One-Dimensional Optical Data Arrays Implemented Within Optical Networks

1. Field of the Invention

The present invention relates to the use of one-dimensional optical data arrays within optical networks.

2. Description of the Related Art

Fiber optics technology has completely penetrated the long-haul telephony network due to its inherent low loss and high bandwidth. In the area of local loop applications, however, financially attractive options have developed more slowly. As of late, substantial research effort has been directed towards developing technology to implement fiber optics within local loop applications (e.g., fiber in the loop). Cost, capacity, and switching problems, however, still must be overcome.

The active double star is an example of fiber optics technology implemented within the local loop. As shown in FIG. 1A, an active double star 10 employs sets of transmitters and receivers T.sub.1, R.sub.1, . . . T.sub.N, R.sub.N, at a central office 20, which transmit and receive downstream and upstream optical signals, respectively, via optical fibers 25.sub.D, 25.sub.U. The central office and fibers act as a primary star. Each optical fiber 25.sub.D, 25.sub.U, is linked to a remote terminal 30, where downstream-directed optical signals are terminated at a receiver R' and then processed (switched) electronically. Downstream optical signals are demultiplexed, re-formed, and launched from transmitters T.sub.1 ' within fibers 35.sub.D to a plurality of optical network units (ONUs) 40 forming a secondary star.

Alternatively, conventional passive optical networks 10' utilize passive optical couplers 32, as shown in FIG. 1B. The passive coupler 32 is located at a remote node 30' to direct downstream optical signals launched from a transmitter 14 at the central office 20. In a passive time division multiplexing or subcarrier multiplexing schemes, optical signals are sent to remote nodes 30' forming the primary star. Each remote node distributes its received optical signal passively, directing similar portions to each of a plurality of optical network units 40 along fibers 35.sub.D to form a secondary star. Filtering means 42 contained at each optical network unit 40 extracts an intended portion of the received signal. During upstream communication, each optical network unit transmits an optical signal within a prearranged time slot or frequency on upstream fiber 35.sub.U. The upstream signals are received at the remote node 30', multiplexed, and directed therefrom to a receiver 16 at the central office 15 via fiber 25.sub.U.

Timing and power budget throughput problems are inherent within PONs operating according to this broadcast distribution. The problems may be avoided, however, utilizing switching methods such as wavelength division multiplexing. A PON network 10" which utilizes wavelength division multiplexing is shown in FIG. 1C. Therein, each transmitter 14' at the central office 20 modulates downstream data at N distinct wavelengths, multiplexing the modulated data signals and transmitting the multiplexed signals onto downstream feeder fibers 25.sub.D. Each feeder fiber 25.sub.D carries the multiplexed signals to an input port P.sub.D of a WDM coupler 32' at remote node 30" for demultiplexing. The demultiplexed signals are passively directed by the WDM coupler to various output ports P.sub.D ' according to wavelength. The output ports direct the signals along downstream fibers 35.sub.D to optical network units 40'. Signals are generated at and transmitted upstream from each optical network unit along fibers 35.sub.U. Said upstream signals are received at ports P.sub.U ', multiplexed within coupler 32', directed to port Pu and fiber 25.sub.U for delivery to the central office. In many PONs, "U" and "D" refer to the same fibers and the same ports.

Remote Interrogation of Terminal Equipment, or RITE-Net.TM., is an emerging passive, WDM-based optical network technology that is disclosed in commonly owned copending U.S. patent application Ser. No. 08/029,724, filed Mar. 1, 1993, and incorporated herein by reference. A RITE-Net.TM. system 10" is shown in FIG. 1D, includes a transmitter 14" and receiver 16" at the central office 20. The transmitter, typically a laser, transmits downstream information, according to wavelength, to a "Dragone" router.sup.1 hereinafter referred to interchangeably as a wavelength division multiplexer/router (WDM/R) 32" or waveguide grating router (WGR) located at remote node 30"'.

The WDM/R 32" splits optical signals received thereat and directs the split signals to individual optical network units 40" according to wavelength. The optical network units return a portion of the received signal (via upstream fiber 350 after first overmodulating it with optical dam. This avoids the need (and cost) for separate transmitters at each optical network unit and the difficulty of registering and tracking the wavelength of said transmitters. At the remote node 30"', the WDM/R 32" multiplexes the overmodulated upstream signals and directs them via port P.sub.U and fiber 25.sub.U, to the central office.

