need for human intervention to correct system faults.
U.S. Pat. No. 6,359,729 to Amoruso, entitled “Optical Communication System And Component Control Architectures And Methods”, provides for control of optical components and network management. A component controller is configured to receive element instructions from an element manager and provides work function instructions to one or more work function controllers. The work function controller controls and monitors the work function pursuant to work function instructions provided by the component controller. Element instructions and other system information can be transmitted through the optical system using either a dedicated service channel or a mixed data channel carrying both communication traffic and system information. Amoruso does not disclose or suggest a system that provides for automatic detection of chromatic dispersion or automatic adjustment of amplifier gain or tuning of components in the transmission system to eliminate the need for human intervention to correct system faults.
Prior art systems suffer from the limitation that external measurement devices must be used by the operator to measure and correct for attenuation and dispersion. Also, the tuning of various components in the transmission system must be manually done by a system's operator. A further limitation of prior art systems is the length of time to restart the system after system failure. In some cases tuning and characterizing and correcting for attenuation and dispersion can take hours or days using the methods of the prior art.
The present invention addresses these problems by allowing transmission systems to perform required fiber characterization and correction, and to initiate, control and tune various components in the transmission system to operate at a desired efficiency in an automated fashion, thereby reducing equipment and personnel cost and providing for higher quality long haul fiberoptic transmission systems. The improvements address the limitations in the art described above, and provide more efficient and economical operation of the optical transport system.
In one aspect of the invention, an optical transport system with autodiscovery capability is taught.
In another aspect of the invention, an optical transport system with autoinventory capability is taught.
In another aspect of the invention, an optical transport system with autotopology capability is taught.
In another aspect of this invention, an optical transport system with autoprovisioning capability is taught.
In another aspect of this invention, an optical transport system with autocharacterization capability is taught.
In another aspect of this invention, an optical transport system with autotuning capability is taught.
The present invention provides an improved fiber optic network architecture which allows automated tuning and fiber characterization and correction of attenuation and chromatic dispersion along the fiber span. It allows for efficient collection and transfer of information regarding components in the transmission system and an optical fiber's attenuation and chromatic dispersion through a network of management controller's at each optical amplifier stage along the fiberoptic transmission line.
Each controller in a segment communicates with the others and with end terminals to forward information about performance measurements and characteristics of the components and optical fiber in the transmission system on an ongoing basis. Communication takes place over a reserved wavelength known as the service channel. The characteristics of each component and fiber are continuously analyzed so that corrections can be made automatically and continuously.
The present invention is an improvement over the prior art because it allows for automatic and remote analysis of and correction for components in the transmission system and for attenuation and remote analysis of chromatic dispersion of the fiber used in the transmission system. Automatically detecting and analyzing chromatic dispersion, attenuation, and components in the transmission system largely eliminates the need for manual attention to the system as required by the prior art. Also the current invention allows the system to be self contained and greatly self correcting thereby drastically reducing the time required to recover for a system failure or initial system startup. The system contains the necessary scheme to bring an amplifier up to an active state. Also, the system can monitor the optical spectrum and adjust the spectrum to bring it into a given band.
A further improvement of this invention over the prior art is that it allows the fiberoptic transmission system to be flexible and responsive to changes in fiberoptic cables and component aging or failure. For instance, if a cable is damaged and needs to be replaced, the optical characteristics of the cable will change slightly. An additional change might be the addition of a cable link or the addition or deletion of another optical component which changes the transmission characteristics of the entire system. The present invention allows the fiberoptic transmission system to be automatically responsive to these changes.
A further improvement of this invention over the prior art is that it allows the fiberoptic transmission system to be flexible and responsive to changes in the amount and character of the traffic that is present on the optical transport system. A change in the amount of traffic on the optical system may be a newly deployed or provisioned channel. The present invention allows the fiberoptic transmission system to be automatically responsive to these changes.
A further advantage of the present invention is that it allows longer spans to be accomplished between regeneration stations or optical amplification stations. This is because of the more frequent and accurate compensation for fiber nonlinearities at each optical amplifier stage.
All of the above advantages allow a long haul fiberoptic system outfitted with the invention to be operated with lower maintenance cost and fewer personnel, and results in operational cost savings to the user.
The invention will be better understood from the following more detailed description taken in conjunction with the accompanying drawings.
While the making and using of various embodiments of the present invention are discussed in detail below, it should be appreciated that the present invention provides many applicable inventive concepts which can be embodied in a wide variety of specific contexts. The specific embodiments described herein are merely illustrative of specific ways to make and use the invention and do not delimit the scope of the invention.
In the descriptions that follow, like parts are marked throughout the specification and drawings with the same numerals, respectively. The drawing figures are not necessarily drawn to scale and certain figures may be shown in exaggerated or generalized form in the interest of clarity and conciseness. Reference of an A-Z signal or downstream direction means from the left side of the drawing to the right side of the drawing while Z-A or upstream means from the right side to the left side. Unless otherwise stated, signals propagate in an A-Z direction. The A-Z, Z-A, upstream or downstream designation is used for illustrative purposes only.
Hardware Architecture
Optical amplifier stations 104 are further comprised of optical amplifier cards 120, dispersion compensation cards 124 and management card 110. Optical amplifier station 104 may also include an optical spectrum analyzer card 132. Optical amplifier station 104 may also include a dynamic gain equalizer card 134. In optical amplifier station 104, optical amplifier cards 120, and dispersion compensation cards 124 are connected to management card 110 by an electronic data bus.