An optical network of the present invention utilizes one dimensional optical dam arrays for multi-channel optical dam transmission and processing. Optical data arrays are linear arrays of optical transmitters or optical receivers, usually fabricated to serve as optical data links. Each element (e.g., laser or receiver) within an optical data array is independent, but driven concurrently with the other elements in the array. Because the optical data may be transmitted/received by/from N optical fibers instead of 1, the electronic and optical requirements for driving the optical data over the N fibers are inherently less stringent than that for driving one fiber with N times the bandwidth or throughput.

Each optical data array requires only one fixturing operation to affix a fiber array for the N channels (instead of N operations for N individual components) resulting in lower cost per component. Additionally, the electronics driving the N elements of the array may operate at a small fraction, i.e., approximately 1/N, of the total throughput rate. In other words, for example, instead of transmitting data along one fiber at 1 Gb/s, the same information may be transmitted along twenty fibers within a system at 50 Mb/s, utilizing arrays of 20 transmitters and arrays of 20 receivers. The much lower clock rate requires lower cost processing electronics as compared to those electronics needed for the higher rate.

In a preferred form, the optical data arrays described herein are implemented within an active network, preferably an active double star. Use of optical data arrays of this invention, however, is not limited to active networks, but may be implemented within various network designs according to the needs of those skilled in the art. Further, while RITE-Net.TM. optical network units are typically associated with passive networks, they may be utilized, as described herein, within active networks.

An optical network of this invention may also provide for substantially continuous and time efficient integrity testing of the optical transmission paths between a remote terminal and each system subscriber, i.e., optical network unit. The resources required to implement prior art integrity testing of each system subscriber's line are thereby minimized. This reduces the cost for system maintenance while increasing the system's reliability and maintainability.

The integrity testing described herein is implementable in a design where each optical network unit is interrogated during normal operation, as exemplified by RITE-Net.TM. systems. More particularly, the means used to return a portion of an optical signal received at each optical network unit is concurrently utilized to provide the integrity check. Further, because the looped-back (returned) portions of the optical signals launched from a remote terminal, e.g., a one dimensional optical dam array, realize minimal attenuation, sensitivity requirements for receivers utilized within the network are minimized, further lending to reduced system cost.

FIG. 1A, 1B, 1C and 1D are schematic diagrams depicting optical networks of the prior art;

FIG. 2 is a schematic diagram depicting a one-dimensional optical data array;

FIG. 3 is a schematic diagram depicting an optical network of the present invention;

FIG. 3A is a schematic diagram of the optical network of FIG. 3 to which an overlay for fault location has been applied;

FIG. 4 is a schematic diagram of a RITE-Net.TM. type optical network unit utilized within the present invention; and

FIG. 5 is a schematic diagram of a variation of a RITE-Net.TM. optical network unit which may be used herein.

One of the main impediments to the introduction of fiber optic communications to local subscribers within telephony networks is the cost. In an effort to control system cost, this invention implements one-dimensional optical data arrays within an optical network to transmit and/or receive optical data along independent optical links synchronously or plesiosynchronously. Consequently, data may be transmitted to each of N receiver elements within an optical receiver array along N fibers at lower rates, i.e., 1/N, of the maximum rate in lieu of transmission via one optical fiber at a high rate. Such operation obviates the need for a multiplicity of high speed processing electronics at the remote terminals. While one-dimensional optical data arrays of transmitters and/or receivers are known to be used within high speed computers to provide optical interconnects, their use within FITL optical networks is novel, and leads to improved and reduced-cost optical-network operation.

The one-dimensional optical data arrays may be implemented within active networks that implement RITE-Net.TM. type communication between remote terminals and optical network units. Consequently, system cost, i.e., system maintenance and reliability cost, may be further reduced where the system's inherent communication characteristic is utilized in integrity testing (to be discussed in greater detail below). The need, therefore, to implement conventional electronics, e.g., optical time domain reflectometry (OTDR), within the system to merely confirm the integrity of each separate subscriber loop is minimized. FIG. 2 depicts an implementation of a one-dimensional optical data array of this invention. The depiction is given for illustration purposes only and is not meant to limit the scope of the invention. In the figure, a one-dimensional optical transmitter array 100 is shown optically linked via an optical ribbon 110 to a one-dimensional optical receiver array 120. Transmitter array 100 includes a driver circuit 112 for driving a set 114 of N lasers or like devices 116. The output from each of the N lasers devices is coupled to each of the N parallel optical fibers comprising ribbon 110 via a connector 118, e.g., a MAC II connecter, manufactured by AT&T Corporation. The downstream ends of the fibers are coupled through a second connector 122 to each of a set 124 of optical detectors 126 within receiver array 120. The detected signals are amplified within an array 130 of amplifiers and transferred to an array of decision circuits and line drivers 132. At this point in the dam flow, the dam may be distributed as electronic signals, or, as described above, further transmitted optically by a second transmitter array (not shown in the figure), similar to array 100.

A second possible optical network implementation of the one dimensional optical data arrays of this invention is shown in FIG. 3. As was the embodiment of FIG. 2 described above, FIG. 3 is for explanation purposes only, and is not meant to limit the scope of the present invention. A central office 210 shown in FIG. 3 transmits an optical signal on each of M upstream and downstream optical fibers extending therefrom, i.e., as a primary star. Each of the M optical fibers 212 (and 214) link an active remote terminal 220 to the central office (forming a secondary star).

At the remote terminal 220, the downstream signals are received at a first receiver element 126 of one-dimensional optical data receiver array 120. The received signals are transformed electrically and directed to a demultiplexer 128 for demultiplexing. Each packet or time slot separated from the received signal is directed to one of a plurality of lasers 116 in the transmitter laser array 100, based on destination address. N-1 of the N lasers 116 within the array 100 are optically linked by fibers 222 to destination optical network units 230, referred to interchangeably herein as optical transceivers, up to N-1 in number. The optical network units are preferably RITE-Net.TM. type units.

Upstream light signals launched along optical fibers 232 from each optical network unit 230 are terminated within the remote terminal at array 120. The upstream signals, which may or may not be modulated, are received at the second through N.sup.th distinct elements 126 of the receiver array 120 and converted thereat to electronic form. A multiplexer 142 processes and merges the N-1 signals to form a composite upstream signal. The upstream information is converted into light signals which are then transmitted upstream from a transmitter, preferably within transmitter array 100, where transmitter T.sub.o (116) optically transmits the signal into fiber 214 to the central office 210.

The first receiver element 126 of the N-element array 120 may receive the data within frames clocked at a clock rate CR. The data may pass through the multiplexer 128 and form N-1 frames at the same clock rate. The data contained within the signals formed by the demultiplexing, however would be driven by the transmitters at 1/(N-1) of the duty cycle of the downstream optical data signal frames. It should be noted, however, that downstream light signals could alternatively be received at the remote terminal 220 within a discrete and independent optical receiver element and directed therefrom to demultiplexer 128. Such an arrangement would allow data to be output from each element 116 of the transmitter array 100 at rate CR/N, thereby lowering the cost of subscriber electronics. Further, upstream signals could be transmitted to the central office 210 via a transmitter element that is independent of array 100. Such system decisions are made on assessment of cost and complexity.

FIG. 4 shows a RITE-Net.TM. based optical network unit 300 which may be utilized in two configurations identified as "a" and "b" in the figure. In the "a" configuration, a light signal received from a remote transmitter at a port 302 is directed to a tap coupler 304. A portion of the received signal is directed by the tap coupler to a receiver 306. An optional wavelength specific filter 305 may be included for identification or diagnostic purposes. Electronics 314 thereat formats the signal for the user and additionally provides upstream data for optical encoding and transfer to the central office. A second portion of the received signal is directed to modulator for remodulation by the upstream electronic data and returned to the central office via port 312. In the "b" mode operation, the modulated portion of the signal is reflected for return to the central office and passed back through the tap coupler 304 before output from port 312. While the FIG. 4 embodiment includes dedicated upstream and downstream fiber, one skilled in the art can appreciate that the embodiment may be easily modified to accommodate a single two-way fiber. Further, while operating in reflective mode, the dotted lines need not exist and the upstream signal could exit port 302.

Monitoring the health and integrity of a telephony network is a necessary but laborious task, whether the system is active or passive. In a conventional passive optical network (FIG. 1B), an integrity check may be performed by interrogating each optical network unit in a separate signaling channel. This is accomplished with a physical probe to verify the optical integrity of the loop, which is verified via the subscriber response. In other words, the ONU must operate in response to the signaling probe. Alternatively, in a WDM passive optical network (FIG. 1C) a wavelength selective OTDR can passively probe individual lines 25D and 35D. However, the finite resolution of OTDR reduces its effectiveness in discriminating between an optical break near the optical network unit and an electronic failure within the optical network unit itself. OTDR is most useful in long-length point-to-point links.