Terminal stations 102 and optical amplifier stations 104 are connected by fiber spans 122. Management card 110 in neighboring stations are connected by an optical service channel. In a preferred embodiment, said optical service channel comprises an optical data link that propagates on fiber span 122. More specifically in a preferred embodiment, said optical data link comprises a 1 Gbps Ethernet optical data link. In a preferred embodiment, said optical data link counter-propagates to the optical traffic moving between terminal stations 102.
In
In
In
In a preferred embodiment electronic memory 410 is realized with programmable memory integrated circuits and random access memory integrated circuits. In a preferred embodiment microprocessor 420 is a microprocessor integrated circuit such as a Pentium microprocessor from Intel. In a preferred embodiment, the microprocessor is operated using the Unix operating system. In a preferred embodiment integrative logic 424 is comprised of field programmable gate array integrated circuitry. In a preferred embodiment power block 440 is comprised of dc-dc power converter integrated circuitry. In a preferred embodiment, user interface 436 is realized through a variety of connections and supporting logic integrated circuitry. Examples of these connections realized in a preferred embodiment include an RJ-45 interface, or an infra-red LED port. These connections enable an external computer to interact with the microprocessor on the local management card 110, or any other management card connected through the optical service channel interface 434. Optical service channel interface 434 is realized in a preferred embodiment by a 1 Gbps Ethernet data link and the supporting logic integrated circuits, including a gigabit interface converter (GBIC). For additional range between adjacent stations, the optical service channel interface may be capable of decoding and encoding forward correcting codes (FEC) such a repeat code or parity checks as known in the prior art. The encoder in the preferred embodiment introduces a redundancy to the input data before transmission. The decoder uses this redundancy to reconstitute the input sequence even in the presence of transmission errors, provided the frequency of errors is less than a threshold defined as the power of the code. The decoder and encoder in other embodiments of the invention may also use a more powerful set of codes which is linear block codes, convolution codes or threshold and sequential codes. In the preferred embodiment, the decoder and encoder are incorporated into the GBIC, as is well known in the art. The communication protocol in a preferred embodiment is TCP/IP. TCP/IP protocol is well known in the art and will not be discussed further here. It should be understood that other operating systems and protocols can be utilized in the invention with equal success.
In a preferred embodiment the optical service channel propagates between adjacent stations on fiber strands 122 at a wavelength distinct from the terminal traffic. In a preferred embodiment the optical service channel counter-propagates to the terminal traffic.
To better understand how the autotuning process works, the architecture of the preferred embodiment will now be described. A high level graphical view of the preferred embodiment of the invention is shown in
The “stream” of the customer traffic data signal is indicated as A to Z; therefore, the amplifiers to the left in the drawing are “upstream” to those to the right, which are referred to as “downstream.” It should be understood that each segment 600 is duplicated for each direction of data flow. Only one direction is shown for simplicity of description.
This preferred embodiment of the architecture of segment 600 includes three upstream amplifier stations 610, one upstream optical amplifier and equalizer station 620, and one downstream optical amplifier and analyzer station 615. Upstream optical amplifier stations 610 and upstream optical amplifier and equalizer station 620 act as “slave” stations in the autotuning process. Downstream optical amplifier and analyzer station 615 acts as a “master” station in the autotuning process.
With respect to the autotuning software process optical amplifier station 610 comprises management card 110, optical amplifier card 120 for amplifying optical signals traveling in the A-Z direction, and dispersion compensation card 124 for compensating the dispersion in fiber spans 122 for optical signals traveling in the A-Z direction. Optical amplifier and equalizer station 620 comprises management card 110, optical amplifier card 120 for amplifying optical signal traveling in the A-Z direction, dispersion compensation card 124 for compensating the dispersion in fiber spans 122 for customer traffic optical signals traveling in the A-Z direction, and further comprises dynamic gain equalizer card 134 for equalizing the power or optical signal to noise ratio in each channel in the optical signal traveling in the A-Z direction. Optical amplifier and analyzer station 615 comprises management card 110, optical amplifier card 120 for amplifying customer traffic optical signal traveling in the A-Z direction, dispersion compensation card 124 for compensating the dispersion in fiber spans 122 for customer traffic optical signals traveling in the A-Z direction, and further comprises an optical spectrum analyzer 132 for measuring the power or optical signal to noise ratio in each channel in the customer traffic optical signal traveling in the A-Z direction.
The architecture of optical amplifier station 610, optical amplifier and analyzer station 615, and optical amplifier and equalizer station 620, are similar, and each may be understood from
Referring now to
In this preferred embodiment, circulator 705 is used to inject an optical service channel signal originating from management card 110 onto fiber span 122 for propagation to the adjacent upstream station. Circulator 750 is used to extract an optical service channel signal from fiber span 122 propagating from the adjacent downstream station. The optical service channel signal from circulator 750 is received at management card 110.
Optical amplifier station 610 will not include dynamic gain equalizer 720, and will not include third stage erbium doped fiber amplifier 745.
Optical amplifier and analyzer station 615 will necessarily include the optical spectrum analyzer 710.
First stage erbium doped fiber amplifier 725 and second stage fiber amplifier 740 are controlled by management card 110 to optimize the gain for the entire amplifier station. The first and second stages necessarily result in “tilt” or linear variance in amplification according to frequency. Each type of optical amplifier has a characteristic tilt. The amplifiers may have their tilt adjusted by systematically changing the gain for each wavelength amplifiers or through variable optical attenuator 795. The amplifiers also demonstrate gain “ripple”. “Ripple” is a nonlinear effect which results from the erbium gain shape in the two stages.
In a preferred embodiment, dispersion compensation module 715 may include a number of sub-modules in which various length spans of dispersion compensation fiber are contained. In this preferred embodiment, dispersion compensation module 715 also includes a programmable read only memory (“EPROM”). Upon installation of VDCM 715, the EPROM is programmed to contain the approximate value of the dispersion compensation contained in the dispersion compensation module. The EPROM may be read by management card 110 to determine the compensation available.
Management card 110 is responsible for initially setting the gain of first stage erbium doped fiber amplifier 725 and setting the gain of second stage erbium doped fiber amplifier 740 on system startup and reading the values of the dispersion compensation available in the EPROM. This information is stored in memory 410 and communicated, via TCP/IP protocol to master station 615. Management card 110 is also responsible for continually updating master station 615 with gain settings and dispersion compensation values periodically during operation of the network system 510. Additionally, management card 110 is responsible for implementing commands from master station 615 with regard to resetting the gain of first stage erbium doped fiber amplifier 725 and resetting the gain of second stage erbium doped fiber amplifier 740. This process will be described in detail later.
Software Architecture
Control sequence 815 initiates the autotuning process.
Topology discovery module 800 provides a fiber wavelength topology view of system 500 and provides status and configuration information for the key components required for proper function of system 500. Topology discovery module 800 also controls the discovery process and builds, propagates and maintains a database called topology tree. The topology tree database is stored in memory 410 of management card 110. In the preferred embodiment, the topology tree contains information required for autotuning by provisioning module 805 such as node type and components on the node as well as other information not related to autotuning such as node name given by the user and software versions being used in the various components of the node. The discovery process is based on open shortest path first (OSPF) topology propagating methods and contains logic to detect specific station changes. The module also can propagate information about changes throughout the network.
Provisioning module 805 contains the logic to support the provisioning of a logical circuit or optical channel from an endpoint. Provisioning module 805 contains the logic to turn up a circuit, or bring a circuit into service, and turn down a circuit, or remove a circuit from service. This includes the initiation and cessation of pseudo random binary sequence (“PRBS”) signals on the circuit. The logic used can determine primary and secondary service states. Primary service states include in service or out of service. Secondary services states include out of service due to testing or out of service due to tuning. While provisioning module 805 is present on all stations, provisioning module 805 only runs on end terminal stations 102.
Provisioning module 805 examines the topology tree created by topology discovery module 800 and stored in memory 410 of management card 110 and other functions such as validating channels and optical paths. Provisioning module 805 also tests various components to ensure they are in a workable order. In addition, provisioning module 805 initiates autotuning module 810.
Autotuning module 810 communicates with variable optical attenuators in multiplexer-demultiplexer cards, variable optical attenuators 795, management cards 110, optical spectrum analyzers 710, optical amplifiers 120, dynamic gain equalizers 134, etc., and determines if a station is a “slave” or “master”. Autotuning module 810 handles and tunes Raman amplifiers when they are necessary to compensate for higher span losses.
Autotuning module 810 contains the logic to execute tuning scenarios based upon topological conditions and as well as bringing up a wavelength or optical channel path between two end points. In addition, autotuning module 810 contains the logic to bring an optical device such as optical amplifier 120 into and out of service.
In order for system 500 to perform autotuning, provisioning module 805 will direct autotuning module 810 to turn up autoprovisioned wavelengths or manually provisioned wavelengths. To control provisioning and autotuning of optical channels, provisioning module 805 is dependent on the set of state and configuration data created by topology discovery module 800. The state and configuration data is stored both in a specific node's database and can be used by a local provisioning module 805 on the specific node where the database is located and in the topology tree database and therefore available to all network elements. In the alternative, the data is stored in a local configuration database in memory 410. From this information the system can recognize if the system has been previously tuned or if it is exercising an initial tune-up. Also, each station can automatically determine its function in the system.
The default behavior of system 500 is for segment 575 to be auto-tuned rather than manually tuned. This allows system 500 to be built and activated without having to be initially manually tuned.
Several triggers initiate control sequence 815 to initiate autotuning. For instance, the discovery process begins in a single AZ direction when the system is first turned on. Each station uses basic Ethernet protocol TCP/IP to determine what type of station it is and where the upstream and downstream neighbors are. This can be accomplished because on every station is a shelf containing at least one management card 110 and on each management card 110 is an identifier identifying the shelf type. Once the shelf type of the station is known, management card 110 does an inventory of the shelf to ensure all the components are in good working order. After management card 110 determines the inventory and shelf type, a modified OSPF signal is sent telling all management cards 110 in the system the shelf type of each station. Once all management cards 110 know what shelf type each station has, initial topology discovery is completed and topology tree is built and stored in memory 410 of each management card 110. A new topology database will not be built until a predefined triggering event takes place.