In an active network, however, OTDRs cannot be utilized at the central office without some network modification. Several modifications have been proposed, including optical bypasses, e.g., U.S. Pat. No. 5,285,305 to Cohen, et al., commonly owned and incorporated herein by reference. However, similar modifications may be accomplished within an active network to enable OTDR testing from the central office. In particular, a simple wavelength division multiplexer (1.times.N coupler) overlay, such as that shown in FIG. 3A, may be utilized at the remote terminal which may be interrogated or switched into the network when needed.

Several overlay configurations are available. For instance, a down/up-up/down configuration is shown in FIG. 3A. There, test signals present on fiber 212 may be directed via tap coupler 251 to/from 1.times.N coupler 250, where the signals may be coupled downstream/upstream to/from the ONUs onto/from upstream fibers 232 via tap couplers 252, 253 and 254. Test signals may be coupled to/from upstream fiber 214 to/from tap coupler 258 to/from a second 1.times.N coupler 259. 1.times.N coupler 259 couples the test signal to/from the appropriate ONU.

In a second overlay configuration, tap coupler 251 may couple test signals to/from second 1.times.N coupler 259, where they can be coupled, through tap couplers 255,256 and 257 to/from fibers 222. Tap coupler 258 may couple test signals to/from fiber 214 to/from first 1.times.N coupler 250 where the data can be coupled through tap coupler 252, 253 and 254 to/from upstream fibers 232. A third overlay configuration may include a "Dragone" type (WGR) router that replaces both first and second 1.times.N couplers 250 and 259. The WGR would be connected to the tap couplers and upstream and downstream fibers in a manner similar to that described above.

In a preferred form, each return signal looped-back from each RITE-Net.TM. optical network unit (whether overmodulated or not) may be utilized as an interrogation signal. That is, each downstream directed signal interrogates the modulator and necessarily checks the optical continuity of the optical circuit. Thus, any break in the loop may be detected at the central office as a matter of course, almost immediately. A lack of response from the ONU over several optical frames, (e.g., 125 .mu.sec.), indicates a line break and the central office takes appropriate action. While detecting a line fault in an active double star network will not aid in determining the exact location of the fault, the assignment of craft personnel and equipment to find the break is expedited since the line fault has been promptly and uniquely identified.

The configuration of FIG. 5 is an optical network unit 400 which may be implemented and may be very useful for extremely long fiber runs. The unit may operate in either an "a" or "b" mode. The optical network unit shown possesses its own laser source (i.e, transmitter 408 or 408') to overcome large attenuation which will tend to occur between the center of the second star and the optical network unit. In "a" mode operation, signals received at input port 402 are directed to coupler 404, which may be a WDM coupler. A portion of the signal is directed from the coupler to receiver 406. Transmitter 408' transmits optically encoded data upstream via coupler 404 through return port 412. If necessary, a filter 405 can isolate receiver 406 from interference from 408'.

In "b" mode operation, coupler 404 directs a return portion of the received signal to second coupler 410. Light from transmitter 408 is also coupled into coupler 410, an output of which is directed to port 412. In such an arrangement, the inclusion of transmitter 408' is not necessary. The use of such ONUs described herein enables the network provider the ability to perform surveillance and measurement on an individual line using transmission tests in lieu of an OTDR as described above. Finally, the use of one-dimensional optical data arrays within active double star networks provides for continuity in the development and implementation of technology to transport switched or broadcast video signals over the network.

While the above-described embodiments took the form of active networks, the scope of the invention is not intended to be limited thereto. For example, a network similar to the network of FIG. 3 is envisioned wherein each optical network unit (i.e., optical transceiver, whether conventional or RITE-Net based) is replaced by an optical coupler which in turn further transmits/receives downstream/upstream optical signals to/from a plurality of optical network units. In today's parlance, the remote terminal is now similar in function to a host digital terminal running conventional passive optical networks.

One-dimensional optical data arrays used within active double star systems provide high capacity and flexibility at low optical component cost, lowering the network cost and simplifying operation system's burdens. Although the invention has been described in connection with an active double star, the invention is not limited to such applications. Those skilled in the art will appreciate that numerous other adaptations and modifications that may be made without departing from the scope and spirit of the invention as set forth in the following claims.