Other events that trigger control sequence 815 to initiate topology discovery module 800 and create a new topology database include change of state or topology, user command, an idle period expiration, insertion or activation of a new optics card module, such as optical amplifier 120, or optical spectrum analyzer 710, optical hardware change, such as a card reset or hardware failure, and changes of optical topology, such as adding or turning up a wavelength or dropping or turning down a wavelength. In addition, the system can periodically refresh itself to ensure it does not miss any topological changes due to temporary disconnects, packet loss, bugs, etc.
Failure of different components can be recognized by control sequence 815 by recognizing different system responses. If the topology changes because of a failure of an optical spectrum analyzer 710, then topology discovery module 800 is triggered and a new topology tree database is built. The failure will propagate through the topology as the next optical spectrum analyzer 710 downstream from the failed optical spectrum analyzer 710 begins receiving signals measurably different from the signals received before the failure. If the failed optical spectrum analyzer 710 is at a terminal station 102 there will be at most degraded tuning available through discrete devices on that final segment. In the event of an optical amplifier 120 failure, tuning will be degraded for that segment 575 by treating the station where the amplifier failed as if it were not adjustable. In the event of a dynamic gain equalizer 720 failure, tuning will significantly degrade for that segment and the only adjustments would be spectrum wide adjustments on all optical amplifiers 120.
Autotuning can be disabled on a per-segment or per station basis by the user. If a slave station is manually turned off, then master station must avoid sending tuning messages to it. If master station is manually turned off, this effectively turns off tuning on all stations on the segment except for the master station. Manual controls allow disabling in a single direction or disabling particular devices or parameters.
The user can explicitly disable autotuning by entering the appropriate commands into the Unix operating system. The user commands are communicated via TCP/IP protocol between management cards 110 in the network using a control data channel or management channel on the optical fiber. A user request to disable autotuning results in the disablement of autotuning module 810 of the segment if the station that receives the request is a master station, otherwise the user requested disablement results in the disablement of autotuning module 810 for the station for those parameters or channels that are manually tuned. Autotuning module 810 will autotune “around” the stations that have been disabled. In other words, adjustments will not be made to a disabled station.
Autotuning module 810 allows for fixed settings on first stage erbium doped fiber amplifier 725, second stage erbium doped fiber amplifier 740, dynamic gain equalizer 720, third stage erbium doped fiber amplifier 745, or other optical device and can tune as well as possible around any settings that have been manually fixed by the user or due to malfunction.
Autotuning module 810 provides events to notify other management subsystems of important activities of the tuning process and any information associated with those activities. These events may include, but are not limited to, tuning finished and tuning successes, failures and canceling, tuning started, wavelength turn up or turn down completed, unable to converge and/or span loss over user configurable thresholds.
All changes made by the autotuning process during normal operation of the system are limited to less than a maximum instantaneous change value, such as 0.25 dB's and some maximum change bandwidth such as 2 dB/sec. for 0.25 dB/change at 10 Hz. Any changes greater than the above may be broken into smaller sub-changes so as to avoid data corrupting excursions and the prior levels reaching terminal station 102. It is preferable to have as little time as possible between the recognition of an optical spectrum issue and initiating the autotuning sequence.
Any failures detected are reported to management card 110. Such failures should include, but are not limited to, excessive span loss, loss of an expected wavelength in the spectrum, excessive power tilt, inability to bring spectrum into an appropriate band, hardware or software failure of a tuning station or tuning communications failure. The method for detecting and reporting of such failures is known in the art and will not be discussed further.
If management card 110 receives a reported failure signal, management card 110 will initiate trigger control sequence 815 to initiate topology discovery module 800 to build a new topology tree database. If management card 110 cannot initiate trigger control sequence 815, then the reported failure signal will be transmitted to a node that can initiate trigger control sequence 815.
The autotuning tuning control system is comprised of several processes running on the line card and management cards. In the preferred embodiment, all such processes communicate via an interprocess communication (IPC) programming interface. Use of IPC programming interfaces are known in the art. A different set of programming interfaces that can handle many user requests at the same time and allows individual program processes to run concurrently in an operating system may also be used. IPC messages are handled both while idle and while tuning. These messages can cause tuning to start and stop at anytime. The messages are handled asynchronously in the same thread as tuning calculations discussed below. For example, the process may include waiting for a message, processing the message and making any necessary tuning changes and starting any necessary tuning tasks, then during any tuning tasks, process any further IPC messages, if necessary canceling the current tuning cycle in order to act on a new message, then going back and waiting for a message.
Autotuning module 810 performs the actual tuning algorithm using the topology tree database as a reference. In order to control the optics for tuning, master station communicates with management card 110 on master station 910 (as will be further described in relation to
As shown in
Management card 110 is the control point for all tuning operations. The autotuning software is stored on management card 110. Upon start up, management card 110 uses topology discovery module 800 and determines its station designation in system 500 which is dependant upon position in the system 500 and the equipment installed on each station where management card 110 is located. Following the determination, management card 110 determines if the station the controller operates on is either a “master station” or “slave station” (as will be further described later).
Below, various tuning procedures are discussed. Headings are used for reference purposes only and are meant to be not limiting. The tuning procedures used are implemented by management card 110 based on the autotuning software located in management card 110.
Initial Erbium Doped Fiber Amplifier Turn Up
The procedure for initially turning up optical amplifier 120 requires knowledge of the correct output power at previous amplifier 120 and approximate knowledge of the amplifier's noise. Previous amplifier 120 is the amplifier directly upstream from the current amplifier being turned up.
The steps for initially turning up an amplifier are shown in
Na=NFavg,linear*10−4.8*BW(nm) Equation 1
Wherein, NFavg,linear is the average noise figure of an erbium doped fiber amplifier, and BW is the noise bandwidth of second stage erbium doped fiber amplifier 740. In the preferred embodiment, the average noise figure of an erbium doped fiber amplifier is approximately 3.5 and the noise bandwidth is approximately 64 nm.
Next, the output power of second stage erbium doped fiber amplifier 740 is measured, step 1235. Then, if the output power of second stage erbium doped fiber amplifier 740 is approximately equal to the input power of first stage erbium doped fiber amplifier 725 plus the linear noise contribution parameter times the span loss, step 1240, tuning is complete, step 1245.
The linear noise contribution parameter is calculated by Equation 1. If the output power of second stage erbium doped fiber amplifier 740 is not approximately equal to the input power of first stage erbium doped fiber amplifier 725 plus the linear noise contribution parameter times the span loss, step 1240, then, variable optical attenuator 795 is adjusted by the calculated output power out of second stage erbium doped fiber amplifier 740 divided by the measured output power of second stage erbium doped fiber amplifier 740, step 1250 and the output power of second stage erbium doped fiber amplifier 740 is again measured.
DGE and Associated Variable Optical Attenuator Tuning
Regarding the tuning of dynamic gain equalizer 720, as well as third stage erbium doped fiber amplifier 745 and associated variable optical attenuator 795 and assuming that the noise in the inactive channels is suppressed by DGE 720, the gain of third stage erbium doped fiber amplifier 745 is adjusted by variable optical attenuator 795 such that third stage erbium doped fiber amplifier 745 output power is comprised of the signal power of inactive channels plus the accumulated noise of these channels plus the noise of the erbium doped fiber amplifier 745. As shown in
Pnoise=NFEDFA.avg*10(GEDFA.avg/10)*BWfilter(nm)*10−4.8 mW Equation 2
wherein Pnoise is the noise power per channel NFEDFA.avg is the average noise figure of an erbium doped fiber amplifier 740, GEDFA.avg is the average gain of erbium doped fiber amplifier 740, BWfilter is the channel bandwidth of DGE 720, step 1325. Typical values for the average noise figure of second stage erbium doped fiber amplifier 740 is approximately 3.5 to 5.4. The average gain is approximately 24 dB, the channel bandwidth of DGE 720 is approximately 0.6-0.64 nm and the noise power per channel is approximately 8.36-13.8 μW.
The required output power is calculated by the following equation:
PEDFA.out req=Nact*(Psig+Pnoise*nEDFA)+na*1018/10 Equation 3
wherein PEDFA.out.req is required output power, Nact is the number of active channels, Psig is the noise power per signal, Pnoise is the noise power per channel, #nEDFA is the number of erbium doped fiber amplifiers 740 in segment 575 and Na is linear noise contribution calculated by Equation 3, step 1330. The noise power per signal is approximately 104/10 mW.
Variable optical attenuator 795 is set to 2 dB, step 1335. The gain of third stage erbium doped fiber amplifier 745 is set to 18 dB, step 1340, and the output power of the third stage erbium doped fiber amplifier 745 is measured, step 1345. Step 1350 shows the tuning of the DGE 720 is finished if:
|(10*log10(PTcur/Pout.req))|<0.1 dB Equation 4
If the tuning of the DGE 720 is not finished, then the value of variable optical attenuator 795 is read, step 1360 and adjusted by 10 times the log of the output power of third stage erbium doped fiber amplifier 745 divided by the calculated required output power, step 1365. Then the output power of third stage erbium doped fiber amplifier 745 is read again, step 1345. If Equation 4 is true, step 1350, then tuning of the DGE 720 is finished, step 1355; if not, then the value of variable optical attenuator 795 is adjusted as above and the process is repeated.
Tuning Transmitter Variable Optical Attenuator
The procedure for tuning transmitter variable optical attenuator located on the multiplexer-demultiplexer card is shown in
Tuning Receiver Variable Optical Attenuator
The procedure for tuning receiver variable optical attenuator located on the multiplexer-demultiplexer card is shown in
Erbium Doped Fiber Amplifier Tuning
|(10*log10(PTcur/Pout.req))|<0.1 dB Equation 5
Wherein PTcur is the measured output power of erbium doped fiber amplifier 725 and Pout.req is the calculated required output of erbium doped fiber amplifier 725.
However, if Equation 5 is not true, then the current value of variable optical attenuator 795 is read, step 1630. Next, step 1635 sets a new variable optical attenuator setting calculated by:
10*log10(PD4cur/Pout.req) Equation 6
If the new variable optical attenuator setting is not less than zero, step 1640, then the tuning of erbium doped fiber amplifier 725 is completed, step 1625. If the new variable optical attenuator setting would be less than zero, step 1640, the variable optical attenuator loss is set to zero, step 1645, and the Raman gain is set to the current Raman gain minus the new variable optical attenuator setting after being adjusted, step 1650. Again, the output power of erbium doped fiber amplifier 725 and the input power of previous erbium doped fiber amplifier 725 are measured, steps 1605 and 1610, and the required output power of erbium doped fiber amplifier 725 is calculated, step 1615. If Equation 5 is true, step 1620, then the tuning of erbium doped fiber amplifier 725 is completed, step 1625, otherwise the process is repeated.
Amplifier Tuning Using OSA Readings
Prequired=Psig+Pnoise*(#nEDFA) Equation 7
The power of active channels is read from optical spectrum analyzer 710 where the bandwith is approximately 0.8 times the channel spacing, step 1720. Next, the average power of active channels is calculated by taking the average of the logarithmic power values in dB, step 1725. In step 1730, the required additional gain is calculated by the following equation:
additional gain=10*log10(Prequired/Pave) Equation 8
Wherein Pave is the calculated average power per channels. Next, in step 1735, the linear power tilt of active channels is calculated using the following equation:
where Npop is the number (count) of populated channels, nch is a channel number (1-40 or 1-80), Pch is the signal power (in dBm) measured in channel nch, and Nsystem is the number (count) of possible channels in the system (40 or 80). If the absolute value of the calculated tilt is less than 0.3 dB and if the absolute value of the required additional gain is less than 0.3 dB or if the maximum iteration count has been reached, step 1740, then tuning of amplifier 120 is completed and optical amplifier 120 is properly tuned, step 1745. Preferably the maximum iteration count is 500 but may be between 50-5000. If in step 1740, the absolute value of the calculated tilt is not less than 0.3 dB or the calculated value of the required additional gain is less than 0.3 dB and the maximum iteration count has not been reached, then the number of amplifiers 120 in segment 575 is determined, step 1750. A new tilt is calculated by dividing the old tilt by the number of amplifiers 120 in current segment 575, step 1755. A new additional gain is calculated by dividing the old additional gain by the number of amplifiers 120 in current segment 575, step 1760. Next, for each amplifier 120 in the tuning segment, the first gain is reduced by the newly calculated tilt, step 1765. The variable optical attenuator 590 is adjusted by the newly calculated tilt minus the newly calculated additional gain, step 1770. To ensure the gain does not go over a pre-set limit such a gain alarm may be used, step 1775. The range of a pre-set limit is dependent on the type of amplifier used. For example, the gain alarm is triggered at 28 dB for a Corning pure gain 5500 EDFA with a gain range of 23-28 dB. The gain alarm would be monitored by management card 110. If the alarm is set and triggered, the first gain is reduced by 0.5 dB and the second gain is increased by 0.5 dB until the alarm is no longer set, step 1780. Then once the alarm is no longer triggered, the power of active channels from optical spectrum analyzer 710 is re-read, step 1720, and the process is repeated.
Full Chain Tuning
The process is repeated until signals are within an acceptable range or a hard limit of iterations is reached. In the preferred embodiment, the acceptable ranges are 0-1024. In the preferred embodiment, the hard limit of iterations is 256 but can be anywhere in the range of 50-5000. In addition, after each of these steps the final optical spectrum analyzer 710 reading from the previous step is reused in the next step in order to reduce unnecessary switching and time use.
DGE Tuning With OSA Readings
To tune dynamic gain equalizer 720 utilizing optical spectrum analyzer 710 readings, the signal spectrum is measured at optical spectrum analyzer 710, step 1800. Then signal channel power excursions are adjusted via the channel by channel dynamic gain equalizer 720, step 1805. A count of active channels and the required values for each channel is acquired from the topology tree stored in memory 410 of management card 110, step 1810. Then, the power of the active channels are read from optical spectrum analyzer 710 preferably at a bandwidth of approximately 0.8 times the channel spacing, step 1815. Next, for each active channel, the required channel power is calculated by using Equation 7, step 1820. Step 1825 and 1830 show tuning of dynamic gain equalizer 720 is completed if the maximum iteration count has been reached or if following equation is true:
|(10*log10(Pmeasured/Prequired))|<0.1 dB Equation 10
Wherein Pmeasured is the measured channel power and Prequired is the calculated required channel power. Preferably the maximum iteration count is 500 but may be between 50-5000. If the maximum iteration count has not been reached or if Equation 9 is not true then, for each active channel, is dynamic gain equalizer 720 attenuation is set to the measured power of that channel n divided by the power required for that channel n, step 1835. Then, the power of the active channels is read from optical spectrum analyzer 710, step 1815, and the process is repeated. Setting the dynamic gain equalizer 720 attenuation may require feedback depending on the dynamic gain equalizer 720 repeatability.
Fiber Tuning
The effects of chromatic dispersion, ripple attenuation, fiber aging, fiber damage or failure, as well as other fiber related effects on the overall optimization of the system are compensated for by equalizing the fiber related effect. The gain ripple of various stages of the erbium doped fiber amplifier and gain tilt from Stimulated Raman Scattering (SRS) during fiber propagation and polarization dependent loss (PDL) lead to variations in the signal power over the spectrum. Without equalization, these variations accumulate and lead to a reduction in the optical signal-to-noise ratio (OSNR) as well as to increased non-linear distortions for some channels.
The power is equalized to yield uniform signal power over the transmission spectrum at the output of every Nth optical amplifier 120. This is monitored via a channel monitoring system such as optical spectrum analyzer 710 with a coarse resolution bandwidth about 0.7 times the channel spacing. The number of spans between equalization sites, or “equalization period” N, should be an odd number.
Referring to
Once the channel power levels are known, the next step is to equalize the linear power tilt by controlling the gain-tilt at the N amplifier 120 sites preceding optical spectrum analyzer 710, step 2010. The gain tilt in each of these amplifiers 120 needs to be adjusted by a uniform amount to yield a flat power spectrum at optical spectrum analyzer 710. Instructions are then sent to each amplifier 120 to control the gain tilt independently from each amplifier 120 internal gain-control algorithm, step 2015. The internal gain-control algorithm of an amplifier is well known.
The next step is to equalize the deviation in power in each channel from the average per-channel power using a dynamic equalizing filter (DEF) such as receiver variable optical attenuator 795, dynamic gain equalizer 720 or transmitter variable optical attenuator, step 2020. The average power per signal channel is measured at optical spectrum analyzer 710. For each channel, the deviation from this average power is calculated. Finally, the attenuation in the DEF for each channel is set to the calculated deviation amount to yield a uniform power at optical spectrum analyzer 710, step 2025. The maximum required power attenuation value depends on the amount of amplifier gain ripple and PDL in each span and on the number of spans between equalization sites.
The attenuation values are described relative to an average attenuation level. To equalize the channel powers, some channels require additional attenuation, while other channels require “negative attenuation” relative to this average attenuation level. To avoid requiring active elements (gain elements) in the DEF, the average attenuation in the DEF needs to be at least as large as the maximum required “negative attenuation” for the weakest channel. The required average attenuation level contributes to the fixed insertion loss specification of the DEF device.
Channel Turn-Up
To begin, topology discovery module 800 marks the channel as turned on in the topology tree, step 1105. Then source of the signal or transceiver is turned on and the signal is allowed to begin propagation in system 500, step 1110. The transceiver may already be on and could be outside system 500. Chain tuning is started and the full chain tuning process described in
Channel Turn Down
As shown in
The sub-routine can also be modified. Settings will be equalized to power during the next autotuning cycle. It may be desirable to remove tilt from variable optical attenuator 795 and adjust it from the amplifiers to maintain an acceptable range.
Token Based Priority Scheme
In order to successfully tune an entire transmission chain, each segment 575 must be tuned by itself in the order of source to destination along the chain. To accomplish this a designated token passing scheme that is fault tolerant, but which allows only one segment 575 to be tuned at a time and provides for tuning segments 575 in the correct order is used.
The topology of transmission system 901 is created by topology discovery module 800 (
In order to conduct tuning and topology discovery in an orderly fashion, the software adopts “token passing”. Referring to
The token is spontaneously regenerated at the start of the chain by originator 900 if the token is somehow “lost”. During typical operation, as the token is passed from master station 910 to master station 910, the station that transmits the token sends a signal to the originator 900 as to where the token is being sent. Once a master station receives a token, it sends a signal to the master station that created the token has been received. If this signal is not received, the token is “lost”. To find the token, a token control update signal is generated by originator 900. It queries each master station 910 to determine if any master station 910 still has the last token transmitted. If no station has the last token transmitted, then originator 900 generates a new token.
If the tuning is triggered by control sequence 815 while a segment 575 is currently being tuned, then originator 900 can start tuning over at the beginning. To do this, originator 900 issues a token control return request to any tokens in transmission system 901. After all the found tokens are returned, originator 900 transmits a new token.
Master station 910 may create a copy token if master station 910 has children 915. If a token is transmitted to master station 910 and master station 910 has children 915, master station 910 will make a copy of the received token. One token stays at master station 910 while the copy token is transmitted to children 915. Before master station 910 with children 915 transmits the token to the next master station 910, master station 910 with children 915 will wait until it has received all the copy tokens it sent out. It is possible master station 910 could have more than one chain or group of children 915 and therefore master station 910 may transmit out more than one copy token. However, only one copy token is transmitted down a single chain of children 915. The copy token that is transmitted down a chain contains a unique identifier to distinguish it from the original token and from other copy tokens.
When a master station 910 sends the token to the next master station 910, the master station 910 that has the token sends a token control update to originator 900 that originated the token. The token control update contains information regarding the tuning process such as whether or not the tuning was successful and where the token is being sent. All stations are identified within the token by their lookback IP address. If no token control updates have been received by the originator in a specified amount of time, then a token return request is sent to each station in the chain requesting a token control update. The specified amount of time or cycles is preferably short enough to prevent the system from becoming stale and long enough to allow for efficient token recovery. In the preferred embodiment, this time is two to thirty minutes, with a preferred time of five minutes. Each time the tuning cycle starts and before reading the first optical spectrum analyzer 710, a token control update message is sent to the token originator. By combing topology information with an awareness of the current token location, the system can respond to token loss and restart the tuning process.
Tuning of segment 575 may act differently depending on various conditions. If a device does not work, then tuning attempts to optimize the signal without changing the damaged device's settings. If a device is manually set, the settings are treated as if the device was damaged and the device's settings are not changed. If an entire slave station 905 is off, then slave station 905 is treated as if all devices on that station have been manually set. If master station 910 is turned off then master station 910 does not tune the segment. If an optimize flag has been set by the user within the token, then tuning will attempt to optimize the signal by changing parameters, even if system measurements are currently within the acceptable range. If an override manual flag is set, then master station 910 will autotune all devices, even if they have been manually set by the user. If an override on/off flag is set, then master station 910 will tune itself and all slave stations 905 in its segment 575, regardless of whether tuning has been turned off by the user.
The token may contain several pieces of information including, but not limited to, the token originator identifier, unique token ID, direction of travel of the station, previous, current and next master stations for the token, token copy identifier, optimize flag, override manual flag, or override on/off flag.
If a channel is split unto multiple parts, then multiple tokens will be generated at the start of each disconnected sub-chain. When the channel is restored, the tuning control process will have to remove the extra tokens. To do this, after master station 910 accepts the token, master station 910 checks the originator ID of the token against the topology table stored in memory 410 of management card 110. If the originator ID of the token does not match the ID of originator 900, a token originator return message is sent to the correct token originator and the token is then destroyed.
Switch Manager Process
The switch manager process is a separate process which handles all data to switched devices. Switched devices operate on one direction of traffic flow at a time. The switch manager allows for simultaneous requests for the attenuation of the device to be handled without generating conflicts or invalid data. In the preferred embodiment, the only switch device is optical spectrum analyzer 710, which is connected via a 4×1 switch to multiple points of the system. However, the system may contain multiple switched devices, including single devices with distributed switch setups. All access to the switched devices are managed through a switch manager.
The switch manager ensures that all data read from switched devices is from the switch position requested. This prevents multiple applications from simultaneously attempting to perform separate operations involving a switch. The communication mechanism is provided through the IPC layer, and messages are provided by the switch manager as a proxy for the DCA OSA card.
The switch manager keeps track of the current status of all optical switches. Then it takes the configuration request and the data request for any device it controls. Each request is queued until no other request is being processed. Then the switch is set to the appropriate position and the request is sent to the device. The switch manager waits the time required to execute the command or until a process is received if one is required, before processing further requests. If it is useful or necessary, the switch manager can optimize the queue to reduce switch cycles or to prefer higher priority request. Note that the switch manager actually prevents direct access to the switches or switch devices. All applications which need these devices will be programmed to work through the switch manager.
Script Controlled Tuning
The choice as to which station becomes director 1000 is based upon shelf type and the serial number of the shelf. The assignment hierarchy for stations is master terminal shelf (MTS), optical add drop master shelf (OMT), and then other shelves. An MTS differs from an OMT in that an OMT has an optical add drop while an MTS does not have an optical add drop. A MTS is preferred over an OMT because there is a greater chance that an MTS is an end terminal and also because generally an MTS has a direct link to the most amplifiers 120. If there are MTS shelves available, the MTS shelf with the lowest serial number becomes director 1000. If an MTS shelf is not available, but there are OMT shelves available, the OMT shelf with the lowest serial number becomes director 1000. If an OMT shelf is not available, the system will pick the lowest serial number of all the shelves to be director 1000. By using this process, all the stations are independently able to determine what station or shelf becomes director 1000. If two stations decide they are director 1000, then a topology error report is generated.
The hierarchy of stations is determined during the topology discovery process 800, and when the topology changes. Once a station decides it is director 1000, the director software becomes active on that station. On all the other stations, the director software is inactive.
Director 1000 is also the interface to the user. The command line interface or user interface communicates to director 1000. To manually tune the system, a user issues the proper command from the command line to instruct director 1000 to run a tuning cycle. Director 1000 builds a script to tune the system. The script is based on topology, communicates the script to script manager 1005, which executes the script. The user can enter a command code on any station. If a command code is entered onto a station that is not director 1000, the director software for the station where the command code was entered sends the command code to director 1000. Then the director software on the station where the command was entered returns to an inactive state.
Director 1000 determines the topology of the system and tuning requirements of the system, generates a list of low level tuning functions, and determines the order in which the tuning functions are to be executed. The list of tuning functions is known as the “script”. It contains the various tuning equations and algorithms previously described. Station manager 1010's function is to accept the script from script manager 1005 and perform the instructions on the script. Station manager 1010 executes the instructions on the script one at a time. For example, the instructions in the script may be to pre-tune a long chain of slave stations 1015, similar to a configuration like segment 575. In another example, script manager 1005 communicates to station manager 1010 how and when to adjust gain functions on slave stations 1015. Script manager 1005 can also do parallel tuning by coordinating and executing the tuning functions which are not interdependent at the same time.
Although the invention has been described with reference to one or more preferred embodiments, this description is not to be construed in a limiting sense. There is modification of the disclosed embodiments, as well as alternative embodiments of this invention, which will be apparent to persons of ordinary skill in the art, and the invention shall be viewed as limited only by reference to the following claims.
This application claims priority to Provisional Application Ser. No. 60/374,373, entitled “Transmission System Architecture Allowing Automated Fiber Characterization”, by Marvin R. Young, et al., filed Apr. 22, 2002, Provisional Application Ser. No. 60/377,160, entitled “Method of Topology Discovery of an Optical Transport System”, by Daniel Nooner, et al., filed Apr. 30, 2002, Provisional Application Ser. No. 60/377,162, entitled “Method of Optical Transport System Provisioning”, by Daniel Nooner, et al., filed Apr. 30, 2002, and Provisional Application Ser. No. 60/377,123, entitled “Method of Tuning an Optical Transport System”, by Daniel Nooner, et al., filed Apr. 30, 2002.
| Number | Date | Country | |
|---|---|---|---|
| 60374373 | Apr 2002 | US | |
| 60377123 | Apr 2002 | US | |
| 60377160 | Apr 2002 | US | |
| 60377162 | Apr 2002 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10421681 | Apr 2003 | US |
| Child | 11514781 | Aug 2006 | US |
